Needing closest parts for Rebuilding a tube U47

[CalavoBob]

I've been looking for a thread that really addresses my situation, but I haven't found one so here we go. About 5 years ago I had the miracle happen that we are all waiting for. Because of my contacts in town, a guy called me up who needed to clear everything out of the house that his recently deceased musical-electronics engineering brother had left behind. The house was filled with all kinds of test equipment and audio gear. He wanted no money for it, and just wanted me to haul away as much as I could. I loaded up oscilloscopes, oscillators, cheap mics, stands, wire, parts, an Altec speaker, a Revox PR99, and an Ampex PR10. As I walked to the door to leave I opened a large old wooden box thinking it might be a turntable and saw the head basket of a U47 on a non Neumann housing. I held my composure, said I'd like this too, and left with a truckload of goodies. The first thing I did when I got home was open up that mic to discover a VF14 tube inside the housing. There is no M on this tube. The other parts did not look original, and the power supply is homemade. I figured out right away that whoever made the supply didn't know what they were doing, so I rewired it correctly, and replaced the electrolytics which were bad, for $33. I have attached pics of the schematic of the mic as found, the capsule which looks a hellofa lot like a M7(not reskinned), the tube and transformer, the capsule interface parts, and the resistors and caps. and the head basket.

The U47 build threads I've found on the groupdiy site are all oriented to not using a VF14, but I'm lucky enough to have one here even though it does not have the M (It actually sounds pretty quiet) Thats why I started this thread. I need opinions about rebuilding a U47 using a VF14. I'd like to assemble parts made by the original manufacturers or something very similar.

You'll notice in the schematic that there is no 1780 ohm resistor to the heater. He made a separate adjustable line from the power supply that sets the 36V which works pretty well. There is also a 0.002 uF disk capacitor to ground from the backplate instead of the correct 0.01uF cap. An extra 0.1 uF cap has been added in parallel to R8 which is 3.5 Mohm instead of the stock 3 Mohm. R7 is 2.25 Mohm instead of 2 Mohm. Finally, a Triad HS-60 transformer set for 150 ohms has replaced the original BV8 transformer.

The additional good news is that it works and sounds pretty good. I don't have another U47 to compare it to but I've used it on vocals, in front of a complete drum set, and as a room mic, and its got a great sound.

What I really want to do is to gather replacement parts, instead of the ones he used, that are as close to the originals as possible, including the transformer. After looking over the forums, I've found 60 Mohm resistors from Alipress, and I see that Moby's BV0.8 transformer is highly recommended for the U47. I noticed the U47 schematic on this site shows (2) 100 Mohm resistors going to the capsule instead of the 100 Mohm and 60 Mohm that I've seen on some of the earlier U47 schematics. Does anyone have strong opinions on what kind of difference that creates? Looking at photos of original U47s I see one capacitor made by Bosch near the output connector. Does anyone know the manufacturers of the original caps and resistors? I've had a hard time finding wirewound 2 Mohm and 3 Mohm resistors on Mouser. The ones in this mic are 2.25 Mohm which is pretty close, and 3.5 Mohm which is a little farther off. Has anyone fooled with the values on these resistors feeding the capsule?

The part that really annoys me is the 0.002 uF disk capacitor to ground from the backplate. Neumann must have used something better than that and this one is much smaller. I've got a dialectric absorption comparison tester that I'll go through the caps I'm going to choose, but it would be cool to find the original manufacturers.

I haven't included a photo of the casing, but it really doesn't look "Neumann". I'm thinking of seeing what micparts has for that, hoping that my head basket will fit.

Lastly, I'd like to rewire the heater voltage circuit to its original layout, but it means finding a 1780 ohm resistor that can handle the 2.67 Watts it uses. (105V-36V)^2/(1780 ohms) = 2.67 Watts. I'd like to use a 5W resistor for that so it runs cooler. There's nothing like that at Mouser. Anyone know where you can get a monster like that?

Well, any opinions, suggestions, and wild ideas about how you think I should tackle this are really appreciated.
In closing this is a reminder that dreams do come true. Treat people right and eat your vegetables and your dreams may also come true.

 

[Drosselmeier]

R7 on the picture in kiloOhms

 

[Murdock]

Nice find!
Concerning the 1780 ohm resistor. There is someone on eBay selling replicas and I think Andreas Stenzel also started to manufacture these. Maybe contact him.

Concerning the difference between 100 Meg and 60 Meg ohm. The lower that resistor is the lower the bass response is. And I think it was the other way around. In the later U47’s they incorporated this bass cut together with lowering the output capacitor from 1uF to 0.5uf. Try both and see what you like more. I for myself liked 60 Meg & 0.5uf more.
The original manufacturer of the output capacitor was Bosch. If you search for “Bosch MP capacitor” you should find them. But I would be cautious when sourcing these. Test for leakage as they are over 50 years old by now…

 

[CalavoBob]

R7 on the picture in kiloOhms

I'm not sure to which picture you are referring. R4 is not present in my pictures or schematic. The valuefor R4 that I see shown on the 1955 Telefunken schematic, the 1959 Neumann schematic, and the 1959 Supersonic Schematic is 1780 ohms.

