Re: [Emc-users] SSR Issues, Thyristor dv/dt

[andy pugh]

That was a funny eBay search.

According to Fotek the DC to DC version is SSR-40DD

eBay search for SSR-40DD shows many DC-rated SSRs for 60VDC output.
But I would be wanting 600V rating, and that ought to be a -H suffix.

SSR-40DD-H as an eBay search seems to show things very unrelated to
SSRs. I have rarely seen two characters change an eBay search so
completely :-)

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is
designed for the especial use of mechanical geniuses, daredevils and
lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1916

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[andy pugh]

On 4 May 2016 at 12:33, Steve Stallings  wrote:
> We have been the victim of counterfeit Fotek SSR units. They
> looked fine, but required over 6 volts input to turn them on.

Have you pulled any apart to see what Triac they use?

You might be able to tell real Fotek from fake by weight, I am pretty
sure that mine are nowhere near the 105g quoted:
http://www.fotek.com.hk/page1e.htm

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is
designed for the especial use of mechanical geniuses, daredevils and
lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1916

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[Steve Stallings]

We have been the victim of counterfeit Fotek SSR units. They
looked fine, but required over 6 volts input to turn them on.
That of course does not work with logic signals driving them.
Now we test all of them before selling them. What a pain.

Steve Stallings
www.PMDX.com
 

> -Original Message-
> From: andy pugh [mailto:bodge...@gmail.com] 
> Sent: Wednesday, May 04, 2016 6:38 AM
> To: Enhanced Machine Controller (EMC)
> Subject: Re: [Emc-users] SSR Issues, Thyristor dv/dt
> 
> On 4 May 2016 at 02:38, Jon Elson <el...@pico-systems.com> wrote:
> > OH WELL, there you have it!  A 16 A triac in a 40 A labeled
> > SSR!  If there's one bogus thing in a device, you can depend
> > on the WHOLE thing having bogus engineering.
> 
> Here is an analysis of one from bigclive:
> https://www.youtube.com/watch?v=DxEhxjvifyY
> Though mine did not have any potting resin in it, either.
> 
> From what I can gather Fotek are a reasonably reputable company, but
> are being heavily counterfeited. The fact that mine were constructed
> differently from the one in the video rather suggests that mine are
> counterfeit, even if the one in the video isn't. The one in the video
> is 100% over-rated, whereas mine was 150% over-rated.
> 
> Not that that actually explains the failure. The devices have not been
> asked to pass more than 2A so far.
> 
> -- 
> atp
> "A motorcycle is a bicycle with a pandemonium attachment and is
> designed for the especial use of mechanical geniuses, daredevils and
> lunatics."
> - George Fitch, Atlanta Constitution Newspaper, 1916
> 
> --
> 
> Find and fix application performance issues faster with 
> Applications Manager
> Applications Manager provides deep performance insights into 
> multiple tiers of
> your business applications. It resolves application problems 
> quickly and
> reduces your MTTR. Get your free trial!
> https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
> 


--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[andy pugh]

On 4 May 2016 at 02:38, Jon Elson  wrote:
> OH WELL, there you have it!  A 16 A triac in a 40 A labeled
> SSR!  If there's one bogus thing in a device, you can depend
> on the WHOLE thing having bogus engineering.

Here is an analysis of one from bigclive:
https://www.youtube.com/watch?v=DxEhxjvifyY
Though mine did not have any potting resin in it, either.

From what I can gather Fotek are a reasonably reputable company, but
are being heavily counterfeited. The fact that mine were constructed
differently from the one in the video rather suggests that mine are
counterfeit, even if the one in the video isn't. The one in the video
is 100% over-rated, whereas mine was 150% over-rated.

Not that that actually explains the failure. The devices have not been
asked to pass more than 2A so far.

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is
designed for the especial use of mechanical geniuses, daredevils and
lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1916

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[Jon Elson]

On 05/03/2016 12:39 PM, andy pugh wrote:
> On 3 May 2016 at 18:16, Jon Elson  wrote:
>> Are these name-brand SSRs
>> or some china noname units you got on eBay?
> Well, the labels say "Fotek" but then the labels are peeling off, and
> I today found web pages talking about fake Fotek SSRs.
>
> Having taken one apart, the triac in my 40-Amp SSR is a BTA16-800B so
> rated at 800V (good) but only 16A (bad)
> Of course, it isn't necessarily even a real triac :-)
>
OH WELL, there you have it!  A 16 A triac in a 40 A labeled 
SSR!  If there's one bogus thing in a device, you can depend 
on the WHOLE thing having bogus engineering.

Jon

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[dave]

This is clearly ancient history but: many years ago in a time before 
time 
a guy needed a CC device to light orchard heaters. Nichrome coiled 
elements in series; must touch off either oil or propane. It used an 
off-the-shelf SCR feeding (ugh!) an 240 v to 440 v transformer.
Indeed the dv/dt's were horrible. Still I got by with plain SCR's not 
the radar spec ones.
On short strings they switched for only about 6 degrees. Amazing what 
one can do (it helps to
be young and dumb and poor) and that was 45 years ago.  The project got 
dumped, I suspect, simply because of liability.
However, one could light off 20 A of orchard in about 5 min.

Dave

On 05/03/2016 10:39 AM, andy pugh wrote:
> On 3 May 2016 at 18:16, Jon Elson  wrote:
>> Are these name-brand SSRs
>> or some china noname units you got on eBay?
> Well, the labels say "Fotek" but then the labels are peeling off, and
> I today found web pages talking about fake Fotek SSRs.
>
> Having taken one apart, the triac in my 40-Amp SSR is a BTA16-800B so
> rated at 800V (good) but only 16A (bad)
> Of course, it isn't necessarily even a real triac :-)
>


--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[andy pugh]

On 3 May 2016 at 18:16, Jon Elson  wrote:
> Are these name-brand SSRs
> or some china noname units you got on eBay?

Well, the labels say "Fotek" but then the labels are peeling off, and
I today found web pages talking about fake Fotek SSRs.

Having taken one apart, the triac in my 40-Amp SSR is a BTA16-800B so
rated at 800V (good) but only 16A (bad)
Of course, it isn't necessarily even a real triac :-)

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is
designed for the especial use of mechanical geniuses, daredevils and
lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1916

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt --> Inrush current?

[Nicklas Karlsson]

> > If you need limiting of inrush current I guess a transistor connected as a 
> > current source would be the perfect solution for this? Or is it against 
> > surge voltage?
> 
> The reason I need inrush current limiting is mainly so that the lights
> and fridge stay on when I turn on the lathe :-)

Yes of course the beer in the fridge is the most important.

> The surge current charging the capacitor is enough to trip the main
> breaker in my house fuse box.

I a current source like this one but probably with only one diode
https://www.google.se/imgres?imgurl=https%3A%2F%2Fwww.st-andrews.ac.uk%2F~www_pa%2FScots_Guide%2Faudio%2Fpart1%2Ffig8.gif=https%3A%2F%2Fwww.st-andrews.ac.uk%2F~www_pa%2FScots_Guide%2Faudio%2Fpart1%2Fpage3.html=XWfLwCKH0X1kXM=raqWXyKWt6SvxM%3A=481=404=887=1280=0ahUKEwjLnOCZuL7MAhXM2SwKHdV3DN0QMwgkKAgwCA=mrc=8
connected to the DC bus should limit the inrush current. It will consume some 
power so it should at least ideally be disabled by a relay once charged. This 
kind of circuit is common for low power but you have high power.

