Re: [Emc-users] Another crazy idea?

[Gene Heskett]

On Monday 11 January 2016 14:44:54 John Kasunich wrote:

> On Mon, Jan 11, 2016, at 02:00 PM, Gene Heskett wrote:
> > On Monday 11 January 2016 13:10:10 John Kasunich wrote:
> > > SSRs can fail shorted, I don't think this buys you much in terms
> > > of reliability and safety.
> > >
> > > I think there is a little over-kill going on.
> > >
> > > Do you really need a fast bleeder (<1 minute to safe)?  Or would a
> > > five minute bleeder be OK?
> > >
> > > My understanding is that you have two 68000uF caps in series,
> > > charged to about 130V. (If this is incorrect let me know - the
> > > thread is long and rambling and refers to other thread.  I'm
> > > basing this on something written in the first message of this
> > > particular thread.)
> >
> > Actually 10 6800 u-f in parallel, working at their labeled 63 volts
> > ratings.
> >
> > But yes, you have the right idea. :)
> >
> > > Series caps divide, so you have 34000uF.  Let's say that "safe" is
> > > 20V.  So you need to discharge from 130V to 20V, that is to 15% of
> > > the original voltage.  Ln(0.15) is -1.9, so it takes 1.9 RC time
> > > constants to get to a safe voltage.  Call it 2.0 for easy math. 
> > > If you want to get there in 5 minutes, then one time constant is
> > > 2.5 minutes = 150 seconds. T = RC, solve for R, gives you R = T/C
> > > = 4411 ohms.  At 130V, that would draw 29mA and burn 3.8 watts. 
> > > Next lower 10% value is 3.9K, that would draw and burn 4.3 watts.
> > > I like to derate power resistors by 50%, so you want a 3.9K 10W
> > > resistor.
> >
> > I actually have, atm, 6 ea 5k 10 watters in parallel clipped onto it
> > right now that I was going to use once the relays get here.  Thats
> > 833 ohms which would drain it even faster than just the one.
>
> The 5K's are across the entire bus, or across each cap?

The bus, since there are 4 toroids and rectifiers, (dictated by the 
toroids I could source for close to nothing,) is arranged as 5 caps per 
group, in a 2 in parallel, stacked two more on top of that so there is 
one 5 k R across each 5 caps. 20 of those 6800x63 caps total. I did not 
have a bar joining the center of the stack, still don't.  So there's a 
possibility of a volt or so's miss-match there.
>
> If across the entire bus, 130V would result in 3.38 watts per
> resistor. Less than the 5W rating, but more than half the rating,
> which is the derate I like to use to keep things from getting too hot
> and ensure long life.

These are 10 watt R's.
>
> I'd suggest 3 of the 5K across each side of the bus.  At 65V across 5K
> each resistor would be running at 0.85W, nice and cool.  Total
> resistance of the 3-parallel 2-series arrangement would be 3.33K, RC
> time constant 113 seconds, bleed-down time from 130V to 20V = 212
> seconds = 3.5 minutes.  And free voltage balancing between the two
> sets of caps.
>
> > And, somewhere
> > on an undermanned galley in the pacific, is a 250 ohm 250 watt
> > resistor which would do the bleed-down even faster, and the current
> > peak would be nominally .5 amps in that case. I can't see that as
> > being capable of welding a relay contact.
>
> Making 0.5A probably won't weld any contacts.  Breaking 0.5A DC on
> a relay designed for AC is another story.  And keep in mind that
> contact bounce = breaking and making.

True, very true.

> I'm just not a big believer in switched bleeders.  Significant hassle
> and risk.  If fixed bleeders can meet your discharge time
> requirements, call it good and move on.

That, once I put in GOOD charge-pump detectors, is exactly what I'm going 
to do.  And when I put the cover back on it, a dymo label that reminds  
me that the minimum safe off time is at least 20 minutes.

> Context - I design large motor drives for a living.  The vast majority
> used permanent bleeders only.  One recent design does augment them
> with a switched bleeder.  But that drive has 90 caps, each rated
> 6800uF 400V, in a 30-parallel by 3-series array, total of 68000uF. 
> Nominal bus voltage is 975V DC.  Stored energy is 32 kilo-joules.  UL
> requires that it discharge below 50V in five minutes.  The fixed
> bleeders would do it in 10.5 minutes, so we added a switched bleeder
> to help it along.
>
> Your unit has 34000uF at 130V, for a stored energy of 287 joules.
>
> > > If you do indeed have two caps in series, they ought to have
> > > balancing resistors anyway, so you could split the bleeder into
> > > two sections. Make each section 2K or 2.2K and 5 watts.
> >
> > That too, I was contemplating, but the 51 ohmmer would need to be
> > dismounted to gain access to the pcb bus connecting them all
> > together. Since I have them already 2 each of the 5k's across each
> > bank would seem to be about right.  They are well discharged atm,
> > and I could do that yet this afternoon.  If my math is correct, at
> > 5k and 63.5 volts=0.80645 watts per, so if I spread them out on the
> > busses, the heat shouldn't hurt the caps.
>
> Agreed.  But why not use all 6 of the 5K?

Re: [Emc-users] Another crazy idea?

[Gene Heskett]

On Monday 11 January 2016 13:33:22 Bertho Stultiens wrote:

> On 01/11/2016 06:01 PM, Gene Heskett wrote:
> > 1. put a small tranny delivering say 6 volts AC in series with the
> > resistor and its shorting SSR to encourage that SSR to turn off when
> > the cap voltage has dropped to 6 volts or less.
>
> How do you connect the 6V AC without it bearing a huge (peak) current?
>
> If you have a DC signal in it too, it will feedback into the primary.
> That is a huge problem. Not to speak of the inductance.
>
Probably, I hadn't though it all the way thru.

> > 2. enforce in hal, 2 delay circuits.  An on delay long enough for #1
> > to have completed its job and the short across the resistor no
> > longer exists thru that pair of SSR's.
>
> Enforcing a software solution is a risk. If the software fails, then
> you get potentially fried? You'd normally have an electronic interlock
> to prevent problems.
>
> > 3. Time the short enable so its 5 to 10 seconds after the relay has
> > closed, and time the short disable to take place a second or so
> > before the relay is de-energized, so that SSR is off by the time the
> > switch of the resistors position in the circuit takes place. By
> > placing another timedelay in the relay control, with longer time
> > delays than the shorting time delay would enforce disconnecting the
> > short whose timedelay is set for 1 second, so the resistor is
> > switched back into the circuit well before its transferred to dump
> > the caps duty.
>
> The real problem of shorting the resistor are the huge peak currents.

Miss-conception John. Its shorted after the caps are nearly charged.  
When the final short comes on, the caps are up to 110-teens already, so 
the remaining charge isn't much.  I put an amprobe on the black wire of 
the AC feed, and its a blond one north of 3 amps when the motor is doing 
2500 revs cutting air.

> You said that you have two 68000uF caps in series, or 34000uF
> effectively on it, which suggests that you are handling an average DC
> current somewhere between 30...60A.

No, I just over estimated the u-f's I'd need for a 1HP motor.  The 
current capacity is there should something short it out, but thats a 
fault condition, nothing approaching that will ever occur in normal 
operation even if I should manage to lock the rotor, the servo amp is 
set for a current limit in the 14.5 amp range, 150% of the motors 
nameplate draw at full song.  The toroids might warm up 30 degrees doing 
it for 30 minutes, but Jon's driver will survive that.

> The current peak from the rectifier, just to maintain that DC current
> after the cap, is somewhere between 120...240A (estimated) and even
> higher currents are possible. That will fuse most normal relays.

Rated 410 amps for these 40 amp CCS relays, but that would not be thru 
the relay contacts ever.

With that resistor in circuit for the first 5 or 6 seconds, the max amps 
is nominally 2.5 rms for the first 1/2 second or so, and tapers to very 
little as the caps charge.

> SSRs would probably survive is you get one and stay well below the
> fusing current (specified in the I^2t parameter) and stay below the
> peak current rating.
> However, SSRs do not like to be operated at high 
> peak currents for long periods.
>
> > Does anyone have an data on how sensitive these SSR's are to a false
> > trigger from dv/dt effects applied to the output terminals?
>
> They all have that specified in the datasheet. Most large ones are
> specified at about 500V/us.

These are SSR-40DA's, and the spec sheet I'm looking at makes no 
reference to that, only the maximum "durated" current, which as I read 
the chinglish translation means 1 cycle duration.  I doubt if they'll 
ever see 10% of that.

