FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups 
 ProfileProfile   PreferencesPreferences   Log in to check your private messagesLog in to check your private messages   Log inLog in 
Forum index » Electronix » design
Driver for IGBT? (-ve off and +ve on from CMOS)
Post new topic   Reply to topic Page 1 of 3 [41 Posts] View previous topic :: View next topic
Goto page:  1, 2, 3 Next
Author Message
Terry Given
electronics forum Guru


Joined: 30 Apr 2005
Posts: 868

PostPosted: Tue Oct 11, 2005 3:33 am    Post subject: Re: Driver for IGBT? (-ve off and +ve on from CMOS) Reply with quote

Ignoramus14838 wrote:
Quote:
On Tue, 11 Oct 2005 10:18:27 +1300, Terry Given <my_name@ieee.org> wrote:

Ignoramus14838 wrote:

On Tue, 11 Oct 2005 09:03:50 +1300, Terry Given <my_name@ieee.org> wrote:


Ignoramus14838 wrote:


On Mon, 10 Oct 2005 11:50:18 -0700, Jamie <jamie_5_not_valid_after_5_Please@charter.net> wrote:



Richard wrote:




I would like to drive an IGBT from a microcontroller (CMOS).
Is there a driver circuit that can outputs around +15v (on) and a -5v (off)
for switching frequencies around 40Khz.
I think -ve voltage supply is avaliable
What is this type of circuit called?

Thanks
Richard





i really don't understand why you need that?, normally pulling the
gate low will do how ever if you insist.


I am actually playing with IGBTs.

I have Toshiba-MG200Q2YS40 IGBT.

It does NOT turn off itself if I simply remove Gate/Emitter voltage.

I have to apply reverse voltage to turn it off. I actually tried it.

a decent short from gate to emitter would do it.



It would, possibly, switch itself off with a small resistor between
gate and emitter, but it is just a hypothesis of mine.


i


what you are loking at is a combination of things - miller effect
mostly. There is a non-linear capacitance between gate and collector,
call it Ccg. Whenever the collector voltage changes value (say from
Vcesat to Vdc when turning off) the voltage slews at some rate, dV/dt.
This causes current to flow thru Ccg, Icg = Ccg*dVce/dt. This current
flows into the gate circuit, charging up Cge. If the gate driver is not
very low impedance, Icg will happily charge Cge up to Vth, and beyond.
when turning the IGBT on, Vce falls so Icg flows out of Cge, discharging
it - IOW trying to turn it back off.

This can greatly increase switching time (and hence losses) and, if the
gatedrive is bad enough, can make the device "latch" in the on-state,
with Vge sitting around Vth - but you need a seriously piss-weak gate
driver to do that. It can even over-voltage the gate, poke a hole thru
the thin gate oxide layer, and KABOOM, one dead IGBT.

a negative gate bias means the switch-off miller current has to supply
more charge to reach Vth - instead of supplying Qg = Vth*Cge, it must
supply (Vth + |Vnegative|)*Cge. This allows the use of higher gate
impedance circuitry. Its not just the resistance, its the inductance
too. bad gate drives have high resistance and high inductance; good gate
drives have low resistance and low inductance.

to give you an idea, I've worked on gatedrives for 0.5kW - 2.2kW drives
that have about -3V reverse-bias, and -15V for 400kW drives (6 300A fuji
IGBTs directly paralleled).

modern IGBT designs have much better packaging - far lower inductance,
both Lce and Lge. with some of these IGBTs it is theoretically possible
to keep the gate impedance sufficiently low that negative bias isnt
necessary. Like I said, IR wrote a pretty good paper on the topic - why
you dont, in theory, need -ve bias for little IGBTs.

In practice, if you screw up the gatedriver, your circuit is going to go
BANG. If you really know what you are doing, and are very good at taking
measurements, its fairly easy to dewsign gatedrivers and make them work.
OTOH if you are not too experienced, its very easy to destroy many, many
IGBTs.

Generally the gatedrive is a lot cheaper than the IGBTs, so its sensible
to throw -ve bias at it. For little IGBTs, it can be as simple as a
zener in series with the gate resistor, with a 100nF cap across it.


Terry, it is great that you participate in this thread and actually
have experience with power switching. I am building a DC -> AC
inverter for a TIG welder and seems like I can learn a lot from you.

This is a 200A constant current, 28 V welding current, 80 v open
circuit voltage tig welder.I have aforementioned IGBTs.

Can you give a little schematic of how go give the gates negative bias
during turn off. I definitely want to turn IGBTs off in the best
possible way. I want the simplest solution.

here's one simple solution: use an optocoupler to drive a FET driver
chip, running from a +10V/-5V supply. connect the supply 0V to the IGBT
emitter, +10V to the driver chip Vcc and -5V to the driver chip gnd.
that way the driver chip "sees" a 15V supply, whereas the IGBT "sees"
+10V and -5V.


That's good to know. I am looking at using an IR21094 chip. Is it true
that it supports that configuration (it seems to be the case to me,
but I want to confirm).

the chip only sees the total supply voltage. as long as that is within
the device ratings, its fine.

your job is to ensure the "0V" point - where you connect the IGBT
emitter - is sufficiently stiff. in practice that means +10V and -5V
supply capacitors that are a lot bigger than Cge, and a good layout.

Quote:


we used to use tens of thousands of UC3842 smps chips as gate drivers.
An opto drove the 3842, which ran from a 15V isolated supply. the 3842
Vref output was connected to the IGBT emitter, giving +10V/-5V gate
drive. We had hefty caps from +10V to 0V to -5V. for little drives. for
big drives, we used +/-15V supplies.


I will double check, but I think that I have a 12V with +-5V power
supply lying around. If not, I can make one from something, even a
dual power supply for starters.

Now, I want to make sure that I understand what you mean.

Question: So, the above should work for a "big drive" (which, I
assume, applies to my 200 A system). What you write below is simply
another option that applies to "smaller drives".

