Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[soapfoot]

I made a microphone using the schematic linked here, on the left:

http://www.tab-funkenwerk.com/id85.html

The tube I used is an EF800.

Sonically, the mic has performed exceptionally well.  The transformer is Oliver's BV8R, the capsule is a Thiersch Blue Line M7, and the tubes I've tried have all been NOS Telefunken EF800. The power supply allows me to trim voltages to ca. 5 VDC for the filament/bias supply and 105V DC for the plate/polarization supply.

I'm thrilled with how it sounds.  However, despite everything seeming 'right,' something is wrong-- tubes tend to fail too often.  I regularly check voltages and all seems right, both within the power supply and the mic itself. I've tried replacing R1 on the schematic (grid leak resistor), and that didn't fix the problem.  Several times I've carefully cleaned the tube socket and all non-conductive parts with 99.953% pure isopropyl alcohol. Some tubes last longer than others-- this past one endured for a solid month before exhibiting the telltale rustling sounds I associate with grid contamination. Others fail within 24 hours. I've tried over 10 tubes... I think somewhere between 12 and 15... and while some were unsuitable right away, others started out fine and then eventually began behaving in a manner consistent with grid contamination.

Where should I look next?  Could a leaky C2 cause this problem?  Could the capsule be defective and leaking DC from the backplate to the diaphragm somehow? Some other problem? Any ideas?

Thanks!

[soapfoot]

In case anyone is watching, I have a lead I'd like to follow up on/ask the good people here.

While looking up possible causes of my noise, I saw one person seem to advise against cleaning on or around polystyrene capacitors with isopropyl alcohol. This thread, post #3: http://www.groupdiy.com/index.php?topic=38658.0

Well, my C2 is polystyrene, and I did indeed clean (directly on) the cap with pure isopropyl.  Could this have caused the cap to go leaky, causing my problem?

[klaus]

To get right to the core of your issue:
You do not mention how many tubes failed, how exactly they failed (blew up? got noisy? frying sound...?) and whether they were all from the same shitty batch, or whether they were widely sourced Telefunken NOS versions. Read this stickie, one of three I wrote, for an idea how hard it is to select microphone tubes: http://repforums.prosoundweb.com/index.php/topic,1141.0.html

Aside of a bad batch of yubes that may have spoiled your enthusiasm, looking at the TAB schematic, and if you implemented it exactly with the values listed there, I see a couple of voltage oddities which I would check with Oliver first to confirm that they would specifically apply for an EF800, rather than an EF14.

For example, I cannot believe that the EF800 penthode would be happy with such severe under-heating as 5V. instead of 6.3V. These low voltages would be ok for an EF14 but I would assume that 5V on an EF800 would cake up the cathode quickly.

I would not be too concerned about super-cleanliness in the high impedance circuit area as advocated by some on the internets. I have learned through the years from intentional personal experimentation and sheer sloppiness that this warning sounds rather professional and gives an aura of expertise to the advisor or such warning as it sounds quite logical; but in my experience, a little dirt here and there has little bearing on the noise floor of a mic. (Crucial exceptions excluded!)

As to polystyrene and 99% isopropyl alcohol: Make tests on a capacitor to see what happens. I am incredulous that alcohol would harm the cap as it is well protected by a hefty layer of plastic before you would be able to dissolve material down to the first film/foil wrap.

Check in with Oliver and post again!

[soapfoot]

To get right to the core of your issue:
You do not mention how many tubes failed, how exactly they failed (blew up? got noisy? frying sound...?) and whether they were all from the same shitty batch, or whether they were widely sourced Telefunken NOS versions. Read this stickie, one of three I wrote, for an idea how hard it is to select microphone tubes: http://repforums.prosoundweb.com/index.php/topic,1141.0.html

Klaus, first of all, thanks for your time.  I'll try to clear up some of the questions.

The tubes were widely sourced.  I had between 12 and 15 fail so far, the majority from very strong batches known to produce many good tubes--All NOS Telefunken. This is why I feel something is wrong with my microphone.  Five were from a reputable person whom I trust (and whom you know personally-- can reveal who in a PM if you want) with a known-good stock who had some good success using this particular batch in microphones-- better than 70% success rate, he said.  Two were from a well-known retailer, quality of stock unknown.  Still others were from Ebay, from an old bulk stock of unknown quantity.  Finally, three were packaged-for-retail versions from two different eras (different box designs; two of the common red/blue design, and one an older white/olive scheme), all still in the old shrinkwrap, untouched since packaged for retail.  These three lasted longest.

Failure mode is always frying/rustling sounds.

I have experimented a bit with the heater voltage.  Like you, I suspected that the underheating may be causing the problems, so I cranked the heater voltage up to 6V.  Oddly, in this case, a higher heater voltage seemed to kill tubes even faster. Then, on the advice of a local friend quite knowledgeable in vacuum tube construction and application (owner of Leeds Radio), I tried the opposite direction-- underheating down to about 5v, or even a touch below.  When underheated in this way, I got a few tubes to last several weeks to a month, but just when I would start to feel really good about it and think I had the problem solved, the rustling/frying sounds would come back.