That doesn't sound appetizing in view of the hard switch the relay would 
make.  So scrub that.

> > Or do I need to use 2 of these relays with a fraction of a second
> > between them, to assure the line side SSR's have time enough to turn
> > off? At 60Hz thats 8. milliseconds after drive has been removed,
> > and the single relay could be faster than that.  All TBD when the
> > relays arrive I guess.  They haven't yet.
>
> Well, it looks like a very complicated and rather risky setup.

Well, with better charge-pump-detectors, what I have after adding those 
dual 5k r's across each cap bank, which is working fairly well, if a 
second or so slow, right now. Off time back to 40 volts is 4 minutes 
even, and is down to 7 volts in about 25 minutes.  I got tired of 
resetting the meter because it was timing out and shutting down.

It was working, and I had taken measurements and sawed off a 4" piece of 
white 1.25" PVC that I am going to bore out to about 1.53" for about 
3/4" on one end for a tight friction fit on the exhaust snout of the 
sander, and turn the OD down to around 1.33" for about an inch on the 
other end, which will serve as the adaptor to hook it up to a vac for 
dust collection.  Where I am right now on these chest lids is a 

Re: [Emc-users] Another crazy idea?

[Erik Christiansen]

On 11.01.16 14:00, Gene Heskett wrote:
> 
> I actually have, atm, 6 ea 5k 10 watters in parallel clipped onto it 
> right now that I was going to use once the relays get here.  Thats 833  
> ohms which would drain it even faster than just the one.  And, somewhere 
> on an undermanned galley in the pacific, is a 250 ohm 250 watt resistor 
> which would do the bleed-down even faster, and the current peak would be 
> nominally .5 amps in that case. I can't see that as being capable of 
> welding a relay contact.

There are alternatives to the electromechanical relay or an
SCR/Triac-based SSR, such as opto-coupled FET switches. They switch off,
even on DC. In my junkbox I (think I still) have some

http://documentation.renesas.com/doc/YOUSYS/document/PN10273EJ02V0DS.pdf

With both FETs in parallel, it'll just handle 0.5A, so adding some ohms
in series with the bleeder would be wise in this case. There are
doubtless beefier units out there by now.

Note: If used to switch AC, the FETs need to be used in series, halving
  the current capacity.

That brief technology diversion aside, I'd just go with fixed slow
bleeders, I think.¹ (Though a small relay buffered by a big MOSFET would
solve all your contact welding concerns, and allow a low value bleeder
in the drain circuit. The MOSFET would only switch, obviating your
concerns with running one in the linear region. OK, in the last few
volts, gate drive would fall off, unless you added a diode to keep
charge on the gate.)

¹ Especially given John's suggestion that a split bleeder can serve for
voltage equalisation across series capacitors. You're not doing that,
instead running the electros at full rated voltage? In the Siemens
design labs we were never allowed to design that way. Half of rated was
the allowed max, for reliability. I like to stick to 60% for my own
stuff, still.

Where I'm happier with a switched resistor is as inrush limiter.

Erik

--
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=267308311=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Another crazy idea?

[Gene Heskett]

On Monday 11 January 2016 21:07:35 John Kasunich wrote:

> On Mon, Jan 11, 2016, at 05:41 PM, Gene Heskett wrote:
> > On Monday 11 January 2016 14:44:54 John Kasunich wrote:
> > > On Mon, Jan 11, 2016, at 02:00 PM, Gene Heskett wrote:
> > > > On Monday 11 January 2016 13:10:10 John Kasunich wrote:
> > > > I actually have, atm, 6 ea 5k 10 watters in parallel clipped
> > > > onto it
> > >
> > > The 5K's are across the entire bus, or across each cap?
> > > If across the entire bus, 130V would result in 3.38 watts per
> > > resistor. Less than the 5W rating, but more than half the rating,
> > > which is the derate I like to use to keep things from getting too
> > > hot and ensure long life.
> >
> > These are 10 watt R's.
>
> My mistake, I somehow read that as 5K 5W.
>
> > > Agreed.  But why not use all 6 of the 5K?
> >
> > I could, if I tied the center points of the stacks together,
> > otherwise I'd need 8 so I could put 2 on each group.
>
> Understood.
>
> Although, since they are 10W, you could put the last two across the
> entire bus to speed up the discharge.

Humm, damn I must be getting slow, I should have thought of that.  Now I 
have to take it all apart again. ;-)  After I collect some ZZ's though.

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 

--
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=267308311=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Another crazy idea?

[John Kasunich]

On Mon, Jan 11, 2016, at 05:41 PM, Gene Heskett wrote:
> On Monday 11 January 2016 14:44:54 John Kasunich wrote:
> 
> > On Mon, Jan 11, 2016, at 02:00 PM, Gene Heskett wrote:
> > > On Monday 11 January 2016 13:10:10 John Kasunich wrote:
> > > I actually have, atm, 6 ea 5k 10 watters in parallel clipped onto it
> >
> > The 5K's are across the entire bus, or across each cap?
> > If across the entire bus, 130V would result in 3.38 watts per
> > resistor. Less than the 5W rating, but more than half the rating,
> > which is the derate I like to use to keep things from getting too hot
> > and ensure long life.
> 
> These are 10 watt R's.

My mistake, I somehow read that as 5K 5W.

> > Agreed.  But why not use all 6 of the 5K?
> 
> I could, if I tied the center points of the stacks together, otherwise 
> I'd need 8 so I could put 2 on each group.
> 

Understood.

Although, since they are 10W, you could put the last two across the
entire bus to speed up the discharge.

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

--
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=267308311=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Another crazy idea?

[Gene Heskett]

On Tuesday 12 January 2016 00:37:55 Erik Christiansen wrote:

> On 11.01.16 14:00, Gene Heskett wrote:
> > I actually have, atm, 6 ea 5k 10 watters in parallel clipped onto it
> > right now that I was going to use once the relays get here.  Thats
> > 833 ohms which would drain it even faster than just the one.  And,
> > somewhere on an undermanned galley in the pacific, is a 250 ohm 250
> > watt resistor which would do the bleed-down even faster, and the
> > current peak would be nominally .5 amps in that case. I can't see
> > that as being capable of welding a relay contact.
>
> There are alternatives to the electromechanical relay or an
> SCR/Triac-based SSR, such as opto-coupled FET switches. They switch
> off, even on DC. In my junkbox I (think I still) have some
>
> http://documentation.renesas.com/doc/YOUSYS/document/PN10273EJ02V0DS.p
>df
>
> With both FETs in parallel, it'll just handle 0.5A, so adding some
> ohms in series with the bleeder would be wise in this case. There are
> doubtless beefier units out there by now.
>
> Note: If used to switch AC, the FETs need to be used in series,
> halving the current capacity.
>
> That brief technology diversion aside, I'd just go with fixed slow
> bleeders, I think.¹ (Though a small relay buffered by a big MOSFET
> would solve all your contact welding concerns, and allow a low value
> bleeder in the drain circuit. The MOSFET would only switch, obviating
> your concerns with running one in the linear region. OK, in the last
> few volts, gate drive would fall off, unless you added a diode to keep
> charge on the gate.)
>
> ¹ Especially given John's suggestion that a split bleeder can serve
> for voltage equalisation across series capacitors. You're not doing
> that, instead running the electros at full rated voltage? In the
> Siemens design labs we were never allowed to design that way. Half of
> rated was the allowed max, for reliability. I like to stick to 60% for
> my own stuff, still.

Yes, 63 volt rated caps, running at 63 volts.  They have formed up to 
such low leakage that without the bleeders, they are still dangerous 2 
days after the shut down.  No clue how long they will last but I suspect 
they were a decade+ old when I pulled the 20 some I used out a bushel 
sized box.  Dave was going to make hisself a 10 kw PA amp that would 
carry the Grateful Dead from his place clear into Ellamore WV, about 5 
miles. GD was the star attraction at a charity concert he threw at his 
place every summer for nearly 25 years while he was working for me at 
the tv station.  It was called, because the stage was across the road 
from the river, River Rocks.  But he's retired, pushing 70 and 
essentially broke, a small vet pension and SS, but his 40 acres is paid 
for so he hasn't done the concert in 3 or 4 years.  He putters about & 
keeps busy.  I guess that would also describe me. :)

> Where I'm happier with a switched resistor is as inrush limiter.
>
> Erik
>
> --
> Site24x7 APM Insight: Get Deep Visibility into Application
> Performance APM + Mobile APM + RUM: Monitor 3 App instances at just
> $35/Month Monitor end-to-end web transactions and take corrective
> actions now Troubleshoot faster and improve end-user experience.
> Signup Now!
> http://pubads.g.doubleclick.net/gampad/clk?id=267308311=/4140
> ___
> 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 

--
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=267308311=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Another crazy idea?