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[Jon Elson]

On 05/03/2016 07:46 AM, andy pugh wrote:
> On 3 May 2016 at 13:37, Gene Heskett  wrote:
>>> And does the "fail to turn off" problem description include not
>>> turning off ever again, even when removed from the circuit and from
>>> all power?
>> Generally yes, its an avalanche mode failure that shorts the switching
>> junction.
> I am not 100% clear on the source of this dv/dt. Is it simply the
> result of turning on the main contactor, with the SSR turned off, when
> the mains supply is in a high part of the sine wave?
>
The mains wires have inductance.  The SSRs should only turn 
off at zero current, so that SHOULD help.  But, your SSRs 
BOTH failed! That is an indication something is wrong.  Is 
this SSR setup between a transformer secondary and the 
rectifier?  Or, are you using it direct off the mains?  If 
there is a transformer, the leakage inductance could be 
playing a part.

Yes, I suppose flipping the mains contactor will give a huge 
dV/dt, practically unlimited.  That's why an RC snubber 
across the SSR outputs would help even more than an MOV, 
which will only clip voltage peaks.

Jon

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[Jon Elson]

On 05/03/2016 06:14 AM, andy pugh wrote:
> I could pick a couple of these up today, though it is an 80-mile round
> trip. I wonder if they would work better?
> http://www.rapidonline.com/kudom-ksim380d25-l-panel-mount-ssr-4-32vdc-48-440vac-25a-load-led-zero-cross-60-1598
> (For a total of 2kW of servo drive a 25A device ought to be
> sufficient, I think? The servo drives will be limited to 10A each (at
> the most) and it would be very rare for both motors to pull max
> current at exactly the same time)
>
I think you want to stay with 40 A or so, and use NAME BRAND 
parts. Who's ever heard of Kudom?
Is that supposed to REMIND you of Crydom?

Jon

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[Jon Elson]

On 05/03/2016 05:42 AM, andy pugh wrote:
it sounds like your derating should have been sufficient.  
The fact that an SSR with a SIZEABLE resistor in series with 
it failed is awfully suspicious!  Are these name-brand SSRs 
or some china noname units you got on eBay?
While the nonlinear nature of the load could cause some 
problems with turn-off of the SSRs, and the direct SSR could 
get damaged by surge if it ever got triggered in the wrong 
sequence, I can't imagine what could have caused the other 
relay (the slow charge one) to fail.  So, it has to be 
either a voltage spike or the relays are actually total junk 
- like 120 V relays relabeled as 240 V units.  I think you 
should get name brand relays and then put a series RC 
snubber across the "contacts".

Jon

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[andy pugh]

On 3 May 2016 at 15:12, John Kasunich  wrote:
> The vast majority of the AC drives on the market don't even have a
> pre-charge contactor.  When power is applied, the bus immediately
> charges thru the resistor.

And that would be perfectly acceptable in my application too, thinking
about it, though I like the idea of being able to power up the servo
drives independent of the main contactor, just so that I can delay it
until the VFD has come up.

I think that I will also look into using this as the main contactor,
as it is cheap and I can buy it on the way home:
http://www.maplin.co.uk/p/30a-spdt-power-relay-24vdc-n25aw

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is
designed for the especial use of mechanical geniuses, daredevils and
lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1916

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[John Kasunich]

On Tue, May 3, 2016, at 10:00 AM, andy pugh wrote:
> If there is no reason not to leave the pre-charge relay closed during
> operation, I could even use the existing mechanical relay that brings
> in the discharge resistor. Is there a reason to switch-out the
> precharge relay once the main contactor is closed? I guess the danger
> is that if the main contactor fails to operate the machine would run
> through the precharge resistor. Though I imagine that that would cause
> an easily-detected undervoltage fault in the drives.

You can certainly leave the pre-charge relay closed after closing the
bypass (main) contactor.

The vast majority of the AC drives on the market don't even have a
pre-charge contactor.  When power is applied, the bus immediately
charges thru the resistor.  The low voltage power supply for the 
electronics runs off the DC bus (great for power dip ride-thru.)  
Once the bus is up, the low-voltage power supply starts and the
microcontroller comes alive.  Then the micro-controller makes the
decision to close the bypass contactor.

If the bypass contactor opens during normal operation, the drive 
typically trips on low bus voltage.  Although the worst case situation
is a motor load that's not quite enough to drag the bus down to the
low bus trip point, but is high enough to cook the pre-charge resistor.
Pre-charge resistors are usually sized for intermittent duty, so they
need to be a type that fails safely under continuous overload (no
arcing or fire).  On large (expensive) drives we sometimes use an
auxiliary contact on the bypass contactor as an interlock.




-- 
  John Kasunich
  jmkasun...@fastmail.fm

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[andy pugh]

On 3 May 2016 at 14:31, John Kasunich  wrote:
>
> Is there any particular reason you are using SSRs instead of
> regular contactors?  I like contactors because they are rugged
> and pretty much immune to overvoltage, dv/dt, and di/dt problems.

Mainly it is a matter of available space (or became such). I found
that I didn't have the room in the existing control cabinet in the
side of the lathe, and I found another space inside a cover behind the
headstock where the feed motor used to live.
>
> Contactors (or electromechanical relays) also provide an air-gap
> between line power and the load, which means I'm comfortable
> using them in an e-stop circuit.

I do have a proper contactor upstream of the PSU that is controlled by
the e-stop circuit. The first SSR is there mainly as part of the
power-off discharge arrangement.

> Also, is there any particular reason you are running the full load
> current through both relays?

Yes, the reason is that I didn't think things through properly :-)
This PSU is a development of one that I have been using in my milling
machine. That one uses a 230V coil relay with a NC contact that
connects the discharge resistor when the power to the PSU goes off. On
that one there was a second relay controlled by a 555 timer circuit.
It was only later that I realised that I could more intelligently
control the system via HAL, as the servo drives report bus voltage
back to HAL. The specific reason to want to do this is to avoid
turning _on_ the PSU while the cap is discharging, as trying to break
the discharge (DC) voltage leads to relay failure. So I now have
things interlocked so that HAL won't turn on the PSU unless bus
voltage is < 2V.
Given that I now have intelligent control of the PSU, I can see that
your system makes a lot of sense.
If there is no reason not to leave the pre-charge relay closed during
operation, I could even use the existing mechanical relay that brings
in the discharge resistor. Is there a reason to switch-out the
precharge relay once the main contactor is closed? I guess the danger
is that if the main contactor fails to operate the machine would run
through the precharge resistor. Though I imagine that that would cause
an easily-detected undervoltage fault in the drives.

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is
designed for the especial use of mechanical geniuses, daredevils and
lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1916

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[John Kasunich]

On Tue, May 3, 2016, at 09:34 AM, andy pugh wrote:
> On 3 May 2016 at 14:27, Erik Christiansen  wrote:
> > Yup, a MOV goes across the triac, to clamp voltage spikes.
> 
> When you say "Across" do you mean across the output terminals of the
> SSR, or between the live terminal of the SSR and neutral?
> 
> The former is easier, and I doubt that there would be much trouble
> caused by dumping the spikes into the main capacitor as long as the
> rectifier wasn't too upset.

A MOV across the SSR protects the SSR (only).

A MOV across the input protects everything.

However, the MOV across the SSR can potentially do a better job
of protecting the SSR because the lead length is shorter and thus
inductance is lower.

At decent power levels (tens of kW and up) the main cap and 
rectifier can clamp most transients, so only the SSR is at risk.
But small rectifiers can be destroyed by the peak currents that
happen when the cap clamps an incoming transient.  At a former
employer they had to beef up the rectifiers on a sub-kW class
drive for exactly that reason.

