Need help with bench PS

[Zeebit]

I just realized pin 7 of U4 wasn't connected to Vout of the 317.

I'll fix this in a sec and see what happens.


*******************************
Finally, it works now! Woo! What a stupid mistake. I hate myself. But hey, they say the way to learning electronics is by failing. Sounds reasonable but it can be frustrating.

Just one problem left. Even though I've changed R9 to 2.2K, the CV led doesn't light up but the CC led does work. I hope nothing goes wrong when I try to fix this.
*******************************

Got the LEDs to light up properly now.

I finally have a working DIY bench supply now. Thanks for all the help guys. I appreciate all the help
*******************************

Edit: Something happened again. I going to try charging a li-ion battery and I was setting the current with my multimeter connected to the output because the panel ammeter wasn't accurate. I set my multimeter to mA mode and I accidentally plugged the lead in the Amps socket. I turned the current knob and all of a sudden the output was gone. I discovered that the 4 amp fuse on the main transformer winding blew.

I don't know why it blew because it was supposed to limit the current to what was set on the pot. A few moment before I was able to set the current to max and my multimeter reads 2.9A.

Does it have something to do with the multimeter set to mA and the lead connected to the Amps socket? But I would still go into CC mode so that's got me puzzled.

 

[Cyparagon]

The LED problem should be an easy fix. I'll ask for voltages again and hope you don't ignore me this time. While in constant voltage mode, What is the voltage across R9, D7, D8, and D4?

The fuse is on the primary side, yes?

Wrong DMM socket shouldn't affect your supply.

 

[Zeebit]

The CV led was fixed. It wasn't properly grounded.

I did try to measure the op amp voltages but when doing that, I discovered that it wasn't powered with the positive supply so I fixed that. I will measure it again tonight.

I have fuses on the primary and secondary side. The only fuse that blew twice already is the one on the main secondary winding. I haven't yet figured out why.

 

[Zeebit]

Cyparagon, I can't measure the voltages right now. I was tidying up the wires including the ones on the transformer. I accidentally shorted one winding and the fuse on the pri side blew. I don't have extra fuses so it'll have to wait till tomorrow.

 

[Zeebit]

Power has been out all day so this will have to wait again until tomorrow.

 

[Zeebit]

4th consecutive post.

I changed the primary side fuse and left all the secondary wires floating. The fuse did not blow when I switched it on but it was making a loud hum.

I've googled it and some say its because there is no load. But it was not humming when I first tested the secondary voltages with no load when I first got this transformer.

I hesitant to test the secondary voltages with all this humming. Is this normal?

Edit: Wrong connections caused the humming. Lesson learned: always triple check every connection before powering. Lucky the transformer didn't cook.

 

[Zeebit]

Here are the voltages of U4. Both adjust pots are set to minimum. The output voltage is 1.269V and the minimum current is 15mA.

Pin 1 --> (-4.55V)
note: CC mode is triggered when test lead contacts this pin 1. When current limit is raised to 49mA the supply stays is CV mode and the voltage at this pin is now (-5.46V).
Pin 2 --> 1.269V
Pin 3 --> 1.269V
Pin 4 --> (-6.04V)
Pin 5 --> (-4.86V)
Pin 6 --> 0.771V
Pin 7 --> 1.274V
Pin 8 --> 0.767V


I also wanted to measure the voltages when the supply is in CC mode. The adjustment pots are still at minimum and I shorted the output with a piece of wire and then, POOF!. The 4A fuse on the main secondary winding blew again.

When I connect my multimeter set on Amps mode to the output, I can raise and lower the current limit all I wan't without the fuse popping. What's happening here? :thinking:

Edit: I've replaced the fuse 4A fuse with the same rating and it blew again while I was setting the current with my multimeter connected.

 

[Cyparagon]

Have a look at this diagram and read up a bit more on rectification. If your average current is 3A, peak current is going to be substantially higher than your 4A fuse will handle.

Remove the secondary fuse from the circuit - the primary fuse should be sufficient.

I've never seen a fuse on the secondary in any consumer gear, probably for this reason.

 

[Zeebit]

Ok I'll try that. I've never thought about that. My only aim was to protect the secondary windings in case something goes wrong.

 

[Cyparagon]

If there's a short on the secondary, the primary current will go way up and blow the primary fuse.

 

[Zeebit]

Not quite in my case. There are two (0-20V) windings and two (0-12V) windings. The 20V are paralleled and used in the main rectifier and the 12V are in series and used as center tap for the aux voltages.

The part where I said the transformer was humming, one of the 20V windings is actually shorted and the rest are floating. The 1A fuse on the primary did not blow.

 

[Zeebit]

I removed the fuse but I wanted to see the current draw so I put my multimeter where the fuse was.

I switched the supply on with the adjust pots set to minimum and nothing is connected to output. It displayed the correct voltage (1.27V) and the CV led was on. As soon as I turned one pot (forgot which one) it did some crazy things again. It went to CC and the voltage was around 21V or so. I shut it off, zeroed the pots, turned it on again and it was the same and then a few seconds later, the voltage went zero and the LEDs are not lit.

During this time the multimeter did not read anything because I forgot to set it to AC. I checked the 10A fuse and its fine.

Then I found out that the pass transistor is dead. The base-emitter junction is open and the base-collector junction is shorted. There is no way that the transistor could have shorted because it was mounted on an anodized heatsink which was then mounted to the back of the chassis with rubber pads and insulated bolts. The two leads were also heatshrunked.

This is rather frustrating. I'm thinking about just getting a cheapie from eBay or use djQUAN's board layout.

 

[Zeebit]

I shouldn't be bumping this in just a few hours but here's an update:

I've replaced the pass transistor and it worked fine again. With both pots at minimum, I connected a 3W LED to the output and it worked fine. I was able to adjust the voltage and current limit without anything blowing up.

However, here are two problems that I saw:
1. The CC and CV LEDs aren't properly lighting up. When it goes to CC mode, the CC led would light up but the CV led was still partly lit. It would only completely turn off if I raise the current limit to a certain level.

2. The supply does not like a dead short across the output. After successfully driving the LED, I shorted the output (pots at min)and the pass transistor died again. This time, its the emitter and collector that are shorted.

 

[djQUAN]

You still did not post pics of the setup. Without it I can't tell which is causing your problem since you are probably having oscillation/instability issues due to long wiring or similar causes.

Are you sure the negative supply for the op amp is connected to the negative supply and not the 0V common? It appears the op amp can't sink below GND that's why current limiting won't work at very low output voltages.

 

[Zeebit]

Sorry. I forgot to post the pics because I was busy this week so I'll do it tonight. And yes, the negative supply pin is connected to -6V.

 

[Zeebit]

Heatsink and transistor are isolated from the chassis

-----------------------------------------------------------------------------------------------------------------------------

-----------------------------------------------------------------------------------------------------------------------------

Main transformer and a smaller one(rectified) powering the fan and the ammeter

Output terminals. Green wire goes to the voltmeter

-----------------------------------------------------------------------------------------------------------------------------

Main power rail bridge rectifier mounted on chassis to keep the heat away from the filter caps. The brass standoffs have a washer on top to isolate it from the pcb.

-----------------------------------------------------------------------------------------------------------------------------

Everything connected in various angles.

-----------------------------------------------------------------------------------------------------------------------------

-----------------------------------------------------------------------------------------------------------------------------
The resistors on the lower left are the shunt (R3 to R7). I've put it there to keep the heat away from the filter caps.

-----------------------------------------------------------------------------------------------------------------------------

Excuse the fugly pcb. These are enamel coated wires