R/E/P Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: 1 2 [All]   Go Down

Author Topic: KM56c problem  (Read 1890 times)

afterlifestudios

  • Newbie
  • *
  • Online Online
  • Posts: 41
  • Real Full Name: John Raham
KM56c problem
« on: December 18, 2016, 12:58:53 am »

Hi all.  My km56c suffered a fall from its shock mount.  Here is the series of events that followed, and the current condition.  Any help in troubleshooting is very welcome and would be greatly appreciated.

The mic fell from a height of 4 feet onto thin carpet over concrete.

Immediately afterward, the output was high in noise and low in signal.

I opened up the mic's amplifier to inspect, them reassembled and powered up the system.  The signal seemed strong for a couple of hours. After that, intermittent "wind" noise with a large low frequency component would creep into the signal.  Also, a crackling sound occurred when handling the mic.  I did the "breath test" and nothing seemed to change.  I powered down the PSU and a prolonged loud burst of noise occurred for 8-10 seconds until the power had drained from the PSU.



I opened the mic up again and reoriented (or replaced where necessary) the plastic "shields" that isolate the amplifier circuitry from the inside of the mic's housing tube.  Then I reassembled and tested.  The crackling noise while handling had stopped.  The signal seemed strong momentarily.  Then, I switched the pattern to omni, and the intermittent noise returned.  I "tapped" the microphone housing and the intermittent wind noise subsided.  But moments later it returned.  I tapped again.  Same thing.  I tapped again and the intermittent noise stopped, but the signal went low again and the noise floor was very high (but steady).  It stayed that way...

So I went deeper, and removed the capsule from the amplifier, and then from its protective grill and housing.  The capsule looked remarkably clean compared to say my 87's or 67, but I believe that there is less voltage on the km56 capsule and therefor it attracts less particulate?  (photos attached)  I couldn't see any obvious breaks in the tiny leads coming from the diaphragm down into the capsule.

So I reassembled again, and tested.  Same low signal and high noise floor.  At this point I also noticed that the actual polar pattern was not changing when switched.  The pattern selector ring correctly moves the pattern switch fork, and you can hear the contacts switching, along with the typical "pop" in the mic's output signal, but the polar pattern remains the same.  The pickup pattern is very much front facing, almost like hypercardioid.

Then I checked the PSU voltages (under mic load). I measured 4.0v and 120v on the correct pins.  I originally panicked when I observed 46v on pins 1 and 5, but for testing I had plugged the mic into the console which has global phantom.  (Should I be really seeing that?) 

Then I examined the km56 cable.  I noticed that pins 3 and 6 (ground and "measurement input") were shorting, but only on ONE end of the cable!  So I inspected further to find that pin 3 on one end was off its wire.   I soldered it back on and have continuity on each pair of pins now, as well as pins 3 and 6 shorted.
 
I plugged the mic back in but did not reconnect all the shields to the connectors etc yet.  Tested the mic with the same results.  As I type this, I realize that maybe I DO NEED to reconnect the shields to connectors etc to properly test the mic...

Anyway, that's where I'm at...  Anyone have any thoughts?  Or any recommendations on who could work on it?  I feel like I'm stretching my skill level here to say the least!

Thank in advance.
John


Logged

Kai

  • Sr. Member
  • ****
  • Online Online
  • Posts: 251
Re: KM56c problem
« Reply #1 on: December 18, 2016, 08:56:11 am »

Be sure the tiny screws on the body and capsule housing give good contact.
Else it might be a problem with the pattern switch, or a broken solder joint, or very unlikely something with tube or capsule.
If you switch to figure of eight, does the noise disappear? Then it is the pattern switch.
Remove the capsule, if the noise disappears, something in the capsule is broken.
Remove the body tube, listen to the output while tapping parts with a little plastic pin to identify bad solder joints or mechanically broken parts.
Measure all internal voltages.
Measure the voltages at the output of the pattern switch.
Here is a schematic of the KM256, which is identical except for the connector pinout.
Logged

afterlifestudios

  • Newbie
  • *
  • Online Online
  • Posts: 41
  • Real Full Name: John Raham
Re: KM56c problem
« Reply #2 on: December 19, 2016, 12:04:59 am »

Thank you, Kai. 
Mine is the "c" version.  So the schematic is slightly different...   (I tried to attach the km56c schematic, but pdf's are not permitted.)

Anyway, I have good contact on the tiny screws on the body and capsule housing.  And the cable is now fully reassembled.  Pins 3 and 6 and shield are all tied together, except pin 6 on the PSU end is not connected.