As an aside, here is a blog posting about the power supply I built
for a CNC project.  There is a link to the complete schematic early
in the text, and photos of the completed supply.
http://jmkasunich.com/cgi-bin/blosxom/shoptask/power-input-01-11-07.html
Note that the EMC filter and inductor after the rectifier are not 
absolutely required.  I had them in my well-stocked junk box so
I used them.  The contactors and relays also came from junk.
I think the only thing I bought was the disconnect switch (from
Automation Direct).

This particular supply is a bit unique in that I designed it to work
as a voltage doubler on 120V, or as a bridge rectifier on 240V.


-- 
  John Kasunich
  jmkasun...@fastmail.fm

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[Gene Heskett]

On Tuesday 03 May 2016 08:46:07 andy pugh wrote:

> On 3 May 2016 at 13:37, Gene Heskett  wrote:
> >> And does the "fail to turn off" problem description include not
> >> turning off ever again, even when removed from the circuit and from
> >> all power?
> >
> > Generally yes, its an avalanche mode failure that shorts the
> > switching junction.
>
> I am not 100% clear on the source of this dv/dt. Is it simply the
> result of turning on the main contactor, with the SSR turned off, when
> the mains supply is in a high part of the sine wave?

I've not been able to confirm that as a destructive turn on. Looking at 
your circuit, the 620 seems too high so its instant surge heating may be 
quite destructive over many cycles of power up, I have a 50 at 200 watt 
wirewound there. But I also have a bit of a di/dt limiter in the form of 
some large toroid power transformers that are series-paralleled to get 
the dc I desired.

With a purely capacitative load, the di/dt would be limited mainly by the 
powerline impedance. If suitable chokes could be had to make a choke 
input filter, thats what I would do as its output voltage regulation is 
quite a bit better than a purely capacitative design.  But suitable 
chokes are lotsa bucks even on the surplus market these days.  A 
suitably current rated "filament" transformer could be used if the 
primary winding could be cut away as its voltage output when using the 
filament winding as a choke could easily top a couple thousand volts 
leading to a short circuit failure.  Then most of the inductance would 
go away.  Unfortunately, most of those are wound with the high voltage 
primary winding laid on the bobbin first.  Winding your own on a good 
sized ferrite core could be done if you can find a core with high enough 
magnetic saturation so it doesn't saturate at the expected current, 
keeping in mind that if it does saturate, you have effectively an air 
cored inductance.

That isn't a desirable thing, and its exactly why you cannot use a vga 
sweep speed CRT monitor on an old computer that outputs pal or ntsc 
sweep speeds.  That doubles the effective on time of the sweep 
transformers drive, the core saturates and everything between it and the 
power source goes by-by by letting out all the smoke and breaking the 
mirrors that makes all this stuff work.

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt --> Inrush current?

[andy pugh]

On 3 May 2016 at 14:00, Nicklas Karlsson  wrote:

> If you need limiting of inrush current I guess a transistor connected as a 
> current source would be the perfect solution for this? Or is it against surge 
> voltage?

The reason I need inrush current limiting is mainly so that the lights
and fridge stay on when I turn on the lathe :-)
The surge current charging the capacitor is enough to trip the main
breaker in my house fuse box.

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is
designed for the especial use of mechanical geniuses, daredevils and
lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1916

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[andy pugh]

On 3 May 2016 at 14:27, Erik Christiansen  wrote:
> Yup, a MOV goes across the triac, to clamp voltage spikes.

When you say "Across" do you mean across the output terminals of the
SSR, or between the live terminal of the SSR and neutral?

The former is easier, and I doubt that there would be much trouble
caused by dumping the spikes into the main capacitor as long as the
rectifier wasn't too upset.

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is
designed for the especial use of mechanical geniuses, daredevils and
lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1916

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[John Kasunich]

On Tue, May 3, 2016, at 06:42 AM, andy pugh wrote:
> Section 6 on page C-343 here:
> https://www.omron.com/ecb/products/pdf/precautions_ssr.pdf shows a
> square-wave with zero time at 0 current. But I don't understand how
> that is possible with a sinusoidal input voltage. Surely the current
> is zero for all the period of the cycle where the input voltage is
> less than the capacitor voltage?

That picture shows an inductive load.  And the text refers to solenoids
which are inductive.  A sufficiently large inductor will draw a nearly
constant current with modest ripple at double the line frequency, which
becomes a more-or-less square wave on the line side of the rectifier.

> I wonder whether I would be better with DC-rated SSRs on the DC side
> of the rectifier?

> In case it helps, this is the input circuit.
> https://imagebin.ca/v/2fs1b7szpsMy The SSRs are controlled by LinuxCNC
> individually

Is there any particular reason you are using SSRs instead of 
regular contactors?  I like contactors because they are rugged 
and pretty much immune to overvoltage, dv/dt, and di/dt problems.

Contactors (or electromechanical relays) also provide an air-gap
between line power and the load, which means I'm comfortable 
using them in an e-stop circuit.  SSRs (or any other semiconductor)
are far more likely to fail shorted than open, and can't safely be
used to disconnect power.

Also, is there any particular reason you are running the full load
current through both relays?  I design large AC motor drives that
rectify the AC line to make a DC bus.  (Large is hundreds of kW).
It would be prohibitively expensive to use two contactors sized
for the full rated power.  The precharge resistors are fed through 
a small contactor, rated (and fused) for the precharge current only
(line voltage divided by resistor value).  The bypass contactor is
rated for full load current and bypasses the resistors.  See the 
attached drawing "dc-supply-threephase.pdf" for an example.
(The three 1.5uF caps are to absorb line-to-ground transients
that couple across the distribution transformer from HV switching.)

The other attached drawing is for a single-phase home-shop
scale project.  For home stuff. I use two poles for both precharge
and bypass.  American 240V consists of two lines, each 120V
from ground, and for safety I need to disconnect both.  I use
the outer two poles of a three-pole contactor or relay.

If British 240V has line and neutral then in theory you could use
only one pole in the line side - but I personally would still use two.
Most contactors have three poles anyway, might as well use them.

At home-shop power levels, only the bypass device needs to be
a contactor.  The charging device can be a simple two-pole 
relay, like the ubiquitous ice-cube or KUP.

-- 
  John Kasunich
  jmkasun...@fastmail.fm


dc-supply-precharge.pdf
Description: Adobe PDF document


dc-supply-threephase.pdf
Description: Adobe PDF document
--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[Erik Christiansen]

On 03.05.16 13:32, andy pugh wrote:
> On 3 May 2016 at 13:29, andy pugh  wrote:
> > Until I googled just now I thought that this was what was meant by
> > MOV. The data sheets for Varistors seem to specify an activation
> > voltage and a power dissipation, but no off-state resistance
> > measurement to calculate the power dissipation. Or is the resistance
> > so low that this can be ignored?
> 
> And I now see I was reading a voltage/current graph as a
> voltage/resistance graph. Ignore that question.

Yup, a MOV goes across the triac, to clamp voltage spikes.
It's high impedance until a spike makes it conduct while there is enough
voltage across it. It has only a limited life if hit with big surges,
IIRC.

A Transorb (transient suppressor diode) is also good for protecting
triacs and MOSFETs, but the application of suppressors is roughly:
Gaseous arrestors on lines going outdoors, for clamping induced voltage
from nearby lightning; MOVs after that, and on mains circuits, for
example; transorbs for final protection of other semiconductors after
the edge of really big spikes has been taken off by primary protection.

Erik

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[Nicklas Karlsson]

> > > If Solid State Relay SSR happen to be a thyristor it will only allow
> > > a limited dv/dt or it will fail to turn off. A common source of
> > > dv/dt is inductive or maybe capacitive load and it happens because a
> > > thyristor turn off at zero current although a reactive load do not
> > > have zero voltage at zero current. Problem is circumvented by a
> > > properly sized capacitor in parallel to limit voltage rise during
> > > turn off.
> >
> A dv/dt failure is usually destructive.

