Neumann NKMa power supply

[NOON]

Today I picked up a Neumann NKMa power supply. It's an older one with the stablyte (or whatever they were called, seem to be NiCad batteries) and they have leaked badly, leading to some pretty nasty corrosion all through the unit. I will be restoring it, so will post photos etc as I go.

Two questions to start with. As this is an older unit and seems fairly original, does anyone want proper photos or other documentation of what components were used, wiring layout etc? Any specifics let me know.

Does anyone out there have experience with these units and have any tips on restoration back to working order, once I get the corrosion under control? Would you recommend fitting new NICad cells or is there a better way to do it now? (e.g. is NIMH a better way to go or some other battery technology, or even supercapacitors)

Back story is I picked up a KM56 a few years ago without a PSU and designed my own, which has been working well. The NKMa is not 'necessary' but it will be nice to pair the mic up with an original. So I'm not in any hurry to get it going and don't mind taking a bit longer to do it properly, especially if there is some kind of 'magic sauce' to be had from e.g. using batteries for the heater reg rather than the modern regulator I am using in my PSU design.

 

[Khron]

if there is some kind of 'magic sauce' to be had from e.g. using batteries for the heater reg rather than the modern regulator I am using in my PSU design.

Doubtful...

 

[NOON]

Doubtful...

Magic sauce is always doubtful, but you never know when it comes to nonlinear things like batteries. I did manage to get the heater rail pretty damn quiet on my design. LM317 didn't come close, can't remember what I ended up using, I'll dig out the schematics later.

 

[RuudNL]

Interesting to see what you did.
I have used a LM317 followed by 3.3 ohm and 6800µF. That was noise-free enough.

Or use the original Neumann Stabilyt replacement.

 

[NOON]

My KM56 doesn't have the cathode bias resistor so it's super sensitive to any noise on the heater. I tried a few things with the LM317 before resorting to a modern very low noise regulator.
Does anyone know why Neumann changed to the solid state regulator? Was it simplicity, cost, reliability or performance?

 

[Tubetec]

Potential for the cells to leak or degrade and the need for maintenance was probably good enough reason .

The Stabilyt cells could sustain the mic through short term mains outages or brownouts , maybe during the cold war era that was seen as an advantage .

I think its definately worth documenting as much as possible and taking photos of the unit as is .


Nicads can provide a very quiet heater supply , voltage will vary to some degree depending on cell capacity and current demand over time .
A 5 cell nicad pack makes a very quiet test bench LT supply for 6.3 volt tubes ,
An RC style microcontrolled battery charger allows you dial any level of undervolting you like .

I keep a handful of Nicad emergency lighting cells packs on standby , 4 cell 4.8V for 5V usb audio device power , 5 cell for 6.3v tubes .

 

[ursonate]

I converted my NKMa with an adjustable voltage regulator that includes an LM317.
To eliminate the high frequency noise, I applied RuudNL's suggestion and installed a 6800 µF capacitor between the voltage regulator and the 2.5 ohm wire resistor. What remained was a humming sound, which I was able to eliminate with a 680 µF tantalum SMD capacitor. This is installed in the connection plug to the microphone between pin 2 and pin 6, or the shield. There is no longer any humming or hissing to be heard. However, I use a Telefunken (Schoeps) M221 microphone and not a KM56, so I don't know if the method works with this microphone.

 

[TLRT]

There is a good replacement board from atomic analog to replace those NiCd cells

 

[NOON]

Interesting to see what you did.

I have used a LM317 followed by 3.3 ohm and 6800µF. That was noise-free enough.

Or use the original Neumann Stabilyt replacement.

I ended up using an LT3045. It has a capacitor multiplier first, experimented with the LM317 but couldn't get it quiet enough. The LT3045 does an amazing job. I used a TL783 for the HT rail regulation.

 

[NOON]

Case photos. Is 1906 an early serial number?

 

[NOON]

Internal shots and board shots, showing the corrosion from the NiCads

 

[NOON]

A few wiring and details shots, mostly for my own reference.

 

[NOON]

Because vintage computers tend to have leaky batteries that destroy traces there are loads of videos on Youtube on how to deal with this kind of damage. Neutralise the alkaline deposits with some weak vinegar, physically scrub the corrosion away, give it all a good wash, then dry it out and continue cleaning with some isopropyl. Then you're in a good position to assess the physical damage and repair as required.

It looks like the corrosion hasn't gone too deep on this one yet, 5 minutes scrubbing with some vinegar and it's already looking a lot better.

 

[NOON]

Progress!