PWM chip that can drive an H bridge?

[Ignoramus10070]

Is there a PWM chip that can properly drive an H bridge? (with dead
time, fast on and off times etc)?

thanks

i

 

[Tim Wescott]

Is there a PWM chip that can properly drive an H bridge? (with dead
time, fast on and off times etc)?

thanks

i

Most of the DSP chips designed for motor drive that I've looked into --
ADI, Freescale, TI -- do this.

Dunno about discrete chips, though.

 

[Ignoramus10070]

Most of the DSP chips designed for motor drive that I've looked into --
ADI, Freescale, TI -- do this.

Dunno about discrete chips, though.

Thanks. You are suggesting something very interesting. I looked at
motor drive chips yesterday and was a little confused. Are they also
good for power switching applications?

Again, I am looking for some easy solution that would have frequency
and duty cycle settable, and would drive an H bridge. I can do that
either with one oscillator chip and two half bridge gate drivers, or,
possibly, with something else.

If you have a suggestion for something simple and inexpensive, it will
be greatly appreciated.

Thanks!

i

 

[Tim Wescott]

Thanks. You are suggesting something very interesting. I looked at
motor drive chips yesterday and was a little confused. Are they also
good for power switching applications?

Again, I am looking for some easy solution that would have frequency
and duty cycle settable, and would drive an H bridge. I can do that
either with one oscillator chip and two half bridge gate drivers, or,
possibly, with something else.

If you have a suggestion for something simple and inexpensive, it will
be greatly appreciated.

Thanks!

i

A DSP may be the most simple and inexpensive solution for your problem,
but only if your problem is big enough to justify having a processor
sitting on your board.

What the PWM circuits in the motor drive DSPs provide is all the
"smarts" -- settable PWM frequency, duty cycle, & vector drive math for
that little bit of extra overhead on a 3-phase brushless system. Their
output is just logic level, so you still have to provide all the power
electronics but your power circuit can be as dumb as a post -- you only
need enough stuff to turn the FETs on and off fast, not all the timing &c.

Take a look at a datasheet, or some whitepapers, and see.

 

[Tim Wescott]

Most of the DSP chips designed for motor drive that I've looked into --
ADI, Freescale, TI -- do this.

Dunno about discrete chips, though.

Oh, I'm stupid.

Some of the more sophisticated switching regulators provide all the
deadband & timing stuff, and sometimes even FET switching. There's no
reason you have to use them to actually regulate, unless you want to.

I'm trying to remember the part number, but I've seen this sort of thing
used for 4-quadrant switching amplifiers with full H bridges. All of
the smarts were analog and on the regulator controller, with the
regulator connected either directly to the FET gates or through snubber
circuits.

But there were some interesting snubber circuits, which would still be
required even with a DSP solution. Something about shoot-through, and
FETs that would scatter bits of chip package all over the lab...

 

[Ignoramus10070]

Oh, I'm stupid.

Some of the more sophisticated switching regulators provide all the
deadband & timing stuff,

I have seen several chips like that, with easily settable frequency
and duty cycle.

and sometimes even FET switching.

I have not seen one like that, and, frankly, that's exactly what I am
looking for.

So, I could find a chip that would switch IGBTs in a full bridge, and
also have settable frequency and duty cycle (using pots or some such),
I would be in heaven.

The trouble is, no one knows if there is such a chip, people say all
kinds of smart things, I spent a couple of hours looking, and did not
find anything.

There's no
reason you have to use them to actually regulate, unless you want to.

I'm trying to remember the part number, but I've seen this sort of
thing used for 4-quadrant switching amplifiers with full H bridges.
All of the smarts were analog and on the regulator controller, with
the regulator connected either directly to the FET gates or through
snubber circuits.

If you could recall the part number, I would be eternally grateful.

But there were some interesting snubber circuits, which would still be
required even with a DSP solution. Something about shoot-through, and
FETs that would scatter bits of chip package all over the lab...

Sure, no doubt about that. That's not really a problem. I am looking
for a simple gate driver solution, but snubber circuits would be
separate.

So, again, if you can recall the part #, that would be wonderful and
most appreciated.