I think it is a failure to turn off.

> > How would one determine "properly sized"?
> 
> That is a big compromise because that capacitor is a leakage across it, 
> enough to knock you around if its over done. A.02 u-f, 1 kv ceramic 
> might help.  But you'll probably feel a buzz if you get across it.

I do not think output should be considered safe anyway.

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues

[andy pugh]

On 3 May 2016 at 13:57, Nicklas Karlsson  wrote:
>> You may need higher voltage rated devices, Andy.  Without ANY switching
>> spikes, or the EMP from a nearby lightning strike, they are seeing the
>> RMS peak voltage of 338.4 volts (240 * 1.41) 100 times a second.
>
> Yes you are correct 240Vac*sqrt(2).

My assumption was that a device rated at 380VAC was rated to 380V RMS.
Are you suggesting that this is a DC rating?

Even the phony SSRs in the link I posted earlier seemed to have a 600V
device in them:
http://docs-europe.electrocomponents.com/webdocs/133a/0900766b8133a375.pdf
So I think that the SSRs are actually intended for the supply voltage
on the label.

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is
designed for the especial use of mechanical geniuses, daredevils and
lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1916

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt --> Inrush current?

[Nicklas Karlsson]

On Tue, 3 May 2016 13:29:42 +0100
andy pugh  wrote:

> On 3 May 2016 at 12:41, Erik Christiansen  wrote:
> >
> > The link to SSR precautions indicates that the SSR already includes a
> > snubber. But page 3 of the notes says that is often not sufficient, and
> > suggests a 440 - 470 volt varistor (MOV) as protection in a 200 - 240
> > volt application. That clearly assumes the use of e.g. a 600v SSR, not a
> > mere 400v unit.
> 
> Part of the reason for this arrangement of relays is that I could not
> find an NTC device  with enough current carrying capacity to work as a
> surge suppressor.

If you need limiting of inrush current I guess a transistor connected as a 
current source would be the perfect solution for this? Or is it against surge 
voltage?

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues

[Nicklas Karlsson]

> On Tuesday 03 May 2016 04:54:32 andy pugh wrote:
> 
> > On 3 May 2016 at 02:51, Jon Elson  wrote:
> > > You should very conservatively derate the SSRs
> > > in many circumstances, but especially where capacitive
> > > inrush surge is possible.
> >
> > The PSU is powered through a 20A breaker and the SSRs are rated at
> > 40A. The anticipated peak motor current is about 12A total.
> > Would you suggest a hiher rating than that?
> >
> > > Do you know what the ACTUAL inrush surge is?
> >
> > I did, until the relay with the job of switching-out the
> > inrush-limiting resistor failed. It should be much less than an amp.
> > (240V AC through a 620 ohm resistor)
> >
> > >  Also, is it
> > > possible that a momentary power dip caused a surge while the
> > > SSRs were in the turned-on mode?  Possibly, there was a
> > > voltage spike while the SSRs were off, and it caused them to
> > > catastrophically break down.  You might think of getting
> > > some MOV surge suppressors to put across the input mains.
> > > Also, maybe get SSRs rated for higher input voltage.
> >
> > The SSRs are rated for 380V and are running on 240V mains. I have not
> > previouslyhad any problems with mains voltage spikes.
> > The PSU seems to have failed either when turned off, or at power-up.
> > This is on a lathe that is not, as yet, working. I only found that the
> > PSU was dead when I came to try to set up the X axis motor having
> > finally finished the mechanical assembly and wiring. (I did not get as
> > far as enabling the servo amps)
> 
> You may need higher voltage rated devices, Andy.  Without ANY switching 
> spikes, or the EMP from a nearby lightning strike, they are seeing the 
> RMS peak voltage of 338.4 volts (240 * 1.41) 100 times a second.

Yes you are correct 240Vac*sqrt(2).

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[andy pugh]

On 3 May 2016 at 13:37, Gene Heskett  wrote:
>> And does the "fail to turn off" problem description include not
>> turning off ever again, even when removed from the circuit and from
>> all power?
>
> Generally yes, its an avalanche mode failure that shorts the switching
> junction.

I am not 100% clear on the source of this dv/dt. Is it simply the
result of turning on the main contactor, with the SSR turned off, when
the mains supply is in a high part of the sine wave?

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is
designed for the especial use of mechanical geniuses, daredevils and
lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1916

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[Gene Heskett]

On Tuesday 03 May 2016 06:40:35 Erik Christiansen wrote:

> On 03.05.16 10:30, andy pugh wrote:
> > And does the "fail to turn off" problem description include not
> > turning off ever again, even when removed from the circuit and from
> > all power?
>
> No, dv/dt triggering is merely spurious initiation of conduction by
> excessively rapid increase of applied voltage, without gate current
> being applied. An SCR (thyristor) is a four-layer diode, electrically
> equivalent to a PNP and NPN transistor stacked into a totem feedback
> pair, and once it starts to conduct, it latches on - while there's
> more than its holding current flowing.
>
> It is not going to happen with rapidly decreasing voltage, AIUI. In
> fact, if you connect two SCRs, each with a resistor in the anode
> circuit, across a DC supply, then interconnect the anodes with an
> adequately sized nonpolarised capacitor, and one SCR is conducting,
> then triggering the other will switch off the first.
>
I've seen that used in my business.  It is actually a pretty dependable 
switch when configured correctly.

> My suspicion is that the device failure is likely due to the 40v
> margin on the mains peak voltage not being adequate. In a machine
> environment, with heavy inductive loads being switched, it might be
> instructive to try a 600v SSR, and compare lifetimes.
>
> If the overvoltage is due to an inductive load on the SSR, then a
> flyback diode will fix it if the circuit is DC, otherwise a snubber
> can absorb the spike energy. (A snubber is just a series RC
> combination, with the resistor low enough in value to limit the spike
> amplitude, but high enough not to run hot normally, and with the
> capacitor large enough to be a low impedance to the spike's dv/dt, but
> high enough at 50 Hz for the resistor not to run hot normally, at
> least that's my grasp of it.)

With the exception that I have seen dv/dt failures that were destructive, 
thats a good understanding.

> Erik
>
> --
> Find and fix application performance issues faster with
> Applications Manager Applications Manager provides deep performance
> insights into multiple tiers of your business applications. It
> resolves application problems quickly and reduces your MTTR. Get your
> free trial!
> https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users


Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[andy pugh]

On 3 May 2016 at 13:29, Nicklas Karlsson  wrote:
> If possible I would measure voltage spike before to make sure there are any 
> otherwise the problems is elsewhere.

I have no real reason to think that I have a  voltage spike problem at
all, really. I don't think I have ever had an appliance fail (granted
those may all be protected).
I would have to check on the voltage that my 'Scope can display, but I
rather fear I will be looking for something that isn't there to be
measured.

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is
designed for the especial use of mechanical geniuses, daredevils and
lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1916

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[Gene Heskett]

On Tuesday 03 May 2016 05:30:54 andy pugh wrote:

> On 3 May 2016 at 10:14, Nicklas Karlsson  
wrote:
> > If Solid State Relay SSR happen to be a thyristor it will only allow
> > a limited dv/dt or it will fail to turn off. A common source of
> > dv/dt is inductive or maybe capacitive load and it happens because a
> > thyristor turn off at zero current although a reactive load do not
> > have zero voltage at zero current. Problem is circumvented by a
> > properly sized capacitor in parallel to limit voltage rise during
> > turn off.
>
A dv/dt failure is usually destructive.

> How would one determine "properly sized"?

That is a big compromise because that capacitor is a leakage across it, 
enough to knock you around if its over done. A.02 u-f, 1 kv ceramic 
might help.  But you'll probably feel a buzz if you get across it.

