Voltage regulator circuit that doesn't regulate voltage. Am I doing it wrong?

[evovi]

I'm currently building myself a 48v phantom power supply for a microphone. I've been using **broken link removed** project posted online for both the powersupply and microphone circuits.

So far I've built this circuit below:

The problem is the voltage regulator is doing absolutely nothing to control the voltage. Obviously the key to this circuit working is the 24v zener diode (D3) working as the reference voltage, however when I take a volt reading across the zener it is not 24v. Wondering if I had a faulty diode I removed it from the board and tested with a 1k resistor with a power supply set to 30vdc and found the zener is doing it's job and dropping the supplied 30vdc to 24vdc. I have replaced the 2 transistors and all the resistors and checked their values. All checked out okay. I even went to the bother of scratch building the voltage regulating components of the schematic and had exactly the same issue. I've triple checked all my connections and made sure nothing is wired incorrectly. I have a 'walkaround' if I can't find a solution to this problem. I'll just use a step up converter using the same transformer and a bridge+cap combo to get around this problem. However that's only a backup plan as I would like to get this circuit working if possible?


While on the subject of reviewing schematics does anyone see a problem with this schematic below before I build it?

 

[ronsimpson]

I will assume the output is low to zero. What is the output voltage?
What is the voltage from the bottom of C4 to the top of C3? 70V?
Voltage across C5?
Voltage on D3? (not 24 does not help much) Probably very low.

 

[evovi]

with 30vac in the voltage after the voltage doubler ( between c4 & c3) is 89vdc. The output after the voltage regulator is around 77vdc. At the moment I've isolated the voltage regulator section of the circuit and powering it with a 54vdc power supply to limit the voltage to a safe level for trouble shooting. With a input of 54vdc the voltage across D3 is 32vdc.

 

[ronsimpson]

the voltage across D3 is 32vdc.

Is D3 HOT?

 

[evovi]

Negative, after 60secs of run time the diode is still room temperature.

 

[ronsimpson]

A 24V diode can not have 32 volts across it, unless the current is very high, which would cause it to explode with heat.
What is the part number?
Do you have two 12V diodes that could be used as a test? Do you have any Zener diodes that are near 24V?

 

[evovi]

Perhaps I have a faulty diode? Though it's odd it works when tested in series with a 1kohm resistor @30vdc. Unfortunately I live out 30mins out from the nearest town and won't be able to get my hands on another 24v zener until friday. I only have 10v zeners on hand at the moment. If I put a 10v zener @ D3 I should get 20vdc out which would prove if the circuit is working correctly?

EDIT: I really appreciate your help btw!!

 

[ronsimpson]

one diode, 10V=20V
two diodes in series, 20V=40V (close)
---edited----
I am collecting "likes". If we get this thing working......

 

[Mikebits]

Is your ground really at ground.

 

[evovi]

Okay so this is a more complex answer than I would have imagined. So i replaced the 24v zener with a single 10v zener. Powered the circuit and was pleased when I saw 9.90vdc across the 10v zener BUT the output voltage of the circuit still remained unchanged. So I then installed two 10v zeners in series. And again was pleased to see 20vdc across the 2 zeners in series but the output voltage still did not change. So I figured if the 24v zener was infact faulty perhaps transistors @ Q2 and Q3 are now damaged? So I replaced them with new BC546's and nope the output voltage still hasn't changed. But here is the kicker....... Now I'm only getting 12vdc across the two 10v zeners (before I replaced Q2 & Q3 it was 20vdc)

 

[Mikebits]

I had to look at the circuit, so I added some notes to explain its operation, have a look and see if it makes sense. Sounds like Ron was helping ya so I don't want to step on his toes. But I made this already so, I don't want it to be for nothing so im posting it. From the symptoms it sounds like the circuit is oscillating. See what Ron says.

 

[evovi]

From the brief time I've known Ron he seems like a really good chap So I'm sure he won't mind? Many thanks to you both for spending the time to figure out what going on with my circuit! Assuming the circuit is indeed osculating. Do you have any suggestions on a modification to the schematic to prevent this from happening?

 

[Mikebits]

Can you take a photo of the circuit you built and post (what I mean is the actual physical board, not schematic.? Your layout may be causing some trouble.

 

[evovi]

This is not the original layout I intended to use. This is the test circuit I built to troubleshoot the voltage problem. I laid the components out in a way that was very similar to the schematic for clarity for myself and others looking at the circuit (not very proud of how it looks, but then it's only for testing purposes and not for intended use in the final project)

 

[Mikebits]

All those long leads could cause trouble, and you want to keep your ground wires short as possible and thick awg. You might try placing a .01uf cap from ground to various points around those transistors, and see if it has any affect. I am assuming you do not have a scope, but if you do then see if you see some oscillations. Considering where you got the circuit I would assume the design is ok. Getting late here, I will check back later.

 

[evovi]

Thanks for your suggestions! First thing tomorrow I'll try placing a cap around the transistors. I have a scope, it's nothing fancy but I'll hook it up and see what's going on. I'll report back tomorrow with my findings.

 

[Mikebits]

Well, I could not sleep as this was bugging me, so I went and tried to simulate your circuit, and it would not simulate as I suspect it oscillates. I went ahead and added a few .01uf caps at a few spots. Reran the simulation and no problems, so I suspect your having oscillations as well. I am attaching my schematic with the added caps. I think in a permanent board you should have copper clad vero board (see link). The layout should be done real neat with short leads and lots of ground plane. Below is the schematic. Maybe others will have better suggestions. Good luck, i'm off to sleep. Oh, ignore R6 100Ω, I just wanted some sort of load on the output.

 

[alec_t]

Are you absolutely sure you have the pinouts of the transistors correct (the BD139 pinout differs from the BC546)?

 

[alec_t]

I went and tried to simulate your circuit, and it would not simulate as I suspect it oscillates

Strange. It simulates fine in LTspice. No sign of oscillation.

 

[Les Jones]

Hi Alec,
The thing that looks very strange on Mike's circuit (Post #17) is the 49 volts RMS at 13 Khz on the output. As there is a 100 uF capacitor across the output it would mean a very large current trough this capacitor. (100 uF at 13 Khz has a reactance of 0.12 ohms. This is probably less than a real capacitors ESR value.) When I first looked at this I thought it was the OP's information from his circuit. I was going to suggest that the 100 uF capacitor was open circuit until I realised it was Mike's simulation.

Les.