I will be driving toshiba IGBTs. I received them (4 on a heatsink)
last night. I was already able to drive them with a wavetek 171, and
see them conduct current when turned ON and not conduct when turned
off. I watched wavetek's on my oscilloscope and it was a fun evening.

i

--

 

[Tim Wescott]

I have seen several chips like that, with easily settable frequency
and duty cycle.




I have not seen one like that, and, frankly, that's exactly what I am
looking for.

So, I could find a chip that would switch IGBTs in a full bridge, and
also have settable frequency and duty cycle (using pots or some such),
I would be in heaven.

The trouble is, no one knows if there is such a chip, people say all
kinds of smart things, I spent a couple of hours looking, and did not
find anything.




If you could recall the part number, I would be eternally grateful.




Sure, no doubt about that. That's not really a problem. I am looking
for a simple gate driver solution, but snubber circuits would be
separate.

So, again, if you can recall the part #, that would be wonderful and
most appreciated.

I will be driving toshiba IGBTs. I received them (4 on a heatsink)
last night. I was already able to drive them with a wavetek 171, and
see them conduct current when turned ON and not conduct when turned
off. I watched wavetek's on my oscilloscope and it was a fun evening.

i

Check out the HIP4080A, by Intersil. It doesn't include the oscillator,
but it has everything else including programmable dead time on the top
and bottom FETs. You could use anything from a 555 on up for the
oscillator (and if you're clever about where you put the PNP you can
even get a linear ramp from a 555).

 

[Pooh Bear]

I have seen several chips like that, with easily settable frequency
and duty cycle.


I have not seen one like that, and, frankly, that's exactly what I am
looking for.

So, I could find a chip that would switch IGBTs in a full bridge, and
also have settable frequency and duty cycle (using pots or some such),
I would be in heaven.

The actual FET / IGBT gate driver ( and level shifter ) is typically a separate
IC from the control chip.

Probably can't be done well in one package due to different processes required I
expect.

Graham

 

[Ignoramus10070]

The actual FET / IGBT gate driver ( and level shifter ) is typically a separate
IC from the control chip.

Probably can't be done well in one package due to different processes required I
expect.

I see. That's unfortunate, but looks like it's the reality that I can
deal with.

I like IR21094 as a gate driver, and also MAX038 as the signal
generator. Both are DIP based. The only issue that I see, for now
(which may not be a big deal) is that MAX038 produces +- 1 volt, and
IR21094 needs 5 or so volts to turn the bridge sides on and off. So I
need to amplify MAX038's output a little bit. That can probably be
done with an appropriate transistor or some such.

So, I would use MAX038 driving two IR21094 gate drivers.

MAX038 has its frequency and duty cycle separately settable with pots.

Does this plan make sense to you?

i

 

[Winfield Hill]

Tim Wescott wrote...

Check out the HIP4080A, by Intersil. It doesn't include the oscillator,
but it has everything else including programmable dead time on the top
and bottom FETs. You could use anything from a 555 on up for the
oscillator (and if you're clever about where you put the PNP you can
even get a linear ramp from a 555).

I'm fond of the HIP4080A and HIP4081A parts. Intersil has app notes
showing how to implement a PWM controller from the '4080, using its
comparator, but I prefer the '4081 with an external PWM generator.
For example, with a phase-shift PWM modulator for the H-bridge FETs.

 

[Ignoramus10070]

Check out the HIP4080A, by Intersil. It doesn't include the oscillator,
but it has everything else including programmable dead time on the top
and bottom FETs. You could use anything from a 555 on up for the
oscillator (and if you're clever about where you put the PNP you can
even get a linear ramp from a 555).

It certainly looks very nice. I like the fact that it is a full bridge
driver. I just printed out its datasheet and will read it on the way
home.

Thanks!

i

 

[Ignoramus10070]

Tim Wescott wrote...

I'm fond of the HIP4080A and HIP4081A parts. Intersil has app notes
showing how to implement a PWM controller from the '4080, using its
comparator, but I prefer the '4081 with an external PWM generator.
For example, with a phase-shift PWM modulator for the H-bridge FETs.

Sounds very nice. Seems like you know what you are talking about.

What PWM generator would you suggest (preferably DIP), using similar
power supply voltage and working from 50-1000 Hz at least, and
providing duty cycle control, at least 15-85%, etc.

Thank you!

i

 

[Pooh Bear]

I see. That's unfortunate, but looks like it's the reality that I can
deal with.

