mosfet gets "stuck on" when using project volatage but not external voltage for dc-ac inverter

Thread Starter

.frank.

Joined Feb 21, 2016
46
Punch-through...a short from drain to source...sometimes temporary due to a reversal of Vcc.

Add a 10 uf aluminum electrolytic or tantalum capacitor and a ceramic capacitor of 100 nf directly to the power and ground traces right at the PIC to protect it from glitches on the power line or the ground line.
Tried it - no difference
 

Thread Starter

.frank.

Joined Feb 21, 2016
46
So if it de-latches when the gate/source is shorted out, the fet is being driven on by the pic.

How about a couple of flyback diodes 1N4001, one across the inverter drain to v+, and one across the fet D/S terminals, also i would drive the gate with a 10r to 1k resistor, and one across gate/source also.
There is a "built in" diode on the fet, it's more of a byproduct of the design, either way I added a diode and upped the voltage to make sure and compensate for the forward voltage drop on the diode - also no change. As for the resistors, why would I add resistors on the gate - wouldn't that just slow this down?
 

Thread Starter

.frank.

Joined Feb 21, 2016
46
I have no idea how to delete an attachment so check my second attachment which is the current circuit - Results are still the same. The stupid leds works fine but once it cycles to the inverter, it turns on but wont turn back off again. What I find interesting is that the sound detector is also stuck in an "on / I detect sound" mode too.
 

Attachments

takao21203

Joined Apr 28, 2012
3,702
i'd try inline a choke start with common kinds i have for instance 33uh 220uh or 1000uh

you may need a TVS diode parallel to the inverter.

Problems can come from spikes fed back or ultrafast surges in the current.

The inductor can deal with both problems it has some DCR and it doesnt permit infinitely fast current rise.

Some modern converters work in the 1MHz range.

So if you dont get good results from a choke you need a TVS diode,
transient voltage suppressor
 

takao21203

Joined Apr 28, 2012
3,702
ive had a large TL494 power supply the MOSFET would latch if I cranked up output voltage to 60 volts under load (free runnining voltage 105 volts).

Essentially no regulation at all.

Ive swapped with a large BJT problem was cured.
 

Thread Starter

.frank.

Joined Feb 21, 2016
46
i'd try inline a choke start with common kinds i have for instance 33uh 220uh or 1000uh

you may need a TVS diode parallel to the inverter.

Problems can come from spikes fed back or ultrafast surges in the current.

The inductor can deal with both problems it has some DCR and it doesnt permit infinitely fast current rise.

Some modern converters work in the 1MHz range.

So if you dont get good results from a choke you need a TVS diode,
transient voltage suppressor
Can I use my scope to somehow catch/measure this? If so, where would I hook my leads.
 

#12

Joined Nov 30, 2010
18,224
why would I add resistors on the gate - wouldn't that just slow this down?
See post #20. This circuit is not megahertz fast so you don't need to optimize for switching speed. More likely you need to protect the PIC from some unknown glitches feeding back from the load.
 

Dodgydave

Joined Jun 22, 2012
11,285
If you're using a Npn transistor put the load to the Collector, Emitter to ground, and feed the base with a 10k,

as for the scope probe the drain and the gate, with the ground as the earth.
 

Thread Starter

.frank.

Joined Feb 21, 2016
46
See post #20. This circuit is not megahertz fast so you don't need to optimize for switching speed. More likely you need to protect the PIC from some unknown glitches feeding back from the load.
The clock out on the pic shows on my scope to be running at 1MHZ
 

Thread Starter

.frank.

Joined Feb 21, 2016
46
Two pictures from the scope with circuit still as is on post#24

  • Both probes have the common clipped to the earth rail
  • Yellow CH1 is hooked to Pin 5/Gate of the IC/Mosfet
  • Blue CH2 is hooked to the drain of the mosfet

The first picture shows the circuit at rest but switched on and waiting to detect sound
Circuit off or leds on.png



The second picture shows the inverter/fet latched to the on state

Inverter on.png
 

takao21203

Joined Apr 28, 2012
3,702
for 1 MHz you can get away with small electrolytic when you dont need precision electronics.
Even at 20 MHz you could but thats the borderline.
The converter turning on of course is a problem.


You should check (hence blinking LED is employed) if the noise just latches the MOSFET or if the controller is actually resetting.

there are several approaches isolating the controller and improving its stability particulary the crystal osc. You could try and supply from a module the clock pulses, have a visual LED so you see a RESET.

On some pics you can catch a RESET but you may just get a corruption which is worse.

You dont need a resistor on the gate, but youll have a glitch when the IO is undefined, as well, if you dont just draw from a controller IO you should have a small inline resistance (the IO port already does have some)

Try to isolate your problem until you can clearly identify the cause.

Another idea is to softstart the converter because it has output capacitor there will be almost a short for an instant.
 

Dodgydave

Joined Jun 22, 2012
11,285
What clock speed is the pic running at, is it set to 8mhz internal,

From the scope trace on the Drain its pulsing at 12khz, thats the inverter back emf oscillations, try putting a 1uf capacitor across the D/S terminals.
 
Last edited:
Top