The Ultimate DIY U47 - Build Thread (And some PSU questions)

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
there was discussion maybe here and definitely at guru fora about 'thermal biasing'--part of the sound of the original comes from the heat; the mic circuit system performs as it does due in part to the components being hot...
That's interesting stuff. I was thing of using the Phaedrus tube replacemnt, and earlier in the thread heat was mentioned as a problem with that device.
 
there was discussion maybe here and definitely at guru fora about 'thermal biasing'--part of the sound of the original comes from the heat; the mic circuit system performs as it does due in part to the components being hot...
IIRC I think it might have been an Oliver post at a forum. I seem to remember it might be about heat in the transformer laminations.
Anyone else remember reading this?

I have tested caps that are used in guitar amps because I was wondering why tube amps change in tone after running for some time(IIRC I read this online and people I know have told me this) I guessed it might be heat.
I placed the power supply Al electrolytic caps in a temp controlled air oven and I tested them at room temp and at higher temps. IIRC the ESR changed. I will look for my notes hopefully I still have them it was about 20 years ago that I did the test.


I have not tested a transformer at different temperatures. I would guess tests would be the BH curve and inductance at different temperatures.
 
there was discussion maybe here and definitely at guru fora about 'thermal biasing'--part of the sound of the original comes from the heat; the mic circuit system performs as it does due in part to the components being hot...
Right ! I've been criticised about my choice for building an MK-47 from Max Kircher because this design has the reputation of making the mic very hot. And after building it, I can say it's getting hot, not so much IMO, but it just sounds so nice... I love it ! So probably the sound and colour of this mic is due to the components getting hot like you mentioned.
 
R4 in the original U47 is a custom made 1780 Ohm flat wirewound resistor that's mounted along the side of the bottom bell using the body as a heat sink. I remember a discussion regarding using a regular resistor instead. This was criticized since the microphone gets too hot inside making VF14 noisier. The heat is also considered bad for the capsule. That's why U47 often are seen hanging upside down. On a genereal basis excessive heat is not good for electronics. That's why we use heat sinks.
 
I have Dans u47 Micbody with GE408 tubes inside.
Yes the bottom bell gets hot, but not as hot, that you can’t touch the bottom bell.
A year ago I experienced with a NOS Telefunken UF14 tube.
That experience was, what I would subscribe as hot.
You couldn’t touch the bottom bell without burning the fingers.
The mic sounds damn good with it, but I had to put R4 into the PSU to get the heat away from resistor inside the mic, so I decide to put back the 408 in it.
 
Older components like carbon resistors and mp caps are affected by extreme heat. On top of that the older components are rated 5 to 10%. So after a few hours of continuous use the values of some components might drift 15-20% of their intended value. Also I have read that the transformer is loosing low end when it gets hot.
R4 after a couple of hours of used can’t be touched with bare hands…it’s hot, so there is quite a bit of heat in the mic.
It would be interesting if someone calculated those drifts and substitute with more stable components with the “drifted values” in clones with separate heater voltage.
 
Does anyone know if a 7pin XLR can safely handle the power from the 105v lead? I know a binder connector is usually the standard, but I was thinking about using a 7pin XLR for the PSU connection since they're easy to come by. Just not sure if the connector pins are beefy enough.
 
Last edited:
I fabricated a metal clamp to secure a few 5W block resistors tightly to the bottom bell of my Chunger u47 body using the two threaded holes.. makes a nice heat sink and the bell gets a little warm using VF14 after many hours, but never too hot. Almost too hot with a nuvistor for my comfort level, so maybe a bigger resistor would do it. YMMV, and I've only heard online examples of ef12/ef14 but in my personal testing I'm also in the nuvistor camp for non VF14 options.
 
This is the solution I’ll be using for my drop-down resistor. Dan’s u47 chassis comes with an aluminum block pre-drilled to mount an RH-5 5w 1500ohm chassis-mount resistor. I'll be using a 250ohm wire-wound resistor in series with it to get the exact voltage I need, as the chassis mount resistors don't come in that specific of a value. I assume I won't need to worry about cooling that second resistor as much since it won't be dissipating the majority of the heat?


E2FF5955-B42E-4B16-B06C-A10DD6B5E67A.jpeg
 
Does anyone know if a 7pin XLR can safely handle the power from the 105v lead? I know a binder connector is usually the standard, but I was thinking about using a 7pin XLR for the PSU connection since they're easy to come by. Just not sure if the connector pins are beefy enough.
7 pin xlr totally fine for 105v
 
I fabricated a metal clamp to secure a few 5W block resistors tightly to the bottom bell of my Chunger u47 body using the two threaded holes.. makes a nice heat sink and the bell gets a little warm using VF14 after many hours, but never too hot. Almost too hot with a nuvistor for my comfort level, so maybe a bigger resistor would do it. YMMV, and I've only heard online examples of ef12/ef14 but in my personal testing I'm also in the nuvistor camp for non VF14 options.
Interesting ! Do you plan to offer some for sell ? Any photos ?
 
Interesting ! Do you plan to offer some for sell ? Any photos ?
Thanks, I fabricated, as in, I dremeled off the edges of a small angle bracket and hammered it to a shape that fit the resistor and screw:LOL:! ... so none for sale. The resistor was a 1k5 and the others in series didn't get as hot, and the Dale RH-5 posted by StormsofVogue is probably a better plan if you can fit it.
 
Thanks, I fabricated, as in, I dremeled off the edges of a small angle bracket and hammered it to a shape that fit the resistor and screw:LOL:! ... so none for sale. The resistor was a 1k5 and the others in series didn't get as hot, and the Dale RH-5 posted by StormsofVogue is probably a better plan if you can fit it.
I'd love to see pics as well!
 
Does anyone know if a 7pin XLR can safely handle the power from the 105v lead? I know a binder connector is usually the standard, but I was thinking about using a 7pin XLR for the PSU connection since they're easy to come by. Just not sure if the connector pins are beefy enough.
The reason I don't like XLR in tube microphones is the increased risk of a pin loosing contact. If you have an unregulated PSU the voltage will rise and damage the tube. I guess this risk is lower on the PSU since you don't move it around as much.

To be honest, I don't know if is a true problem. Has anyone encountered this?
 
I used Binder for my U47 and M49 clone since 3 years and XLR since 3 years for my C12 clone.
Never had any problem with booth of them.
 
nice mic project ,
..thinking to get 1
but is any audio comparison with the original 47 somewhere ?
thanks
 
Hey guys! Just a quick update. I decided I wanted to replicate the look below from one of Dan's builds. I contacted Dan about purchasing a steel mount plate, and possibly the circuit boards, but the one in his example was just a one-off and he didn't have any of those parts available. Of course, that wasn't going to deter me at all, so I got ahold of my brother who is a machinist and he built a custom stainless steel transformer bracket for me. As for the circuit boards, I've designed PCBs in Kicad before, so those should be pretty straight-forward to have manufactured.

Dan's transformer/PCB assembly:
unnamed.png

My custom mount bracket:
IMG_1756.jpg

IMG_1760.jpg

Also, I finished mounting the LCD readout in my variac. It works great! Viewing angle is a bit limited, and the readout updates in 1.5 second intervals, so it behaves a little weird as you're turning the dial since it's not giving real-time feedback. Caveats aside, it should be a useful tool for this and other projects.

IMG_1594.jpg
 
Last edited:

Latest posts

Back
Top