200A is a fairly big IGBT, so the zener circuit may not be the best
option. but that depends on switching frequency too. My take? its an
expensive part, I'd use a pair of supplies, +12V and -5V sound good.

your IGBT could have up to 100nF of gate capacitance (I have some 600A
powerex/bitsashitty parts with about 120nF Cge), so you will want at
least 1uF across the zener, and tens of uF of low-ESR caps across the
supply rail(s).

ware the ESR of "hobbyist" electrolytics - I have seen a 10uF 16V smt
part with 27 Ohms ESR, at room temperature. Personally, I never use an
electrolytic unless its got 105C stamped on it....

Quote:

Is that correct?


for really little drives (< 2kW) we used the series zener trick. get a
15V FET driver, with a 15V supply. bung a 5V zener in series with the
output, with a 100nF cap across it. When the gatedrive output is high,
5V is dropped across the zener, charging the cap - the cathode end of
the cap is 5V below the anode end, and the gate gets 10V.


OK.


when the gatedrive output switches to 0V, the anode end of the cap is at
0V. The cathode end of the cap is 5V more negative than the anode end,
so the gate sees -5V. The 100nF cap needs to be at least 10x the gate
capacitance. For larger IGBTs, 1uF would be better.


I think that I understand.


this is probably the easiest option for you, and works with *any* gate
driver. Make sure you use a stompy zener though - you dont want its
series impedance to be large compared to your gate resistor. Think BZT03
or suchlike.

gatedrives are tricky, and until they work you kill a lot of IGBTs. why
not just buy one from Semikron.


Which one would you suggest?

buy the semikron driver. its not dirt cheap, but it will work, and is
fully isolated. those IGBTs wont be cheap.

and *KEEP THE INDUCTANCE LOW*

Actually the "three R's" of drive design are:

Reduce Inductance
Reduce Inductance
Reduce Inductance



Quote:

Another question, would a resistor between gate and emitter solve ths
issue? Or is that not the case?

yes but no. You would need a seriously low valued resistor (< 10R)
to even attempt to hold the IGBT off wrt Cmiller, which would quite
successfully bugger up the gatedrive. not to mention the 15V^2/10R =
22W or more power dissipation when the gatedrive is ON.


Not very feasible... I was hoping to be able to get away with a much
smaller resistor...

Thanks a lot...

i


thanks a lot for your time!

i


Cheers
Terry
Back to top
Ignoramus24489
electronics forum beginner


Joined: 12 Oct 2005
Posts: 14

PostPosted: Wed Oct 12, 2005 2:50 pm    Post subject: Re: Driver for IGBT? (-ve off and +ve on from CMOS) Reply with quote

On 12 Oct 2005 14:39:27 GMT, Robert Latest <boblatest@yahoo.com> wrote:
Quote:
["Followup-To:" header set to sci.electronics.design.]
On Wed, 12 Oct 2005 03:19:44 GMT,
Glen Walpert <gwalpert@notaxs.com> wrote
in Msg. <hmsok1t5i1im9rdqq3p35ck1oi0r4ur484@4ax.com

I haven't used the Hobart Cybertig, but I doubt very much if it puts
out *exactly* what is specified regardless of arc resistance. Do you
have the owners manual for it? If not I highly recommend you get one.

http://www.hobartwelders.com/owners_manuals/

I don't have a TIG welder, but if I (hopefully) ever have a good enough
excuse to buy one, it'll be a Hobart.

I simply love companies that publish manuals on the Net. It's also a
great service not only to the unfortunate owner who lost his manual,
but also to the would-be customer who can see exactly what he's gonna
get.

Hobart and Miller manuals are available from
millerwelds.com. Actually, it seems that it is Miller that distributes
Hobart manuals. Hobart and its rights were bought in some complicated
business transactions.

I uploaded manuals for my cybertig to

http://igor.chudov.com/projects/Welding/00-Hobart-CyberTig-Welder/

Quote:
That said -- your Ignoramus's welder looks like a massive EMP device. No
electronics is going to survive this thing.

What if the bridge shorts instead of opens during th switching event?
Would that help?

i
Back to top
Spehro Pefhany
electronics forum Guru


Joined: 01 May 2005
Posts: 2326

PostPosted: Wed Oct 12, 2005 3:25 pm    Post subject: Re: Driver for IGBT? (-ve off and +ve on from CMOS) Reply with quote

On 12 Oct 2005 14:27:17 GMT, the renowned Robert Latest
<boblatest@yahoo.com> wrote:

Quote:
["Followup-To:" header set to sci.electronics.design.]
On Wed, 12 Oct 2005 12:53:49 GMT,
Glen Walpert <gwalpert@notaxs.com> wrote
in Msg. <2j0qk15270s6an346h9vh7sjc92st2ad6q@4ax.com

Just because your unit is rated at 200 A does not mean it will not put
out well over that into a short, and just because it is rated at 70 V
open circuit does not mean that the "inductive kick" of that big
"reactor" will not push the voltage well over 1200 V when your IGBT
attempts to turn off (probably with a big bang, one time only) - you
had better provide someplace for that "reactor" energy to go when the
IGBTs turn off, if you want them to turn off more than once!

That's why he must make sure his bridge shorts rather than opens during
switching.

robert

And you have to make sure that short doesn't ever (even once, during
testing) persist for too long...


