power MOS advice

[:: Marcolino ::]

Hi!
I ask you an advice witch MOS to use for pwm (20Khz) signal for my load
(Peltier cell) on 15V 7A
I run it with IRLZ14 but crashed in 2 weeks.

Thanks,

Marco


p.s.
sorry for my english, but I'm foreign

 

[colin]

Hi!
I ask you an advice witch MOS to use for pwm (20Khz) signal for my load
(Peltier cell) on 15V 7A
I run it with IRLZ14 but crashed in 2 weeks.

Thanks,

Marco


p.s.
sorry for my english, but I'm foreign

couldnt find the specs for that device off hand, but although replacing it
with a biger heftier device may solve the problem, maybe it would be better
to check that the mosfet is being driven on and off quickly and hard enough
and the peak switching voltage is controlled.

Colin =^.^=

 

[Ken Smith]

Hi!
I ask you an advice witch MOS to use for pwm (20Khz) signal for my load
(Peltier cell) on 15V 7A
I run it with IRLZ14 but crashed in 2 weeks.

What sort of failure did you have?

Can you discribe the circuit?

If it ran for 2 weeks it could be that it wasn't the MOSFETs fault. Some
other part may be the first failure or a design error may from time to
time cause bad things to happen.

 

[Spehro Pefhany]

Hi!
I ask you an advice witch MOS to use for pwm (20Khz) signal for my load
(Peltier cell) on 15V 7A
I run it with IRLZ14 but crashed in 2 weeks.

Thanks,
Marco

http://www.irf.com/product-info/datasheets/data/irlz14.pdf

That's a pretty wimpy old MOSFET, only 10A Id rating (7A at 100'C),
and 200m Ohm Rds(on) (around 300 hot). Something like the IRLZ44
(50A/60V/28mOhm cold) or the Fairchild RFP30N06LE (30A/60V/47mOhm
cold) would be better, and not expensive.

http://www.irf.com/product-info/datasheets/data/irlz44.pdf
http://www.fairchildsemi.com/ds/RF/RFP30N06LE.pdf

What does the actual current through the MOSFET look like?

I do hope you're not going directly into the Peltier with raw PWM.

sorry for my english, but I'm foreign

Nto a prblem.

 

[:: Marcolino ::]

Hi, this is circuit:
http://www.gallicisalpini.it/img/peltier.jpg

What does the actual current through the MOSFET look like? 6A

I do hope you're not going directly into the Peltier with raw PWM.

see the image if correct?


Marco

 

[:: Marcolino ::]

What sort of failure did you have?
plastic explose

Can you discribe the circuit?

http://www.gallicisalpini.it/img/peltier.jpg


Thanks, Marco

 

[Fred Bartoli]

On Tue, 12 Oct 2004 14:36:37 GMT, the renowned ":: Marcolino ::"


Nto a prblem.

Foreign fingers Spehro ?

 

[[email protected]]

http://www.gallicisalpini.it/img/peltier.jpg

I just had an wild idea.. maybe the capacitor C6 or a reverse peltier effect
(heat diff -> reverse potential) is causeing a "back emf". And could be solved
by haveing a regular diode as protection for the mosfet.

Sufficient cooling is an absolute must however. How have you solved this?

And you could possible investigate the use of "IBGT" semiconductor to switch
the peltier.

Beware thermal stresses applied to the peltier through pwm.

 

[Spehro Pefhany]

Hi, this is circuit:
http://www.gallicisalpini.it/img/peltier.jpg


see the image if correct?

You are probably getting a lot of unnecessary I^2*R heating in the
Peltier. If the Peltier is actually designed for, say, 9 Vmax and
you're hitting it with 15V pulses, for example, the average current
(as measured by an inexpensive multimeter) may look okay but the
efficiency will be less than it should be because of the current
spikes (much higher RMS current), and can even cause thermal runaway
in a closed-loop system. If you add a series inductor, shunt capacitor
(low impedance type) and a fast catch diode you can reduce the peak
current in both the MOSFET and the Peltier device. Usually module
manufacturers recommend < 10% ripple.