> And does the "fail to turn off" problem description include not
> turning off ever again, even when removed from the circuit and from
> all power?

Generally yes, its an avalanche mode failure that shorts the switching 
junction.

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[andy pugh]

On 3 May 2016 at 13:29, andy pugh  wrote:
> Until I googled just now I thought that this was what was meant by
> MOV. The data sheets for Varistors seem to specify an activation
> voltage and a power dissipation, but no off-state resistance
> measurement to calculate the power dissipation. Or is the resistance
> so low that this can be ignored?

And I now see I was reading a voltage/current graph as a
voltage/resistance graph. Ignore that question.


-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is
designed for the especial use of mechanical geniuses, daredevils and
lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1916

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[Nicklas Karlsson]

> On 03.05.16 12:14, andy pugh wrote:
> > I could pick a couple of these up today, though it is an 80-mile round
> > trip. I wonder if they would work better?
> > http://www.rapidonline.com/kudom-ksim380d25-l-panel-mount-ssr-4-32vdc-48-440vac-25a-load-led-zero-cross-60-1598
> 
> Well, 620v peak ought to give some margin on average spikes on 240 Vac
> (340 Vp), but a 440v MOV would do a lot to extend the life of the SSR.
> 
> Erik

If possible I would measure voltage spike before to make sure there are any 
otherwise the problems is elsewhere.

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[andy pugh]

On 3 May 2016 at 12:41, Erik Christiansen  wrote:
>
> The link to SSR precautions indicates that the SSR already includes a
> snubber. But page 3 of the notes says that is often not sufficient, and
> suggests a 440 - 470 volt varistor (MOV) as protection in a 200 - 240
> volt application. That clearly assumes the use of e.g. a 600v SSR, not a
> mere 400v unit.

Part of the reason for this arrangement of relays is that I could not
find an NTC device  with enough current carrying capacity to work as a
surge suppressor.
Until I googled just now I thought that this was what was meant by
MOV. The data sheets for Varistors seem to specify an activation
voltage and a power dissipation, but no off-state resistance
measurement to calculate the power dissipation. Or is the resistance
so low that this can be ignored?

> IIUC, there's also a bridge rectifier between the SSRs and the capacitor.
> But the inductance being switched is the power transformer on the input.

This PSU is connected to the mains without an isolation transformer.

> It is tempting to move to the DC side, and use a 600v DC SSR, or just a
> high voltage MOSFET. (N-channel, in the -ve lead has lower RDSon, and is
> easier to drive.) The section 6 that you mention recommends using a
> MOSFET SSR on the DC side. Do you see a 600v one of those on ebay?

The closest is a 400V unit, only 20A and £135
http://www.ebay.co.uk/itm/SSR-20A-400VDC-Part-DC400D20-/351592581782?hash=item51dc8d0a96:g:S9EAAOxyd8NSa8CE

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is
designed for the especial use of mechanical geniuses, daredevils and
lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1916

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues

[Gene Heskett]

On Tuesday 03 May 2016 04:54:32 andy pugh wrote:

> On 3 May 2016 at 02:51, Jon Elson  wrote:
> > You should very conservatively derate the SSRs
> > in many circumstances, but especially where capacitive
> > inrush surge is possible.
>
> The PSU is powered through a 20A breaker and the SSRs are rated at
> 40A. The anticipated peak motor current is about 12A total.
> Would you suggest a hiher rating than that?
>
> > Do you know what the ACTUAL inrush surge is?
>
> I did, until the relay with the job of switching-out the
> inrush-limiting resistor failed. It should be much less than an amp.
> (240V AC through a 620 ohm resistor)
>
> >  Also, is it
> > possible that a momentary power dip caused a surge while the
> > SSRs were in the turned-on mode?  Possibly, there was a
> > voltage spike while the SSRs were off, and it caused them to
> > catastrophically break down.  You might think of getting
> > some MOV surge suppressors to put across the input mains.
> > Also, maybe get SSRs rated for higher input voltage.
>
> The SSRs are rated for 380V and are running on 240V mains. I have not
> previouslyhad any problems with mains voltage spikes.
> The PSU seems to have failed either when turned off, or at power-up.
> This is on a lathe that is not, as yet, working. I only found that the
> PSU was dead when I came to try to set up the X axis motor having
> finally finished the mechanical assembly and wiring. (I did not get as
> far as enabling the servo amps)

You may need higher voltage rated devices, Andy.  Without ANY switching 
spikes, or the EMP from a nearby lightning strike, they are seeing the 
RMS peak voltage of 338.4 volts (240 * 1.41) 100 times a second.

Here, the common $10 one is 480 volt rated, somewhat better than your 380 
devices, but on a 240 volt circuit I'd use this one from mpja.com:



at $30 a copy.  You could probably remove the heat sink shown.

You need at least 600 volt rated devices, and some MOV's across the line 
that will absorb the stuff above 400 volts. I don't see any such 
critters on the mpja site though. The higher the joules rating, the 
better, with 100 being a recommended minimum from grandpa Gene.  They 
will need to be fused in the event they get too hot and fail of course, 
but you could just let the circuit breaker handle it in that event.  I 
would install them someplace where they won't do any further damage 
should they explode, and where they are easily seen if you want to check 
them for any discoloration or damage.

On this side of the pond, most, but not all of the commonly available 
multi-outlet switched strips that sell for more than a 15 dollar bill 
have at least a 50 joule device in them, and here in the coyote.den I 
have a big one that ties power, cable tv cable and telco lines all to a 
common point with a 250 joule rating for the whole works.  I haven't 
blown anything that was plugged into that "master strip" in about 15 
years now. Its on a 1500WA UPS but my brother color laser printers power 
up inrush was too much for the UPS so it was plugged into the wall, and 
it got blown about 6 weeks back.  I just replaced it with a MFC inkjet, 
that seems to be permanently set in super ink saver mode, black never 
gets any darker than about an 18% grey card.  Photos WILL be washed out. 
But thats another story...

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[Nicklas Karlsson]

> On 03.05.16 11:42, andy pugh wrote:
> > On 3 May 2016 at 10:41, Nicklas Karlsson  
> > wrote:
> > > In other words i/C should be less than maximum allowed du/dt.
> > 
> > I am not sure how to determine the i for this calculation. I guess I
> > can find the dv/dt from the component numbers on this page:
> > http://www.instructables.com/id/The-inner-workings-of-Counterfeit-FOTEK-SSRs/
> > The servo drives have so far been limited to 4A, but whether that
> > translates to 4A at the input to the rectifier I am not sure.
> > Section 6 on page C-343 here:
> > https://www.omron.com/ecb/products/pdf/precautions_ssr.pdf shows a
> > square-wave with zero time at 0 current. But I don't understand how
> > that is possible with a sinusoidal input voltage. Surely the current
> > is zero for all the period of the cycle where the input voltage is
> > less than the capacitor voltage?
> 
> Spot on, but the circuit in section 6 doesn't include a capacitor, that
> I can see.
> 
> > I wonder whether I would be better with DC-rated SSRs on the DC side
> > of the rectifier?
> 
> The link to SSR precautions indicates that the SSR already includes a
> snubber. But page 3 of the notes says that is often not sufficient, and
> suggests a 440 - 470 volt varistor (MOV) as protection in a 200 - 240
> volt application. That clearly assumes the use of e.g. a 600v SSR, not a
> mere 400v unit.

The way to find out if snubber capacitor is large enough is to make the 
calculations or simply measure at turn off.