Best regards,
Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
speff@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com
Back to top
Glen Walpert
electronics forum Guru Wannabe


Joined: 04 May 2005
Posts: 134

PostPosted: Wed Oct 12, 2005 4:53 pm    Post subject: Re: Driver for IGBT? (-ve off and +ve on from CMOS) Reply with quote

On Wed, 12 Oct 2005 11:25:16 -0400, Spehro Pefhany
<speffSNIP@interlogDOTyou.knowwhat> wrote:

Quote:
On 12 Oct 2005 14:27:17 GMT, the renowned Robert Latest
boblatest@yahoo.com> wrote:

["Followup-To:" header set to sci.electronics.design.]
On Wed, 12 Oct 2005 12:53:49 GMT,
Glen Walpert <gwalpert@notaxs.com> wrote
in Msg. <2j0qk15270s6an346h9vh7sjc92st2ad6q@4ax.com

Just because your unit is rated at 200 A does not mean it will not put
out well over that into a short, and just because it is rated at 70 V
open circuit does not mean that the "inductive kick" of that big
"reactor" will not push the voltage well over 1200 V when your IGBT
attempts to turn off (probably with a big bang, one time only) - you
had better provide someplace for that "reactor" energy to go when the
IGBTs turn off, if you want them to turn off more than once!

That's why he must make sure his bridge shorts rather than opens during
switching.

robert

And you have to make sure that short doesn't ever (even once, during
testing) persist for too long...

Indeed. And he won't find detailed instructions on doing that in any
data sheet or app note I know of, since they all assume or explicitly
show some reasonable approximation of a constant voltage supply and
try to avoid "shoot-through". i is exploring uncharted territory here
and I think he should at the very least get an oscilloscope so he can
measure the on overlap time, start at the lowest possible current
setting and wear safety glasses and a face shield at all times his
equipment is energized. Buying a few dozen spare IGBTs and gate
drivers would be a good idea too.

OTOH, I think i might want to reconsider his entire approach,
especially if he wants to wind up with a really useful result.
Polarity reversals on his CC supply will let him manual TIG weld
aluminum in the event he can pull it off, but to duplicate the very
useful "foreground/background" pulse capabilities of a good inverter
welder he needs to go in and convert it to constant voltage by
providing new SCR controls to let the existing 3-phase SCR bridge
charge a new capacitor bank to an adjustable voltage (which would
normally be set between 40 and 80 volts or so), which would feed his
new IGBT bridge, which would drive the arc through the existing
"filter reactor", which I am guessing is on the order of 10 uH but he
should measure it since the manual provides no value for it. This of
course means he needs to create a fairly complex PWM controller to
make his IGBTs control current and might not obviate the need for the
few dozen spares.

While figuring out how to do one or the other of these he could also
probably learn to weld aluminum with DC Smile.
Back to top
Glen Walpert
electronics forum Guru Wannabe


Joined: 04 May 2005
Posts: 134

PostPosted: Wed Oct 12, 2005 5:14 pm    Post subject: Re: Driver for IGBT? (-ve off and +ve on from CMOS) Reply with quote

On 12 Oct 2005 14:39:27 GMT, Robert Latest <boblatest@yahoo.com>
wrote:

Quote:
["Followup-To:" header set to sci.electronics.design.]
On Wed, 12 Oct 2005 03:19:44 GMT,
Glen Walpert <gwalpert@notaxs.com> wrote
in Msg. <hmsok1t5i1im9rdqq3p35ck1oi0r4ur484@4ax.com

I haven't used the Hobart Cybertig, but I doubt very much if it puts
out *exactly* what is specified regardless of arc resistance. Do you
have the owners manual for it? If not I highly recommend you get one.

http://www.hobartwelders.com/owners_manuals/

I don't have a TIG welder, but if I (hopefully) ever have a good enough
excuse to buy one, it'll be a Hobart.

I simply love companies that publish manuals on the Net. It's also a
great service not only to the unfortunate owner who lost his manual,
but also to the would-be customer who can see exactly what he's gonna
get.

That said -- your Ignoramus's welder looks like a massive EMP device. No
electronics is going to survive this thing.

I would base my selection on features rather than brand if I were ever
to replace my ancient Miller Dialarc HF with a modern inverter, they
all come with manuals. And while making electronics work reliable in
a high EMI (ElectroMagnetic Interference) environment is not easy, it
has been done many times, especially in military and medical gear. I
sometimes do a crude EMI immunity test by wrapping a couple of turns
of welding cable around the DUT (Device Under Test) and firing up a
300 A AC arc right next to it, with the high frequency set to
continuous. Being careful not to melt the DUT. Sometimes it
continues working:-).
Back to top
Glen Walpert
electronics forum Guru Wannabe


Joined: 04 May 2005
Posts: 134

PostPosted: Wed Oct 12, 2005 5:24 pm    Post subject: Re: Driver for IGBT? (-ve off and +ve on from CMOS) Reply with quote

On Wed, 12 Oct 2005 14:50:16 GMT, Ignoramus24489
<ignoramus24489@NOSPAM.24489.invalid> wrote:

Quote:
On 12 Oct 2005 14:39:27 GMT, Robert Latest <boblatest@yahoo.com> wrote:
["Followup-To:" header set to sci.electronics.design.]
On Wed, 12 Oct 2005 03:19:44 GMT,
Glen Walpert <gwalpert@notaxs.com> wrote
in Msg. <hmsok1t5i1im9rdqq3p35ck1oi0r4ur484@4ax.com

I haven't used the Hobart Cybertig, but I doubt very much if it puts
out *exactly* what is specified regardless of arc resistance. Do you
have the owners manual for it? If not I highly recommend you get one.

http://www.hobartwelders.com/owners_manuals/

I don't have a TIG welder, but if I (hopefully) ever have a good enough
excuse to buy one, it'll be a Hobart.

I simply love companies that publish manuals on the Net. It's also a
great service not only to the unfortunate owner who lost his manual,
but also to the would-be customer who can see exactly what he's gonna
get.

Hobart and Miller manuals are available from
millerwelds.com. Actually, it seems that it is Miller that distributes
Hobart manuals. Hobart and its rights were bought in some complicated
business transactions.

I uploaded manuals for my cybertig to

http://igor.chudov.com/projects/Welding/00-Hobart-CyberTig-Welder/

That said -- your Ignoramus's welder looks like a massive EMP device. No
electronics is going to survive this thing.