[Gene Heskett]

On Friday 08 January 2016 23:44:43 Gene Heskett wrote:

> On Thursday 07 January 2016 05:27:09 Gene Heskett wrote:
> > On Thursday 07 January 2016 05:07:25 Gene Heskett wrote:
> > > On Thursday 07 January 2016 04:50:35 andy pugh wrote:
> > > > On 7 January 2016 at 03:32, Gene Heskett  
wrote:
> > > > >  > > > >4P DT -1
> > > > > 4-Pin-10A-250VAC-With-Socket-Base-/120897376163?hash=item1c260
> > > > >b8 fa 3:g
> > > > >
> > > > >:3~kAAMXQgPhRkf79>
> > > > >
> > > > > That gives me a 4pdt,
> > > >
> > > > It doesn't mention being force-guided.
> > >
> > > That is a relay term I am not familiar with, please define?
> >
> > I looked it up on wikipedia. It does seem to be a good safety
> > feature. Perhaps I should interlock its operation by assuring it is
> > switched at no load by turning off a series SSR? I'll see what I can
> > cobble up in the hal file.
>
> As an update, sorta. The parts I ordered are trickling in, the
> schottky diodes and tantalum caps have arrived, as has a small 6.3V ct
> filament transformer in case I need to encourage a faster turn off the
> bleeder, but they aren't much use until the smallish hexfets to drive
> relays and SSR's with get here.
>
> I've changed my mind a bit since I'll have a 4pdt relay. The swinging
> contacts of 2 poles will connect both ends of that 51 ohm 200 watt
> resistor in series with the line power to the toroids when enabled,
> and switch it across the caps for a bleeder when de-energized.  The
> 3rd set of contacts will parallel the resistor with an SSR, shorting
> it when the on timeout has expired.
>
> This will also, by interrupting that path as the relay opens, remove
> any possibility of the resistor being shorted when the relay closes.
> The 4th set will feed back to hal, hitting a short timer to make sure
> the softstart done SSR is off before it can be re-enabled, enforcing a
> 1 or 2 second delay in that event.  I haven't located an input pin on
> that BoB, but as thats on the P2 connector of a 5i25, there are
> several pins available on it yet.
>
> I also have sourced a 100 foot roll of 26 conductor ribbon, and a bag
> of db25 connectors of both genders so I can make cables that don't hit
> the wall behind the computer, trying to knock the 5i25 out of its
> socket. The cable has arrived, but the connectors are still on a
> rowboat someplace...
>
> Thanks everybody.
>
> Cheers, Gene Heskett

Continued...

Andy's comment about that relay not being force guided, meaning it 
wouldn't be prevented from a misscue if a contact welded, got me 
thinking.  Which can be entertaining at times...

What if I used it essentially for a dry contact, which has its own 
closure problems, but used the contacts to steer SSR's?

1 set of spdt contacts would swing a 25 volt source thru 2 SSR's control 
terminals. That pair would be wired to connect the line power to one end 
of that resistor, while the other end of the resistor would be connected 
to the toroid power trannies by the 2nd SSR.  When they are off, the 
resistor is isolated, so...

The back, nc side of that same spdt section would connect the resistor 
across the cap bank with 2 more SSR's to serve as a suitable bleeder 
when the relay was de-energized.

The problem with that is that there needs to be a method of turning the 
SSR's off because the 2nd condition is DC and I've no clue as to the 
holding current they need to stay on.

And I will still be shorting the resistor after a turn on delay to charge 
the caps to about 90% of full voltage.

So how about 2 things:

1. put a small tranny delivering say 6 volts AC in series with the 
resistor and its shorting SSR to encourage that SSR to turn off when the 
cap voltage has dropped to 6 volts or less.

2. enforce in hal, 2 delay circuits.  An on delay long enough for #1 to 
have completed its job and the short across the resistor no longer 
exists thru that pair of SSR's.

3. Time the short enable so its 5 to 10 seconds after the relay has 
closed, and time the short disable to take place a second or so before 
the relay is de-energized, so that SSR is off by the time the switch of 
the resistors position in the circuit takes place. By placing another 
timedelay in the relay control, with longer time delays than the 
shorting time delay would enforce disconnecting the short whose 
timedelay is set for 1 second, so the resistor is switched back into the 
circuit well before its transferred to dump the caps duty.

All proved out on the scope before hooking up the power source of course.

That will take one more SSR than I have, but I got those 4 from MPJones 
fairly fast and relatively cheap, like about $9 USD a copy.

Does anyone have an data on how sensitive these SSR's are to a false 
trigger from dv/dt effects applied to the output terminals?

Or do I need to use 2 of these relays with a fraction of a second between 
them, to assure the line side SSR's have time enough to 

Re: [Emc-users] Another crazy idea?

[John Kasunich]

SSRs can fail shorted, I don't think this buys you much in terms of
reliability and safety.

I think there is a little over-kill going on.

Do you really need a fast bleeder (<1 minute to safe)?  Or would a
five minute bleeder be OK?

My understanding is that you have two 68000uF caps in series, charged to about 
130V.
(If this is incorrect let me know - the thread is long and rambling and refers 
to other 
thread.  I'm basing this on something written in the first message of this 
particular
thread.)

Series caps divide, so you have 34000uF.  Let's say that "safe" is 20V.  So you 
need to
discharge from 130V to 20V, that is to 15% of the original voltage.  Ln(0.15) 
is -1.9, so
it takes 1.9 RC time constants to get to a safe voltage.  Call it 2.0 for easy 
math.  If you
want to get there in 5 minutes, then one time constant is 2.5 minutes = 150 
seconds.
T = RC, solve for R, gives you R = T/C = 4411 ohms.  At 130V, that would draw 
29mA 
and burn 3.8 watts.  Next lower 10% value is 3.9K, that would draw and burn 4.3 
watts.
I like to derate power resistors by 50%, so you want a 3.9K 10W resistor.  

If you do indeed have two caps in series, they ought to have balancing 
resistors anyway,
so you could split the bleeder into two sections.  Make each section 2K or 2.2K 
and 5 watts.

Something like this can be chassis mounted and wired to the caps:
http://www.digikey.com/product-detail/en/RH0052K000FE02/RHRA-2.0K-ND/1166266
Cost is $5 each, total $10

Even cheaper would be to parallel up some 1W leaded resistors:
http://www.digikey.com/product-detail/en/PR01000101002JR500/PPC10KW-1CT-ND/597177
$2.06 for ten - five in parallel will make 2K 5W, put one set of five across 
each cap.
Might be a nuisance to mount.

If you have can type caps with screw terminals, use power resistors with solid 
axial leads,
crimp/solder them to ring lugs, and mount right on top of the caps:
http://www.digikey.com/product-detail/en/SQP10AJB-2K0/2.0KW-10-ND/18789
$0.64 each, total $1.28

I realize that Digikey might not be your supplier of choice due to shipping 
costs, but my
point is that an always-on bleeder is simple, safe, and at this power level, 
cheap.  As
you get to larger energy levels (more volts or more uF) or if you need a much 
faster
discharge time you might be able to justify a switched fast bleeder.  But that 
will ALWAYS
be less reliable and less safe than an always-on bleeder.  Even if you have a 
switched
bleeder it would be wise to have an permanent bleeder with maybe a 10 minute 
time
constant.  At 10 minutes, the power and cost are negligible, and it serves 
double duty
as the voltage balancing resistors for your series caps.

Inrush limiting is a separate issue.  Separate resistor, bypassed by a 
contactor or relay
for normal operation.  Put it on the AC side so you don't have any issues with 
DC contact
ratings.  The main risk is that if you try to run with the bypass relay open 
you can fry your
resistor.  One way to avoid that is to make the "resistor" a 100W incandescent 
light bulb.

Using bulbs for bleeders is risky because they run all the time, and when they 
burn 
out they leave the caps charged with no indication of the risk.  But using a 
bulb for 
charging is different.  The bulb only lights for a few seconds when you first 
apply 
power.  If it is burned out the caps won't charge.  If you try to run with the 
bypass 
open the light will glow to let you know.

