[Emc-users] OT: good tiny edm infos

[TJoseph Powderly]

Nicklas Sarah Bari et al,
I stumbled on a few _very_ good sites for tiny edms

(many) using real on/off time fet edm power supplies ( spark generators).
some french some german, but easy to understand the schematics and 
builds and movies.

most homebuilts use r/c capacitive discharge, which triggers wg-hen it 
wnats to
these (some) use real microsecond on and offtime oscillators

most use window comparator servo drives ( classic servo system )

http://www.usinages.com/threads/systeme-delectro-erosion-fait-maison.78941/print?page=1

http://www.hcp-hofbauer.de/indexerodier.htm

http://www.dl1dow.de/artikel/mikro-edm/index.htm

thought i'd drop this note so others can find it in the swamp called 
'interweb'.
tomp
tjtr33

--
Announcing the Oxford Dictionaries API! The API offers world-renowned
dictionary content that is easy and intuitive to access. Sign up for an
account today to start using our lexical data to power your apps and
projects. Get started today and enter our developer competition.
http://sdm.link/oxford
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] axis position cmd lost

[Jeff Epler]

this is a bug in the UI library of AXIS, Tkinter.  it is believed to be fixed
in master branch, and not affect the debian and ubuntu versions that we build
2.7 debs for.

Specifically, a certain radiobutton uses the string "y" to identify the "y"
axis, but because Tkinter is based on a string-typed language (Tcl), the value
"y" can on occasion be turned into the value True or 1 when transferred to 
Python
(which is a strongly but dynamically typed language), which causes AXIS not to
know which axis the +/- buttons are intended to move.

iirc machinekit has a hacky fix for this (explicit check for 1/true in a
specific spot), perhaps you'd like to find it and apply it to your local tree.
Since the fix is a hack and as far as we know it happens to be fixed by other
code changes in master branch for ja, we haven't incorporated it in
linuxcnc.org's software.

Jeff

--
Announcing the Oxford Dictionaries API! The API offers world-renowned
dictionary content that is easy and intuitive to access. Sign up for an
account today to start using our lexical data to power your apps and
projects. Get started today and enter our developer competition.
http://sdm.link/oxford
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] damn radio shack

[Gene Heskett]

On Monday 06 March 2017 18:23:35 Ben Potter wrote:

> > From: Gene Heskett [mailto:ghesk...@shentel.net]
> >
> > I bought what I thought were 3 momentary contact spst pushbutton at
> > the Bridgeport shack Saturday. I opened the plastic bag to measure
> > them and machine the holes in the apron to hold them Saturday night.
> >
> > Putting the thing back together slowly today, I wondered if my dvm
> > could show me which end was which on the led in the button but at
> > 1.7 volts, no, so I moved the probes to check the switch, pushed it
> > once, goes to zero ohms and stays there until I push it again!  No
> > place on the bag does it
>
> say
>
> > alternate action. And I don't care how gently you push it, no hint
> > of an alternate action can be felt.
> >
> > And fleabay is no help, all are wrong mounting, everthing is nuts on
> > the backside of the panel, without a long enough of a thread to work
> > in a
>
> panel
>
> > that despite the front being countersunk 5mm's, is still 8+mm's
> > thick.
>
> Ditto
>
> > digikey, Newark-Element14, and Mouser. And no one makes it easy to
> > spec the desired diameter and mount. Somebody, and it sure isn't
> > going to be
>
> me,
>
> > could spend weeks finding a front of the panel, pushes into a 12mm
> > hole to mount momentary contact switch.
>
> Apem do an IL series that is fairly long - I think the max is about
> 10mm panel thickness.
>
> http://www.apem.com/files/apem/brochures/switch-pushbutton-IL-ENG.pdf
>
> Digikey seem to have at least some of the range in stock.

Some but no mount options so you have to look thru the pdf's to see what 
size they are.  After about 30 of them this afternoon I was getting 
crosseyed.

