Strange problem with simple power supply

[RuudNL]

A year ago I made a power supply for a tube condenser microphone.
I used a 'Chinese' voltage converter that can supply 6.3 V and a variable 150-280 V output.
Because I needed +120 V and I couldn't adjust the voltage lower than about 141 V, I added a series resistor and two zener diodes in series. (62V and 56V) As an extra precaution I added a 47 µF capacitor over the output.
This works perfectly, without any noise on the output of the microphone.
(Schematic is the top version in the attachment.)

Because of this success, I decided to create a second power supply for a U67, using the same DC converter module.
I wanted some extra filtering, so I added a series resistor of 4.7K and a 47 µF capacitor.
But... for some strange reason I can't get a higher output voltage than about 12 Volts!
Without the resistor and capacitor I can get 210 V. output, but as soon as I connect the RC combination, the voltage drops significantly. (Lower schematic)
When I load the output (without the capacitor) with two 10 K resistors in series, the output stays at 210 V. (So 210/20K = 10,5 mA)

Now, what would make the difference that the upper schematic works perfectly, but the lower schematic doesn't?
Just curious!

 

[moamps]

Maybe you need to connect some load to the output. Try connecting a 20k resistor across the capacitor.

 

[NOON]

Does it have a minimum load? Try putting a load resistor across the second one. First one is drawing some current through the zeners, second is drawing nothing (depending whether it sees a charging cap as enough of a load) If that doesn't work then I'd say faulty module or possibly faulty cap.

 

[ruffrecords]

Agreed, there is no dc load in the second case.

Cheers

Ian

 

[RuudNL]

Without a load I get 210 V and with a 20K load (so >10 mA) I still get 210 V!
So that can't be the problem.
But as soon as I connect a capacitor the voltage drops.
I tried a couple of (new) capacitors and it is unikely that all new capacitors are faulty...

 

[ruffrecords]

Just to be clear, are you saying that with two zeners and a parallel cap as a load it works, but with a resistor load and parallel cap it does not work?

Cheers

Ian

 

[gyraf]

..damaged cap? ..that's about the only simple answer..

 

[Walrus]

The 4.7k isn't an incorrectly marked 470k is it?

 

[abbey road d enfer]

A year ago I made a power supply for a tube condenser microphone.
I used a 'Chinese' voltage converter that can supply 6.3 V and a variable 150-280 V output.
Because I needed +120 V and I couldn't adjust the voltage lower than about 141 V, I added a series resistor and two zener diodes in series. (62V and 56V) As an extra precaution I added a 47 µF capacitor over the output.
This works perfectly, without any noise on the output of the microphone.
(Schematic is the top version in the attachment.)

Because of this success, I decided to create a second power supply for a U67, using the same DC converter module.
I wanted some extra filtering, so I added a series resistor of 4.7K and a 47 µF capacitor.
But... for some strange reason I can't get a higher output voltage than about 12 Volts!

Do you mean with the lower schemo you get only 12V at the capacitor? Without additional filtering?
That means about 8W in the 4.7k resistor. Does it get hot?

 

[RuudNL]

The output voltage of the DC converter drops, so no excessive dissipation in the resistor.

I have swapped the two DC converters; the one I used for the +210V now only delivers 6.3 V and vice versa.
To my surprise this solved the problem. So it seems one DC converter had a problem.

Still it is strange that the 'faulty' DC converter delivered 210 V without a load, and did the same with a 10 mA load, but dropped the voltage when an R-C load was conneted. Well, problem solved it seems...

 

[Ike Zimbel]

I have seen power supplies appear to o/p the correct voltage under no load conditions, only to tank when a load is connected. In at least one case this was due to depleted/defective reservoir capacitors so you could check/replace the electrolytics on the bad one. In another case this was due to a wrong value resistor in the o/p path. I forget the exact numbers but it was something like 10x or 100x the correct value, like 47k or 470k in place of a 4R7.

 

[RuudNL]

Now it is working fine with the same capacitor as I used before.
The resistor is indeed 4700 ohm, I have checked this.

 

[Tubetec]

Is the fillament voltage unloaded on one of the pcbs?
maybe it has to be loaded for the HT to work properly .

You often see that in multi rail smps if the 5 volt line is unloaded the higher voltage rails arent regulated properly .

Its nice and neat with the bloc transformer , I remember you saying both units require a seperate winding and rectifier due to the shared HT and LT ground on the boards , was the pcb that failed then same one as was in used for a year ?
Maybe those electrolytics are suspect .

 

[Hubbub]

Anytime you work with a switching supply do verify the operation as proper and clean using an oscilloscope, and maybe an adjustable load. Doesn't need to be a fancy 'scope, but at least 1 MHz bandwidth. I have a quite usable digital 'scope/signal generator/transistor-diode tester that was $55 US (with a probe!). Once that result looks good then use your voltmeter to check the output level more accurately. Some voltmeters behave unpredictably when faced with high frequency noise on the DC input.

 

[rock soderstrom]

Now it is working fine with the same capacitor as I used before.

Hi Ruud, I also ordered two of the PSU boards, they are just too tempting. The size and the features are convincing for small tube projects. Are you still happy with them? I'm really curious how clean B+ is?

 

[mrclunk]

Hi Ruud, I also ordered two of the PSU boards, they are just too tempting. The size and the features are convincing for small tube projects. Are you still happy with them? I'm really curious how clean B+ is?

I'd be interested too..
If Ruud says they are clean i imagine they are fine?

I've some old Schoeps tube SDC that need 160VDC psu's.
Was going to make a pcb for this charge pump circuit, with some additional filtering, but one of those boards is a lot simpler.
https://frogpedals.com/wp-content/u...harge-pump-power-supply-Documentation-1.1.pdf

 

[abbey road d enfer]

Strange calling that a charge-pump, when it's a standard boost converter...
Charge pumps store energy in capacitors, smps in inductors.

 

[rock soderstrom]

I'd be interested too..
If Ruud says they are clean i imagine they are fine?

I think so too. I've tried something similar before, not quite as developed, and I had problems. If this works with mics, it should also work with smaller tube FX devices and preamps. The operation with a standard 9V DC wall wart is very tempting.

 

[rock soderstrom]

I've some old Schoeps tube SDC that need 160VDC psu's.
Was going to make a pcb for this charge pump circuit, with some additional filtering, but one of these boards is a lot simpler.
https://frogpedals.com/wp-content/u...harge-pump-power-supply-Documentation-1.1.pdf

This one is similar to the one I once built, but I didn't find it so convincing; I expect more from Ruud's suggestion.

 

[mrclunk]

Ruud, What DC-DC IC does your board use?
I found an interesting blog here about a HV design. Might try rolling my own...
https://jan.rychter.com/high-voltage-power-supply-for-nixie-tube-projects

A year ago I made a power supply for a tube condenser microphone.
I used a 'Chinese' voltage converter that can supply 6.3 V and a variable 150-280 V output.
Because I needed +120 V and I couldn't adjust the voltage lower than about 141 V, I added a series resistor and two zener diodes in series. (62V and 56V) As an extra precaution I added a 47 µF capacitor over the output.
This works perfectly, without any noise on the output of the microphone.
(Schematic is the top version in the attachment.)