A 100W 120V bulb has a hot resistance of 144 ohms.  With 34000uF the charging
time constant is 4.9 seconds.  As the bus charges the voltage drop across the
bulb goes down, it cools down, and its resistance decreased - that is good, it 
make
the "long tail" of the charging curve go faster.  The inrush during the first 
cycle or
so will be based on the cold resistance of the bulb, but if your switch and 
diodes
can handle a 100W light bulb you know it can handle any size cap bank fed thru
that same 100W light bulb.


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

--
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=267308311=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Another crazy idea?

[Gene Heskett]

On Monday 11 January 2016 13:10:10 John Kasunich wrote:

> SSRs can fail shorted, I don't think this buys you much in terms of
> reliability and safety.
>
> I think there is a little over-kill going on.
>
> Do you really need a fast bleeder (<1 minute to safe)?  Or would a
> five minute bleeder be OK?
>
> My understanding is that you have two 68000uF caps in series, charged
> to about 130V. (If this is incorrect let me know - the thread is long
> and rambling and refers to other thread.  I'm basing this on something
> written in the first message of this particular thread.)

Actually 10 6800 u-f in parallel, working at their labeled 63 volts 
ratings.

But yes, you have the right idea. :)
>
> Series caps divide, so you have 34000uF.  Let's say that "safe" is
> 20V.  So you need to discharge from 130V to 20V, that is to 15% of the
> original voltage.  Ln(0.15) is -1.9, so it takes 1.9 RC time constants
> to get to a safe voltage.  Call it 2.0 for easy math.  If you want to
> get there in 5 minutes, then one time constant is 2.5 minutes = 150
> seconds. T = RC, solve for R, gives you R = T/C = 4411 ohms.  At 130V,
> that would draw 29mA and burn 3.8 watts.  Next lower 10% value is
> 3.9K, that would draw and burn 4.3 watts. I like to derate power
> resistors by 50%, so you want a 3.9K 10W resistor.

I actually have, atm, 6 ea 5k 10 watters in parallel clipped onto it 
right now that I was going to use once the relays get here.  Thats 833  
ohms which would drain it even faster than just the one.  And, somewhere 
on an undermanned galley in the pacific, is a 250 ohm 250 watt resistor 
which would do the bleed-down even faster, and the current peak would be 
nominally .5 amps in that case. I can't see that as being capable of 
welding a relay contact.


> If you do indeed have two caps in series, they ought to have balancing
> resistors anyway, so you could split the bleeder into two sections. 
> Make each section 2K or 2.2K and 5 watts.
>
That too, I was contemplating, but the 51 ohmmer would need to be 
dismounted to gain access to the pcb bus connecting them all together.  
Since I have them already 2 each of the 5k's across each bank would seem 
to be about right.  They are well discharged atm, and I could do that 
yet this afternoon.  If my math is correct, at 5k and 63.5 volts=0.80645 
watts per, so if I spread them out on the busses, the heat shouldn't 
hurt the caps.  And my existing SSR lashup would be fine when the rest 
of the parts to make a decent charge-pump-detector to drive the SSR's 
with do arrive.

I can do that yet this afternoon if I can get high enough on a ladder to 
turn the unit face up so the soldering is easy. There's a 4 footer 
currently parked 3 feet away. :) The problem is the floor real estate to 
spread it out.

> Something like this can be chassis mounted and wired to the caps:
> http://www.digikey.com/product-detail/en/RH0052K000FE02/RHRA-2.0K-ND/1
>166266 Cost is $5 each, total $10
>
> Even cheaper would be to parallel up some 1W leaded resistors:
> http://www.digikey.com/product-detail/en/PR01000101002JR500/PPC10KW-1C
>T-ND/597177 $2.06 for ten - five in parallel will make 2K 5W, put one
> set of five across each cap. Might be a nuisance to mount.
>
The cap busses are nominally a foot long, 3/8" wide 2oz copper

> If you have can type caps with screw terminals, use power resistors
> with solid axial leads, crimp/solder them to ring lugs, and mount
> right on top of the caps:
> http://www.digikey.com/product-detail/en/SQP10AJB-2K0/2.0KW-10-ND/1878
>9 $0.64 each, total $1.28
>
> I realize that Digikey might not be your supplier of choice due to
> shipping costs, but my point is that an always-on bleeder is simple,
> safe, and at this power level, cheap.  As you get to larger energy
> levels (more volts or more uF) or if you need a much faster discharge
> time you might be able to justify a switched fast bleeder.  But that
> will ALWAYS be less reliable and less safe than an always-on bleeder. 
> Even if you have a switched bleeder it would be wise to have an
> permanent bleeder with maybe a 10 minute time constant.  At 10
> minutes, the power and cost are negligible, and it serves double duty
> as the voltage balancing resistors for your series caps.
>
> Inrush limiting is a separate issue.  Separate resistor, bypassed by a
> contactor or relay for normal operation.  Put it on the AC side so you
> don't have any issues with DC contact ratings.

That is what I am doing right now with the SSR's, one to switch on the 
power thru that 51 ohm resistor, and a second SSR about 5 seconds later 
to short out the resistor.

> The main risk is that 
> if you try to run with the bypass relay open you can fry your
> resistor.  One way to avoid that is to make the "resistor" a 100W
> incandescent light bulb.

Which rigid tapping would quickly make history.  The spindle turnaround, 
even if I have it rate restricted, still pumps the supply high enough to 
blow a C7 bulb in 4 or 5 pecks.  

Re: [Emc-users] Another crazy idea?

[Bertho Stultiens]

On 01/11/2016 06:01 PM, Gene Heskett wrote:
> 1. put a small tranny delivering say 6 volts AC in series with the 
> resistor and its shorting SSR to encourage that SSR to turn off when the 
> cap voltage has dropped to 6 volts or less.

How do you connect the 6V AC without it bearing a huge (peak) current?

If you have a DC signal in it too, it will feedback into the primary.
That is a huge problem. Not to speak of the inductance.


> 2. enforce in hal, 2 delay circuits.  An on delay long enough for #1 to 
> have completed its job and the short across the resistor no longer 
> exists thru that pair of SSR's.

Enforcing a software solution is a risk. If the software fails, then you
get potentially fried? You'd normally have an electronic interlock to
prevent problems.


> 3. Time the short enable so its 5 to 10 seconds after the relay has 
> closed, and time the short disable to take place a second or so before 
> the relay is de-energized, so that SSR is off by the time the switch of 
> the resistors position in the circuit takes place. By placing another 
> timedelay in the relay control, with longer time delays than the 
> shorting time delay would enforce disconnecting the short whose 
> timedelay is set for 1 second, so the resistor is switched back into the 
> circuit well before its transferred to dump the caps duty.

The real problem of shorting the resistor are the huge peak currents.
You said that you have two 68000uF caps in series, or 34000uF
effectively on it, which suggests that you are handling an average DC
current somewhere between 30...60A.

The current peak from the rectifier, just to maintain that DC current
after the cap, is somewhere between 120...240A (estimated) and even
higher currents are possible. That will fuse most normal relays.

SSRs would probably survive is you get one and stay well below the
fusing current (specified in the I^2t parameter) and stay below the peak
current rating. However, SSRs do not like to be operated at high peak
currents for long periods.


> Does anyone have an data on how sensitive these SSR's are to a false 
> trigger from dv/dt effects applied to the output terminals?

They all have that specified in the datasheet. Most large ones are
specified at about 500V/us.


> Or do I need to use 2 of these relays with a fraction of a second between 
> them, to assure the line side SSR's have time enough to turn off? At 
> 60Hz thats 8. milliseconds after drive has been removed, and the 
> single relay could be faster than that.  All TBD when the relays arrive 
> I guess.  They haven't yet.

Well, it looks like a very complicated and rather risky setup.


-- 
Greetings Bertho

(disclaimers are disclaimed)

--
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=267308311=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Another crazy idea?