What if the bridge shorts instead of opens during th switching event?
Would that help?

Yes, see other posts. But it won't help with providing adequate EMI
immunity for your gate drive circuits, something the Semikron driver
design probably gave some consideration to.
Back to top
Ignoramus24489
electronics forum beginner


Joined: 12 Oct 2005
Posts: 14

PostPosted: Wed Oct 12, 2005 5:30 pm    Post subject: Re: Driver for IGBT? (-ve off and +ve on from CMOS) Reply with quote

On Wed, 12 Oct 2005 17:24:36 GMT, Glen Walpert <gwalpert@notaxs.com> wrote:
Quote:
On Wed, 12 Oct 2005 14:50:16 GMT, Ignoramus24489
ignoramus24489@NOSPAM.24489.invalid> wrote:

On 12 Oct 2005 14:39:27 GMT, Robert Latest <boblatest@yahoo.com> wrote:
["Followup-To:" header set to sci.electronics.design.]
On Wed, 12 Oct 2005 03:19:44 GMT,
Glen Walpert <gwalpert@notaxs.com> wrote
in Msg. <hmsok1t5i1im9rdqq3p35ck1oi0r4ur484@4ax.com

I haven't used the Hobart Cybertig, but I doubt very much if it puts
out *exactly* what is specified regardless of arc resistance. Do you
have the owners manual for it? If not I highly recommend you get one.

http://www.hobartwelders.com/owners_manuals/

I don't have a TIG welder, but if I (hopefully) ever have a good enough
excuse to buy one, it'll be a Hobart.

I simply love companies that publish manuals on the Net. It's also a
great service not only to the unfortunate owner who lost his manual,
but also to the would-be customer who can see exactly what he's gonna
get.

Hobart and Miller manuals are available from
millerwelds.com. Actually, it seems that it is Miller that distributes
Hobart manuals. Hobart and its rights were bought in some complicated
business transactions.

I uploaded manuals for my cybertig to

http://igor.chudov.com/projects/Welding/00-Hobart-CyberTig-Welder/

That said -- your Ignoramus's welder looks like a massive EMP device. No
electronics is going to survive this thing.

What if the bridge shorts instead of opens during th switching event?
Would that help?

Yes, see other posts. But it won't help with providing adequate EMI
immunity for your gate drive circuits, something the Semikron driver
design probably gave some consideration to.

I am confused. My plan for my inverter circuit is to place it BEFORE
the high voltage, high frequency arc starter circuit. I believe,
rightly or wrongly, that the HF unit is isolated from the power supply
in the sense of conducting interference. If I am mistaken, or am
thinking about a wrong issue, I will appreciate being corrected.


i
--
Back to top
Ignoramus24489
electronics forum beginner


Joined: 12 Oct 2005
Posts: 14

PostPosted: Wed Oct 12, 2005 5:58 pm    Post subject: Re: Driver for IGBT? (-ve off and +ve on from CMOS) Reply with quote

On Wed, 12 Oct 2005 16:53:36 GMT, Glen Walpert <gwalpert@notaxs.com>
wrote:

Quote:
Indeed. And he won't find detailed instructions on doing that in
any data sheet or app note I know of, since they all assume or
explicitly show some reasonable approximation of a constant voltage
supply and try to avoid "shoot-through". i is exploring uncharted
territory here and I think he should at the very least get an
oscilloscope so he can measure the on overlap time,

I have two oscilloscopes, Tek 475 and Tek 2465, IIRC. Both actually
work. You see, I sell military surplus stuff on ebay and accumulated
some equipment. I will probably sell Tek 2465 eventually, but Tek 475
is my workhorse for testing stuff.

That said... How do you get to plot rare events such as my switchings,
on the oscilloscope, so that I see voltage spikes.

Also, would high voltage not endanger the oscilloscope itself?

Quote:
start at the lowest possible current setting

I will actually conduct first tests with a small regulated DC power
supply, current limited etc. Not even with the welder.

Quote:
and wear safety glasses and a face shield at all times his equipment
is energized. Buying a few dozen spare IGBTs and gate drivers would
be a good idea too.

haha

Quote:
OTOH, I think i might want to reconsider his entire approach,
especially if he wants to wind up with a really useful result.
Polarity reversals on his CC supply will let him manual TIG weld
aluminum in the event he can pull it off, but to duplicate the very
useful "foreground/background" pulse capabilities of a good inverter
welder

Can you elaborate on these capabilities?

Are you talking about different absolute values of voltage in negative
vs. positive cycle?

Or are you talking about pulsing output where current is varied
through time, say changing every X seconds from high to low current?

If you are talking about the latter, then my tig welder supports
pulsing, it has a few controls for it.


Quote:
he needs to go in and convert it to constant voltage by
providing new SCR controls to let the existing 3-phase SCR bridge
charge a new capacitor bank to an adjustable voltage (which would
normally be set between 40 and 80 volts or so), which would feed his
new IGBT bridge, which would drive the arc through the existing
"filter reactor", which I am guessing is on the order of 10 uH but
he should measure it since the manual provides no value for it.

IOW, you are suggesting to place the inverter before the "reactor"?

Quote:
This of course means he needs to create a fairly complex PWM
controller to make his IGBTs control current and might not obviate
the need for the few dozen spares.

While figuring out how to do one or the other of these he could also
probably learn to weld aluminum with DC Smile.

I think that it's kind of a fun project. I am not incredibly worried
about frying the electronics that I build, I mean, it is not that much
money. I would be upset, sure, but it is not the end of the world.

What I would be more upset is if somehow, this switching cound damage
the welder itself.

i
Back to top
Winfield Hill
electronics forum Guru


Joined: 05 May 2005
Posts: 1996

PostPosted: Wed Oct 12, 2005 9:31 pm    Post subject: Re: Driver for IGBT? (-ve off and +ve on from CMOS) Reply with quote

Ignoramus24489 wrote...
Quote:

That said -- your Ignoramus's welder looks like a massive
EMP device. No electronics is going to survive this thing.
[ snip ]
I am confused. My plan for my inverter circuit is to place it
BEFORE the high voltage, high frequency arc starter circuit.