I measure 38v and 79v at the pattern selection switch output on the cardioid and figure of eight respectively.  That seems pretty close.

I also measured the pins where the capsule connects to the amplifier (with capsule in place) and on one pin I measure 4v in fig 8, 2v in cardioid and 0v in omni.  So the pattern switch IS changing the voltages going to the capsule, but I'm not sure if those voltages I'm seeing are correct?  And I'm still not getting any actual change in the polar pattern.

I measure 3.9v right at the the heater on the tube.
I measure 1.6v at "k" on the tube.
121v on the plate.

The low signal, but high noise does not disappear when switched to figure eight.  The noise appears perhaps slightly higher in both omni and figure 8 compared to cardioid.

So just to recap, the intermittent wind noise is gone, but the output of the mic is still low and thin and noisy.  (about 24db down from a km84, and very thin and noisy).  And the polar pattern does not seem to change when switched.  Handling noise etc seems like "normal" levels similar to before the mic fell.

To me, it's looking more and more like the capsule itself?  Anyone have one they want to send me to try out? ;) ;)

Or could someone measure at the capsule connection pins and let me know what they get?  Or can anyone think of other helpful measurements I could take and post here?

Thanks again,
John





Logged

panman

  • Jr. Member
  • **
  • Online Online
  • Posts: 56
Re: KM56c problem
« Reply #3 on: December 19, 2016, 06:25:04 am »

Just measure the capsule! Both sides should read 45pF. If it is something else or not at least very close, then the capsule is damaged.
All the best!
Esa
Logged
Esa Tervala

Kai

  • Sr. Member
  • ****
  • Online Online
  • Posts: 251
Re: KM56c problem
« Reply #4 on: December 19, 2016, 05:05:25 pm »

Measure the amplifier, it would be best if you have some AC meter handy  to measure absolute values:

How to measure the electrical frequency response of a condenser mic's amplifier:
Feed in a sine wave generator's signal where the capsule is connected, via a ca. 1nF cap (exact value not critical, no elco, use a foil or ceramic)
Capsule can stay in place, no soldering necessary.
Level should be about 40mV.
Connect the mic to it's usual record path and use whatever meter is availabe, while tuning the generator frequency from 20Hz-20kHz.
If:
- the response is within a few dB from 40Hz to 15kHz,
- no audible distortion is present (most mic's can take up to 200-800mV without distortion, check specs),
- the output level is about what you get from a very loud voice from 20 cm distance from a known working microphone,
the amplifier is likely not broken.
Most condenser mic's amps have a gain of about 1, this means what goes in comes out.
So if you have the possibility to measure mV voltage you can check for that.
Logged

uwe ret

  • Newbie
  • *
  • Offline Offline
  • Posts: 34
  • Real Full Name: Uwe Sattler
Re: KM56c problem
« Reply #5 on: December 19, 2016, 08:14:41 pm »

Quote
I measure 38v and 79v at the pattern selection switch output on the cardioid and figure of eight respectively.  That seems pretty close.

I also measured the pins where the capsule connects to the amplifier (with capsule in place) and on one pin I measure 4v in fig 8, 2v in cardioid and 0v in omni.  So the pattern switch IS changing the voltages going to the capsule, but I'm not sure if those voltages I'm seeing are correct?  And I'm still not getting any actual change in the polar pattern.

I measure 3.9v right at the the heater on the tube.
I measure 1.6v at "k" on the tube.
121v on the plate.

To me, it's looking more and more like the capsule itself?  Anyone have one they want to send me to try out? ;) ;)


Correct, except for the plate measurement, these voltages are close enough.

The voltage measurement at the capsule is determined by the ratio of the resistance of your volt meter (typically 10MΩ) and the sum of this resistance with R1 + R2 (=10MΩ  + 150MΩ + 30MΩ = 190MΩ)!

Filament and cathode voltages are correct. However, with around 0.4mA current through the 200kΩ anode resistor R9, the plate voltage at the tube should be close to 40V! Make sure to take this measurement at the appropriate side of R9.

Double check for correct wiring of the capsule. It is between rather difficult to unlikely to find a working KK56 or KK88.

Attached please find the correct schematic diagram for the KM56c.
Logged

afterlifestudios

  • Newbie
  • *
  • Online Online
  • Posts: 41
  • Real Full Name: John Raham
Re: KM56c problem
« Reply #6 on: December 19, 2016, 08:35:36 pm »

Thanks gentlemen. My mistake. I am getting 121v at the connector pin #4, and 40v at the tube. 

I'm not sure how to measure the capsule for the 45pf.  (Measure on the pins?  Capsule connected?  System powered on?) I'm also not sure how to check that the capsule wiring is correct. However, the system was working quite well until the fall from the shock mount, and I haven't rewired anything...