I make assumption SSR is a thyristor/triac or similar. Varistor protect against 
overvoltage but do nothing against dv/dt. If SSR turn off at zeror current 
there should be very little voltage snap. For resistive load both current and 
voltage are at zero at the same time. For reactive load there are voltage at 
zero current and dv/dt depend speed of turn off or how fast snubber capacitor 
is charged.

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[Erik Christiansen]

On 03.05.16 12:14, andy pugh wrote:
> I could pick a couple of these up today, though it is an 80-mile round
> trip. I wonder if they would work better?
> http://www.rapidonline.com/kudom-ksim380d25-l-panel-mount-ssr-4-32vdc-48-440vac-25a-load-led-zero-cross-60-1598

Well, 620v peak ought to give some margin on average spikes on 240 Vac
(340 Vp), but a 440v MOV would do a lot to extend the life of the SSR.

Erik

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[Erik Christiansen]

On 03.05.16 11:42, andy pugh wrote:
> On 3 May 2016 at 10:41, Nicklas Karlsson  wrote:
> > In other words i/C should be less than maximum allowed du/dt.
> 
> I am not sure how to determine the i for this calculation. I guess I
> can find the dv/dt from the component numbers on this page:
> http://www.instructables.com/id/The-inner-workings-of-Counterfeit-FOTEK-SSRs/
> The servo drives have so far been limited to 4A, but whether that
> translates to 4A at the input to the rectifier I am not sure.
> Section 6 on page C-343 here:
> https://www.omron.com/ecb/products/pdf/precautions_ssr.pdf shows a
> square-wave with zero time at 0 current. But I don't understand how
> that is possible with a sinusoidal input voltage. Surely the current
> is zero for all the period of the cycle where the input voltage is
> less than the capacitor voltage?

Spot on, but the circuit in section 6 doesn't include a capacitor, that
I can see.

> I wonder whether I would be better with DC-rated SSRs on the DC side
> of the rectifier?

The link to SSR precautions indicates that the SSR already includes a
snubber. But page 3 of the notes says that is often not sufficient, and
suggests a 440 - 470 volt varistor (MOV) as protection in a 200 - 240
volt application. That clearly assumes the use of e.g. a 600v SSR, not a
mere 400v unit.

> > Problem might also be because you discharge a voltage into a capacitor and 
> > get a very large inrush current but this require a different solution.
> One of the failed SSRs is meant to be that solution, it is used to
> bypass a current-limiting resistor once the bus voltage reaches 300V.
> 
> In case it helps, this is the input circuit.
> https://imagebin.ca/v/2fs1b7szpsMy The SSRs are controlled by LinuxCNC
> individually.

IIUC, there's also a bridge rectifier between the SSRs and the capacitor.
But the inductance being switched is the power transformer on the input.
If the rectifier were not there, you'd have an LC circuit which would
oscillate at its resonant frequency if the current were interrupted
mid-cycle, rather than at a zero crossing. In such a situation, several
times the 340v mains peak voltage could appear in the circuit, to the
distinct detriment of the SSR. The bridge rectifier prevents that
resonance.

It is tempting to move to the DC side, and use a 600v DC SSR, or just a
high voltage MOSFET. (N-channel, in the -ve lead has lower RDSon, and is
easier to drive.) The section 6 that you mention recommends using a
MOSFET SSR on the DC side. Do you see a 600v one of those on ebay?

Regards,
Erik



--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[andy pugh]

I could pick a couple of these up today, though it is an 80-mile round
trip. I wonder if they would work better?
http://www.rapidonline.com/kudom-ksim380d25-l-panel-mount-ssr-4-32vdc-48-440vac-25a-load-led-zero-cross-60-1598
(For a total of 2kW of servo drive a 25A device ought to be
sufficient, I think? The servo drives will be limited to 10A each (at
the most) and it would be very rare for both motors to pull max
current at exactly the same time)

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is
designed for the especial use of mechanical geniuses, daredevils and
lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1916

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[andy pugh]

On 3 May 2016 at 11:42, andy pugh  wrote:
>
> In case it helps, this is the input circuit.
> https://imagebin.ca/v/2fs1b7szpsMy The SSRs are controlled by LinuxCNC
> individually.

That left out an important part of the circuit
https://imagebin.ca/v/2fs60d4YbDyo is better.



-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is
designed for the especial use of mechanical geniuses, daredevils and
lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1916

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[andy pugh]

On 3 May 2016 at 10:41, Nicklas Karlsson  wrote:
> In other words i/C should be less than maximum allowed du/dt.

I am not sure how to determine the i for this calculation. I guess I
can find the dv/dt from the component numbers on this page:
http://www.instructables.com/id/The-inner-workings-of-Counterfeit-FOTEK-SSRs/
The servo drives have so far been limited to 4A, but whether that
translates to 4A at the input to the rectifier I am not sure.
Section 6 on page C-343 here:
https://www.omron.com/ecb/products/pdf/precautions_ssr.pdf shows a
square-wave with zero time at 0 current. But I don't understand how
that is possible with a sinusoidal input voltage. Surely the current
is zero for all the period of the cycle where the input voltage is
less than the capacitor voltage?

I wonder whether I would be better with DC-rated SSRs on the DC side
of the rectifier?

> Problem might also be because you discharge a voltage into a capacitor and 
> get a very large inrush current but this require a different solution.
One of the failed SSRs is meant to be that solution, it is used to
bypass a current-limiting resistor once the bus voltage reaches 300V.

In case it helps, this is the input circuit.
https://imagebin.ca/v/2fs1b7szpsMy The SSRs are controlled by LinuxCNC
individually.
(actually, the first SSR is controlled by a small mechanical
force-guided relay which also switches-in a discharge resistor when
the PSU is turned off, but I don't think that circuit is relevant to
the SSR circuit)

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is
designed for the especial use of mechanical geniuses, daredevils and
lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1916

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[Erik Christiansen]

On 03.05.16 10:30, andy pugh wrote:
> And does the "fail to turn off" problem description include not
> turning off ever again, even when removed from the circuit and from
> all power?

No, dv/dt triggering is merely spurious initiation of conduction by
excessively rapid increase of applied voltage, without gate current
being applied. An SCR (thyristor) is a four-layer diode, electrically
equivalent to a PNP and NPN transistor stacked into a totem feedback
pair, and once it starts to conduct, it latches on - while there's more
than its holding current flowing.

It is not going to happen with rapidly decreasing voltage, AIUI. In
fact, if you connect two SCRs, each with a resistor in the anode
circuit, across a DC supply, then interconnect the anodes with an
adequately sized nonpolarised capacitor, and one SCR is conducting, then
triggering the other will switch off the first.

My suspicion is that the device failure is likely due to the 40v margin
on the mains peak voltage not being adequate. In a machine environment,
with heavy inductive loads being switched, it might be instructive to
try a 600v SSR, and compare lifetimes.

If the overvoltage is due to an inductive load on the SSR, then a
flyback diode will fix it if the circuit is DC, otherwise a snubber can
absorb the spike energy. (A snubber is just a series RC combination,
with the resistor low enough in value to limit the spike amplitude,
but high enough not to run hot normally, and with the capacitor large
enough to be a low impedance to the spike's dv/dt, but high enough at 50
Hz for the resistor not to run hot normally, at least that's my grasp of
it.)

Erik

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[Nicklas Karlsson]

> On 3 May 2016 at 10:14, Nicklas Karlsson  wrote:
> > If Solid State Relay SSR happen to be a thyristor it will only allow a 
> > limited dv/dt or it will fail to turn off. A common source of dv/dt is 
> > inductive or maybe capacitive load and it happens because a thyristor turn 
> > off at zero current although a reactive load do not have zero voltage at 
> > zero current. Problem is circumvented by a properly sized capacitor in 
> > parallel to limit voltage rise during turn off.
> 
> How would one determine "properly sized"?