That probably won't help much, the EMP comes from the arc, and
its dI/dt variations blast right back into the welder's circuits,
creating high-frequency V = L dI/dt spikes, where L is your big
rqeactor, plus lots of other wiring inductance, etc.


--
Thanks,
- Win
Back to top
Glen Walpert
electronics forum Guru Wannabe


Joined: 04 May 2005
Posts: 134

PostPosted: Thu Oct 13, 2005 12:04 am    Post subject: Re: Driver for IGBT? (-ve off and +ve on from CMOS) Reply with quote

On Wed, 12 Oct 2005 17:30:22 GMT, Ignoramus24489
<ignoramus24489@NOSPAM.24489.invalid> wrote:

Quote:
On Wed, 12 Oct 2005 17:24:36 GMT, Glen Walpert <gwalpert@notaxs.com> wrote:
On Wed, 12 Oct 2005 14:50:16 GMT, Ignoramus24489
ignoramus24489@NOSPAM.24489.invalid> wrote:

What if the bridge shorts instead of opens during th switching event?
Would that help?

Yes, see other posts. But it won't help with providing adequate EMI
immunity for your gate drive circuits, something the Semikron driver
design probably gave some consideration to.

I am confused. My plan for my inverter circuit is to place it BEFORE
the high voltage, high frequency arc starter circuit. I believe,
rightly or wrongly, that the HF unit is isolated from the power supply
in the sense of conducting interference. If I am mistaken, or am
thinking about a wrong issue, I will appreciate being corrected.

The HF is somewhat isolated from the existing SCR switches in the
inverter by the reactor, which also smoothes out the nasty switching
transients from the SCRs - while they are off the freewheeling diode
and the reactor inductance keeps the arc current going (but decaying,
to be boosted when the SCRs are on, thus the AC ripple on the output).
So before the reactor you see SCR switching transients, and after it
you see HF starting noise (if it is on, so don't turn it on) and arc
noise (unavoidable).

The electromagnetic fields from all those high currents being switched
will couple into every piece of wire in the general vicinity,
regardless of which side of the reactor you are on, unless you take
precautions to prevent it, like using shielded boxes with all I/O
differential with good common mode rejection. Books have been written
on the subject.
Back to top
Ignoramus24489
electronics forum beginner


Joined: 12 Oct 2005
Posts: 14

PostPosted: Thu Oct 13, 2005 12:44 am    Post subject: Re: Driver for IGBT? (-ve off and +ve on from CMOS) Reply with quote

On Thu, 13 Oct 2005 00:04:40 GMT, Glen Walpert <gwalpert@notaxs.com> wrote:
Quote:
On Wed, 12 Oct 2005 17:30:22 GMT, Ignoramus24489
ignoramus24489@NOSPAM.24489.invalid> wrote:

On Wed, 12 Oct 2005 17:24:36 GMT, Glen Walpert <gwalpert@notaxs.com> wrote:
On Wed, 12 Oct 2005 14:50:16 GMT, Ignoramus24489
ignoramus24489@NOSPAM.24489.invalid> wrote:

What if the bridge shorts instead of opens during th switching event?
Would that help?

Yes, see other posts. But it won't help with providing adequate EMI
immunity for your gate drive circuits, something the Semikron driver
design probably gave some consideration to.

I am confused. My plan for my inverter circuit is to place it BEFORE
the high voltage, high frequency arc starter circuit. I believe,
rightly or wrongly, that the HF unit is isolated from the power supply
in the sense of conducting interference. If I am mistaken, or am
thinking about a wrong issue, I will appreciate being corrected.

The HF is somewhat isolated from the existing SCR switches in the
inverter by the reactor, which also smoothes out the nasty switching
transients from the SCRs - while they are off the freewheeling diode
and the reactor inductance keeps the arc current going (but decaying,
to be boosted when the SCRs are on, thus the AC ripple on the output).
So before the reactor you see SCR switching transients, and after it
you see HF starting noise (if it is on, so don't turn it on) and arc
noise (unavoidable).

The electromagnetic fields from all those high currents being switched
will couple into every piece of wire in the general vicinity,
regardless of which side of the reactor you are on, unless you take
precautions to prevent it, like using shielded boxes with all I/O
differential with good common mode rejection. Books have been written
on the subject.

Well, in fact, I will place the box with the breadboard, heatsink and
everything electronic kind of far away from the reactor and HF
circuit. The welder is separated by a wall in the middle, and the
circuit will be beyond the wall. It is needed anyway for proper
cooling.

That said, would not a capacitor and a resistor protect me from
transient spikes? I thought that that's what snubbers are for.

i
Back to top
Glen Walpert
electronics forum Guru Wannabe


Joined: 04 May 2005
Posts: 134

PostPosted: Thu Oct 13, 2005 12:44 am    Post subject: Re: Driver for IGBT? (-ve off and +ve on from CMOS) Reply with quote

On 12 Oct 2005 17:58:00 GMT, Ignoramus24489
<ignoramus24489@NOSPAM.24489.invalid> wrote:

Quote:
On Wed, 12 Oct 2005 16:53:36 GMT, Glen Walpert <gwalpert@notaxs.com
wrote:

Indeed. And he won't find detailed instructions on doing that in
any data sheet or app note I know of, since they all assume or
explicitly show some reasonable approximation of a constant voltage
supply and try to avoid "shoot-through". i is exploring uncharted
territory here and I think he should at the very least get an
oscilloscope so he can measure the on overlap time,

I have two oscilloscopes, Tek 475 and Tek 2465, IIRC. Both actually
work. You see, I sell military surplus stuff on ebay and accumulated
some equipment. I will probably sell Tek 2465 eventually, but Tek 475
is my workhorse for testing stuff.

