U47 Clone Distorts Easily

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Phrazemaster

Well-known member
Joined
Oct 2, 2006
Messages
2,010
Location
Southern CA
Hi all,

Just completed a brand new PTP U47 clone, with Redline capsule, EF12 tube, and power supply by Andreas Grosser.

Mic is dead quiet, and has a full lush tone. However with moderate SPL's from singers it begins to sound buzzy or distorted, even before clipping. And it also seems to clip easily.

Could this be a sign of a bad EF12 tube, or some other error in build or capsule problem?

Just wondering your thoughts.

Thanks,

Mike
 
what transformer? what anode current? what bias? what anode resistor value? what grid resistor value?there are many ways to get too much distortion. i have built a couple of EF12's U47 with normal distotrtion. 
 
Xformer: HAUFE T-8094 = 7 : 1 ~ NEUMANN BV-11

Plate voltage is supposed to be ~ 75V (per schematic of known clone design). I need to check this with my meter when I'm in my studio next, but heater B+ and A+ voltage coming into the mic are spot on, which I did check.

Heater voltage is -6V;  A+ is ~120V.

Thanks for your kind thoughts Om3isha.

Mike
 
Of course, you may try another Ef12( all tubes differ).
If all values and voltages are normal, I would try to "play" with
1. reducing polarisation voltage
2. changing the anode current ( higher current =less distortion)
3. Increasing ( to 125-130VDC) anode feeding
 
Hey terrific! Gives me something to go on!

I'll be at my studio again this week or this weekend, so I'll try your suggestions and report back. I did just change the EF12 tube to a "premium" hand selected one, but I don't recall the original tube distorting so easily. I'll have to do some more testing, and I'll report back here finding.

Thanks!

Mike
 
I do have another question. I used an original Bosch MP capacitor, 1uF for output capacitor. I tested it; it tests roughly 1uF. Doesn't mean it's good, of course.

But could an old capacitor cause distortion too? I have another I can use in its place instead...

Thanks,

Mike
 
Old capacitors can leak, but I don't think this will lead to increased distortion. The most simple - solder a capacitor of a small capacitance in parallel with your great. If no significant changes will occur, something with a big capacitor is all right.
 
1:clean all the HiZ parts first with isopropyl alcool
2:measure the capsule polarization voltage , should be around 60V
3: measure bias , should  be around  -2,2v  (before 220MOhms)
4:measure anode voltage after 47K : should be around  75V with EF12

ifyou have bad measure, then  :

2:HT is the culprit, or you have to avoid 80V (1/2,2mohms) if you choose it before ... return to 1/1Mohms, or 0,01uF is leaking...
3:heater psu is the culprit (-6v should be stable) , or divider bridge contains wrong resistor values , or 100nF is leaking...
4:tube is bad if psu is good

check also twice the transformer wiring
but 7:1 ratio is perhaps a bit low for an EF12 ...You could try to wire primary sections in paralel

regards
Fred
 
Just measured some things. B+ and H+ are spot on, at -6.00V and 120.00V.

If I'm measuring correctly, it seems my bias is off. It seemed to read about -3.1 or -3.2. I'm wondering how to lift it to the requisite -2.2. If I read the transfer graph correctly, this makes a huge difference in output. I'll post back when I have more to share, but it looks like I'm  going to need to install a trimmer temporarily to find the right value to affect the bias point. Maybe I can post a partial schematic, but I hesitate as it's not my schematic to give away. I'll see what I can do.

Thanks for the help guys.

Mike
 
A bad capacitor can indeed cause this sort of distortion.  I usually build first with new caps, then experiment after it's up and running.

Good luck,
Paul
 
Replaced the old Bosch with a brand new cap.

Also discovered a bias resistor had failed open (had tested prior to installation). Replaced it too.

Still bad distortion. Tried 3 different EF12k's.

