JSR04, +12V, +5V (from the negative side)

[builder brad]

I have just tried turning the wiper P3 down and have managed to reduce the output down to 13.6v which remains stable with the 47 ohm resistor on the output!

i have 470R on r47, 1k5 on r43, and 1k on r41 and r45

this 1/2 of the power supply will be used to replace a SLA battery so 13 volts should not be a problem, out of interest if I wanted to reduce the output to 12 volts which value/ resistors should I change?

RE. 5 volt output (from altered 12 neg side)

what do I need to do to get the 5v section of this PS working, I tried a 2k2 resistor across r32 and this did not change the output voltage

[peranders]

The gain of the LM431 should be 6/2.5 = 2.4

This means R43/R45 = 1.4 If you also use P3 and 41 (which is option) you must include them in "R45".

One way to determine that the LM431 is working properly is to measure the voltage between anode and ref, should be very near 2.5 volts.

Do you have a stable output voltage now but at a bit more than 12 volts?

Is the output voltage at pin 6 around 5-9 volts?

Can you tell me how the 5 volts side is built?

Can you tell me which voltages you have?

[builder brad]

everything looks good on the 12v + side

I have 2.5 v across the input and reference on the LM431

I do not see a change to the output voltage with a 47 ohm resistor connected across the outputs

Pin 6 measures 6.5 volts

The output does drop from 13.6 volts to just under 12 volts when I put 22 ohms across the terminals - is this correct?

Re the 5v 1/2 of the board

The components have all been reversed with the exception of the pre regulator which is an exact mirror of the 12v positive side including LM317T pre regulator and pcb re-working.

[peranders]

If you load under the current limitation limit and you have a constant voltage, then you have a working regulator.

if you have 1 ohms as shunt you'll get approx. 600 mA out.

For the 5 volt section, make sure your LM431 is properly flipped and when it works you will have 2.5 volts across REF and ANODE.

Make sure all electrolitic caps are turned right.

Check also the opamp.

If you use LM317 bear in mind that LM337 isn't pin kompatible, check how the difference is.

[builder brad]

Re The 5v section

Hi P-A,

this is getting really confusing...

i have taken a lot of time to check the pcb and components and found 1 error with the designations on the pcb screen print. R44 and R46 are transposed which meant that the ref out from the LM431 was being taken to ground via the 470 ohm resistor. I have now corrected for this but have not observed any changes to the output voltage. The voltages measured on the LM431 are 1.37v on both K and ref.

Would the regulator have been damaged by taking the ref output to ground via the 470 ohm resistor?

What do I need to check now?

[builder brad]

I have also just noticed that R12 gets hot very quickly - definately too hot to touch and that is without a load! R12 drops 2.2 volts - does this help diagnose the problem?

[peranders]

Have you GND on pin 7 and minus something on pin 4?

Changed all transistors?

LM337 patched due to different pinning?

[builder brad]

Hi P-A

The op amp has been rewired so that pin 7 is + supply, directly from R20 and pin 4 is gnd. I have checked both of these and they are OK.

I have used an LM317T as the preregulator and have carefully reworked the traces so that I have an exact replica of the 12v positive 1/2 of the board. I am getting a good 9-10v before the pre-regulator.

I have reversed all of the elec capacitors and the transistors and ICs except the LM431 which required some additional work to get the connections as per your sugestions. The LM431 has the centre pin to ground and the reference output linked back to the Cathode via R44 @ 470 ohms.

HELP!

[peranders]

9 V in is too little if you have a preregulator. 10 V or more is better.

I'm afraid I have done the sketch a bit wrong. It may work with R20 used but it will work better with R22. Exclude C26 (or have it), put a wire in R18 and D10.

How much voltage do you have out?

[builder brad]

hi P-A

I have altered the supply for the op-amp which is now taken from the post pre regulator output via r22 @ 47 ohms. the op amp is getting 1.9v on pin 7 now.

The dc output is now 1.07 volts!!!


I am measuring 7 volts before R12, which drops 2.4 volts across R12! - R11 on the 12 volt 1/2 of the board is getting 26 volts and only drops 0.4v - R12 gets very hot very quickly - could this help diagnose the problem, surely R12 should not do this when there is no loading on the power supply.

I have again double checked all of the components on the pcb. IC2 is a re wired LM317T with R14 and R16 values of 1k0. the following parts have simply been turned around on the pcb to work with the + supply T2,T8,T10, the LM431/ IC10 has been reworked with the centre pin to ground and the reference output tied back to the Cathode via a 470 ohn resistor. I have used a BD139 for T4 as the pining is identical to the + half of the supply and the power supply for the op-amp has been reversed.

The led was illuminated but this has now stopped working as well.

do you have any more ideas, I keep checking the pcb, looking for any mistakes ect, but have just come to a complete standstill.

I do wonder about the minimum voltage for the pre regulator, but believe that the 0-9-18 transformer used should be giving at least 10 volts on the 1st winding as the 2 connected in series are producing 26v.

[peranders]

You might exclude or reduce R12 due to low raw voltage.

In order to get on track:

Use R20, short R18, short D10. Omit R22

Remove DZ1 and solder in a PNP like T10. Connect a pot to the base in series with 1 k (protection) so you can get a variable voltage. You can use the raw voltage to drive the pot.

Adjust the base voltage between 0-5 volts. Do you get 0-5 V out? If you do get a variable voltage, this section is working. If the opamp is still consuming more current than expected, remove T10. Better now?

If the power section is working, how much voltage do you have at the opamp inputs? Around 2.5 volts? 0.5 times the output at pin 2 (feedback) and exactly 2.5 volts at pin 3 (reference voltage).

H2 must be lit when the output section is working.

[builder brad]

Hi P-A,

thankyou,

If you remember I have fitted op-amp ic connectors, so it is easy for me to swap or remove the op-amps. When I remove the op-amp on the 5v section I get 8.6 volts out and the input voltage increases to 12v. with 0.12v drop across R12. I have also tried to power the 5v op amp with the supply from the 12v side which is over 13v and only observed a 0.4 volt increase to the out put and one single flash from the led.

Before I do as you sugest is there anything else worth trying, I think that the problem may not be with the op amps power consumption because there was no improvement when using the 12v op amp supply.

Is it worth making an exact copy of the 12v working side ie. resistor values ect. and get that working first, and then make the sdjustments to get 5cv dc out?

[peranders]

Of cource you can make a 12 volt section but since you have trouble it's important to rule out the different sections. When you are troubleshooting you must have a strategy.

[builder brad]

Hi P-A

Finally got to around to having a closer look at this project again......

The faulty component was the T4 transistor and I now have 4.13 volts regulated output which drops to 2.7 volts when I connect a 10 watt 47 ohm load resistor. The H2 LED is not illuminated and only has .3 volts across its terminals. R32 is supplying 2.61 volts to the LM431 regulator.

I just need that final push to get this PS up and running.

Cheers

Brad

[peranders]

I'll get back to you in two weeks. I'm on vacation.