I like IR21094 as a gate driver, and also MAX038 as the signal
generator. Both are DIP based. The only issue that I see, for now
(which may not be a big deal) is that MAX038 produces +- 1 volt, and
IR21094 needs 5 or so volts to turn the bridge sides on and off. So I
need to amplify MAX038's output a little bit. That can probably be
done with an appropriate transistor or some such.

So, I would use MAX038 driving two IR21094 gate drivers.

MAX038 has its frequency and duty cycle separately settable with pots.

Does this plan make sense to you?

I don't understand why you want to use a Max038.

The MAX038 is a high-frequency, precision function generator producing accurate, high-frequency
triangle, sawtooth, sine, square, and pulse waveforms with a minimum of external components. The
output frequency can be controlled over a frequency range of 0.1Hz to 20MHz

Why not use a classic PWM controller ?

Graham

 

[Ignoramus10070]

I don't understand why you want to use a Max038.

The MAX038 is a high-frequency, precision function generator producing accurate, high-frequency
triangle, sawtooth, sine, square, and pulse waveforms with a minimum of external components. The
output frequency can be controlled over a frequency range of 0.1Hz to 20MHz

There is no reason other than seeing that it could somehow do what I
need. If you have a better suggestion, I will appreciate.

Why not use a classic PWM controller ?

Would you have a specific chip that you could suggest?

thanks

i

 

[Ignoramus10070]

I don't understand why you want to use a Max038.

The MAX038 is a high-frequency, precision function generator producing accurate, high-frequency
triangle, sawtooth, sine, square, and pulse waveforms with a minimum of external components. The
output frequency can be controlled over a frequency range of 0.1Hz to 20MHz

Why not use a classic PWM controller ?

Pooh, I found a very simple chip XR2206, it is not perfect because
duty cycle and frequency are not independently selected, but it is
simple.

i

 

[Robert Latest]

I'm fond of the HIP4080A and HIP4081A parts. Intersil has app notes
showing how to implement a PWM controller from the '4080, using its
comparator, but I prefer the '4081 with an external PWM generator.
For example, with a phase-shift PWM modulator for the H-bridge FETs.

Don't forget to look at IR's 2110 half-bridge driver and friends.
Two of them are actually cheaper than the HIP full-bridge driver,
and they go to 600V. The HIPs have some nice features though,
like UV lockout, charge pumps, shoot-through protection, and
higher speed.

robert

 

[Robert Latest]

The actual FET / IGBT gate driver ( and level shifter ) is typically a separate
IC from the control chip.

Probably can't be done well in one package due to different processes required I
expect.

And there are too many different output stage requirements. There
isn't really much point in integrating the gate driver with the
control chip. ST makes integrated power half and full bridges
with logic compatible inputs. I what you're after is low parts
count you might want to check into those, too.

robert

 

[Robert Latest]

Pooh, I found a very simple chip XR2206, it is not perfect because
duty cycle and frequency are not independently selected, but it is
simple.

Ignoramus, this is about the time you'd better start telling us
what it is that you're trying to accomplish, because as the
thread wears on you're getting more and more off track.

If your problem falls into the typical range of PWM applications
(motor/solenoid control or power conversion) there is a host of
good PWM control chips out there, and the XR2206 isn't one of
them.

robert

 

[Ignoramus30105]

Ignoramus, this is about the time you'd better start telling us
what it is that you're trying to accomplish, because as the
thread wears on you're getting more and more off track.

I am trying to build a tig inverter. To that end, I want to have gate
driver controlling a full H bridge. To send signals to gate
drivers, I want to find an appropriate chip that would send square
wave signal to gate drivers and have its frequency and duty cycle
separately settable. Frequency should vary between 30 and 1000 Hz, at
least, and duty cycle between 15% and 85%, or better.

Do you have any suggestions, such as specific chips.

If your problem falls into the typical range of PWM applications
(motor/solenoid control or power conversion) there is a host of
good PWM control chips out there, and the XR2206 isn't one of
them.

Would you have a specific suggestion for the application that I
described?

That would be most appreciated.

i

 

[Pooh Bear]

Pooh, I found a very simple chip XR2206, it is not perfect because
duty cycle and frequency are not independently selected, but it is
simple.

EXAR !

You're missing the point.

These chips aren't designed as PWM controllers.

No dead time for starters.

Graham