Getting frustrated and possibly at the limit of my skills.
 
granger.frederic said:
1:clean all the HiZ parts first with isopropyl alcool
2:measure the capsule polarization voltage , should be around 60V
3: measure bias , should  be around  -2,2v  (before 220MOhms)
4:measure anode voltage after 47K : should be around  75V with EF12

ifyou have bad measure, then  :

2:HT is the culprit, or you have to avoid 80V (1/2,2mohms) if you choose it before ... return to 1/1Mohms, or 0,01uF is leaking...
3:heater psu is the culprit (-6v should be stable) , or divider bridge contains wrong resistor values , or 100nF is leaking...
4:tube is bad if psu is good

check also twice the transformer wiring
but 7:1 ratio is perhaps a bit low for an EF12 ...You could try to wire primary sections in paralel

regards
Fred

Just address the points, Fred...

1) cleaned all Hi-Z multiple times very well.
2) polarization V is around +77V or so. Schematic calls for ~80V
3) Bias. Hard to tell on this. Measuring immediately prior to the 220M resistor (I am using a 200M instead) I have about -1.1V. After a 10M resistor I have -2.1V. Not sure which is the correct point to measure, but I can't make the point just before the 200M resistor to be -2.2V without drastically changing the values of the voltage divider resistors. Makes me think something must be wrong, or I am measuring it correctly after all.
4) Anode voltage is around +72V
5) What is HT? I am using a voltage divider per schematic of 1M/2.2M. The .01uFuF is brand new Mundorf
6) PSU puts out a very stable -6V; it's built by a 3rd party vendor. Divider bridge has right resistors. 100nF is a brand new V-cap teflon.
7) Have tried 3 different tubes; all the same problem: distortion even at low amplification levels. It actually sounds like a kazoo! Like the capsule membrane is overloaded, yet even at low (below clipping) preamp levels. Normal to soft speaking/singing sound OK, but not stellar.

Regarding Om's suggestions:

[quote author=O3misha]Of course, you may try another Ef12( all tubes differ).
If all values and voltages are normal, I would try to "play" with
1. reducing polarisation voltage
2. changing the anode current ( higher current =less distortion)
3. Increasing ( to 125-130VDC) anode feeding[/quote]

All values are normal except I'm not sure if I got the bias voltage right; all resistor values are right and all the rest of the voltages are right.

1) I tried reducing polarization voltage. Schematic calls for 80v; I reduced to 72V no difference - still distortion.
2) I tried the opposite - reducing anode voltage, which didn't help. I'll see what I can do about trying a higher value
3) I was afraid I might fry something to go this high, so haven't yet tried 125-130V.

Thanks for any more thoughts. I have reached out to A.Grosser to see if he has any other thoughts as well. I may have to send this to someone, but I'll keep trying any suggestions that come in. Thanks.
 
Also, your tube socket may cause distortion. I had same problem with chip chinese socket. Try to use Preh or Lenko sockets they are much better. Or, if you use chinese socket, then change contacts  ( use silver coated ones).
If reducing polarisation to 70VDC is not working- forget it. Try to contact Andreas Grosser. If you built the mic very carefully and accordingly to A.Grosser schematic, then I am shure, that  it is contact problem.
-2.2 after divider before 10M -is ok.
72VDC anode voltage-is ok,
60-80VDC for Red Line Thiersch- is ok
 
Thanks O3m!! Much appreciated!

Andreas emailed me some instructions to try, as well as explained I need an electrostatic meter to test in the bias around the high Meg resistors.

I requested to buy another tube socket from him as well, per your suggestion (thanks!). My socket did get quite loose after a few times working in/out the tube. I accidentally broke the FleA one! I had put the socket in/out a few times in the past, but didn't do it prior to the build. Then it was SO tight that I had to use excessive force (mistake!) and broke it. I had another cheap one that I used, so your thoughts are timely. Andreas emailed me about getting a new socket from him, an old German one with steel springs that will not get loose immediately...gotta love the old German stuff!

Will update on progress as time goes on...
 
i doubt it 's the socket

i think it's the bias ( too low) or the out put transformer

you CAN measure the bias before the 200MOhms with a good standard multimeter.

if the bias is ok then the 7:1 is two low for an EF12 ( never tried ) or worse : the primary is a bit shorted and the ratio is in fact  even lower !
try to input 1V sinus 1khz and measure the output voltage with a 1k load resistor at the output, then you will know the real ratio

in the same time, you can de-magnetize it : increase the input voltage  (100Hz sinus) till saturation , wait 30sec then slowly decrease to 0v.

cheers
Fred
 

Latest posts

Back
Top