Neumann U67 Clone : D-U67 Tube Microphone Build Thread.

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
So just to reiterate:

On the IEC plug, GND should be connected to chassis
No connection from IEC GND to 0v on pcb
Use a switch SPST or one side of DPDT to lift the screen (pin3) from mic body from GND (pin 7) so it goes to chassis

and the other point was to change R2 from 22R to 25R 5W and lower the H- voltage to 5.8v from 6.3v

could someone confirm this is all correct or correct me please?
 
Spencerleehorton said:
So just to reiterate:

On the IEC plug, GND should be connected to chassis
No connection from IEC GND to 0v on pcb
Use a switch SPST or one side of DPDT to lift the screen (pin3) from mic body from GND (pin 7) so it goes to chassis

and the other point was to change R2 from 22R to 25R 5W and lower the H- voltage to 5.8v from 6.3v

could someone confirm this is all correct or correct me please?
Hard to understand without a drawing.  I would modify it to match the Neumann schematic being careful that 0v only connects to shield/chassis in 1 place, in the mic. Then have the SHLF jumper connect a secondary connection of 0v to chassis/shield in the PSU.  The SHLF jumper should be out during normal operation. The mic body and PSU chassis should always have continuity to the IEC ground.
Easiest way to do this with Poctop's boards would be take pin 3 from the mic and go right to chassis in the PSU. Then take this connection to the Pin3 pad on the PCB.
The next thing I would do is wire a SPST switch between the pin1 XLR out and the chassis, to make a true ground lift switch. In the Neumann schematic this is always lifted and in Poctop's design it is always connected to 0V. This is the most likely mistake causing the problem, IMO
I have to try this stuff out though before I can say for sure it is the solution.
 
Spencerleehorton said:
I had a connection from the IEC to the 0v on the Pcb as well as the IEC to chassis.
Could this have been my problem?
Yes, that gives you two connections between chassis and 0v, one in the mic and one in the psu. You only want a single connection when the SHLF jumper is out.
If you build it according to the plan and have only a single connection from 0v to chassis, the mic will be quiet and operate great in 'nice' environments.  However, if you record in different environments, you'll probably find it has hum in some situations. That's what I've found and some other people here.
I was recording at a friend's place where the mic was plugged in upstairs and the preamp was plugged in down in the basement and the mic had a bad hum. I am thinking lifting pin1 on the output XLR would have killed the ground loop.
 
Thanks dmp.....

And what about mic cable "ground" connections?

I found this on Neumann forum:
"The twisted pair is best used for pin1 + 2. If the Mogami should have any thicker wires they should be used for 3 + 4, i.e. for the filament current of 100 mAmps. Wires 3 + 7(shield) do run parallel, and should be connected at both ends of the cable. Any metal housing parts of the connectors should be connected with shield/ground as well, to avoid RF interference. The inner pin is #8, which is not really used.
Best regards, Martin Schneider Neumann Mic. Development"


 
And what about mic cable "ground" connections?

That is how I wire it. In the build guide, the cable shield is used for p7 and the black wire is left unused.

mrerdat posted a lot of good stuff from experiments in a 'bad' electrical environment, so be sure to read through all his posts. I have yet to try out any of the proposed solutions. Hopefully we can come up with a final solution for the U67 wiring that works robustly.
 
Im going to rewire both my U67's PSU this morning, will test continuity before powering on, will try the 5.8v and see how it sounds, maybe this is the same trick as for the U47's, slightly under power the valve to get a nicer bias point?
 
better to ask rather than blow myself up:

should i have continuity between pin 7 and the chassis? this is just measuring the PSU?

Is it:

Chassis of mic pin 3 (screen) is connected to chassis with the switch, and you can connect to 0v (pin 7) + chassis?

I am getting continuity from chassis to pin 7 on one way of switch, then the other way pin 7 is then disconnected from chassis.
so this seem to match what the schematic says?
I've then tested continuity from chassis of PSU to chassis of microphone and have good continuity in both switch positions.

So to my mind i have done this correctly, please let me know if i haven't!!!
 
My other observations are that in these PSU cases needs the paint to be scratched off to connect  the top part otherwise you only have GND on bit you connect to the chassis.
I wonder how much of a difference this makes, i would suspect a lot!!! if its meant to be in a fully 360 degree isolated box and your only isolating part of it, this could then be prone to interference.
Im going to look at my second one now and then hopefully power them up and have a listen.
 
wired them up and tested, one works, the other i've wired up wrong as it don't work!!!
will need to do another day when have more time and less snow and so cold!!!!

cant tell if the one thats works is better yet, i think the noise floor sounds better but until i get a vocalist in, i cant really tell.

 
Spencerleehorton said:
better to ask rather than blow myself up:

should i have continuity between pin 7 and the chassis? this is just measuring the PSU?

pin7 is 0v, so with the mic unplugged it should not have continuity to the chassis. The switch in the Neumann schematic connects 0V to chassis within the PSU optionally. With the mic plugged in, the 0v to chassis connection in the mic should give you continuity.



Is it:

Chassis of mic pin 3 (screen) is connected to chassis with the switch, and you can connect to 0v (pin 7) + chassis?

I am getting continuity from chassis to pin 7 on one way of switch, then the other way pin 7 is then disconnected from chassis.
so this seem to match what the schematic says?
I've then tested continuity from chassis of PSU to chassis of microphone and have good continuity in both switch positions.

So to my mind i have done this correctly, please let me know if i haven't!!!
Do you have pin 3 connected to the PSU chassis directly?

My other observations are that in these PSU cases needs the paint to be scratched off to connect  the top part otherwise you only have GND on bit you connect to the chassis.
Definitely! This is something you have to check with nearly all DIY builds!
I also had a mic build once where I had to scrape a little paint off the part that screws the body together to make sure the mic body formed a good faraday shield.
 
And do you have a noisy environment to test it in?

Unfortunately, my buddy moved so I can't go try it in the house where I had the problem. My U67 works fine in my normal studio space.
 
My next option today to investigate is this R2 change from 22R 2W to 25R 5W.
And lowering H- from 6.3v to 5.8v.
See if this lowers the noise.

I also have a 20v 12va which I can test rather than a 6va.
As am i right in thinking a 6va only gives 300ma and a 12va gives 600ma?

Will report back.
 
right, i've plugged in my 2nd version of the U67 and the GND lift does seem to now work, where as my first version doesn't?
will need to check in a bit.

first of all i have wound my own psu txf and have 201v and 20v, the actual txf does hum audibly a bit, i didn't clam yet or varnish but i have ordered another toroidal and i have another 20v 12v here to change to.

I can hear a fair amount of hiss when i set gain and bring the level up, this might be that its not got a great capsule in it, hiss does go after about 10 minutes i've noticed? valve warming in?

My main concern is when i touch the pattern switch i can hear that i am then grounding it!! will try connecting it to ground.

the changes so far seem to have improved the mic, will continue to investigate.
 
On another test, my first version of the U67 is much quieter than my second version.
Same output within a nats! same tone, i'm putting through API-312 and 1176 rev D and it sounds great.
But my first version doesn't seem to lift the GND? and doesn't have the GND problem with the pattern switch?
 
Spencerleehorton said:
just to simplify, does this GND lift function just take the GND off the output pin1 XLR?

A ground lift could do that - have you tried it?

The SHLF on the pcb disconnects the 0v chassis connection on the PSU leaving 1 connection in the mic.

I think if the mic grounding is correct you will just hear tube noise. It is only if you have a ground loop in a hostile environment that this stuff matters.
 
Back
Top