> If you can go up to 15 or 22mm diameter there are a whole bunch more
> options.
>
15mm could be arranged, thats what the counterbore is now.  But any of 
those is another trip to the mills table with the clean up apron 
to "make them fit". So if I do anything, it may well be making my own 
from microswitches. I can most certainly do that quicker than I can get 
something from China by China Post. And have something that will outlast 
me.

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 

--
Announcing the Oxford Dictionaries API! The API offers world-renowned
dictionary content that is easy and intuitive to access. Sign up for an
account today to start using our lexical data to power your apps and
projects. Get started today and enter our developer competition.
http://sdm.link/oxford
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] damn radio shack

[Gene Heskett]

On Monday 06 March 2017 18:14:12 andy pugh wrote:

> On 6 March 2017 at 22:39, Gene Heskett  wrote:
> > Ditto digikey, Newark-Element14, and Mouser. And no one makes it
> > easy to spec the desired diameter and mount.
>
> RS (Radio Spares, not Radio Shack) let you specify the panel cutout
> size:
> http://uk.rs-online.com/web/c/switches/push-button-switches-accessorie
>s/push-button-switches/#esid=4294351089&applied-dimensions=4293214630,4
>293402607,4293404312,4294350769
>
Those prices, with one or 2 exceptions, are scary, particular once the 
price is converted to dollars.

> Then look up the part number at Newark. For example:
> http://www.newark.com/omron-industrial-automation/a3ct-7011/pushbutton
>-switch/dp/89C3647?ost=A3CT7011&selectedCategoryId=&categoryNameResp=Al
>l&searchView=table&iscrfnonsku=false But that is very expensive, and
> seems to need an LED and button cap to be added.

Which pretty well rules that source out.  Surely you aren't paying those 
prices on your side of the pond?

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 

--
Announcing the Oxford Dictionaries API! The API offers world-renowned
dictionary content that is easy and intuitive to access. Sign up for an
account today to start using our lexical data to power your apps and
projects. Get started today and enter our developer competition.
http://sdm.link/oxford
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] damn radio shack

[Ben Potter]

> From: Gene Heskett [mailto:ghesk...@shentel.net]
> 
> I bought what I thought were 3 momentary contact spst pushbutton at the
> Bridgeport shack Saturday. I opened the plastic bag to measure them and
> machine the holes in the apron to hold them Saturday night.
> 
> Putting the thing back together slowly today, I wondered if my dvm could
> show me which end was which on the led in the button but at 1.7 volts, no,
> so I moved the probes to check the switch, pushed it once, goes to zero
> ohms and stays there until I push it again!  No place on the bag does it
say
> alternate action. And I don't care how gently you push it, no hint of an
> alternate action can be felt.
> 
> And fleabay is no help, all are wrong mounting, everthing is nuts on the
> backside of the panel, without a long enough of a thread to work in a
panel
> that despite the front being countersunk 5mm's, is still 8+mm's thick.
Ditto
> digikey, Newark-Element14, and Mouser. And no one makes it easy to spec
> the desired diameter and mount. Somebody, and it sure isn't going to be
me,
> could spend weeks finding a front of the panel, pushes into a 12mm hole to
> mount momentary contact switch.

Apem do an IL series that is fairly long - I think the max is about 10mm
panel thickness.

http://www.apem.com/files/apem/brochures/switch-pushbutton-IL-ENG.pdf

Digikey seem to have at least some of the range in stock.

If you can go up to 15 or 22mm diameter there are a whole bunch more
options.

