R/E/P Community
R/E/P => Klaus Heyne's Mic Lab => Topic started by: Koen on September 26, 2015, 09:41:54 AM
-
A thing that bugs me regarding NU67's:
I employ 3 U67's and consequently use 3 original Neumann NU67 power supplies. Two of them are of the same (later?) type, with serial n°s in the 10000's. Now: when recording, using the very same microphone, but switching between the power supplies, the output differs by more the 3-4 db. I do have a pair that come quite close, but another one is really way quieter. How is this possible? And, is it at all possible to 'match' them a bit more?
Thanks for your input,
Koen.
-
Start by measuring the supply voltages of all three mics with their respective power supplies (you should always keep mic and power supply as a permanent team, and not switch around: different tubes can have quite a bit differing current draw, and once the power supply is calibrated for the specific mic, it should stay with it).
Then substitute the mic/power supply combo that gave you lower gain, and measure that combo. You may find that the mic is voltage starved.
Let us know what you find!
-
Thanks,
Will do later this week and post back.
210V polarisation voltage and 6,15V filament, right?
Greetings,
Koen.
-
Right! And measured at the mic!
-
A thing that bugs me regarding NU67's:
I employ 3 U67's and consequently use 3 original Neumann NU67 power supplies. Two of them are of the same (later?) type, with serial n°s in the 10000's. Now: when recording, using the very same microphone, but switching between the power supplies, the output differs by more than 3-4 db. I do have a pair that comes quite close, but another one is really way quieter. How is this possible? And, is it at all possible to 'match' them a bit more?
I would just suggest you to start by measuring supply voltages from three mics, and then substitute combo and measure it as well.
____________
Katerin, professional consultant on essay trust (http://www.essaytrust.org)
-
I cannot emphasize enough to :
1. calibrate the power supply voltage to nominal 210VDC/6.15VDC, which adjusts the supply to a specific mic's (actually its tube's) current draw.
2. not switch supplies after having done the calibration, in order to maximize tube life, minimize diaphragm stress, and optimize the mic's noise floor
-
Will do, thank you!
-
I concur with Klaus. All my PSUs are assigned and color coded to go with a respective mic, and calibrated to that mic. You should never mix and match, unless you want to shorten the life of your tubes.
-
Mystery solved. Well, sort of.
The polarisation voltage of said NU67 supply measures 299V (!) (under load), the filament heater 6,11V.
B+ is way off in other words...
Also, it's a early serial number and the schematic and layout look somewhat different from the later ones (this is type N-084-03, my other ones are N-084-07). I can't find a schematic for this version unfortunately. Any ideas?
regards,
K.A. Gisen.
-
Another question:
at this voltage, wouldn't it be wise to make some sort of 'dummy' load instead of attaching the microphone itself?
I mean, couldn't these voltages damage the capsule?
If at all possible, anyone any clues for me on how to device such a load?
Thanks,
K.A.G.
-
Please check R7 (160 kΩ / ¼ W), which is typically located across the final filter capacitor C4 (25 µF). I suspect it may be open. With this resistor and a load resistor of 240 kΩ / ¼ W (equivalent to a normal U 67) the voltages against ground should be within 5 % as follows:
at the junction of the rectifier Gr1, first electrolytic filter capacitor C1 and R1 = +243 V;
at the junction of R1, second electrolytic filter capacitor C2 and R3 = +225 V;
and at the junction of R2, third and final filter capacitor C3 and R7 = +210 V (= across R7 and equivalent load or U 67).
-
Many thanks Uwe,
will try asap!
Regards,
Koen.
-
Uwe, if you're reading this:
I'm afraid this PSU has a different schematic and layout than the later ones (as I pointed out, this is a N-084-03) the codes of the large caps for example already differ from your schematic, eg. C5, C6 and C7 are 500uF here and C1, C2 & C3 are 25 uF cans. C4 is, as far as I can see, a Wima 0,1 axial cap. Confusing without a good schematic...
-
Uwe, if you're reading this:
I'm afraid this PSU has a different schematic and layout than the later ones ...Confusing without a good schematic...
HEADLINE:
NEVER MANIPULATE OR TOUCH ANYTHING INSIDE OF SUCH A PSU WHILE THE POWER PLUG IS CONNECTED.
EVEN AFTER DISCONNECT LETHAL VOLTAGES ARE STILL PRESENT IN THE CHARGED CAPACITORS FOR A LONGER TIME.
SPECIALLY IF THE THING IS BROKEN THESE VOLTAGES MAY NOT BE DISCHARGED FOR HOURS.
IF YOU ARE NOT SKILLED GET SOMEONE DOING IT FOR YOU!!!
First check if the line voltage selector is set to the correct value.
Even check the secondary voltages at the transformer to confirm.
The 25uF caps are the high voltage (B+) filter stages.
POSSIBLE REPAIR:
Check all resistors in-circuit (without desoldering) for if the value is in a reasonable range (about what their color coding shows).
Usually they are either working or fully open, so the faulty one should be be locatable.