C*du/dt = i <=> du/dt = i/C

In other words i/C should be less than maximum allowed du/dt.

Problem might also be because you discharge a voltage into a capacitor and get 
a very large inrush current but this require a different solution.

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[andy pugh]

On 3 May 2016 at 10:14, Nicklas Karlsson  wrote:
> If Solid State Relay SSR happen to be a thyristor it will only allow a 
> limited dv/dt or it will fail to turn off. A common source of dv/dt is 
> inductive or maybe capacitive load and it happens because a thyristor turn 
> off at zero current although a reactive load do not have zero voltage at zero 
> current. Problem is circumvented by a properly sized capacitor in parallel to 
> limit voltage rise during turn off.

How would one determine "properly sized"?

And does the "fail to turn off" problem description include not
turning off ever again, even when removed from the circuit and from
all power?

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is
designed for the especial use of mechanical geniuses, daredevils and
lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1916

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues, Thyristor dv/dt

[Nicklas Karlsson]

If Solid State Relay SSR happen to be a thyristor it will only allow a limited 
dv/dt or it will fail to turn off. A common source of dv/dt is inductive or 
maybe capacitive load and it happens because a thyristor turn off at zero 
current although a reactive load do not have zero voltage at zero current. 
Problem is circumvented by a properly sized capacitor in parallel to limit 
voltage rise during turn off.


Nicklas Karlsson



On Tue, 3 May 2016 09:54:32 +0100
andy pugh  wrote:

> On 3 May 2016 at 02:51, Jon Elson  wrote:
> 
> > You should very conservatively derate the SSRs
> > in many circumstances, but especially where capacitive
> > inrush surge is possible.
> 
> The PSU is powered through a 20A breaker and the SSRs are rated at 40A.
> The anticipated peak motor current is about 12A total.
> Would you suggest a hiher rating than that?
> 
> > Do you know what the ACTUAL inrush surge is?
> 
> I did, until the relay with the job of switching-out the
> inrush-limiting resistor failed. It should be much less than an amp.
> (240V AC through a 620 ohm resistor)
> 
> >  Also, is it
> > possible that a momentary power dip caused a surge while the
> > SSRs were in the turned-on mode?  Possibly, there was a
> > voltage spike while the SSRs were off, and it caused them to
> > catastrophically break down.  You might think of getting
> > some MOV surge suppressors to put across the input mains.
> > Also, maybe get SSRs rated for higher input voltage.
> 
> The SSRs are rated for 380V and are running on 240V mains. I have not
> previouslyhad any problems with mains voltage spikes.
> The PSU seems to have failed either when turned off, or at power-up.
> This is on a lathe that is not, as yet, working. I only found that the
> PSU was dead when I came to try to set up the X axis motor having
> finally finished the mechanical assembly and wiring. (I did not get as
> far as enabling the servo amps)
> 
> -- 
> atp
> "A motorcycle is a bicycle with a pandemonium attachment and is
> designed for the especial use of mechanical geniuses, daredevils and
> lunatics."
> — George Fitch, Atlanta Constitution Newspaper, 1916

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues

[andy pugh]

On 3 May 2016 at 03:04, Gene Heskett  wrote:
> You may have been tapped by the EMP of a nearby lightning strike, Andy.

The machine has been disconnected from the mains pretty much all the
time since it last worked.
In fact the PSU has only been wired-in for a few weeks, and I don't
think we have had any lightning since then.

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is
designed for the especial use of mechanical geniuses, daredevils and
lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1916

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues

[andy pugh]

On 3 May 2016 at 02:51, Jon Elson  wrote:

> You should very conservatively derate the SSRs
> in many circumstances, but especially where capacitive
> inrush surge is possible.

The PSU is powered through a 20A breaker and the SSRs are rated at 40A.
The anticipated peak motor current is about 12A total.
Would you suggest a hiher rating than that?

> Do you know what the ACTUAL inrush surge is?

I did, until the relay with the job of switching-out the
inrush-limiting resistor failed. It should be much less than an amp.
(240V AC through a 620 ohm resistor)

>  Also, is it
> possible that a momentary power dip caused a surge while the
> SSRs were in the turned-on mode?  Possibly, there was a
> voltage spike while the SSRs were off, and it caused them to
> catastrophically break down.  You might think of getting
> some MOV surge suppressors to put across the input mains.
> Also, maybe get SSRs rated for higher input voltage.

The SSRs are rated for 380V and are running on 240V mains. I have not
previouslyhad any problems with mains voltage spikes.
The PSU seems to have failed either when turned off, or at power-up.
This is on a lathe that is not, as yet, working. I only found that the
PSU was dead when I came to try to set up the X axis motor having
finally finished the mechanical assembly and wiring. (I did not get as
far as enabling the servo amps)

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is
designed for the especial use of mechanical geniuses, daredevils and
lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1916

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues

[Dave Cole]

FWIW, this is the authorized Automation Direct distributor in the UK.

http://www.lamonde.com/acatalog/Timer_Switches_Timer_Relays.html

These are interesting also but I have only used a few of them. So far so 
good.
http://www.lamonde.com/acatalog/WEG-Electric-Miniature-Contactors--3-Pole-.html

The costs appear to be slightly higher there than here but not by a lot.

I have also used a bunch of their SSRs and they work fine as well but 
only to control heaters etc.

I order parts from AD in the US and get things in two days with free 
shipping if I buy $50 worth of gear.I have no idea how Lamonde does 
it there.

Dave



On 5/2/2016 9:09 PM, andy pugh wrote:
> On 3 May 2016 at 02:05, andy pugh  wrote:
>> knows a good place to ask.
> Addendum: Good places would be places where the question gets answered
> with reasoned arguments, not statements that "that is not to code"
> (unless backed up with an explanation about what the code is based
> on).
>
> (Sorry, had a bad time on another forum recently where any suggestion
> of actually experimenting with mains voltage prompted strangely
> vehement vituperation)
>

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues

[Gene Heskett]

On Monday 02 May 2016 19:17:41 andy pugh wrote:

> The PSU on my lathe is something I made myself. On the input side are
> two SSRs, one for power on/off and the other switches-out a soft-start
> resistor.
> There is also s discharge capacitor that is switched in when the main
> input SSR is switched off.
>
> Today things went a bit strange, blowing the breaker and then the
> discharge resistor. This was with the PSU powered up, but not turned
> on.
>
> It turns out that I have an unanticipated failure mode, if both input
> SSRs fail closed-circuit.
>
> I have tested the SSRs on the bench, running a light bulb, and they
> both light the bulb with nothing connected to the control terminals
> (And with the control terminals shorted together).
>
> So, I am wondering if there is a problem running a rectifier input
> with SSRs. I found something in an Omron document suggesting that
> turn-off might be unreliable, but that doesn't seem to be the issue
> here, the relays are now incapable of turning off a (filament) light
> bulb.
>
> Did I get unlucky, or are SSRs a bad choice for this application?

You may have been tapped by the EMP of a nearby lightning strike, Andy.

Considerations about that is why my lashup has a hard metallic switch for 
the final off in the form of a small rocker/thermal breaker in the end 
of the outlet strip that powers my rig.  When thats on, I am doing 
similar to you in that I have 2 40 AMP SSR's, each pumped on by a charge 
pump signal, the first one in series with the switched primary power but 
with a 50 ohm 200 watt R in series, controlled by the 2nd machine on 
button. 5 seconds later, after the 50 ohm 200 watt resistor has had a 
chance to charge up the excessive microfarads, another SSR bypasses the 
resistor, so that charge pump gate also turns the SPPWR button green to 
tell me its safe to fire up the spindle motor.  At powerdown, I unclick 
that second button to put the resistor back into the circuit so an odd 
half cycle power that might occur as the charge pump bucket fades, is at 
least controlled so it doesn't trip the wall breaker going off.