Thanks again,
John
Logged

uwe ret

  • Newbie
  • *
  • Offline Offline
  • Posts: 34
  • Real Full Name: Uwe Sattler
Re: KM56c problem
« Reply #7 on: December 19, 2016, 09:50:58 pm »

With all the voltages correct, it is most likely that the capsule is at fault. To measure the capacitance (at both halves) of the capsule will require a capacitance meter and the capsule disconnected from the circuit, at which point it only matters for safety whether it is powered on or off (OFF!!!).
Logged

afterlifestudios

  • Newbie
  • *
  • Online Online
  • Posts: 41
  • Real Full Name: John Raham
Re: KM56c problem
« Reply #8 on: December 21, 2016, 03:34:57 am »

So far I've only been able to check capsule capacitance with my cheap DMM. It only shows two decimal places of "nf" so I was hoping to see 0.04nf or 0.05nf. But I got 0.00nf or 0.01nf across any pin combination.  I also get 0.01nf measuring my finger, or measuring... the air!

I am borrowing a capacitance meter shortly and will let you know my findings.

But if the capsule is "dead", what died?  And what can I do about it?!  Could it be a crack in a lead from the diaphragm?  I know that Thiersch "reskins", but does anyone "rebuild"?

Thanks again for all the help so far.

John


Logged

Jim Williams

  • Hero Member
  • *****
  • Online Online
  • Posts: 547
Re: KM56c problem
« Reply #9 on: December 21, 2016, 11:35:35 am »

It's possible only one side is damaged. If so the working side can be used in a cardiode only setting. Some have done that with damaged capsules, they stuck them in cheap Chi-com bodies. I did one for Ty Ford years ago, I put a 1/2 working U-89 capsule in a donor mic, it sounded wonderful.
Logged

klaus

  • Moderator
  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1426
Re: KM56c problem
« Reply #10 on: December 21, 2016, 02:16:59 pm »

But if the capsule is "dead", what died?  And what can I do about it? I know that Thiersch "reskins", but does anyone "rebuild"?

Neumann's nickel capsules cannot be repaired, only re-diaphragmed, with OK, not spectacular, results from Thiersch, who is the only guy that does it.
Logged
Klaus Heyne
German Masterworks®
www.GermanMasterworks.com

afterlifestudios

  • Newbie
  • *
  • Online Online
  • Posts: 41
  • Real Full Name: John Raham
Re: KM56c problem
« Reply #11 on: December 22, 2016, 09:57:42 pm »

I got my hands on a capacitance meter...  I'm not sure I'm going about this correctly, but I measured capacitance between each pair of pins on the bottom of the capsule housing and I get:

A.) 1.07nf
B.) 1.08nf
C.) 0.00pf (And I mean ZERO.  Normally, with the probes just dangling in mid air, the meter reads a few pf.  But when I measure between this particular pair of pins, it reads 0.00pf.

I see a 1000pf cap on the schematic (in the capsule), so that would account for my 1.08nf readings, I guess?

Am I measuring in the correct place?



Logged

panman

  • Jr. Member
  • **
  • Online Online
  • Posts: 56
Re: KM56c problem
« Reply #12 on: December 23, 2016, 04:55:17 am »

From those figures it looks like one side of the capsule is shorting(C). A and B would then be the sum of 1000pF and the other side. That`s the only way I can explain those measurements. It should actually read 1045pF, but the cap is never exactly 1000pF especially after so many years. It`s been a while that I have taken a k56 apart, so I cannot start guessing, what could cause the short.
Esa
Logged
Esa Tervala

afterlifestudios

  • Newbie
  • *
  • Online Online
  • Posts: 41
  • Real Full Name: John Raham
Re: KM56c problem
« Reply #13 on: December 24, 2016, 03:11:06 am »

So looking at the schematic of the capsule, I belive the short must lie somewhere between pins 2 and 3 (if we call pin 1 the uppermost pin on the schematic page.). This would explain the 1000pf extra I'm seeing between pins 1 and 2. Because under normal conditions I would expect to see just the 45pf, right?  I would expect to see 1000pf plus 45pf between pins 1 and 3. But only 45pf between pins 1 and 2, as well as between pins 2 and 3. 

Does anyone have the time and courage to measure their k56 pins with a capacitance meter and share their results?

Also, the absolute zero pf reading between pins 2 and 3 is still a little baffling. There's no other material I can measure that gives me 0.00pf on the meter. Not even just touching the probes together...  Is this any kind of clue?