That said... How do you get to plot rare events such as my switchings,
on the oscilloscope, so that I see voltage spikes.

Trigger on the gate drive for instance.

Quote:
Also, would high voltage not endanger the oscilloscope itself?

Perhaps, stay within its ratings using a voltage divider if necessary.

Quote:
start at the lowest possible current setting

I will actually conduct first tests with a small regulated DC power
supply, current limited etc. Not even with the welder.

Very good idea.

Quote:
and wear safety glasses and a face shield at all times his equipment
is energized. Buying a few dozen spare IGBTs and gate drivers would
be a good idea too.

haha

Not joking, I have seen large plastic packages explode with the noise
of a shotgun blast and similar dispersal of small bits. How much do
you value your eyesight?

Quote:
OTOH, I think i might want to reconsider his entire approach,
especially if he wants to wind up with a really useful result.
Polarity reversals on his CC supply will let him manual TIG weld
aluminum in the event he can pull it off, but to duplicate the very
useful "foreground/background" pulse capabilities of a good inverter
welder

Can you elaborate on these capabilities?

Are you talking about different absolute values of voltage in negative
vs. positive cycle?

Or are you talking about pulsing output where current is varied
through time, say changing every X seconds from high to low current?

If you are talking about the latter, then my tig welder supports
pulsing, it has a few controls for it.

No, I am talking about pulsing the current without changing the
polarity from a low current (background) to a higher current
(foreground) at rates from around 10 pulses per sec (for a wide arc)
to several hundred pulses per second (for a narrow arc). Changing
polarity is useful only for manual TIG of aluminum IMO, and when you
do that it is useful to set the reverse polarity time less than the
fwd polarity time to get more heat into the work.

Some inverter welders also allow you to switch to constant voltage
operation for MIG welding, which is the only reasonable way to do most
aluminum welds at reasonable production rates, and to adjust the
"stiffness" or droop in either CC or CV modes.

Quote:
he needs to go in and convert it to constant voltage by
providing new SCR controls to let the existing 3-phase SCR bridge
charge a new capacitor bank to an adjustable voltage (which would
normally be set between 40 and 80 volts or so), which would feed his
new IGBT bridge, which would drive the arc through the existing
"filter reactor", which I am guessing is on the order of 10 uH but
he should measure it since the manual provides no value for it.

IOW, you are suggesting to place the inverter before the "reactor"?

Yes, that was part of my suggestion, but it was a bad suggestion for a
beginner as it requires a rather good understanding of switch mode
power supply design. Sorry I mentioned it.

Quote:
This of course means he needs to create a fairly complex PWM
controller to make his IGBTs control current and might not obviate
the need for the few dozen spares.

While figuring out how to do one or the other of these he could also
probably learn to weld aluminum with DC Smile.

I think that it's kind of a fun project. I am not incredibly worried
about frying the electronics that I build, I mean, it is not that much
money. I would be upset, sure, but it is not the end of the world.

What I would be more upset is if somehow, this switching cound damage
the welder itself.

Not too likely if you keep your bridge downstream of the reactor, like
outside of the welder box. It is a pretty tough machine. And if it
does break you have another fun project on your hands Smile.
Back to top
Terry Given
electronics forum Guru


Joined: 30 Apr 2005
Posts: 868

PostPosted: Thu Oct 13, 2005 2:35 am    Post subject: Re: Driver for IGBT? (-ve off and +ve on from CMOS) Reply with quote

Glen Walpert wrote:
Quote:
On 12 Oct 2005 17:58:00 GMT, Ignoramus24489
ignoramus24489@NOSPAM.24489.invalid> wrote:


On Wed, 12 Oct 2005 16:53:36 GMT, Glen Walpert <gwalpert@notaxs.com
wrote:


Indeed. And he won't find detailed instructions on doing that in
any data sheet or app note I know of, since they all assume or
explicitly show some reasonable approximation of a constant voltage
supply and try to avoid "shoot-through". i is exploring uncharted
territory here and I think he should at the very least get an
oscilloscope so he can measure the on overlap time,

I have two oscilloscopes, Tek 475 and Tek 2465, IIRC. Both actually
work. You see, I sell military surplus stuff on ebay and accumulated
some equipment. I will probably sell Tek 2465 eventually, but Tek 475
is my workhorse for testing stuff.

That said... How do you get to plot rare events such as my switchings,
on the oscilloscope, so that I see voltage spikes.

carefully. usually with a DSO, so you can catch the damned transient
event that snots your IGBTs. I once spent 3hrs in the lab, blanket
draped over me & the PM3394 scope in analogue mode, brightness cranked
up, to catch a glimpse of something I suspected was there, but couldnt
make the digital scope trigger on. I eventually saw one glitch, me & a
buddy then reasoned out the cause in about 3 minutes flat, and had a fix
running about 1/2 an hour later. metastability causing a race condition
causing runt pulse causing very intermittent self-destruct. yay for ISP
CPLDs.

Quote:


Trigger on the gate drive for instance.


Also, would high voltage not endanger the oscilloscope itself?


Perhaps, stay within its ratings using a voltage divider if necessary.


then, when it gets destroyed, replace the scope and get a few P5200 diff
probes.

Quote:

start at the lowest possible current setting

I will actually conduct first tests with a small regulated DC power
supply, current limited etc. Not even with the welder.


Very good idea.


and wear safety glasses and a face shield at all times his equipment
is energized. Buying a few dozen spare IGBTs and gate drivers would
be a good idea too.

haha


Not joking, I have seen large plastic packages explode with the noise
of a shotgun blast and similar dispersal of small bits. How much do
you value your eyesight?