 

[Drosselmeier]

 

[Khron]

I can't help but wonder just how massive of a difference you expect (?) the component values might be making.

Sure, the capsule bias voltage is more like 60v rather than the original's 50v-ish, but apart from ~1.6dB increase in sensitivity, you'll be hard pressed to hear anything else.

The difference in cutoff frequency between 100meg / 2nF and 100meg / 10nF is arguably "academic" - 0.8Hz versus 0.16Hz.

And as long as the supply voltages are clean enough, and the cathode bias voltage is what it's supposed to be, how do you imagine it matters HOW exactly that bias voltage is derived? Or actually, in both cases it's "part of" the heater supply, but then what does it matter if the heater is supplied from the B+, or separately?

 

[ubxf]

how close is a HS-60 to a BV8

 

[OneRoomStudio]

To echo those above, I think you should first decide what you’re wanting to accomplish. If all you want is a really good sounding mic, you’re probably 90% (maybe all the way) there. Don’t worry too much about the specific components.

If, on the other hand, you’re hoping to restore/recreate an authentic U47, I think you should talk to @Ericbazaar (Andreas Stenzel). He has reconstructed and restored many vintage Neumann mics/parts. This will not be an inexpensive project though.

 

[nielsk]

I have a bunch of small turret boards that are pretty much indistinguishable from the ones originally used, you are welcome to a pair for the cost of shipping.
I have used them in a bunch of U-47 builds

 

[CalavoBob]

Thanks for all the insightful replies.

R7 on the picture in kiloOhms

You are correct Drosselmeier. I should have made R7 in Mohms on my schematic. As far as the resulting sound difference, though, I don't think the small difference in R7 being 2.25 Mohms instead of 2 Mohms is going to be noticeable. Even the 3.5 Mohm for R8 is pretty close to the original 3 Mohms.

I am interested in the turret boards that nielsk has offered me for the cost of shipping. I'm new enough on this site that I'm not sure of the proper protocol to arrange the exchange of our addresses or emails, but I'm up for getting a hold of those. The build that the original owner did putting this mic together was fairly sloppy point to point. I'd like to make it look a little better.

Concerning the 1780 ohm resistor. There is someone on eBay selling replicas

I did find that ebay listing but he's wanting almost $600 for that resistor. I still need to contact Stenzel about those.

 

[CalavoBob]

And as long as the supply voltages are clean enough, and the cathode bias voltage is what it's supposed to be, how do you imagine it matters HOW exactly that bias voltage is derived? Or actually, in both cases it's "part of" the heater supply, but then what does it matter if the heater is supplied from the B+, or separately?

I agree with Khron about the fact that the tube's cathode shouldn't care about where the voltage is coming from, but there's another thing to consider. In the original design, the cathode's 36V is derived locally, inside the metal chassis of the mic using a string of voltage divider resistors from the only incoming voltage which is 105V. In my version of the mic the 36 Volts is derived inside the mic's power supply box with a voltage divider and pot, but this lower 36 Volts is then sent down the very long cable to the mic. This lower long distance voltage is going to be more sensitive to picking up interference than the original 105V in the design. It may adversely affect the s/n ratio of the system.

Also Khron's point about the cutoff frequency difference of the capacitor choice on the back plate is well taken. That breakpoint is way out of the audio spectrum and even beyond the 10:1 frequency spacing from 20 Hz needed to minimize phase shift. Ain't gonna hear that one.

Concerning the difference between 100 Meg and 60 Meg ohm. The lower that resistor is the lower the bass response is. And I think it was the other way around. In the later U47’s they incorporated this bass cut together with lowering the output capacitor from 1uF to 0.5uf. Try both and see what you like more. I for myself liked 60 Meg & 0.5uf more.

I like Murdock's thoughts about the value of the original 60 Megohm vs 100 Megohm and the value of the output dc blocking cap. I have seen both values for these on different Neumann schematics. These would be easy to do some comparison listening tests with.

how close is a HS-60 to a BV8

ubxf asks a question that I'll do a little more research on. Besides the resistor/cap comparisons discussed above, I'd really like to find out if the person who put the Triad HS-60 in this mic did his homework. At least I can do some bench testing of the HS-60 since I have it here. I'll have to rely on what I can find on the web to see what the known specs are for the BV8. I think Moby has probably put some good work into the design of his BV0.8 transformer.

