[SOLVED]Split Power Supply - trouble adjusting positive output

I am building a +/- 5V power supply for a project. The circuit is a variation of the standards found in the datasheet. It was designed using Eagle CAD, DRC and ERC checks passed, the PCB manufactured and I have now populated the first PCB.

Using trimpot R3, I have had no trouble adjusting the -5V output to be correct. However I can't get the positive voltage to be anything less than 7.7V using trimpot R1 (it does adjust, just not enough). I have check the board for solder bridges and defects but can't see any.

Being a relative newbie on the hardware side, I would appreciate any pointers on how to 'debug' this situation. Logically the circuits are mirror images and if one works, so should the other.

Attached are the circuit and PCB design and a photo of the board.
Any thoughts?

Split Supply Circuit & PCB.pdf (122 KB)

You say adjustment isn't enough.
Some details of what that adjustment figures are might help identify the problem, otherwise we are just guessing.
My guess is that your R2 is not the 240ohms indicated. I therefore suggest you check its value using a test meter.

Hi, I agree with jackrae, R2 in your case is 220R but I don't think that should be a problem, but I'd measure it just in case.
Also check the orientation of D5 and D6, I notice that D5 on the schematic is D6 on the PCB, similarly D6 is D5.
Otherwise try using 330R for R2 incase this LM317 is a little off spec.

With the pot completely to gnd, you should have the 1.25V ouput that the LM317 regulates to.

Also measure the voltage across R4 it should be 1.25v, same for R2 but opposite polarity.

By the way nice layout, easy to follow and uncluttered.

Tom...... :slight_smile:

R2 & R4 are both measured 220 Ohm.

Good pick up on the numbering of the diodes. It was a screen printing text issue. The direction of the diodes is correct as far as I can see (assuming the diode body labelling and the bar on the schematic are congruent, as is my understanding). D7 and D8 had the same issue of labelling. Both now corrected, thanks.

The voltage across R4 is -1.26 volts and the voltage across R2 is over 7V, in fact the same as the input voltage. This leads me to think there is an issue with the diodes (D5?). The circuit should work without the diodes, so will try to disconnect D5/D6 and see if that makes the situation better.

Thanks for getting me off to a good start.

Removed the diodes from the circuit - no change. Replaced the LM317 - now it works fine.

So the question is whether the LM317 was faulty or whether it was made faulty by the diodes bypassing it. Can someone please verify that the direction for the diode current flow is correct in the schematic?

If R2 is dropping 7 volts then either C3 is defective (short circuit) or your trimmer pot isn't 5k
Again I ask, what range of voltage change do you get from the trimmer pot.

Replacing the LM317, voltage across R2 is now 1.25 volts. Trimpot now works as expected. Output can be regulated to 5V easy.

Question is whether to reconnect the diodes in the same configuration?

D5 and D7 directions are correct. They are there to prevent excessive reverse voltage appearing across the regulator chips when the power is turned off.