1,000J will destroy many plastic packages. A Semikron rep once proudly
announced their "rupture-proof" 1/2-bridge modules. So we bolted some to
our 100kW prototype, and blew them up. The first one didnt rupture, but
the drive leaped about 1' into the air. So we replaced the IGBT, and sat
an anvil on top of a sheet of nomex on top of the unit, and the module
blew itself to tiny pieces. He was suitably embarassed, but bought us
lunch so that was OK. IIRC it was only a kJ or so....

Quote:


OTOH, I think i might want to reconsider his entire approach,
especially if he wants to wind up with a really useful result.
Polarity reversals on his CC supply will let him manual TIG weld
aluminum in the event he can pull it off, but to duplicate the very
useful "foreground/background" pulse capabilities of a good inverter
welder

Can you elaborate on these capabilities?

Are you talking about different absolute values of voltage in negative
vs. positive cycle?

Or are you talking about pulsing output where current is varied
through time, say changing every X seconds from high to low current?

If you are talking about the latter, then my tig welder supports
pulsing, it has a few controls for it.


No, I am talking about pulsing the current without changing the
polarity from a low current (background) to a higher current
(foreground) at rates from around 10 pulses per sec (for a wide arc)
to several hundred pulses per second (for a narrow arc). Changing
polarity is useful only for manual TIG of aluminum IMO, and when you
do that it is useful to set the reverse polarity time less than the
fwd polarity time to get more heat into the work.

Some inverter welders also allow you to switch to constant voltage
operation for MIG welding, which is the only reasonable way to do most
aluminum welds at reasonable production rates, and to adjust the
"stiffness" or droop in either CC or CV modes.


he needs to go in and convert it to constant voltage by
providing new SCR controls to let the existing 3-phase SCR bridge
charge a new capacitor bank to an adjustable voltage (which would
normally be set between 40 and 80 volts or so), which would feed his
new IGBT bridge, which would drive the arc through the existing
"filter reactor", which I am guessing is on the order of 10 uH but
he should measure it since the manual provides no value for it.

IOW, you are suggesting to place the inverter before the "reactor"?


Yes, that was part of my suggestion, but it was a bad suggestion for a
beginner as it requires a rather good understanding of switch mode
power supply design. Sorry I mentioned it.


This of course means he needs to create a fairly complex PWM
controller to make his IGBTs control current and might not obviate
the need for the few dozen spares.

While figuring out how to do one or the other of these he could also
probably learn to weld aluminum with DC Smile.

I think that it's kind of a fun project. I am not incredibly worried
about frying the electronics that I build, I mean, it is not that much
money. I would be upset, sure, but it is not the end of the world.

What I would be more upset is if somehow, this switching cound damage
the welder itself.


Not too likely if you keep your bridge downstream of the reactor, like
outside of the welder box. It is a pretty tough machine. And if it
does break you have another fun project on your hands Smile.
Back to top
Glen Walpert
electronics forum Guru Wannabe


Joined: 04 May 2005
Posts: 134

PostPosted: Thu Oct 13, 2005 1:52 pm    Post subject: Re: Driver for IGBT? (-ve off and +ve on from CMOS) Reply with quote

On Thu, 13 Oct 2005 00:44:21 GMT, Ignoramus24489
<ignoramus24489@NOSPAM.24489.invalid> wrote:

Quote:
On Thu, 13 Oct 2005 00:04:40 GMT, Glen Walpert <gwalpert@notaxs.com> wrote:
On Wed, 12 Oct 2005 17:30:22 GMT, Ignoramus24489
ignoramus24489@NOSPAM.24489.invalid> wrote:

On Wed, 12 Oct 2005 17:24:36 GMT, Glen Walpert <gwalpert@notaxs.com> wrote:
On Wed, 12 Oct 2005 14:50:16 GMT, Ignoramus24489
ignoramus24489@NOSPAM.24489.invalid> wrote:

What if the bridge shorts instead of opens during th switching event?
Would that help?

Yes, see other posts. But it won't help with providing adequate EMI
immunity for your gate drive circuits, something the Semikron driver
design probably gave some consideration to.

I am confused. My plan for my inverter circuit is to place it BEFORE
the high voltage, high frequency arc starter circuit. I believe,
rightly or wrongly, that the HF unit is isolated from the power supply
in the sense of conducting interference. If I am mistaken, or am
thinking about a wrong issue, I will appreciate being corrected.

The HF is somewhat isolated from the existing SCR switches in the
inverter by the reactor, which also smoothes out the nasty switching
transients from the SCRs - while they are off the freewheeling diode
and the reactor inductance keeps the arc current going (but decaying,
to be boosted when the SCRs are on, thus the AC ripple on the output).
So before the reactor you see SCR switching transients, and after it
you see HF starting noise (if it is on, so don't turn it on) and arc
noise (unavoidable).

The electromagnetic fields from all those high currents being switched
will couple into every piece of wire in the general vicinity,
regardless of which side of the reactor you are on, unless you take
precautions to prevent it, like using shielded boxes with all I/O
differential with good common mode rejection. Books have been written
on the subject.

Well, in fact, I will place the box with the breadboard, heatsink and
everything electronic kind of far away from the reactor and HF
circuit. The welder is separated by a wall in the middle, and the
circuit will be beyond the wall. It is needed anyway for proper
cooling.

That said, would not a capacitor and a resistor protect me from
transient spikes? I thought that that's what snubbers are for.

The EMI issue is mostly about your low level oscillator and gate drive
circuits which must be protected by design and layout, not snubbers.
You could in theory protect your IGBTs with snubbers, but when you do
the calculation on how big they need to be to handle your reactor
energy you will appreciate the approach of never letting both sides of
your bridge turn off at the same time in order to avoid the need to
snub that much energy. This approach completely screws desaturation
protection however because if the desat circuit turns off your IGBTs
they will see the full reactor energy. Have you measured your reactor
inductance yet? That is the first thing you need to know in order to
design a workable snubber, otherwise you may just add an exploding
snubber cap to your exploding IGBT!

