Testing, Troubleshooting an Inverter Circuit - Discussion

Continued from HERE

Hi Swagatam, one more question. I don't have any 10K Dual Pot, but I do have a 50K Dual Pot, is that OK??


Hi Ali,

Just connect 22k fixed resistors across each transistor base and ground and then you may connect the 50K Dual pot leads across these resistors, that should imitate a 10k pot variations quite closely.

Regards.

Hi Swagatam, so just to be clear, you want me to remove the 10K resistors connected to the Base of each Transistor and replace with 22K resistors and connect another two 22K resistors from the Emitters  to Ground . Then connecting the 50K Dual Pot across those connections. Is this correct?

Sincerely Ali

Hi Ali,

The 10K resistors which are connected from the transistor base to the diode junctions should not be disturbed.

If you have connected 10K resistors from base to ground, then replace them with 22K resistors and connect the POT terminals across this resistor.

Regards.

Swagatam, was I supposed to have 10K resistors from base to ground ?? Because I don't .

Sincerely Ali

Swagatam, when I have soldered the Pot into place, I  want to make sure I've soldered it correctly before I place it into the Inverter.  When I power up the circuit, and I turn the Pot, should the voltage at the Output of TR1 & TR2 be going Up or Down?

in a earlier e-mail I told you that the Dual Pot I had was 50K Ohms, I was mistaken. It is a 15K ohms Dual Pot. Do I still need the 22K ohms resistors, or a different value ?

Thanks

Hi Ali,

Since you say that with my PWM circuit the current consumption of the inverter shoots to a dangerous 14Amps, therefore I just thought, probably we can solve this by reducing the drive voltage to the gates of the mosfets.
This can be simply done by controlling the base voltage of the buffer transistors. And that is exactly what we are trying to do by connecting a variable resistor across the base and the ground of each transistor.

A 15K pot will do nicely here, so now you don't need anything across the base and the ground except the pots.

As discussed earlier, use lamp and ammeter set-up with the inverter and adjust the 15K pot until the current consumption becomes equal to your square wave design.

After this, hopefully things should get simpler.

Hi Swagatam, I was doing a simple power calculation, my original square wave was pulling 2 amps at 12V so that comes to 24W.

Does this indicate the total power output of the inverter or just what the inverter uses to run at idle.?

Sincerely Ali

Hi Ali,

Surely it is the power consumed by the inverter with no loads (idle) and will proportionately increase as its output gets loaded.

Regards. 

Hi Swagatam, I did what you said and reinstalled the PWM circuit  with Pot modification, annnnddd Nothing blew up and its running.

  We're getting closer !!!

Sincerely Ali

Hi Ali,

I hope this time it starts working.
Regards.

Modified sine wave inverter circuit

Hi Swagatam, I performed that test with the headlight, Amp meter and PWM circuit, it was drawing 2.1 Amps, but when I tried to adjust the Pot , nothing really happened ,the Amps did Not increase or decrease, only when i turned the Pot almost all the way left, it would kill the wave form ,literly stopping the inverter, then turning the Pot back to the right the wave form would come back and you can hear the inverter run. The whole time the Headlight did not change in brightness.

I was thinking what if the Two PWM pulses from the 4017's were slightly overlapping. Would that not cause a problem?

My laptop Oscilloscope is only Software ,with a homemade probe. Unfortunately most laptop Mic's Input is in Mono, one channel. So I can't test both wave forms at the same time to see if there overlapping.

Another observation I made, I've been studying a few other circuits that are all using the 4017 Chip. I've noticed that none of them have those buffer transistors that are used in  my original square wave circuit. What if I remove them from the circuit altogether. I've noticed that in all the other circuits the Clock Input goes directly to the 4017, then out of the two Outputs that they are using , directly to the gate of the mosfet through a small resistor.

I'm grasping for straws. lol

Sincerely Ali

Hi Ali

As long as mosfets are involved I won't be able to help you. I have made good inverters, but using transistors, so I am totally out of clue.

The LED test that we performed earlier proved that the two ICs are not overlapping, and that's exactly why we did the test.

So friend now it's totally up to you, and now it has to be completed only through some trial and error.

You may try different configurations taking the help of other circuits but always keep the series bulb connected, it will prevent the mosfets from blowing off if something goes wrong.

Regards.

Hi Swagatam, I placed 22K resistors across the base and ground of both transistors, this time when I did the Headlight, Ampmeter to Inverter in series Draw test, I got a reading of 1.80 A, Exactly the same as the Square Wave circuit, it drew 1.82A 

Now that sounds good, but when I checked the Wave form at the Gate of both banks of Mosfets, the wave form on the Gate that is driven by IC2 is corrupt, it is missing a pulse, I've sent two screen shots. The screen shot that is missing a pulse comes from the Gate of the Mosfet driven by IC2 and the other screen shot that is not missing a pulse is comes from the Gate of Mosfet driven by IC1.