Thanks again for your continued help on this...

Logged

afterlifestudios

  • Newbie
  • *
  • Online Online
  • Posts: 41
  • Real Full Name: John Raham
Re: KM56c problem
« Reply #14 on: December 29, 2016, 07:19:20 pm »

Still trying to solve this one...  I have two questions:

1.  How can one determine the fault within the capsule?  I don't really want to re diaphragm if the problem lies elsewhere within the capsule... 

2.  Would a capsule from a km88i be an exact "drop in" replacement for the k56 in a km56c?

 
Logged

Kai

  • Sr. Member
  • ****
  • Online Online
  • Posts: 251
Re: KM56c problem
« Reply #15 on: December 30, 2016, 05:28:12 am »

I would ask you to redo the measurements  after making your self familiar with your meters:
First identify the pins, according to the schematic above.
Let's name the pins 1 to 3 from top to bottom.
Now you need to measure capacitance and resistance of each pair.
1-2 C= 45pF, R= >10GOhm
2-3 C= 45pF , R= >10GOhm
1-3 C= 1025pF (+/- 10%), R= >5GOhm
As your meters probably cannot measure higher than several hundred MOhm, simply check if they read "no conductance" in the highest  resistance range.
If there is an Ohm reading value on any pin combination your capsule mount is faulty.
As this does not necessarily mean the capsules themselves are broken, you should consult a very skilled person to fix it.
I would strongly suggest not to mess around by yourself, as these precious diamonds are not repairable, it's very likely you finally break them if you try.
Logged

afterlifestudios

  • Newbie
  • *
  • Online Online
  • Posts: 41
  • Real Full Name: John Raham
Re: KM56c problem
« Reply #16 on: December 31, 2016, 06:13:56 pm »

Thank you, Kai.  Here are the results of my capsule measurements according to your pin out nomenclature.

pin 1-2 C = 1079pf R = no conductance 
pin 2-3 C = 0pf, R = 3.8 kOhm
pin 1-3 C = 1079 pf, R = no conductance

Clearly there is a problem.  I would love to send it to a "very skilled person" to fix it.  As far as I know from comments made earlier in this thread by Klaus, Thiersch only reskins diaphragms for these capsules.  Based on the photos of my diaphragms and the readings I'm getting on the capsule pins, does  it really seem like a diaphragm problem?  If so, great, I'll send it to Thiersh.  If not, does anyone have a recommendation for a "very skilled person" willing work on it?
Logged

uwe ret

  • Newbie
  • *
  • Offline Offline
  • Posts: 34
  • Real Full Name: Uwe Sattler
Re: KM56c problem
« Reply #17 on: January 01, 2017, 02:36:12 pm »

It looks as if your capsule head has an internal problem between the back plate and the rear diaphragm. Kai was close with the expected measurement values. The actual values between either diaphragm and back plate should be 70 pF (35 pF in parallel with the series combination of 1000 pF and 35 pF), and the capacitance between the two diaphragms should be 1018 pF (1000 pF in parallel with the series capacity of the two 35 pF capsule capacitance). Resistance should be infinite for measurements between any of the 3 terminals. As is the capsule could still be used in cardioid only mode.
Logged

afterlifestudios

  • Newbie
  • *
  • Online Online
  • Posts: 41
  • Real Full Name: John Raham
Re: KM56c problem
« Reply #18 on: January 01, 2017, 09:46:51 pm »

Thank you, Ewe.  Indeed, that's what it seems.  But I don't have a healthy cardioid "as is".  Perhaps if I completely lifted the rear diaphragm from the circuit, then the front side would be normal cardioid?  (I'm not ACTUALLY going to do that!)  But right now it definitely is cardioid only, but very low level.  Handling or tapping mechanically on the mic housing gives very full, loud clear results.  Just not from the capsule. 

Could the fact that the rear side is shorted to the back plate be causing that problem with the front side?

John

 
Logged

afterlifestudios

  • Newbie
  • *
  • Online Online
  • Posts: 41
  • Real Full Name: John Raham
Re: KM56c problem
« Reply #19 on: April 07, 2017, 07:45:33 pm »

Just a happy follow up here...

Once it had been determined that the problem was capsule related, not amplifier, I sent the capsule to forum member "Panman".  He found the short between capsule and backplate and corrected it.


I got the capsule back from him today and it sounds marvellous in all 3 patterns. Output level in cardioid is generally on par with my km84's.  Noise floor is about 5db higher than the 84's.

Most importantly, it sounds great!  I thought she was a goner...

Thank you, Panman!

John
Logged
Pages: 1 2 [All]   Go Up