> I am not about to sit and read 40,000 pdf's
> just to find out how big it is and how it mounts. I wouldn't mind a spring
> loaded to off toggle, but everything I tried recently needed an extra 2"
on
> the handle to toggle it w/o cutting fingers on the bat handle. I've dealt
with
> pushbutton toggles before and they have uniformly been a high rate of
> replacement items, the toggle gives up in <1000 pushes.
> 
> Do I make brackets for minimicroswitches and machine my own buttons to
> push the rollers?
> 
> The whole world is on backorder, and until its not, we are not going to
make
> America Great again.
> 
> 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 
> 
>

--
> Announcing the Oxford Dictionaries API! The API offers world-renowned
> dictionary content that is easy and intuitive to access. Sign up for an
account
> today to start using our lexical data to power your apps and projects. Get
> started today and enter our developer competition.
> http://sdm.link/oxford
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users


--
Announcing the Oxford Dictionaries API! The API offers world-renowned
dictionary content that is easy and intuitive to access. Sign up for an
account today to start using our lexical data to power your apps and
projects. Get started today and enter our developer competition.
http://sdm.link/oxford
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] damn radio shack

[andy pugh]

On 6 March 2017 at 22:39, Gene Heskett  wrote:
> Ditto digikey, Newark-Element14, and Mouser. And no one makes it
> easy to spec the desired diameter and mount.

RS (Radio Spares, not Radio Shack) let you specify the panel cutout size:
http://uk.rs-online.com/web/c/switches/push-button-switches-accessories/push-button-switches/#esid=4294351089&applied-dimensions=4293214630,4293402607,4293404312,4294350769

Then look up the part number at Newark. For example:
http://www.newark.com/omron-industrial-automation/a3ct-7011/pushbutton-switch/dp/89C3647?ost=A3CT7011&selectedCategoryId=&categoryNameResp=All&searchView=table&iscrfnonsku=false
But that is very expensive, and seems to need an LED and button cap to
be added.


-- 
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

--
Announcing the Oxford Dictionaries API! The API offers world-renowned
dictionary content that is easy and intuitive to access. Sign up for an
account today to start using our lexical data to power your apps and
projects. Get started today and enter our developer competition.
http://sdm.link/oxford
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] damn radio shack

[Gene Heskett]

I bought what I thought were 3 momentary contact spst pushbutton at the 
Bridgeport shack Saturday. I opened the plastic bag to measure them and 
machine the holes in the apron to hold them Saturday night. 

Putting the thing back together slowly today, I wondered if my dvm could 
show me which end was which on the led in the button but at 1.7 volts, 
no, so I moved the probes to check the switch, pushed it once, goes to 
zero ohms and stays there until I push it again!  No place on the bag 
does it say alternate action. And I don't care how gently you push it, 
no hint of an alternate action can be felt.

And fleabay is no help, all are wrong mounting, everthing is nuts on the 
backside of the panel, without a long enough of a thread to work in a 
panel that despite the front being countersunk 5mm's, is still 8+mm's 
thick. Ditto digikey, Newark-Element14, and Mouser. And no one makes it 
easy to spec the desired diameter and mount. Somebody, and it sure isn't 
going to be me, could spend weeks finding a front of the panel, pushes 
into a 12mm hole to mount momentary contact switch. I am not about to 
sit and read 40,000 pdf's just to find out how big it is and how it 
mounts. I wouldn't mind a spring loaded to off toggle, but everything I 
tried recently needed an extra 2" on the handle to toggle it w/o cutting 
fingers on the bat handle. I've dealt with pushbutton toggles before and 
they have uniformly been a high rate of replacement items, the toggle 
gives up in <1000 pushes.

Do I make brackets for minimicroswitches and machine my own buttons to 
push the rollers?

The whole world is on backorder, and until its not, we are not going to 
make America Great again.

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 

--
Announcing the Oxford Dictionaries API! The API offers world-renowned
dictionary content that is easy and intuitive to access. Sign up for an
account today to start using our lexical data to power your apps and
projects. Get started today and enter our developer competition.
http://sdm.link/oxford
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Bridgeport servo troubleshooting

[Ben Potter]

Thanks to everyone who has given advice on this thread. I've learned a heck
of a lot about older drives.