[John Kasunich]

On Mon, Jan 11, 2016, at 02:00 PM, Gene Heskett wrote:
> On Monday 11 January 2016 13:10:10 John Kasunich wrote:
> 
> > SSRs can fail shorted, I don't think this buys you much in terms of
> > reliability and safety.
> >
> > I think there is a little over-kill going on.
> >
> > Do you really need a fast bleeder (<1 minute to safe)?  Or would a
> > five minute bleeder be OK?
> >
> > My understanding is that you have two 68000uF caps in series, charged
> > to about 130V. (If this is incorrect let me know - the thread is long
> > and rambling and refers to other thread.  I'm basing this on something
> > written in the first message of this particular thread.)
> 
> Actually 10 6800 u-f in parallel, working at their labeled 63 volts 
> ratings.
> 
> But yes, you have the right idea. :)
> >
> > Series caps divide, so you have 34000uF.  Let's say that "safe" is
> > 20V.  So you need to discharge from 130V to 20V, that is to 15% of the
> > original voltage.  Ln(0.15) is -1.9, so it takes 1.9 RC time constants
> > to get to a safe voltage.  Call it 2.0 for easy math.  If you want to
> > get there in 5 minutes, then one time constant is 2.5 minutes = 150
> > seconds. T = RC, solve for R, gives you R = T/C = 4411 ohms.  At 130V,
> > that would draw 29mA and burn 3.8 watts.  Next lower 10% value is
> > 3.9K, that would draw and burn 4.3 watts. I like to derate power
> > resistors by 50%, so you want a 3.9K 10W resistor.
> 
> I actually have, atm, 6 ea 5k 10 watters in parallel clipped onto it 
> right now that I was going to use once the relays get here.  Thats 833  
> ohms which would drain it even faster than just the one. 

The 5K's are across the entire bus, or across each cap?

If across the entire bus, 130V would result in 3.38 watts per resistor. 
Less than the 5W rating, but more than half the rating, which is the derate
I like to use to keep things from getting too hot and ensure long life.

I'd suggest 3 of the 5K across each side of the bus.  At 65V across 5K
each resistor would be running at 0.85W, nice and cool.  Total resistance
of the 3-parallel 2-series arrangement would be 3.33K, RC time constant
113 seconds, bleed-down time from 130V to 20V = 212 seconds = 3.5
minutes.  And free voltage balancing between the two sets of caps.

> And, somewhere 
> on an undermanned galley in the pacific, is a 250 ohm 250 watt resistor 
> which would do the bleed-down even faster, and the current peak would be 
> nominally .5 amps in that case. I can't see that as being capable of 
> welding a relay contact.

Making 0.5A probably won't weld any contacts.  Breaking 0.5A DC on
a relay designed for AC is another story.  And keep in mind that contact
bounce = breaking and making.

I'm just not a big believer in switched bleeders.  Significant hassle and
risk.  If fixed bleeders can meet your discharge time requirements, call
it good and move on.

Context - I design large motor drives for a living.  The vast majority 
used permanent bleeders only.  One recent design does augment them
with a switched bleeder.  But that drive has 90 caps, each rated 6800uF
400V, in a 30-parallel by 3-series array, total of 68000uF.  Nominal bus
voltage is 975V DC.  Stored energy is 32 kilo-joules.  UL requires that
it discharge below 50V in five minutes.  The fixed bleeders would do it
in 10.5 minutes, so we added a switched bleeder to help it along.

Your unit has 34000uF at 130V, for a stored energy of 287 joules.
 
> > If you do indeed have two caps in series, they ought to have balancing
> > resistors anyway, so you could split the bleeder into two sections. 
> > Make each section 2K or 2.2K and 5 watts.
> >
> That too, I was contemplating, but the 51 ohmmer would need to be 
> dismounted to gain access to the pcb bus connecting them all together.  
> Since I have them already 2 each of the 5k's across each bank would seem 
> to be about right.  They are well discharged atm, and I could do that 
> yet this afternoon.  If my math is correct, at 5k and 63.5 volts=0.80645 
> watts per, so if I spread them out on the busses, the heat shouldn't 
> hurt the caps.  

Agreed.  But why not use all 6 of the 5K?

> >
> > Inrush limiting is a separate issue.  Separate resistor, bypassed by a
> > contactor or relay for normal operation.  Put it on the AC side so you
> > don't have any issues with DC contact ratings.
> 
> That is what I am doing right now with the SSR's, one to switch on the 
> power thru that 51 ohm resistor, and a second SSR about 5 seconds later 
> to short out the resistor.
> 
> > The main risk is that 
> > if you try to run with the bypass relay open you can fry your
> > resistor.  One way to avoid that is to make the "resistor" a 100W
> > incandescent light bulb.
> 
> Which rigid tapping would quickly make history.  The spindle turnaround, 
> even if I have it rate restricted, still pumps the supply high enough to 
> blow a C7 bulb in 4 or 5 pecks.  Jon's pwm servo amp is effectively a 
> full 4 quadrant controller, so it dumps 

Re: [Emc-users] Another crazy idea?

[Gene Heskett]

On Thursday 07 January 2016 05:27:09 Gene Heskett wrote:

> On Thursday 07 January 2016 05:07:25 Gene Heskett wrote:
> > On Thursday 07 January 2016 04:50:35 andy pugh wrote:
> > > On 7 January 2016 at 03:32, Gene Heskett  wrote:
> > > >  > > >DT -1
> > > > 4-Pin-10A-250VAC-With-Socket-Base-/120897376163?hash=item1c260b8
> > > >fa 3:g
> > > >
> > > >:3~kAAMXQgPhRkf79>
> > > >
> > > > That gives me a 4pdt,
> > >
> > > It doesn't mention being force-guided.
> >
> > That is a relay term I am not familiar with, please define?
>
> I looked it up on wikipedia. It does seem to be a good safety feature.
> Perhaps I should interlock its operation by assuring it is switched at
> no load by turning off a series SSR? I'll see what I can cobble up in
> the hal file.

As an update, sorta. The parts I ordered are trickling in, the schottky 
diodes and tantalum caps have arrived, as has a small 6.3V ct filament 
transformer in case I need to encourage a faster turn off the bleeder, 
but they aren't much use until the smallish hexfets to drive relays and 
SSR's with get here.

I've changed my mind a bit since I'll have a 4pdt relay. The swinging 
contacts of 2 poles will connect both ends of that 51 ohm 200 watt 
resistor in series with the line power to the toroids when enabled, and 
switch it across the caps for a bleeder when de-energized.  The 3rd set 
of contacts will parallel the resistor with an SSR, shorting it when the 
on timeout has expired.

This will also, by interrupting that path as the relay opens, remove any 
possibility of the resistor being shorted when the relay closes. The 4th 
set will feed back to hal, hitting a short timer to make sure the 
softstart done SSR is off before it can be re-enabled, enforcing a 1 or 
2 second delay in that event.  I haven't located an input pin on that 
BoB, but as thats on the P2 connector of a 5i25, there are several pins 
available on it yet.

I also have sourced a 100 foot roll of 26 conductor ribbon, and a bag of 
db25 connectors of both genders so I can make cables that don't hit the 
wall behind the computer, trying to knock the 5i25 out of its socket. 
The cable has arrived, but the connectors are still on a rowboat 
someplace...

Thanks everybody.

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 

--
Site24x7 APM Insight: Get Deep Visibility into Application Performance
APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month
Monitor end-to-end web transactions and take corrective actions now
Troubleshoot faster and improve end-user experience. Signup Now!
http://pubads.g.doubleclick.net/gampad/clk?id=267308311=/4140
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Another crazy idea?

[Gene Heskett]

On Thursday 07 January 2016 05:07:25 Gene Heskett wrote:

> On Thursday 07 January 2016 04:50:35 andy pugh wrote:
> > On 7 January 2016 at 03:32, Gene Heskett  wrote:
> > >  > >-1
> > > 4-Pin-10A-250VAC-With-Socket-Base-/120897376163?hash=item1c260b8fa
> > >3:g
> > >
> > >:3~kAAMXQgPhRkf79>
> > >
> > > That gives me a 4pdt,
> >
> > It doesn't mention being force-guided.
>
> That is a relay term I am not familiar with, please define?
>
I looked it up on wikipedia. It does seem to be a good safety feature.  
Perhaps I should interlock its operation by assuring it is switched at 
no load by turning off a series SSR? I'll see what I can cobble up in 
the hal file.

> Cheers, Gene Heskett


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 

--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Another crazy idea?

[andy pugh]

On 6 January 2016 at 02:05, Gene Heskett  wrote:

> What do you think? Is that a safe and workable idea?

I have seen NC SSRs. I don't know how they do it, and they are expensive.

I have been using normally-closed mechanical relays to do this, with a
240V coil connected to incoming power.

