TL494: Trouble with TL494 gate drive

Hi Tom,

There is a DTC pin 4 on the device. With 0V on this pin you will get a max duty cycle on the output. And with about DTC = 3.3V you will get no output pulses.
This is sometimes used to set the max duty cycle. You should check the voltage on DTC. It should be less than 3.3 or open circuit.

You should also check the output voltage of the error amplifiers pin 3.
If the voltage on pin 3 is greater than the voltage on pin 5 (the sawtooth) then there will be no output pulses. Check that V(pin3) is less than about 3.3V

Hope this helps

Regards

John

All,

I just checked pin 4 and it shows 17mV.  So no problem there.  Different story on pin 3.  The sawtooth wave is 3.26v, exactly as expected.  But the error amplifier output on pin 3 is 4.8v which clearly exceeds.  The original design of the inverter has an "adjustment" lead that goes through 100k to pin 1.  No idea what sort of voltage is expected there but I think it is to adjust the output and with it the display brightness.  Grounded or tied to VCC makes no difference in the 4.8v level and of course no output.  Tying that ADJ lead to an independent 6v supply (positive ground) raises the voltage on pin 3 to 5.6, so we are going in the wrong direction there.  Switching it to a negative ground or referencing to VCC does nothing. There is also an "N/L" lead that seems to want to be tied to VCC for any power at all to be sent to the TL494; whether that is an intended connection or not we are looking at.  I should add that for a brief moment in all this connection exercise the pin 3 voltage momentarily dropped to 2.6v which kicked everything into operation and I read over 400VAC on the output.  But it was a brief event.

A little additional information:  The "N/L" connection as we suspected appears to be a simple enable connection.  About 1.5V on that line snaps all the voltages right up to full VCC.  And it doesn't seem to make a lot of difference whether that is tied to full VCC, or the lesser amount.  When I put a scope on either of the TL494's emitter outputs to the bridge, on power up, there is about one second of full output, then it goes to right back to zero.  So that explains the brief voltage excursion.  After that brief moment something in the error circuitry kicks in and shuts it down so once powered up nothing seems to allow the chip to restart operation.  Nothing seem to drop the 4.8v I am reading on pin 3 that is stopping things.

Rather inglorious end to this exercise.  I could not do anything to the '494 to make the pin 3 voltage less than the sawtooth amplitude, so no output.  No idea how they wired it or what sort of voltage could be ever be attached to pin 1 to drop the error amplifier voltage.  So I attached a 10k pot to pin 3 and ground and just dialed it down to about 300 ohms and sure enough the voltage finally dropped.  And with that the pulse output came back on- pulse width varying smoothly with the pot- and I was reading about 600 volts AC across the transformer secondary.  Kind of a sloppy solution, but it was working until there came a puff of smoke and blown fuse- no more voltage.  So it looks like pin 1 does something after all and pin 3 doesn't like being grounded externally as a solution.

Thanks to all who have read this.  Time for a new design.