I'm not going to attempt to repair, I'm just going to pull all three drives
and replace with something a little more modern - which will probably double
the space I have in that cabinet.

> From: Jon Elson [mailto:el...@pico-systems.com]

> On 03/05/2017 07:38 AM, Ben Potter wrote:
> > Interesting - I have to admit I'm still a little bit confused as to
> > how these drives work, running 360V AC to the chokes and somehow
> > turning that into 140V DC seems... odd.
> Well, does it run off 3 Phase power?  If not, then it gets hard to drive
motors
> with any constant torque or speed.

Single phase. Well, 180-0-180 on a autotransformer.

> Phase angle control from a 3-phase source takes 6 SCRs, and due to the
> overlapping waveforms of the 3-phase mains, can give fairly steady power
to
> the motor.  Still, the servo bandwidth will be pretty low.
> 
> With single-phase power, the bandwidth will be really low to avoid the 120
Hz
> pulsing from driving everything crazy.

Your 120 Hz remark has just made it click in my head. The drive works by
controlling the phase offset at which the thyristors trigger - which will
control the voltage present when they fire. Typically I'm guessing they
should fire 180 degrees out of phase to one another.
That was probably obvious to everyone else as soon as I said thyristor - but
it took a bit of getting my head around. I'm more used to thinking about
MOSFET drives with switching frequencies in the kHz range.

> 
> it seems a bit odd that a signal capacitor caused the failure, but maybe
that
> fouled up the current control logic to get the SCRs turned on in the wrong
> sequence.

The cap was part of the opamp feedback driving the control voltage for one
phase controller. Assuming it hasn't blown the phase controller completely,
that'd be enough to confuse the drive.

> 
> Jon
> 


--
Announcing the Oxford Dictionaries API! The API offers world-renowned
dictionary content that is easy and intuitive to access. Sign up for an
account today to start using our lexical data to power your apps and
projects. Get started today and enter our developer competition.
http://sdm.link/oxford
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Bridgeport servo troubleshooting

[Ben Potter]

> From: Greg Bentzinger [mailto:skullwo...@yahoo.com]
> My mid 80's Hurco manuals have a service chart which wants the motor and
> tach brushes checked and brush dust blown out with clean dry air every

> hours use.
> I have actually done this (more than once) because the servos were
> accessible.

I've pulled the suspect motor from the machine, I'll vac and retest (with an
insulation tester) before I go any further.

> 
> I suspect the 35Mega Ohm is a short at one brush holder caused by the
brush
> dust.

Makes sense, thank you.
> 
> I admit that the monster FANUC yellow cap motors on my Hitachi Seiki
> Turning center have never been touched since the machine was delivered in
> 86. That machine was designed to run production, and God help you if you
> ever need to do service or repairs.
> 
> Greg
> 
> Out yonder in Yoder, CO
> 
>

--
> Check out the vibrant tech community on one of the world's most engaging
> tech sites, SlashDot.org! http://sdm.link/slashdot
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users


--
Announcing the Oxford Dictionaries API! The API offers world-renowned
dictionary content that is easy and intuitive to access. Sign up for an
account today to start using our lexical data to power your apps and
projects. Get started today and enter our developer competition.
http://sdm.link/oxford
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] axis position cmd lost

[Nicklas Karlsson]

> On 5 March 2017 at 13:40, Nicklas Karlsson  
> wrote:
> >> What happens if you give an MDI move in Y?
> > Then examined again it actually works in MDI but not then jogging.
> 
> Keyboard jogging, wheel-jogging or button/switch/joystick?

Keyboard jogging is OK, MDI is also OK but mouse buttons on "+"/"-" button does 
not work. I have no idea what might be wrong.

--
Announcing the Oxford Dictionaries API! The API offers world-renowned
dictionary content that is easy and intuitive to access. Sign up for an
account today to start using our lexical data to power your apps and
projects. Get started today and enter our developer competition.
http://sdm.link/oxford
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] I am gonna need more wire.