Mostly snubbers are used to protect from the energy stored in
relatively small wiring inductances, not big-ass reactors. If you
simply copy someones wiring inductance snubbers you will be needing
those safety glasses and face shield the first time your bridge turns
off at high current, either due to design error or when EMI kills your
control circuit!
Back to top
Ignoramus25589
electronics forum beginner


Joined: 13 Oct 2005
Posts: 13

PostPosted: Thu Oct 13, 2005 2:58 pm    Post subject: Re: Driver for IGBT? (-ve off and +ve on from CMOS) Reply with quote

On Thu, 13 Oct 2005 13:52:22 GMT, Glen Walpert <gwalpert@notaxs.com> wrote:
Quote:
On Thu, 13 Oct 2005 00:44:21 GMT, Ignoramus24489
ignoramus24489@NOSPAM.24489.invalid> wrote:

On Thu, 13 Oct 2005 00:04:40 GMT, Glen Walpert <gwalpert@notaxs.com> wrote:
On Wed, 12 Oct 2005 17:30:22 GMT, Ignoramus24489
ignoramus24489@NOSPAM.24489.invalid> wrote:

On Wed, 12 Oct 2005 17:24:36 GMT, Glen Walpert <gwalpert@notaxs.com> wrote:
On Wed, 12 Oct 2005 14:50:16 GMT, Ignoramus24489
ignoramus24489@NOSPAM.24489.invalid> wrote:

What if the bridge shorts instead of opens during th switching event?
Would that help?

Yes, see other posts. But it won't help with providing adequate EMI
immunity for your gate drive circuits, something the Semikron driver
design probably gave some consideration to.

I am confused. My plan for my inverter circuit is to place it BEFORE
the high voltage, high frequency arc starter circuit. I believe,
rightly or wrongly, that the HF unit is isolated from the power supply
in the sense of conducting interference. If I am mistaken, or am
thinking about a wrong issue, I will appreciate being corrected.

The HF is somewhat isolated from the existing SCR switches in the
inverter by the reactor, which also smoothes out the nasty switching
transients from the SCRs - while they are off the freewheeling diode
and the reactor inductance keeps the arc current going (but decaying,
to be boosted when the SCRs are on, thus the AC ripple on the output).
So before the reactor you see SCR switching transients, and after it
you see HF starting noise (if it is on, so don't turn it on) and arc
noise (unavoidable).

The electromagnetic fields from all those high currents being switched
will couple into every piece of wire in the general vicinity,
regardless of which side of the reactor you are on, unless you take
precautions to prevent it, like using shielded boxes with all I/O
differential with good common mode rejection. Books have been written
on the subject.

Well, in fact, I will place the box with the breadboard, heatsink and
everything electronic kind of far away from the reactor and HF
circuit. The welder is separated by a wall in the middle, and the
circuit will be beyond the wall. It is needed anyway for proper
cooling.

That said, would not a capacitor and a resistor protect me from
transient spikes? I thought that that's what snubbers are for.

The EMI issue is mostly about your low level oscillator and gate drive
circuits which must be protected by design and layout, not snubbers.

I see. The gate driver would isolate them from load electrically, to
some extent. Are you concerned about EM waves inducing currents in my
circuitry, or about interference arriving through power cables?

If it is the former, I will place this item in a separate compartment
of the welder as opposed to where the HF unit is. If it is the latter,
I will try to make sure that there is sufficient isolation.

Quote:
You could in theory protect your IGBTs with snubbers, but when you do
the calculation on how big they need to be to handle your reactor
energy you will appreciate the approach of never letting both sides of
your bridge turn off at the same time in order to avoid the need to
snub that much energy.

Sure. I mean, I would not mind installing a large capacitor, but that
could worsen starting conditions. I will indeed try to select gate
resistors so that there is a little bit of cross conduction.


Quote:
This approach completely screws desaturation protection however
because if the desat circuit turns off your IGBTs they will see the
full reactor energy. Have you measured your reactor inductance yet?
That is the first thing you need to know in order to design a
workable snubber, otherwise you may just add an exploding snubber
cap to your exploding IGBT!

I have not yet, have not had time, but I do want to do that indeed.

Quote:
Mostly snubbers are used to protect from the energy stored in
relatively small wiring inductances, not big-ass reactors. If you
simply copy someones wiring inductance snubbers you will be needing
those safety glasses and face shield the first time your bridge turns
off at high current, either due to design error or when EMI kills your
control circuit!

I agree 100%. I think that there is a calculation of snubber size
based on inductance, in the Fuji manual.

I started drawing schematic of my circuit, and will calculate all
needed things in Gnumeric (equivalent of excel). All resistances,
capacitances etc.

i
Back to top
Google

Back to top
Display posts from previous:   
Post new topic   Reply to topic Page 1 of 3 [41 Posts] Goto page:  1, 2, 3 Next
View previous topic :: View next topic
The time now is Sat Sep 23, 2017 5:45 am | All times are GMT
Forum index » Electronix » design
Jump to:  

Similar Topics
Topic Author Forum Replies Last Post
No new posts USB to serial driver for FT232R in QNX 6.4.1 coolKitty design 0 Tue May 18, 2010 11:32 am
No new posts Sales:IGBT Module and IC lima-sz design 0 Fri Oct 17, 2008 2:34 pm
No new posts H BRIDGE, driver for a dc motor nuiseng191 Basics 0 Mon Jun 11, 2007 1:42 pm
No new posts PSpiceMOSFET/IGBT Modeling mlwalters cad 0 Sun Oct 15, 2006 7:29 pm
No new posts DC Motor Driver Regenerative Braking Humphrey B. Bear design 10 Sat Jul 15, 2006 9:53 pm

Copyright © 2004-2005 DeniX Solutions SRL
Other DeniX Solutions sites: Unix/Linux blog |  Unix/Linux documentation |  Unix/Linux forums |  Medicine forum |  Science forum  |  Send and track newsletters


Powered by phpBB © 2001, 2005 phpBB Group