TL494 trouble shooting help

Thread Starter

rapidarc

Joined Jul 5, 2009
3
Hello all,
I am trying to resurrect an old PC AT 200W power supply. It's a Morex MXC 200P.
The problem being the output voltages are a bit low. The 5V line measured 4.5V and the 12V line is about 11V.
I couldn't find a circuit diagram on line but was able to identify the mode of control by examining the traces.
Voltage is regulated primarily by the error amplifier inputs on pin 1 and 2. Pin1 receives feedback from the 5v and 12v outputs, while pin2 receives the 5v reference voltage from pin 14 (through 2x 4.7k resistor divider to 2.5V). When the pin 1 voltage is above 2.5V, switching is paused.
I notice the pin 14 reference voltage is not steady, it fluctuates from 4.1 V to 5V.
I thought the reference voltage should be kept pretty much constant internally at 5V inside the TL494 and should not fluctuate like it does. It would make sense that when the reference voltage is below 5V, the output voltages will also down regulate below the supposed 5v and 12v. Does it smell like a defective TL494?

To prove the point, I even cut the pin 14 right off the chip, and supplied a regulated 5v to the board where the pin 14 attached before the cut. Sadly the chip shut down, with pin 1 measured 0.6v and pin 2 at 2.5V.

I have 2 questions for your opinion.

Firstly, does the pin 14 do more than just outputting a regulated 5V, that the internal of the chip will not work if it is not connected to the external circuit?

Secondly, the fact that the reference voltage from pin 14 is more than 15% off the supposed 5V, does it smell like a bad TL494?

Thanks for reading this.
 

ebp

Joined Feb 8, 2018
2,332
First and most important, unless the circuit uses a base or gate drive transformer, TL494 and everything connected to it should be considered to be at AC mains "hot" voltage. If the main 5 V output is not galvanically isolated from the 494, then there must be isolation between the 494 output and the switches - which is convenient because it makes the circuit much safer to work on - BUT THIS MUST BE CONFIRMED!

It appears that the reference is just an output.

The 494 has to get its power from somewhere, It is likely coming from a standby power supply that is either an independent switcher or a supply using a conventional iron core transformer. If the reference is unstable, I would strongly suspect this supply. There is a small chance that the 494 is powered in some other way, but I doubt it in this case.

It is possible but unlikely that the IC has "gone bad" - it happens, but very rarely.
All electrolytic capacitors are suspect.
 

Thread Starter

rapidarc

Joined Jul 5, 2009
3
Hi ebp, thanks for your safety warning. It does have a gate drive transformer. The power of the 494 comes from the transformer as well, about 20V, and yet the reference voltage pin 14 is only <4.5V. Funny thing is the reference voltage changes from 4.2 to 5.7 randomly during the 10 seconds or so when I was measuring its voltage. If the chip is not faulty, then such voltage fluctuation will have to come from the circuit path that pin 14 is connected to. Without a schematics it is hard to follow, particularly to explain the 5.7V observed. I could see how it might be lower than 5 but not above.....??
I did not check all electrolytic caps, only the output filter caps, and they are all good.
Any other thoughts?
.
 

ebp

Joined Feb 8, 2018
2,332
If the supply to the 494 is good and the reference output is unstable, about the only possibility is that a capacitor on the reference is causing problems or the reference itself is bad. 5.7 is certainly too high even for the sloppy spec of the reference. It is a series reference (i.e. like a three terminal regulator) so it is likely possible to pull it above nominal voltage if something else is sourcing current into it, which would be very odd. Usually in these circuits there is no mechanism, other than IC failure, that would allow sourcing current into the reference.


From schematics in Dodgydave's link
  • Some circuits have a separate free-running switcher that provides the power for the control IC.
  • Some have a separate circuit that appears to provide control power at start-up, but then running control power from the main transformer.
  • Some look like they must actually free-run the main switcher using an auxiliary winding on the base drive transformer. This presumably is designed so that during free running the voltages are low and as the control IC takes over the voltage will rise to set levels and regulate properly. Failure of this sequence might account for what you are seeing, but in that case I would expect the Vcc pin of the controller to unstable too. Any "major" faults I can think of (bad rectifiers, etc) would prevent startup, not just interfere with proper control.
 
Top