[Gene Heskett]

On Monday 06 March 2017 09:04:48 John Thornton wrote:

> Some paint screws...
>
> JT

And then they chip off and look ugly if you have to put a socket wrench 
on them to remove them so I leave nice shiny plated bolts unpainted so 
they are part of the "decor".

Those 6 are holding a pair of 1/4"x3/4" about 6" long steel strips on the 
front face of the apron, with 1/8" wide and deep rabbit cuts on the 
inside faces to hold a sheet of 1/8" alu for the x motor mounts belt 
tightening adjustment of the sliding motor mount plate they trap. I 
might paint the steel before I reassemble things if I can find a box of 
thin washers for the 4mm hex head bolts that go in those 6 holes. I have 
some but haven't stumbled over them recently in my midden heap of a 
garage.

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 

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, SlashDot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Bridgeport servo troubleshooting

[Gene Heskett]

On Monday 06 March 2017 04:06:49 Erik Christiansen wrote:

> On 04.03.17 12:29, Jon Elson wrote:
> > On 03/04/2017 12:02 PM, Gene Heskett wrote:
> > > Hexfets can be hard to deal with. Bear in mind they are generally
> > > shipped with a fine wire wrapped around all 3 legs
> >
> > I have never received a transistor with wire around it.  I
> > have purchased roughly 3000 FET power transistors for Pico
> > Systems production.
>
> Nor down here, closer to the penguins. But 100 off 2N7000 small
> MOSFETs came last month on metallised cardboard strip, the IRLZ44s
> came in anti-static plastic tubing, and years ago a solid handful of
> MOSFETs came in a black conductive plastic bag with big yellow ESD
> handling warning.
>
> > The capacitance of a power transistor's gate is so huge it greatly
> > mitigates the ESD susceptibility.
>
> Hadn't thought of that ... sounds good vs the 30 pF of human body
> capacitance ... but even at 3 nF, that's 10v/kv on 30 pF, so even 20
> kv from crossing carpet in dry air becomes 200v = dead MOSFET. I think
> I'll stick with suppliers who adhere to established antistatic norms,
> thanks.
>
> Back at Siemens, we weren't allowed into the PCB assembly area without
> conductive shoe-shorting straps, and even then were confined between
> painted yellow lines bracketing the aisles, out of reach of trouble.
> (And that wasn't just for engineers - other visitors received the same
> treatment. ;-)
>
> > But, I'm pretty sure the servo amps in these older machines
> > were bipolar junction transistors.  However, after a fault
> > that blew a fuse, it is quite likely the base drive
> > circuitry may also be damaged.  When I get FETs blown out, I
> > don't even TEST the gate driver, I just replace it every
> > time.  I think at least the base drive circuits should be
> > checked.
>
> It's well worth that to avoid having to pull it apart again - after
> having ruined a valuable workpiece. Apropos that, ESD damage can be
> latent - the device continues to function for a while, then dies
> later, just like the stressed gate driver, above. I wouldn't install
> MOSFETs which had not received proper ESD-safe handling from
> manufacturer to board.
>
> Erik

Same here.  Having found a more robust type in an old computer psu, the 
first thing I did was tangle its legs up in some stripped 30 gauge 
wrapping wire before applying the hot air to remove it. And it wasn't 
removed until it had been planted solidly in the target board. That was 
7 or 8 years ago. I was concerned that the increased gate capacitance, 
about double the OEM bugs, would be hard on the driver, but I surveyed 
the board while it was running that spindle motor in an HF micro-mill at 
full song and found nothing more than 5F over ambient. I also raised the 
fuse from 2 amps to 3 amps which got rid of the blown fuse syndrome as 
it was formerly blowing fuses without ever slowing the motor. My blown 
fuse budget is so low now that I am still running on the first box of 5 
I bought.

