Power supply circuit problem

[proid]

I made a regulated supply for my dac, using the schematic of millet minimax ps

To get 9v output, RR1 = 120ohm, RR2 = 681 ohm, RR3 = 1k ohm trimpot, all the rectifier diode are 1n4004, i use a 9v 1000ma ac adapter. My power supply circuit use exactly the millet minimax ps layout:




 
Then when i first plug my adapter, the adapter itself become extremely hot, also the wire from the adapter, the heat sink got no heat at all and when i measure output, it's only 1V DC. After plugging the adapter for about 5 minute, i think it was death, i couldn't measure anything from it output. I need some help, plz!

 

[tangent]

Are you certain you got the pinout on the regulator right? Some datasheets are tricky this way, doing weird stuff like giving the pinout as viewed from the bottom side of the PCB.

 

[KimLaroux]

Did you perform a continuity test with a multimeter? You might have a short circuit somewhere between two traces. Some shorts are so small you can't see them with the naked eye. 
 
Also check the voltages at some points on this circuit, see if it makes sens. Is there a normal DC voltage after the rectifier bridge? Is there a normal DC voltage at the capacitors? At the LM317? Is the reference voltage present between the output and the adjust pins of the LM317?
 
And yes, check if you got the pinout of the regulator right. I just built one myself and got the Out and Adj reversed... it didn't fry the chip, but the output voltage was the same ad the input... Anyways, it's very easy to get the pinout wrong on these. Here's a picture that makes it clear :

 
Note that the tab is tied to the Output. So make sure the eat sink is not touching anything else. I took the good practice of isolating the eat sinks from the device. Makes things safer.

 

[proid]

@Kimlaroux: thanks for your advice, i have just measured between pin out and adj of the lm317 and it was 0.5V so i think the problem i cause by the LM317?

 

[KimLaroux]

Quote:

@Kimlaroux: thanks for your advice, i have just measured between pin out and adj of the lm317 and it was 0.5V so i think the problem i cause by the LM317?

 
Highly improbable. Having only 0.5v is just telling you that something is wrong. Have you measured the voltage at the IN pin? Have you verified that you wired the LM317 correctly? Having the LM317 pinout wrong can explain why you only get 0.5v as Vref.
 
But seriously I'm almost certain you have a short circuit somewhere. The first thing I'd do is make sure the rectifying diode bridge is working properly. Then go up the circuit from there. Performing a continuity test on a PCB is a very important thing to so before powering it up. Heck by the sound of things, I'd bet you have a short circuit even before the rectifying diodes.

 

[tangent]

Quote:

i have just measured between pin out and adj of the lm317 and it was 0.5V so i think the problem i cause by the LM317?

 
It could just mean the regulator is fried now. It doesn't tell you why the regulator got fried.

 

[proid]

I measured the AC1 and AC2 and it was only 4.5v. When the adapter didn't connect to the circuit, it was fine and output 13V AC, i'm so confuse 

 

[proid]

Here is how i solder everything together, hope some one can help me:


The orange point is the lead of the component, blue point is the lead that was grounded.

 

[KimLaroux]

Everything seems to be connected correctly. I cannot see the rings on two of the diodes, have you made sure they were all inserted correctly? Most multimeters have a mode to test diodes. You can use this to check if the current flows in the right direction trough the diode rectifier bridge. It can also detect short circuits between the traces.
 
I can't test for short circuits for you. It's a very elementary thing to do, and I'm also certain that's your problem.

 

[proid]

I have finished my power supply, it's working fine but there is a problem: when plug it to my amp i can hear big hum, it's very annoying. Does anyone know how to fix this?

 

[tangent]

You have a ground loop. The proper way to fix it depends on the circuits in question. Generally speaking, you should Google ground loop breakers. These are sometimes called ground lifters. It can be as simple as adding a 5 W 10 ohm resistor between the two problem ground points.
 
You may also have done the grounding outright improperly. For instance, maybe you have a circuit in the mix that requires an isolated power supply, but you've tied PS ground to a metal case, thereby tying something to PS ground that oughtn't be so tied.

 

[proid]

Quote:

You have a ground loop. The proper way to fix it depends on the circuits in question. Generally speaking, you should Google ground loop breakers. These are sometimes called ground lifters. It can be as simple as adding a 5 W 10 ohm resistor between the two problem ground points.
 
You may also have done the grounding outright improperly. For instance, maybe you have a circuit in the mix that requires an isolated power supply, but you've tied PS ground to a metal case, thereby tying something to PS ground that oughtn't be so tied.

Thanks for your advice. I solder a 10ohm 10w resistor pararrel with a 100pf capacitor to the ground of the power supply board, the hum seem to be solved but now another problem: RR1 = 120ohm, RR2 = 2k2 ohm, 1k trimpot so in theory, DC output mean is about 24V but i only got 18V with the trimpot all the way up and 14V all the way down. Then i relace the trimpot with a 5k trimpot. When i slowly adjust the trimpot and get near 24V, it go up all the way to 35V and when at 35V, i adjust the trimpot again, it go down to 14 or 18V 

 

[tangent]

Several things can cause that:
 

1. bad 5K pot; doesn't explain the 1K, though; power down, drain all power caps, then measure resistance across the trim pot while adjusting it; is it linear?
2. marginal implementation, so the regulator is oscillating; check with AC measurement at regulator output with a good meter (not a $10 cheapie!)
3. carbonized flux, solder whisker, etc.; power down, discharge all caps, and check resistance between all the points around the regulator; if it all looks good, do it again at several trim pot settings