Ideally it needs to be a force-guided relay, with the main power feed
(or surrogate thereof) through a second set of terminals so that if
the cap-drain contacts weld shut you can't power up the system with
the drain resistor in circuit.

You have to ensure that the PSU isn't powered up until the caps are
discharged. I do this in HAL because my drives feed-back bus voltage
to HAL.

-- 
atp
If you can't fix it, you don't own it.
http://www.ifixit.com/Manifesto

--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Another crazy idea?

[Gene Heskett]

On Wednesday 06 January 2016 05:13:34 andy pugh wrote:

> On 6 January 2016 at 02:05, Gene Heskett  wrote:
> > What do you think? Is that a safe and workable idea?
>
> I have seen NC SSRs. I don't know how they do it, and they are
> expensive.
>
> I have been using normally-closed mechanical relays to do this, with a
> 240V coil connected to incoming power.
>
> Ideally it needs to be a force-guided relay, with the main power feed
> (or surrogate thereof) through a second set of terminals so that if
> the cap-drain contacts weld shut you can't power up the system with
> the drain resistor in circuit.
>
> You have to ensure that the PSU isn't powered up until the caps are
> discharged. I do this in HAL because my drives feed-back bus voltage
> to HAL.

That thought has crossed my mind too, but the power relays I have, with 
an aux set of contacts that could feed the interlock, are all NO types. 
NC versions seem rare, and pricy. $100 USD range.  The $9 SSR will work, 
but I've no solid tally back to make sure its recovered other than 
enforcing a time delay that gives it plenty of time to recover at 
startup.  As it takes time to home it, and do any touch-offs that may be 
needed to setup the job, I don't see that a 2 minute delay to a usable 
spindle would be a huge handicap.

Checking the pdf of another brand, Fotek, also a Chinese part, fails to 
disclose a minimum holding current, just states its a zero crossing 
control. That gives me the idea of using a low voltage AC as the shutoff 
mechanism.  One clue may also be the minimum applied AC voltage of 24 
volts. Stated as: 24-380VAC.

Using the LV AC in series with the load r's as a shutoff brings up the 
need to restrict the current to about the rated current of that 
transformers low voltage delivery because any higher than that might 
saturate the core with destructive primary currants flowing.  So that 
transformer should probably be larger that the teeny 200ma rated one 
thats in that box furnishing control power now if I elect to use a lower 
value resistor. With these resistors, the peak flow would be 
0.15240061 amps. Now if T=RC, then T is 56.44 seconds to 
drain it to 42 volts if my math is good, and another 57 seconds to 14 
volts, so even with a 12vac in series, it looks like 3 minutes to be 
safe. Thats a bit long.  Looks like I'd better go kick fleabays tires & 
see what might be available.

That, and go get the morning coffee under construction if my lady hasn't, 
I just heard a cabinet door close and a pan rattle in the kitchen. No 
clue what she is up to though as she likes to have a variety, might be 
making tea. I get sent to the store for oddish stuff they never have 
quite often. After 26 years, I still don't have her 100% understood. :)

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 

--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Another crazy idea?

[Gene Heskett]

On Wednesday 06 January 2016 02:36:57 Bertho Stultiens wrote:

> On 01/06/2016 03:05 AM, Gene Heskett wrote:
> [snip]
>
> > The idea being that when that drain goes low, it will charge the
> > .22, thru the diode, essentially maintaining that charge state.  But
> > when LCNC is stopped, that drain will go high (to 25 volts) as the
> > supply is turned off, and as it goes high, so will the far end of
> > that .22, carrying the + terminal on the third SSR far enough to
> > trigger it.
>
> [snip]
>
> That means you want to use the SSR to act as a bleeder, which is
> activated when the main power is shut off.
>
> This is a good idea, but the use of SSR is a problem. An SSR has a
> TRIAC or an SCR as switching element, which cannot be turned off. The
> device is in the on-state once a current flows until the current
> reaches under a holding threshold. This feature is no problem in AC
> systems, but is generally bad in DC systems.
>
> When you turn off the machine and the SSR starts to bleed, and then
> turn on the machine before the caps are completely empty, then you
> will have the bleeder load on there permanently.

Exactly the reason for the forced time delay in the turn on as I 
described. That, and the low voltage AC in series with the load 
resistors to encourage its turn off when the DC volts gets down to the  
6 volt range if it does not recover from lack of sufficient holding 
current, but that may be several T=RC's depending on the minimum holding 
current of the individual device.  This applies to the spindle psu only, 
and other initializations of the machine, like homing & touch offs, can 
continue while this is taking place.

These toroids have a quite low level hum, difficult to hear over the 
other background noise, as there is a 20" box fan with a 20" HEPA filter 
on the back of it, trying to keep the air clean enough that I can air 
brush finish without having to knock off the dust nibs after its setup, 
so I may duplicate the at-speed led and tie it to the output of the soft 
start resistor bypass control signal, thereby giving be a visual 
indicator that the spindle is ready to rock & roll.  All that "wiring" 
is of course in the .hal file(s).

I'll prowl fleabay though and see if I can source a power resistor in the 
300 ohm range, which should shorten the T=RC time a bit if it takes too 
long with this small batch of 10 watt'ers that are equ to 833.33 Ohms. 
Spares I had bought for something else since long forgotten.  Might as 
well use them for something. :)

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 

--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Another crazy idea?

[andy pugh]

On 6 January 2016 at 14:01, Gene Heskett  wrote:
>> Ideally it needs to be a force-guided relay, with the main power feed
>> (or surrogate thereof) through a second set of terminals

> That thought has crossed my mind too, but the power relays I have, with
> an aux set of contacts that could feed the interlock, are all NO types.
> NC versions seem rare, and pricy. $100 USD range.

Where are you looking?

I am using one of these:
http://uk.rs-online.com/web/p/non-latching-relays/0548022/
But I paid £10 for it from eBay, with a screw-terminal base included.

It is _not_ rated to switch 300V DC, so I just ensure it is never
asked to switch when there is 300V across it. Or, in fact, more than
5V.

-- 
atp
If you can't fix it, you don't own it.
http://www.ifixit.com/Manifesto

--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Another crazy idea?

[Gene Heskett]

On Wednesday 06 January 2016 09:21:48 andy pugh wrote:

> On 6 January 2016 at 14:01, Gene Heskett  wrote:
> >> Ideally it needs to be a force-guided relay, with the main power
> >> feed (or surrogate thereof) through a second set of terminals
> >
> > That thought has crossed my mind too, but the power relays I have,
> > with an aux set of contacts that could feed the interlock, are all
> > NO types. NC versions seem rare, and pricy. $100 USD range.
>
> Where are you looking?
>
> I am using one of these:
> http://uk.rs-online.com/web/p/non-latching-relays/0548022/
> But I paid £10 for it from eBay, with a screw-terminal base included.
>
> It is _not_ rated to switch 300V DC, so I just ensure it is never
> asked to switch when there is 300V across it. Or, in fact, more than
> 5V.

Is it big enough to not weld closed forever?

Anyway, I just picked up a 250 Ohm 250 watt resistor and a 120v primary, 
6.3 volt ct tranny on fleabay.  That should get the recovery time under 
2 minutes I'd think.  And that resistor shouldn't do anything but warm 
the garage 50 watts worth even if the SSR doesn't recover. :)

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 

--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Another crazy idea?

[andy pugh]

On 6 January 2016 at 14:43, Gene Heskett  wrote:
>> It is _not_ rated to switch 300V DC, so I just ensure it is never
>> asked to switch when there is 300V across it. Or, in fact, more than
>> 5V.
>
> Is it big enough to not weld closed forever?

It is rated at 6A, so has no problem with the current. But it is only
rated to switch 24V DC.

-- 
atp
If you can't fix it, you don't own it.
http://www.ifixit.com/Manifesto

--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Another crazy idea?

[Gene Heskett]

On Wednesday 06 January 2016 10:33:58 andy pugh wrote:

> On 6 January 2016 at 14:43, Gene Heskett  wrote:
> >> It is _not_ rated to switch 300V DC, so I just ensure it is never
> >> asked to switch when there is 300V across it. Or, in fact, more
> >> than 5V.
> >
> > Is it big enough to not weld closed forever?
>
> It is rated at 6A, so has no problem with the current. But it is only
> rated to switch 24V DC.