To clarify my intentions with this mic, I don't expect to turn this into a mirror image of an original complete U47. I do want to explore possible obvious sonic improvements, and even small sonic improvements if they don't break the bank. I am looking at some of the clone parts sites to see if there is an affordable body that can make the outside look a little more like the original. After all, it is an iconic image we all know. I still can't get over my excitement when I opened that dusty box and recognized that head basket. Unbelievable.

Again, I want to thank those on the forum for their insight. It is really appreciated. I think you can all understand how much fun this project is for me.

 

[Khron]

In my version of the mic the 36 Volts is derived inside the mic's power supply box with a voltage divider and pot, but this lower 36 Volts is then sent down the very long cable to the mic. This lower long distance voltage is going to be more sensitive to picking up interference than the original 105V in the design. It may adversely affect the s/n ratio of the system.

"May"; won't. But if you're really concerned, stick a(nother) capacitor to filter the heater supply, inside the mic. 50v or 63v electrolytic, 105C rated, and as large as you can comfortably fit in there - plenty of choice available, even from "just" the reliable brands (Rubycon, Panasonic, Elna, Nichicon, United Chemi-Con).

 

[CalavoBob]

Hi Khron,
That's a good idea, especially after finding out that the1780 ohm 5 Watt resistor needed to create the original bias method is going for almost $600. A nice filter capacitor inside the mic is a much more cost effective solution. Maybe that's why he did it this way. It makes me wonder if that extra filter cap, C4 that's on R8 was intended to be on the 36V line. C1 filters the DC bias on the backplate while C4 filters the DC at the cathode. I'll try both and see if I can detect a s/n difference. Judging from the condition of the mic, the custom made power supply and the well worn wooden box, it looks like a diy piece from a bygone era.

 

[Khron]

It makes me wonder if that extra filter cap, C4 that's on R8 was intended to be on the 36V line. C1 filters the DC bias on the backplate while C4 filters the DC at the cathode.

Well, C4 also plays a part in filtering the backplate DC bias. And on the other hand, 100nF for a high-current load like the heater won't be doing much filtering, maybe only at (very) high frequencies (ie. well above the audio range).

PS: Neumann's choice to use that high-power dropper resistor was a heavy-handed engineering compromise, and nothing else. Insistence on going that route for (ahem) "vintage accuracy" is only proof of ignorance, from people with more attitude(?) than sense / knowledge / experience. Since now, in the 21st century, we're well free of the "this needs to work off of mains OR 105V DC worth of batteries" parameter that led to this particular design, we can use our brains and the far wider range of tools and components we have access to, to get ancient designs to now be "all they could (or even should) have been in the first place". But that's just me...

 

[tomas.borgstrom]

I'm pretty sure the capsule is an original Neumann Berlin M7. I would buy a body from Flea to use with the headbasket. I can't see the finish of the headbasket but if it's matte you should aim for short body otherwise go with long body. The mechanical kits from Flea are perfect and you can easily mount the Moby BV.08 and all the other components just like the original. I can help you with original resistors and capacitors if you are interested. They changed over time but I would recommend carbon film and Wima TFM found in the latest production runs. Skip the Bosch MP. I have ten NOS and they measure all over the place. There are other alternatives if you want the sound of MP. I would keep the separate filament supply. You can always install the 1780 ohm resistor later if you wan't the microphone nice and warm. You can actually wire it yourself but it's not easy. There's a thread about it here on GroupDIY.

 

[tomas.borgstrom]

PS: Neumann's choice to use that high-power dropper resistor was a heavy-handed engineering compromise, and nothing else. Insistence on going that route for (ahem) "vintage accuracy" is only proof of ignorance, from people with more attitude(?) than sense / knowledge / experience. Since now, in the 21st century, we're well free of the "this needs to work off of mains OR 105V DC worth of batteries" parameter that led to this particular design, we can use our brains and the far wider range of tools and components we have access to, to get ancient designs to now be "all they could (or even should) have been in the first place". But that's just me...

Iv'e built a 1:1 clone with my own hand wound resistor. Not because it's better than using a PSU with separate heater supply but simply because it was a fun challenge.

 

[rock soderstrom]

There are other alternatives if you want the sound of MP

What do you suggest? Siemens MP, Hydra or RFT?

 

[tomas.borgstrom]

What do you suggest? Siemens MP, Hydra or RFT?

The sealed flat Siemens, RFT, Bosch, SEL work fine but then you have to build the U47 like the early models with the transformer upright between the capacitors. Some axial Siemens I've used were within the specifications but they are hard to find. There's a lot of sealed axial MP of different brands but they are almost always too large.

 

[Drea]

Lot of mp condensators hasn’t the gritty crunch sound and compression (almost tape)and you get that super clean direct sound. Nothing for me.

 

[gyraf]

You're joking, right?