U47 Clone Distorts Easily

[Phrazemaster]

Hi all, I got this advice from a venerable mic tech:

about distortion
please send 0dBu over 75pF to the grid
measure the output level with a load of 1K
and control the waveform with a scope
then inform me please

So I bought a 75pf capacitor, and 1K resistor. I found an internet calculator that shows 0dbU is about .77v RMS or 2.19V p-p.

I thought I would get my signal generator, set it for a 1K sine wave at .77V RMS, and attach its output to the grid point of my EF12K capsule. Other wire of signal generator would go to ground?

Where would I attach the 1K resistor then? Not quite sure, and I don't want to bug this very kind and venerable mic tech with stupid questions.

And would I connect the mic to a preamp before measuring the output on my scope? How do you measure it? Across what?

Sorry if this is basic. Thanks for any thoughts.

Best,

Mike

 

[Phrazemaster]

Wow Fred, you posted almost the same suggestions just as I posted them! Thanks!

The transformer was specifically suggested by Andreas, so I trust it's right for the EF12.

Thanks for any help in translating how to do this, as in my previous post.

Mike

 

[o3misha]

BTW: record the sample of voice or voiceover . I'd like to listen your mic and distortion . 

 

[Marik]

If your mic has a high output your distortions also might come from overloading your pre input. Try a different preamp, or even better, mixer with variable input gain.

Best, M

 

[Phrazemaster]

Thanks O3m and Marik.

O3m I'll try to get a sample up in the next couple days; maybe even today. I'm trying to do Andreas' troubleshooting suggestions; hopefully I can do that today or tomorrow. Any help in that regard is welcome.

Marik, I've tried 5-6 different mic pres. If I turn the gain down, it still has the problem, so I do think it's in the mic. Also, the distortion happens before 0dbfs, well below. Normal speaking doesn't do it; it's louder sounds.

Thanks guys.

Mike

 

[Pip]

It's starting to sound like the output transformer. Is there any DC voltage on the primary and are all your ground references tight? It wouldnt be the first time either a cap or a transformer was bad out of the box.

 

[Phrazemaster]

Thank-you Pip. How can I test for DC on the primary?

I performed the distortion test Andreas suggested, inputting a 1K sine wave across a 75pF capacitor. Checked output loaded with a 1K resistor.

Head scratcher...waveform is pristine! I mean perfect. I have to crank it waaay up to 1.4v RMS, or 5.2 dBu before the wave distorts on my oscilloscope.

Now I'm not sure I did this right.

I took the head off the U47 clone, and put the 75pF into the contacts which normally go to the capsule. I took the output of my signal generator and attached to each leg of the capacitor.

I put a 1K resistor onto XLR pin2 output in series with my scope positive probe, and used pin1 XLR Output for my scope ground.

I turned on the mic PSU and took a reading.

I'll post the pic here in a few mins.

By the way is the signal supposed to come out amplified after going through the mic? It looks the same or close to the same.

So what else could this be? Capsule problem/connections?

Thanks,

Mike

 

[Phrazemaster]

My methodology...

Did I set this up right?

 

[Phrazemaster]

I just totally tore down my top head basket, and super cleaned everything. I noticed the original FleA connectors had gotten tarnished. So I took everything apart, and used silver polish on everything until it was all bright and shiny. Then I redid the capsuled connections, managing to fudge up my back capsule! $%%$#%$!!!!

An errant fingernail barely scraped it, causing major damage. Thought I was being so careful.  :

Contacting Thiershe for a reskin of the back capsule. Actually the more I think about it, I'm going to have him reskin the whole thing. I suspect the capsule may be the culprit here.

Pic is of the head assembly prior to tarnish removal; note the crappy solder joints to the tarnished silver connections. Rosin didn't remove it; only silver polish did the trick.

In the meantime, I put her all back together, as front side of capsule is fine, and still in cardioid, same problem!  :-\

Here's a sample of speaking:

https://soundcloud.com/phrazemaster/kazoo-like-buzzz-dry

What does this sound like to you?

 

[Pip]

Thank-you Pip. How can I test for DC on the primary?

Too bad about capsule! The way to see if there is DC on primary is simple just measure from any transformer primary leg to ground you should see nothing as the output cap .5MFD should only let AC pass. But that wave form looks solid so it looks like it is the capsule region that is causing problems. Are you using IOaudios body and if so are you using his magnetic reed switch design?