I expect it can withstand more than that as long as there isn't a big 
capacitative sink on the load side to generate a huge inrush when it 
makes contact. I'd be tempted to use any extra contacts as series wired 
so there would be more than one break. I have seen 3 and 4 pole 50 amp 
relays wired so that all 4 sets of contacts were in series in order to 
assure a break for decades of use. I had to replace a 30 yo coil once 
but that was it, it just worked.

Most transmitter bleeders are wired thru a door switch.  Our old GE had 
two layers to that switch, one to crowbar the 7200 volt supply thru a 
few kilowatts of resistors on the cabinet ceiling, and one set of 
contacts to crowbar the 3 phase power, tripping that cabinets main 
breaker, and cascading  back to kill the AK-225 main plate contactor 
thru its undervoltage relay.  Unknown to us, an adjuster bolt had 
rattled out so it couldn't knock the linkage knee out from under the 
closer latch, so when a new operator opened the door to see what a 
transmitter looked like, but closed the door when all hell broke loose.  
But the chain reaction was started because the dump switch maintained 
the arc when it was pushed apart, and the AK-225 remained closed.  
Cooked everything between the door interlock including a 4400 lb plate 
transformer, the 750mcm cables to the substation pole, tried to blow the 
fuse links in the flag switches on the pole, but the ceramic tubes that 
contain the fuse wire were so metalized from previously blown fuses that 
they didn't totally interrupt the 14.4kv lines, caught fire, and kept 
the lights on dim and flickering, and the dripping material from those 
flags caught fire and dripped into the grass, and I'm out there after I 
arrived on the scene, stomping out a smoldering grass fire that was by 
then about 15 feet in diameter.  With no way in hell for a fire truck 
tanker/pumper to get to it. Stomp it out or let the whole mountain burn, 
3 sq miles or more of it.  You do what you have to do...

We'd bought a spare transmitter just like ours, taken out of service at a 
station in Texas before I took the job in '84, so we were only off the 
air about 2 days total by having a fork lift from a scrap metal yard and 
its driver come up and move the transformers.  Big fork lift, sides 
cleared the garage door about 4" on each side, and less than that after 
he was inside and had to skid steer 90 degrees to get turned to where 
the transformer was located.  That driver was good, never touched a 
thing but what he was asked to do.  Otherwise it would have been around 
9-12 days for Peter C. Dahl to make us another, which due to the 
improvements in transformer steel in 40 years , would probably only 
weighed a ton.  And would still have ran cooler than the original.

In the process of the repairs after power was restored, I managed to get 
across the 3 phase line while looking for an interlock switch part I had 
dropped as I was running on pure coffee by then, gave me 2nd degree 
burns on both arms & my chest where I fell across it.  But I managed to 
kick myself off it and I'm still here to pester you kind folks.  The 
shock trauma did give me a case of the shingles that put me in bed for 
about 2 weeks, redefining my pain threshold upward several notches.  But 
I'm a survivor, having been in similar situations before.  But I did get 
it up and running before I went to see how many times the clock went 
around while I was catching up on sleep & food. Took about 3 days for 
the shingles pox to start showing & several months for them to heal.

One of my several war stories.

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 

--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Another crazy idea?

[Bertho Stultiens]

On 01/06/2016 10:13 PM, Gene Heskett wrote:
[snip]
>> This setup is probably also cheaper than using SSRs.
> Perhaps, but your parts count is many times what I have in mind.

But then again, the components are small and cheap. The setup you
described would probably use more space.


[snip]
> So if after a time, it seems your circuit is getting slower, that is 
> probably the failure mode.  A slower rate of failure at 130 volts might 
> be expected, but I have doubts its completely hideable by the passage of 
> time.
[snip]

That is where you are wrong afaik.

The failure mode you describe is attributed to high dU/dt applications
(aka fast switching). In that case, you have many problems with the
output capacitance, which can cause the mosfet to burn the isolation and
the channel(s) (hexfet is just a particular physical construction of a
mosfet).

The elegant part of using a mosfet as a bleeder in current limited mode
is the explicit lack of high dU/dt, it is actually very slow. As long as
you don't let it heat up too much it will hold forever.

The circuit cannot get "slower" because the speed is not dependent on
low Rdson, but on constant (low) current, which regulates Rds in the
linear region. The mosfet I used in the diagram is just (a cheap) one
that I have on the shelf. You can use one with higher Rdson without
problem, as long as it has a high enough Vds max.


There are few things that you need to ensure with mosfets (regardless
physical construction):
- Never ever exceed Vgs max. This is typically +/- 20V (or lower) and
you need to have a protection zener mounted on the mosfet if you have a
noisy environment or inductive leads up to the gate. Modern small
mosfets have integrated TVS protection built into the package. Otherwise
you would just mount an 18V zener over the gate-source leads.
- Never exceed the Vds max. This is often a cause of trouble in noisy
lines, where high dU/dt noise exceeds the limit and causes failure. In
this specific application it is easily mitigated using a small ceramic
cap over drain-source to short the noise.


> Thank you Bertho, you made me think...

That is always good, that thinking thing ;-)


-- 
Greetings Bertho

(disclaimers are disclaimed)

--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Another crazy idea?

[Gene Heskett]

On Wednesday 06 January 2016 17:05:27 Bertho Stultiens wrote:

> On 01/06/2016 10:13 PM, Gene Heskett wrote:
> [snip]
>
> >> This setup is probably also cheaper than using SSRs.
> >
> > Perhaps, but your parts count is many times what I have in mind.
>
> But then again, the components are small and cheap. The setup you
> described would probably use more space.
>
For the resistor, yes, its a 200 watter.
>
> [snip]
>
> > So if after a time, it seems your circuit is getting slower, that is
> > probably the failure mode.  A slower rate of failure at 130 volts
> > might be expected, but I have doubts its completely hideable by the
> > passage of time.
>
> [snip]
>
> That is where you are wrong afaik.
>
> The failure mode you describe is attributed to high dU/dt applications
> (aka fast switching). In that case, you have many problems with the
> output capacitance, which can cause the mosfet to burn the isolation
> and the channel(s) (hexfet is just a particular physical construction
> of a mosfet).
>
> The elegant part of using a mosfet as a bleeder in current limited
> mode is the explicit lack of high dU/dt, it is actually very slow. As
> long as you don't let it heat up too much it will hold forever.
>
> The circuit cannot get "slower" because the speed is not dependent on
> low Rdson, but on constant (low) current, which regulates Rds in the
> linear region. The mosfet I used in the diagram is just (a cheap) one
> that I have on the shelf. You can use one with higher Rdson without
> problem, as long as it has a high enough Vds max.
>
>
> There are few things that you need to ensure with mosfets (regardless
> physical construction):
> - Never ever exceed Vgs max. This is typically +/- 20V (or lower) and
> you need to have a protection zener mounted on the mosfet if you have
> a noisy environment or inductive leads up to the gate. Modern small
> mosfets have integrated TVS protection built into the package.
> Otherwise you would just mount an 18V zener over the gate-source
> leads. - Never exceed the Vds max. This is often a cause of trouble in
> noisy lines, where high dU/dt noise exceeds the limit and causes
> failure. In this specific application it is easily mitigated using a
> small ceramic cap over drain-source to short the noise.
>
> > Thank you Bertho, you made me think...
>
> That is always good, that thinking thing ;-)

Which got me trolling on fleabay & found this:



That gives me a 4pdt, and I can use 2 of them to move that 51 ohm 
resistor from in series with the AC input when its enegized, and to 
across the caps when its de-energized. One set to tally back to the hal 
file to run a status led. Maximum closing current would be about 2.6 
amps RMS for coming on or 2.6 peak and fading in DC, which those 10A  
contacts should handle, and since it would already be closed, it should 
be able to handle the full 10 amps to the motor when the motor needs it.  
The third set of contacts would be placed in series with the SSR 
crowbaring the resistor so it can only be crowbared when active.

So I just ordered 3.  Never can tell where I might need more 4pdt's. :)

One of these first days I need to order me some more 14 gauge wire in 
assorted colors.

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 

--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Another crazy idea?

[Bertho Stultiens]

On 01/06/2016 01:54 PM, Gene Heskett wrote:
>> When you turn off the machine and the SSR starts to bleed, and then
>> turn on the machine before the caps are completely empty, then you
>> will have the bleeder load on there permanently.
> 
> Exactly the reason for the forced time delay in the turn on as I 
> described. That, and the low voltage AC in series with the load 
> resistors to encourage its turn off when the DC volts gets down to the  
> 6 volt range if it does not recover from lack of sufficient holding 
> current, but that may be several T=RC's depending on the minimum holding 
> current of the individual device.  This applies to the spindle psu only, 
> and other initializations of the machine, like homing & touch offs, can 
> continue while this is taking place.