If this doesn't help, check the voltage of each filter step with no load condition.
Voltage should significantly drop from stage to stage. Where it doesn't you've found the faulty stage.
If the voltages are all the same the output shunt R is open (like Uwe mentioned).
If this doesn't help, unsolder the resitors at one end, one by one, and check their value.
POSSIBLE MOD:
It might even be that someone or Neumann has modified the PSU for whatever reason.
The different model No. point's to Neumann doing it, maybe for a different mic version.
You would need to replace some parts to get the correct output voltage then.
The basic circuit is quite simple, shouldn't be hard to (re-)modify it to N-084-07 circuit version.
Do you see something unusual like a transistor that might have been inserted for regulation?
Throw out everything (in the high voltage path only) except the caps and put in the 6K8, 8K2 and 100K resistors that are standard.
Play with the 1st stage's R's value (6K8) if the voltage is still a bit off under load (240K Ohm refering to Uwe) condition.
Not a big deal.
Regards
Kai
-
Thanks Kai,
this is very helpful!
-
Because of the dangerously high voltages present inside the power supply, none of the experimenting described or recommended in these posts should be done by a lay person. Take the supply to a qualified technician instead!
-
Yes, I forgot to mention:
NEVER MANIPULATE OR TOUCH ANYTHING INSIDE OF SUCH A PSU WHILE THE POWER PLUG IS CONNECTED.
EVEN AFTER DISCONNECT LETHAL VOLTAGES ARE STILL PRESENT IN THE CHARGED CAPACITORS FOR A LONGER TIME.
SPECIALLY IF THE THING IS BROKEN THESE VOLTAGES MAY NOT BE DISCHARGED FOR HOURS.
IF YOU ARE NOT SKILLED GET SOMEONE DOING IT FOR YOU!!!
All the best
Kai
-
Hello KOEN,
Please update the forum: how have you solved the issue?
-
Hello to everyone here.
Although we didn't have the right schedule at hand (in the meantime, Martin from Neumann has send me one), we think we managed to get the PSU to work as it should.
We first measured all the stages in the circuit and found out that the transformer put out a hefty 227V. With a load in place, the polarisation voltage was a staggering 299V! Firstly we tried to attach the 240V tap of the transformer (it is an older model, so 220V only on the circuit board, but there actually is a 240V tap located on the transformer. This brought down the voltage to about 280V, but the filament went down to about 4,8V and we were unable to bring that back up to the required 6,15V. Eventually we went back to the 220V tap and replaced the diode bridge. We subsequently augmented the dropping resistors (R1 and R2 on the newer schematic, but R2 and R3 on our circuit board) to 18 Kohm and 10 Kohm respectively. We also saw that there was no 160 Kohm resistor around R7 (R8 on our circuit) and put a 100 Kohm in there (as per an other, slightly different schematic we found).
In the end, we managed to get a polarisation voltage of 209Vdc. We also discovered the variable resistor on the filament rail was 'stuck' at 26 ohm and there we strapped a fixed value in parallel (actually 3 x 8 ohm resistors) untill we reached 6,18V under load. The microphone has been switched on and in use since three days and it performs flawlessly.
In the meantime I received a schematic for the NU67 with the 03 suffix on the circuit board. It seems no one had tampered with the PSU as I suspected at first but surely the transformer must have been way out of spec. Anyway, it now works as it should. Many thanks for your detailed input Klaus, Kai, JJ Blair and Uwe!
I'll try to attach the early schematic here, it's interesting, with the extra small filter caps etc. compared to the later ones. (I hope I don't break Neumann's forum rules here). Hope it works...
Greetings from Belgium,
Koen.
-
Schematic
-
Congratulations!
You have successfully redesigned the PSU :)
Jokes aside - I suppose Neumann service or more likely anybody else has replaced the transformer with the wrong type once during the long time this PSU has lived.
They must have used what they could get and didn't care about the result, ending up with a mix of different version's parts and a wrong voltage.
E.g. other versions of the NU67 used a single diode rectifier in the high voltage branch, compared to the full wave bridge rectifier in your schematic. This, besides different values of R7 aka R4 (the HV shunt resistor) results in different voltages.
Now this all doesn't matter too much any more - within a reasonable range the resistor values can be adapted to what transformer is there to achieve proper voltages. That's what I proposed and what you did.
The only thing that makes me think "what ..." is your mentioning of the variable resistor being "stuck".
In that case it should either be replaced or unsoldered and cleaned (that's what I usually do if possible) to make it work again.
Cancelling the option to adjust heater voltage might turn out unhandy in the future, e.g if you switch cables. An exact heater voltage is essential to optimize tubelife.
All the best
Kai
-
Most of the trim potentiometers in the NU67 heater voltage rail are stuck because some of the sealing lacquer used to lock the setting in place after factory calibration seeped into the shaft of the switch that turns it.
A simple solution: dab a few drops of acetone into the cup formed by the switch shaft and lock nut. This will usually free the mechanism again.