Is this idea possible

[H2OWiz]

I want to be able to hook an IRFZ34N MOSFET to a 12 volt battery in a car and use it as a 30 amp pulse width modulator to drive an electrolytic cell. I am using 2, 555 timer chips to create the frequency and pulse control. I take pin 3 of the second 555 and connect it directly to the gate of the MOSFET. The Drain of the MOSFET goes to one side of the electrolyzer and the source goes to ground. The other side of the electrolyzer goes to positive of the battery. This will work for a few seconds then the MOSFET burns out. I am using a very large heat sink. 6" x 6" x 2". Any ideas ? Thanks

 

[audioguru]

I couldn't find a datasheet for an IFR34ZN Mosfet.

What limits the current to 30A? Maybe the current is hundreds of Amps.

 

[OutToLunch]

you don't have any idea how much current is passing through the MOSFET when it is turned on, so how would you know what the average current through your electrolyzer is?

I would suggest you move away from blind PWM generation and use a control method that will turn the FET off when the current through it reaches a certain level (peak current control). With a fixed frequency, you will be able to at least approximate the average current (I do not know what the electrical characteristic of an electrolyzer are, so I could not guess on what the current waveform will look like)

 

[H2OWiz]

I thought by controlling the voltage to the gate of the MOSFET this in turn controls or limits the amps. I do have an amp meter in circuit so i can watch the draw. The voltage control to the MOSFET is by PWM.

 

[audioguru]

PWM does not control voltage. It controls the duty-cycle.

PWM is supposed to fully turn on the Mosfet then fully turn it off. This happens at a high frequency. The Mosfet doesn't heat much and hardly any power is wasted.

If the PWM sets the duty-cycle at 50% then the load gets an average current of half the maximum.

Nothing limits the max current except the resistance of the load and its wiring.

You forgot to correct the Mosfet's part number and forgot to attach your schematic.

 

[H2OWiz]

Here is the schematic on my idea.**broken link removed**

 

[Nigel Goodwin]

Here is the schematic on my idea.**broken link removed**

And what use is a link to the C:\Temp directory on YOUR computer?.

 

[H2OWiz]

Can not figure out how to post the schematic. Can I post it through a reply or do I need to go back to the origional message and post it there?

 

[Nigel Goodwin]

Click on 'Go Advanced' and then 'Manage Attachments'.

 

[H2OWiz]

Thank you for the help about posting. The schematic is now uploaded.

 

[audioguru]

You attached a breadboard layout, not a schematic.
It shows an IRFZ34 Mosfet which is not what you said it was.

With 30A going through it, the breadboard will have a high voltage drop so the source voltage of the Mosfet approaches its gate voltage and it doesn't turn on hard enough to stay cool. I am surprised that the breadboard doesn't smoke and burn.

 

[H2OWiz]

I used what i had for the schematic as it was already drawn. The power MOSFET actually sits on a 6" x 6" x 2" heat sink and is off the breadboard.

 

[OutToLunch]

an ammeter is not what you need to control the current through the MOSFET. The ammeter will only tell you the average current. A current sensing resistor tied between the source of the FET and Ground can be used to feedback instantaneous current information to the controller.

If you are measuring 30A via the ammeter before everything smokes, you still have no clue what kind of current is going through the MOSFET. What type of load is an electrolyzer? Is it strictly resistive? Or does it have a significant amount of inductance to it?

Audioguru has a good point about the breadboard as well. Another point to mention with all these long thin wires being used is the inductive kicks that are playing their role in killing the FET as well.

If you want to process power, you need to use a board with lots of copper for the current to flow through.

 

[philba]

This is probably a waste of time but...

Go the current sense approach as suggested earlier. Use a low value resistor to sense the current and feed that into a comparator. The voltage drop across the resistor is proportional to the current through the resistor. The other comparator input - reference voltage - should be variable (use a potentiometer to set). the output of the comparator is used to turn the FET on/off. Set it up so the comparator is on when sensed voltage is below the reference voltage and off when the opposite. Changing the reference voltage will set the trigger current level. Trigger current will be approximately 1/2 average current.

edit: I'd build in some hysteresis. check out schmitt triggers to understand how that works.

 

[H2OWiz]

I think the electrolyzer is simply a water capacitor. I measured the resistance across it and it measured 177 with the meter set to 2000k scale. It also measured 17.6 with the meter set to 200k scale. The resistance did start lower and raised as it charged. I belive the resistance will change with the quality of water. I am using well water but am sure lake water would measure differently. Also I am electrically challenged so I am not sure what you mean by controller. I would need to know connect point A to point B. If this idea can be made to work what would I need to do to make it work. If not should I start looking for other solutions? If PWM can not be used to control the MOSFETs current output can line 3 form the 555 to the gate also be voltage controled. I need the variable PWM to study the effects on gas production.