I'd do it differently and use an active bleeder setup.

See atached diagram and simulation result.

The setup uses the voltage from the cap an energy source for bleeding
the cap. The bleeder kicks in at about 40ms after the power is turned
off and is disabled instantly when the AC source is enables.

The cap is discharged using a constant current bleeder. This reduces the
need to have a high-power resistor in the circuit. Most of the power is
dissipated in the mosfet, which should be mounted on some cooling fin to
dissipate the instantaneous heat (but it is not too bad or much).

This setup is probably also cheaper than using SSRs.

-- 
Greetings Bertho

(disclaimers are disclaimed)


bleeder_1.pdf
Description: Adobe PDF document
--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Another crazy idea?

[Gene Heskett]

On Wednesday 06 January 2016 15:10:55 Bertho Stultiens wrote:

> On 01/06/2016 01:54 PM, Gene Heskett wrote:
> >> When you turn off the machine and the SSR starts to bleed, and then
> >> turn on the machine before the caps are completely empty, then you
> >> will have the bleeder load on there permanently.
> >
> > Exactly the reason for the forced time delay in the turn on as I
> > described. That, and the low voltage AC in series with the load
> > resistors to encourage its turn off when the DC volts gets down to
> > the 6 volt range if it does not recover from lack of sufficient
> > holding current, but that may be several T=RC's depending on the
> > minimum holding current of the individual device.  This applies to
> > the spindle psu only, and other initializations of the machine, like
> > homing & touch offs, can continue while this is taking place.
>
> I'd do it differently and use an active bleeder setup.
>
> See atached diagram and simulation result.
>
> The setup uses the voltage from the cap an energy source for bleeding
> the cap. The bleeder kicks in at about 40ms after the power is turned
> off and is disabled instantly when the AC source is enables.
>
> The cap is discharged using a constant current bleeder. This reduces
> the need to have a high-power resistor in the circuit. Most of the
> power is dissipated in the mosfet, which should be mounted on some
> cooling fin to dissipate the instantaneous heat (but it is not too bad
> or much).
>
> This setup is probably also cheaper than using SSRs.

Perhaps, but your parts count is many times what I have in mind.

Some History:

My experience in using hexfet's for analog pass transistors in a 275 volt 
bias circuit, where they had to sink the surplus grid current, up to 
several hundred milli-amps during vertical synch, was that they had a 
fairly short lifetime, even though I was buying 1 kilovolt, 200 watt 
rated hexfets, individually heat sinked and driven in parallel with 
current sharing resistors in the source leads.  They did not get "can't 
touch the heat sink" warm either, perhaps a 20F rise over ambient.

I was eventually forced to put the 1955 design power supply that used a 
pair of vacuum tube 7094's for the pass element back into our old GE 
transmitter.  Because they did not well handle the surge during vertical 
synch time, there was about a 5% sag in peak power during that time. The 
hexfet circuit only sagged perhaps 1% when they were fresh.

But since those internally are actually made off many junctions bonded 
out in parallel, they would punch thru and blow the bond wires removing 
themselves from the circuit, gradually removing gain until it was time 
to replace them.  As that was about an every 45 to 60 day need, I 
eventually ordered some fresh sockets to fit the 7094's and put the 50 
yo supply back in. The tube sockets heater contacts were oxidized and 
had been overheated so many times only fresh sockets would make it 
dependable again. That was in about 2002, and AFAIK they were still in 
and working when we pushed the off button the last time at midnight June 
30, 2008 for the digital switch over here in the US.

I don't know if there is a major diff in internal construction between a 
hexfet and a mosfet, but while hexfets make fantastic digital switches, 
they are not recommended for use in an analog circuit by IR as I called 
them on the phone and discussed my high failure rate with their 
engineers.  If they had a corresponding mosfet that could have done the 
job, I am sure they would have mentioned it, but they did not.

This failure mode is similar in nature to the SOA ratings of normal 
bipolar junction transistors.

So if after a time, it seems your circuit is getting slower, that is 
probably the failure mode.  A slower rate of failure at 130 volts might 
be expected, but I have doubts its completely hideable by the passage of 
time.  If the 100 ohm r was replaced by an inductor and a current sensor 
of an ohm or so, driving a comparator, essentially converting it into a 
switch mode current regulator running at 20 some kilohertz, I would 
expect close to infinite life from a hexfet used as the pass switch. I 
have no clue where the heat energy (law of conservation of energy) would 
go given good switching drive and a low ohms inductor.  Might be 
interesting to build such a critter and see just how fast it could be 
made to work by raising the current regulation point.

Since I have all sorts of hexfet's in BIG packages laying around in 
failed computer psu's that could do that rather nicely, and inductors 
from the same source, its a should be investigated idea. I am sure there 
are circuits for switch mode current regulators out in the wild.

Thank you Bertho, you made me think... :)

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 

Re: [Emc-users] Another crazy idea?

[Bertho Stultiens]

On 01/06/2016 03:05 AM, Gene Heskett wrote:
[snip]
> The idea being that when that drain goes low, it will charge the .22, 
> thru the diode, essentially maintaining that charge state.  But when 
> LCNC is stopped, that drain will go high (to 25 volts) as the supply is 
> turned off, and as it goes high, so will the far end of that .22, 
> carrying the + terminal on the third SSR far enough to trigger it.
[snip]

That means you want to use the SSR to act as a bleeder, which is
activated when the main power is shut off.

This is a good idea, but the use of SSR is a problem. An SSR has a TRIAC
or an SCR as switching element, which cannot be turned off. The device
is in the on-state once a current flows until the current reaches under
a holding threshold. This feature is no problem in AC systems, but is
generally bad in DC systems.

When you turn off the machine and the SSR starts to bleed, and then turn
on the machine before the caps are completely empty, then you will have
the bleeder load on there permanently.


-- 
Greetings Bertho

(disclaimers are disclaimed)

--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] Another crazy idea?

[Gene Heskett]

Greetings everybody;

I have the soft start working, but have found that the capacitors are 
forming up to a very low leakage since they are operating at about half 
to 1 volt above their labeled voltage, which if it was turned on every 
day, the inrush probably would not trip a 15 amp breaker.  Turned off at 
5 ish because I needed to go see about a feedbag for us, and not turned 
on till nominally 1:30 in the afternoon of the next day, the normal 
operating voltage of 127 had only leaked down to about 115 volt.  Since 
there's the equ of 2, 68,000 uF caps in series, that still represents a 
considerable shock hazard if I decide to do something that needs the + 
line touched.

So, here's my wild idea. Take another of those SSR's, ground the - 
control terminal, put a diode across the + to - terminals to absorb 
any - voltage, and charge a .22 uF cap connected to the + terminal and 
the drain of the SSR doing the original turn on thru that diode.

The idea being that when that drain goes low, it will charge the .22, 
thru the diode, essentially maintaining that charge state.  But when 
LCNC is stopped, that drain will go high (to 25 volts) as the supply is 
turned off, and as it goes high, so will the far end of that .22, 
carrying the + terminal on the third SSR far enough to trigger it.

Nice idea,  but these caps can also act as a battery because of their 
dielectric absorbtion, maintaining a small discharge current that when 
the bleeder is removed, potentially over the next hour bring the back of 
caps back up to as high as 20 volts.

Because I don't know what the recovery characteristics of these SSR's 
are, the temptation is to feed the load r's normally grounded terminal, 
to ground thru a small tranny delivering 6.3 volts AC, which would seem 
to guarantee a shutoff at somewhere near the T=RC time where the R is 6 
5k 10 watt resistors in parallel, and the C is the .068 Farad of the 
filter caps.

With a startup delay 1.5 times that T, then I could be well assured that 
should I forget to do something, and turn it back on immediately, that 
time delay would prevent me from re-applying power before the third SSR 
has reset to the off state, which would leave the now hot bleeders still 
connected & using around 25 watts. Just from the CCS heat alone, thats 
not a great idea.

What do you think? Is that a safe and workable idea?  Or do I need to 
find a lower R than the 833.etc ohms that 6 of those 5K's in 
parallel represents?

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 

--
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users