 

[legg]

Hi, this is circuit:
http://www.gallicisalpini.it/img/peltier.jpg


see the image if correct?

I'd suspect the PIC, oscillator and drive methods first.

To pwm from a PIC, you need to be very sure that SW or clock faults
result in a switch that is off and a program that can recover or
self-reset into a safe to run state.

Check the drive, at the gate, and look for well defined drive voltage
levels.

The PIC outputs can go Hi-Z, if I'm not mistaken, so perhaps a smaller
pull down resistor is in order, to avoid long glitches at supply
turn-on.

If the mosfet is any distance from the PIC, a local drive buffer might
be advisable at the gate.

RL

 

[Ken Smith]

plastic explose

http://www.gallicisalpini.it/img/peltier.jpg

I can't see anything in the schematic that would account for the
transistor failing.

Things to check:

Does the input power have spikes on it?

Can the PIC ever rattle the line up and down very fast or go into a
tri-state condition such that the MOSFET's gate gets some half way voltage
for a long period of time?

 

[Tim Wescott]

I'd suspect the PIC, oscillator and drive methods first.

To pwm from a PIC, you need to be very sure that SW or clock faults
result in a switch that is off and a program that can recover or
self-reset into a safe to run state.

Check the drive, at the gate, and look for well defined drive voltage
levels.

The PIC outputs can go Hi-Z, if I'm not mistaken, so perhaps a smaller
pull down resistor is in order, to avoid long glitches at supply
turn-on.

If the mosfet is any distance from the PIC, a local drive buffer might
be advisable at the gate.

RL

Where I've done this we've put in dead-processor detectors that shut off
the PWM if it stops -- this prevents you from driving it 100% on, but it
also keeps you from letting the smoke out. Unfortunately I did not
design the circuits, so I can't recommend anything.

This should just be a convenience at software design time -- if you
release to production while you're still having software faults in the
amount of code that you can fit into a PIC then you _really_ have QA
problems.

 

[Spehro Pefhany]

I'd suspect the PIC, oscillator and drive methods first.

To pwm from a PIC, you need to be very sure that SW or clock faults
result in a switch that is off and a program that can recover or
self-reset into a safe to run state.

Check the drive, at the gate, and look for well defined drive voltage
levels.

The PIC outputs can go Hi-Z, if I'm not mistaken, so perhaps a smaller
pull down resistor is in order, to avoid long glitches at supply
turn-on.

If the mosfet is any distance from the PIC, a local drive buffer might
be advisable at the gate.

RL

Once the PWM module is enabled, it should only output 0V/Vdd unless
it's disabled in software or reset. It's conceivable that the WDT
could be resetting the chip and thus tristating the output, but even
making generous assumptions, I'm not sure that ~half a joule
worst-case is all that bad, unless it was happening very frequently.

What's the case temperature on the MOSFET in normal operation?

If it's running @50% duty cycle, 6A average, the power dissipation
could be in the 20W neighborhood without even considering the
switching losses.

 

[Spehro Pefhany]

Where I've done this we've put in dead-processor detectors that shut off
the PWM if it stops -- this prevents you from driving it 100% on, but it
also keeps you from letting the smoke out. Unfortunately I did not
design the circuits, so I can't recommend anything.

There's a 6A fuse in series with the load in his circuit. Presumably
the zener and resistor are there to keep a MOSFET with a G-D short
from taking out the PIC.

 

[Joerg]

Hi Marco,

I ask you an advice witch MOS to use for pwm (20Khz) signal for my load
(Peltier cell) on 15V 7A
I run it with IRLZ14 but crashed in 2 weeks.

One of the posters already hinted at it: If you run a PWM to obtain an
average current of 7A (6A fuse?) and the peak current is above the FET's
rating the FET might slowly burn up inside. As Spehro suggested, try a
bigger FET.

As said before, a Peltier might not like to be operated in this way. You
may want to think about a low pass before it, along with a faster PWM to
save on the L and C.

sorry for my english, but I'm foreign

Hey, we are all foreigners. At least in most places of the world.

Ciao, Joerg