That cheap single pwm control is so stiff I had to put an ammeter in the 
power circuit to tell me how hard it was working, you could not tell 
from the sound or running speed.  Someday I should put an encoder on it, 
but little room, and the gears are all white nylon or delrin, nothing to 
sense IOW. I have a belt drive about half made for it, using the 2x more 
powerfull motor I took out of TLM, but ran out of round tuits. Something 
with guts enough to turn a 10-32 tap, so low gear would be slower than 
the 1250 it runs now.  More like 250 maybe?

> (Who experienced the abrupt suicidal tendencies of NMOS devices on
> mild ESD exposure in the early '70s. Thank the pixies that MOS is
> somewhat more robust now.)
>
Somewhat, but not 100% bulletproof yet. And definitely not in an analog 
circuit. BTDT in the old ntsc transmitter. Dismal failure over time 
frames just long enough to bore one.

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 

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, SlashDot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] I am gonna need more wire.

[John Thornton]

Some paint screws...

JT


On 3/5/2017 6:19 PM, Gene Heskett wrote:
> On Sunday 05 March 2017 18:06:59 andy pugh wrote:
>
>> On 5 March 2017 at 22:21, Gene Heskett  wrote:
>>> . I'd picked up two rattle guns
>>> of higher temp stuff,
>> Can I repeat my opinion that brush-painting is more effective for
>> machine parts?
> You may Andy, but how to you plug up the screw holes so you can put the
> screws back in.  Brushes have this nasty habit of plugging the threaded
> hole full of paint.
>
> Cheers, Gene Heskett


--
Check out the vibrant tech community on one of the world's most
engaging tech sites, SlashDot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Bridgeport servo troubleshooting

[Erik Christiansen]

On 04.03.17 12:29, Jon Elson wrote:
> On 03/04/2017 12:02 PM, Gene Heskett wrote:
> > Hexfets can be hard to deal with. Bear in mind they are generally shipped
> > with a fine wire wrapped around all 3 legs
>
> I have never received a transistor with wire around it.  I 
> have purchased roughly 3000 FET power transistors for Pico 
> Systems production. 

Nor down here, closer to the penguins. But 100 off 2N7000 small MOSFETs
came last month on metallised cardboard strip, the IRLZ44s came in
anti-static plastic tubing, and years ago a solid handful of MOSFETs
came in a black conductive plastic bag with big yellow ESD handling
warning.

> The capacitance of a power transistor's gate is so huge it greatly
> mitigates the ESD susceptibility.

Hadn't thought of that ... sounds good vs the 30 pF of human body
capacitance ... but even at 3 nF, that's 10v/kv on 30 pF, so even 20 kv
from crossing carpet in dry air becomes 200v = dead MOSFET. I think I'll
stick with suppliers who adhere to established antistatic norms, thanks.

Back at Siemens, we weren't allowed into the PCB assembly area without
conductive shoe-shorting straps, and even then were confined between
painted yellow lines bracketing the aisles, out of reach of trouble.
(And that wasn't just for engineers - other visitors received the same
treatment. ;-)

> But, I'm pretty sure the servo amps in these older machines 
> were bipolar junction transistors.  However, after a fault 
> that blew a fuse, it is quite likely the base drive 
> circuitry may also be damaged.  When I get FETs blown out, I 
> don't even TEST the gate driver, I just replace it every 
> time.  I think at least the base drive circuits should be 
> checked.

It's well worth that to avoid having to pull it apart again - after
having ruined a valuable workpiece. Apropos that, ESD damage can be
latent - the device continues to function for a while, then dies later,
just like the stressed gate driver, above. I wouldn't install MOSFETs
which had not received proper ESD-safe handling from manufacturer to
board.

Erik
(Who experienced the abrupt suicidal tendencies of NMOS devices on mild
ESD exposure in the early '70s. Thank the pixies that MOS is somewhat
more robust now.)

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, SlashDot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users