DON'T USE DC-DC step down/buck switching converter circuits for H+

[Roman Beilharz]

I have just added this neat little DC-DC step down converter to my passive PSU for my U47-stlye/EF806 tube mic's H+ voltage. It is really convenient to tweak the H+ voltage using this little gadget even w/o taking out my DMM. Plus it stabilizes my heater DC regardless of altering component resistance prior to it when switching relays, LEDs or the like. Got this on Ebay for around 10 Euros including shipping.

I thought some of you might also find this helpful:



There are other more advanced versions which allow you to display your H+ current, throttle it etc.

 

[Drosselmeier]

Not a good idea to use dcdc in heater. IMO. It is better to use LM 317 and a digital voltmeter

 

[Roman Beilharz]

Why? What kind of issues do you see for the heater voltage?

Plus as I just read for the LM317 to deliver stabil power the input voltage needs to be >3V higher than the output voltage. As I am just reducing my PSU's 6.5V to 5.2-5.7V, the efficiency and stability seem questionable.

 

[Roman Beilharz]

Alright, I've learnt it the hard way: At least this step-down-converter chip/circuit introduces too much HF noise through H+. I could get it down to a reasonable amount of noise with heavy (R)C filtering, but in the end this has not been worth it, indeed.

I had to precicely drop the voltage in true cardioid mode with a resistor to make up for my relay in MP mode and could achieve an almost 5dB better SNR.

Conclusion: This was a bad advice. Don't do it.

 

[Khron]

What does "heavy" mean? More often you'll see LC filtering (due to steeper filter slop).

 

[rock soderstrom]

Alright, I've learnt it the hard way: At least this step-down-converter chip/circuit introduces too much HF noise through H+. I could get it down to a reasonable amount of noise with heavy (R)C filtering, but in the end this has not been worth it, indeed.

At the risk of sounding like my old mother, I saw it coming.

When I first read your post I wanted to warn you...my few experiences with similar circuits have been just as negative. The effort to get rid of the HF garbage was considerable.

My favorite for a clean and flexible solution is the LM317 with a trimmer and subsequent RC filters as needed. If only 6.3V AC is available, then I also use a voltage doubler. Winner Team!

 

[Roman Beilharz]

Well, when I put at least 1000uF to ground in front of the DC-DC converter, I could reduce the noise induced by the bugger especially in the 5-10kHz area, but still: w/o the circuit my SNR was way better, so why spoil it with another "improvement"? But sure, there for sure are better ways to get rid of the buck switching noise. I even considered to throw in another ready made PCB like this one here:
EMI-Filter 50V/2A [EMI-FILTER1]

but I ran out of space in my PSU case and well, it didn't really make sense.

It was nice to be able to experiment with the step-down converter inserted and find the sweet spot for the heater voltage - and that I didn't have to care about changing draw from other components on the H+ rail, though. I found that my Telefunken NOS EF806 sounded great and it's noise had the least disturbing shape when choosing 5.27V precisely.

 

[Roman Beilharz]

At the risk of sounding like my old mother, I saw it coming.

When I first read your post I wanted to warn you...my few experiences with similar circuits have been just as negative. The effort to get rid of the HF garbage was considerable.

My favorite for a clean and flexible solution is the LM317 with a trimmer and subsequent RC filters as needed. If only 6.3V AC is available, then I also use a voltage doubler. Winner Team!

I got that, I got that. Will never do it again, I promise...

But wait: THERE'S MORE CHINESE RUBBISH TO PUT INTO MY CHAIN!!!

 

[rock soderstrom]

 

[living sounds]

DC-DC converters work quite well in the Haiki Trafoless Vari-Mu Compressor.

 

[rock soderstrom]

Well, a vario vario mu compressor is not a microphone. There is a difference, the two 6SK7 in Heikki's ingenious design deal with line signals. A microphone, on the other hand, is amplified a lot until it lands on tape/HD.

Don't get me wrong, you can certainly do this successfully with a DC-DC converter. But you have to pay attention to a few things and really be aware of what is in the signal at the end. That's certainly possible...

The question is why? I don't really see the advantages over a classic LM317 PSU.

 

[scott2000]

DC-DC converters work quite well in the Haiki Trafoless Vari-Mu Compressor.

Understated for sure

 

[Khron]

The question is why? I don't really see the advantages over a classic LM317 PSU.

Far less heat dissipation, at least? Also depends what voltage one has available to work with...

 

[rock soderstrom]

Far less heat dissipation, at least? Also depends what voltage one has available to work with...

Yes, it depends on the voltages available. If you do it right with a conventional voltage regulator solution and distribute the thermal load well, nothing in the PSU will get more than hand-warm. It is only one small tube, 150mA to max. 300mA for the usual suspects.

 

[RuudNL]

I have used a Chinese DC/DC converter (output 6.3 V and 150-280 V) for a tube microphone without any problems.

 

[Khron]

I have used a Chinese DC/DC converter (output 6.3 V and 150-280 V) for a tube microphone without any problems.

Do the chips have any visible markings?

 

[Ilya]

I’ve used step-down converter for H+ of my V76 mic pre. Not a ready-made module, but designed around some switcher chip (don’t remember the number off the top of my head).
Input and output were filtered (CM and DM chokes on the input and LC at the output). It was also designed to run in CCM, and switching frequency was around 100k.
No traces of noise whatsoever. And very efficient compared to LM317.

 

[RuudNL]

Do the chips have any visible markings?

I would have to take a peek inside.

EDIT: There is 'something' on the chips, but my eyes are not good enough to read what is printed on them.
(Not even with a magnifying glass... The print has very low contrast.)

EDIT 29/08; Today I looked with 'better eyes', the chips are CXW8509 and MAX1771.

https://datasheetspdf.com/pdf-file/908779/CoreMicroTechnology/CX8509/1
https://www.alldatasheet.com/datasheet-pdf/pdf/1326088/MAXIM/MAX1771.html

 

[rock soderstrom]

I have used a Chinese DC/DC converter (output 6.3 V and 150-280 V) for a tube microphone without any problems.
View attachment 113292

Interesting, good enough for a U67 style mic?

 

[Khron]

Interesting, good enough for a U67 style mic?

Well, the U67 filters the grid bias through a 60meg/10nF filter. If a -3dB frequency of 0.26Hz (and a 6dB/octave slope from there onwards) isn't enough...