 

[Phrazemaster]

Yeah thanks Pip; I'll get the capsule reskinned  :.  Ironically I think the capsule may be the culprit anyway, so as I mentioned in another post I'll probably have him reskin both the front and the back. Although the front looks pristine (as did the back a few hours ago).

I am using a FleA body. No magnetic reed switch in this case. I do have Max's kit waiting for attention, but I've had these parts for this clone for literally 2-3 years or a bit longer. Hence the tarnish on the connectors in the head basket! They were silver plated, and I didn't realize it when I first installed the capsule. I just couldn't figure out why it was so hard to solder to the connectors! It was much easier after the silver polishing.

I was also scared to try a PTP since I was more of a newbie then, although I certainly still am! But I figured, what's the worst that could happen? Damage a $400 part? Ha! No problem! I'd rather take the risk and screw something up than leave the parts in the bin. DIY takes guts sometimes!

I am not at my studio right now - I've done enough damage for one day! But I will make the test you suggested to find out if there's any DC on the primary. Probably tomorrow!

Thanks for your support Pip, Marik, O3m, Fred, Ph, Matador, and everyone else who's been chiming in! I'll keep you guys updated. This is practically becoming OUR U47 clone lol!

Thanks again,

Mike

 

[o3misha]

what is the bias after divider but  before 10 M resistor( not before 200M!).

 

[Phrazemaster]

-2.1V

What does the buzzing/distortion sound like to you?

 

[granger.frederic]

a good pro multimeter can measure after the 10Mohm , but a cheap one should measure before the 10M

the capsule is maybe the culprit but it's rare ... try with another capsule even a crap chinese K87 one...

i think that the transformer ratio is maybe mismatched: the output impedance of an EF12k in a U47 circuit is near 20k thus you should use a least a 10:1

http://www.moxtone.com/mU47_U47_part3.html

i thought a BV11 was around 10:1 but you said that yours is 7:1 , please measure and calculate the real ratio like i told before ...

 

[Phrazemaster]

Thanks Fred.

The data sheet shows the windings are 680R/17R

As I recall my transformer was about the same.

Maybe tomorrow I can try your test as you suggested.

This is a proven design and the transformer recommended by Andreas for this circuit. It is a haufe and I also bought it from him, so I'm confident it is right for the EF12--at least as implemented in this design.

Perhaps there is a problem with the transformer however, or the capsule?

I am going to get the capsule reskinned to find out.

In the meantime your suggestion of using a cheap capsule is a good one; I appreciate it and maybe I can try it tomorrow.

I'm waiting to hear back from Mr. G on his thoughts which I will share.

Thanks!

 

[Matt Nolan]

I always use a cheap capsule in my mic builds until everything is working well. Only then do I put an expensive capsule in. I would recommend this approach to others!

 

[o3misha]

I used that Haufe transformer two times. Forget about wrong mismatch etc, -Andreas gave you the best. What I heard in your sample: wrong bias point. In the bias divider solder one pot instead of any resistor and  increase bias to -2.3- or even to -2.4. What happens? Did you solder the transformer correctly? The gain of microphone is normal? This kind of problem caused by capsule or tube( not optimal bias, too high or too low anode current or bad tube at all).
Two my microphones are running with -2.3 and -2.4 VDC bias without problems. If you used new Red Thiersch M7, properly installed, Andreas's transformer, tested EF12 tube, good socket( good contacts!) you must build good mic with normal distortion.

 

[granger.frederic]

i'm still on the transformer : did you heat too much when soldering ?
if it's the case you've perhaps damaged the coils and thus change the ratio to something even worse than 7:1 for an EF12
thus you should measure it .
with all the respect  , a near 20k output impedance tube should need at least a 10:1 ratio if you whant to perform optimaly ...
try to put a BV11 after a 6072A (20K also) and you'll have the same issues : loss of extremes ,loss of dynamic and early distortion...
That's why everybody uses a T14 with a 12:1 ratio ...

 

[Phrazemaster]

Thank-you Fred. I'll try to make the test you suggested.

I don't think I overheated the transformer.

I'll report back soon.

Andreas said he didn't think the capsule was an issue, presumably after listening to the vocal sample.

Thanks!

 

[o3misha]

Fred, I have no problem with 6.5:1 or 7:1 ratio and EF12 tube with proper schematic,  proper anode resistor, proper bias etc. "with all respect"- I made two microphones with EF12 without any problem. Andreas, Gunter Wagner, Flea.....of course, you may continue think what you want, but ...And you are wrong about Ri=20k0hm.
http://www.radiostation.ru/tubes/ef12_tf.pdf