My bleeder is quite a bit higher R than yours, but its only bleeding 3 or 
4 watts full time, so discharge time is several minutes.

It can and has gotten quite noisy here several times in the last 2-3 
days, and I am thinking the hard switch has probably saved me from some 
of those bumps.  Yet that hard switch is no more complex than a small 
rocker button on the 6 socket power strip that runs it all. I think its 
also a slow circuit breaker but if it is, its never tripped on my watch, 
the 30 amp in the service box has beat it every time.  But with this 
setup now, I've reduced the service breaker to a 20 and have had no 
further tripages.

While these SSR's are cheap enough I bought spares, I'd have 2nd thoughts 
about having them exposed to line power on a 24/7 basis.  Could be just 
my paranoia, but...  And IIRC your line voltage is 2x mine. Mine is 
around 127 RMS over on this side of the salt water pond. I also suspect 
yours have been in service a lot longer than mine.

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues

[Eric Keller]

On Mon, May 2, 2016 at 9:05 PM, andy pugh  wrote:
> I don't think I have a collapsing field. Downstream of the relay is a
> rectifier and a big capacitor.


I would like to see a good reference on this.  I have some homebuilt
equipment with SSR's switching power supplies and never had any
problem.  With a rectifier and capacitor, inrush might be a problem
(Jon beat me to it).  Seems like if you carefully analyze the
conditions they were experiencing, it may reveal the failure mode.

The governing equation for the capacitor i = c dv/dt suggests that you
can get big currents if the voltage drops precipitously.  It's
possible  that current went through the SSRs.

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues

[Dave Cole]

I've had problems with SSRs shorting out in certain applications.
Because of that I limit the use of SSRs to simply switching power to a 
load and that is it.
For instance; they are very useful for driving heaters for time 
proportioned outputs schemes to control heat output.
I also use them when switching heavy DC heavy loads (of course the SSRs 
must be rated for DC to do that).

Other than that, I avoid them and use plain old contactors and relays.  
If a relay is properly sized, and not cycled beyond specifications, they 
will last a very long, and
predictable time.  Avoid the super cheap relays and contactors. I have 
used hundreds of the Automation Direct relays and contactors over the 
last few years and I can only think of one that was bad out of the 
box.   I don't know if you can get Automation Direct products in the UK 
or not?

Dave


On 5/2/2016 9:09 PM, andy pugh wrote:
> On 3 May 2016 at 02:05, andy pugh  wrote:
>> knows a good place to ask.
> Addendum: Good places would be places where the question gets answered
> with reasoned arguments, not statements that "that is not to code"
> (unless backed up with an explanation about what the code is based
> on).
>
> (Sorry, had a bad time on another forum recently where any suggestion
> of actually experimenting with mains voltage prompted strangely
> vehement vituperation)
>

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues

[Jon Elson]

On 05/02/2016 06:17 PM, andy pugh wrote:
>
> Today things went a bit strange, blowing the breaker and then the
> discharge resistor. This was with the PSU powered up, but not turned
> on.
>
> It turns out that I have an unanticipated failure mode, if both input
> SSRs fail closed-circuit.
>
> I have tested the SSRs on the bench, running a light bulb, and they
> both light the bulb with nothing connected to the control terminals
> (And with the control terminals shorted together).
OK, so the SSRs failed shorted.  That is their usual failure 
mechanism.  You should very conservatively derate the SSRs 
in many circumstances, but especially where capacitive 
inrush surge is possible.
> So, I am wondering if there is a problem running a rectifier input
> with SSRs. I found something in an Omron document suggesting that
> turn-off might be unreliable, but that doesn't seem to be the issue
> here, the relays are now incapable of turning off a (filament) light
> bulb.
>
> Did I get unlucky, or are SSRs a bad choice for this application?
>
Do you know what the ACTUAL inrush surge is?  Also, is it 
possible that a momentary power dip caused a surge while the 
SSRs were in the turned-on mode?  Possibly, there was a 
voltage spike while the SSRs were off, and it caused them to 
catastrophically break down.  You might think of getting 
some MOV surge suppressors to put across the input mains.  
Also, maybe get SSRs rated for higher input voltage.

Jon

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues

[andy pugh]

On 3 May 2016 at 02:05, andy pugh  wrote:
> knows a good place to ask.

Addendum: Good places would be places where the question gets answered
with reasoned arguments, not statements that "that is not to code"
(unless backed up with an explanation about what the code is based
on).

(Sorry, had a bad time on another forum recently where any suggestion
of actually experimenting with mains voltage prompted strangely
vehement vituperation)

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is
designed for the especial use of mechanical geniuses, daredevils and
lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1916

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues

[andy pugh]

On 3 May 2016 at 01:11, Eric Keller  wrote:
> I can guarantee that an
> undergraduate mechanical engineering student isn't even going to
> suspect that there is going to be an issue.  What I realized is that I
> know how to deal with a collapsing field with a DC relay, but not with
> an AC circuit.

I don't think I have a collapsing field. Downstream of the relay is a
rectifier and a big capacitor.

It is interesting to learn that you had the same problem, I am
wondering if anyone knows what the failure mode is, or knows a good
place to ask.

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is
designed for the especial use of mechanical geniuses, daredevils and
lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1916

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] SSR Issues

[Eric Keller]

I recently ran into a similar problem in a design class I was
teaching.  As a mechanical engineer, I have to admit that I don't
always think these things through, and I can guarantee that an
undergraduate mechanical engineering student isn't even going to
suspect that there is going to be an issue.  What I realized is that I
know how to deal with a collapsing field with a DC relay, but not with
an AC circuit.  Seems to me that your application would be better off
with cascading a mechanical relay after the SSR. Even then, there is a
missing protection circuit, and I haven't tried to figure out what
that is.

The students blew a SSR that turned a power supply on and off.  We
solved the problem by replacing as needed. Not sure what shorts the
power side of the circuit, there is probably a gate/collector max
voltage violation or something like that.

On Mon, May 2, 2016 at 7:17 PM, andy pugh  wrote:
> The PSU on my lathe is something I made myself. On the input side are
> two SSRs, one for power on/off and the other switches-out a soft-start
> resistor.
> There is also s discharge capacitor that is switched in when the main
> input SSR is switched off.
>
> Today things went a bit strange, blowing the breaker and then the
> discharge resistor. This was with the PSU powered up, but not turned
> on.
>
> It turns out that I have an unanticipated failure mode, if both input
> SSRs fail closed-circuit.
>
> I have tested the SSRs on the bench, running a light bulb, and they
> both light the bulb with nothing connected to the control terminals
> (And with the control terminals shorted together).
>
> So, I am wondering if there is a problem running a rectifier input
> with SSRs. I found something in an Omron document suggesting that
> turn-off might be unreliable, but that doesn't seem to be the issue
> here, the relays are now incapable of turning off a (filament) light
> bulb.
>
> Did I get unlucky, or are SSRs a bad choice for this application?
>
> --
> atp
> "A motorcycle is a bicycle with a pandemonium attachment and is
> designed for the especial use of mechanical geniuses, daredevils and
> lunatics."
> — George Fitch, Atlanta Constitution Newspaper, 1916
>
> --
> Find and fix application performance issues faster with Applications Manager
> Applications Manager provides deep performance insights into multiple tiers of
> your business applications. It resolves application problems quickly and
> reduces your MTTR. Get your free trial!
> https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users

--
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial!
https://ad.doubleclick.net/ddm/clk/302982198;130105516;z
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users