Curious?

What do you think?

Sincerely Ali

Hi Ali,

You will also have to check the waveform at the diode junctions to make sure whether its the fault of the PWM section or is it happening due to the lamp and the base resistor.

I guess it's because of the lamp that's restricting the formation of the waveform.

We can perform one more simple test by reducing the output load to the Mosfets. For this we don't need to connect a lamp in series.

Replace the present transformer with a much lower power transformer, say a 12-012/1 Amp transformer, after this you may proceed with your waveform checks and also by connecting smaller loads at the output (like a 10watt 120V lamp) and see the results.

A smaller transformer will definitely stop the mosfets from blowing off and also give us a clue regarding a transformer mismatch in the present design.

Remember we don't need a headlight lamp in the above modified test design.

Regards.

Hi Swagatam, I checked the wave form at the diode junctions and the wave form is good there. I have a small transformer out of an old 12V 2Amp battery charger ( not sure if this trans is OK to use for test) , and I connected it altogether with No headlight in series. I turn on the circuit and it runs but when I check the frequency at the Gate of Mosfet #2 driven by IC2 it reads over 360Hz, and the same frequency on the Gate of Mosfet #1 but the wave forms are different, not the same. I,ve sent you two screen shots. Each shot is labelled accordingly. This has to be happening because of this transformer , I think. What do you think ?

The screen shots I sent you the other day were much better. They were on the original transformer, with the headlight in series.

I am going to try the PWM circuit in my Inverter setup  without the Headlight in series, with a couple of sacraficial Mosfets I have, that I know work with the Original settup. I'll let you know my results.

Sincerely Ali

Hi Ali,

OK let's see what happens...........I'm totally helpless without being able to check the practical set-up myself.

I was just thinking to engage the couple of idle gates with the diode junctions and connect their outputs with the transistor base.

This will nicely buffer the PWM pulses and stop them from disorientations.

But then the diode junctions will have to be connected with a resistor of any value across them and ground.

Regards.

Hi Swagatam, I tried what I said and it worked for three seconds then a mosfet blew up. No Go.

When you said "I was just thinking to engage the couple of idle gates with the diode junctions " which Idle Gates were you talking about? Unused gates on the 4017's ?? Not sure what you mean.

Sincerely Ali

Hi Ali,

I think three gates from IC 4069 are lying idle, we can pick two of them and use as buffers at the outputs of the diode junctions.

I mean the input of each gate will go to the respective diode junctions and their outputs will go to the transistor bases.

But very importantly the diode junctions will need to be grounded through a resistor, any value will do, preferably above 1K.

Regards.

Hi Swagatam, I'm not sure of how you want me to make the connections. I've sent a diagram depicting the connections you want , are they connected OK ???

Sincerely Ali

Hi Ali,

Please check the attached file.

Regards.

Hi Swag, I had two 22k resistors across the base to ground on both transistors, should I remove them or keep them in the circuit.?

Hi Ali,

With the present configuration using gates at the diode junctions, 22k resistors can be removed, they are meaningless now.
Do everything just as shown in the diagram, which I gave you yesterday (all 10k resistors are important).

Regards.

With HeadLight

Without Headlight

Swagatam here are the screen shots I took of the waveform at the two stages, with headlight and without.

Hi Swagatam, here the revised circuit schematics for our circuit. This schematic begins from the Two Diode banks 1 and 2 of our circuit.

 Ignore the part numbers in the schematic they are not what I used.

 R1,2,3,4 are 1K Ohms resistors 1/4 W , R5,6,7,8 are 21K Ohms resistors 1/4 W, and Mosfets Q1,2,3,4 are IRFZ44 , the Transformer is a Center Tap Transformer 12-0-12 on the Primary side.

 This setup seems to work  with four Mosfets run in 2 banks of 2 mosfets in parallel.

P.S. An observation I made was in trying to rewinding a microwave Transformer, unsuccessfully. When I rewond the New primary side for 12-0-12 the Voltage Output on the now secondary side was not what I expected, It was 55V AC, not the 120V AC I expected.

The reason for this ,I think .Correct me if I'm wrong but on this microwave Transformer I have, has it's coils wound  in Parallel , side by side. The other Two working Transformers I have, have it's Primary coil wrapped over top the Secondary Output coil. Swagatam is my observation valid about the transformer coils configuration ?

 

Tweet