I have been in touch with Oliver about this mic and he's been helpful-- it was he who suggested I try replacing R1 (which did not solve the problem, though I thought it had). I apprised he and Joe of all voltages I'm measuring inside the mic and the power supply, and neither has voiced any concerns about the ~5v heater supply.  I do believe they're quite busy over there right now, so if I can ever get a minute to chat with them again about it, I'll certainly ask about that again.

Thanks so much for giving your advice. If anything else comes to mind, I'm certainly listening. I have read your tube selection FAQ several times, and it's great info that I live by-- it's just that this time, I feel like something else must be wrong.

[klaus]

P.S.: I also cannot share your initial suspicion that there may be possibly something wrong with the value of the grid-to-ground resistor. A value of 100MΩ to 1Gig Ω is quite common in this particular circuit application.

How about this: take one of these tubes from the batch that has already produced several duds and install it in another mic, to verify its quality.

[soapfoot]

Unfortunately, I don't have another mic that uses the EF800.

Today I replaced C2, found/cleaned a bit of flux residue on some other components that I had missed, touched up some solder joints, put in a fresh EF800, trimmed voltages, and put it back together.  It's up and running now, but for how long?

With the fresh tube, I measured plate voltage and cathode voltage with respect to ground, and made a note of those.  If the tube starts to fail again, I can at least compare measurements and see what changed. I also nudged the heater voltage up a few tenths of a volt.  It's still underheated, but not quite as dramatically.

[klaus]

With an EF800, what cathode voltage do you measure on average in that circuit? Should be somewhere between 1.2 and 1.8 VDC.

[soapfoot]

I measured 1.215

[klaus]

What is Oliver's opinion on this? He is the designer of the kit, and has been using EF900 extensively in his mics.

[soapfoot]

His opinion was that I should try changing the grid leak resistor, which I did.  I haven't been able to get in touch with him to inform him of the newest developments. I believe they're quite busy over there.  I'll try calling or e-mailing again next week, but I don't want to become a pest.  They've already helped me out a lot.

P.S., I was told that my microphone malfunctioned again today, in very short order.  So to recap what has been done so far:

R1 and C2 have been changed.  Several NOS Telefunken tubes have been tried.  The mic has been carefully cleaned of all fingerprints and flux residue with 99.953% pure isopropyl, except for the new polystyrene C2, which I was advised not to clean in such a manner.  All solder joints in the top portion of the mic (between capsule and tube) have been inspected and, if necessary, reflowed.  Tube pins were carefully cleaned.

Not sure what my next steps are.  Tomorrow I'll be back at the studio and I'll be able to get some measurements of the tube in its current (malfunctioning) state to compare with the (working) measurements I took yesterday.  Perhaps this will give me some clues.

Hopefully I can speak with Oliver or Joe again soon, too.

[klaus]

The problem for me is that this circuit is somewhat unusual in its layout and its values used.
I would probably have started out by using a n EF86 with a 6.3 V heater, a cathode bypass resistor which would dial in about 1.6 to 1.8 volts, and a plate resistor/voltage divider that gets me around 75V plate voltage.  I would even try that set up with an fresh and proven healthy EF800. I am not familiar with a situation where any of the components you are using could deteriorate and destroy the tube, as you have been experiencing. It would help an awful lot if you had a friend who also is using this unusual tube in a condenser mic.

A friend of mine, Scott Hampton (scott@hamptone.com) experiments with DIY condenser mics and is quite knowledgeable. You may want him to read this thread.

Best of luck,
KH

[soapfoot]

The problem for me is that this circuit is somewhat unusual in its layout and its values used.
I would probably have started out by using a n EF86 with a 6.3 V heater, a cathode bypass resistor which would dial in about 1.6 to 1.8 volts, and a plate resistor/voltage divider that gets me around 75V plate voltage.  I would even try that set up with an fresh and proven healthy EF800. I am not familiar with a situation where any of the components you are using could deteriorate and destroy the tube, as you have been experiencing. It would help an awful lot if you had a friend who also is using this unusual tube in a condenser mic.

A friend of mine, Scott Hampton (scott@hamptone.com) experiments with DIY condenser mics and is quite knowledgeable. You may want him to read this thread.

Best of luck,
KH

Thanks, Klaus.  I'll certainly give Scott an e-mail and see if he has time to read this thread over.  Much appreciated,

--brad

[ryanstreber]

I'm following this thread with interest, as I'm basically done with one of the AMI 47 kits that I built with the intention of using an EF800.  I'm now just waiting for a PSU that I ordered from AMI to fire it up, so we'll see how it goes; but I'm also wondering if I'm in store for problems with tube reliability as well.

I had already picked up a number of EF800s because I have a Wunder CM7-gt that also uses that tube.  I haven't opened that Wunder up to investigate yet, but I'm curious to see how similar / dissimilar it is to Oliver's circuit. 

In any case, Brad if you know anyone with a CM7 GT, that would be a possible other mic to test some tubes in. I have one just north of the city if it's at all useful to you to check out...

[soapfoot]

Thanks, Ryan, for chiming in!

Yes, I'd be very interested to hear a few things about the CM7GT if you're the curious type-- namely, plate voltage, heater voltage, and bias voltage.  If there's any chance for you to check those things at some point, I'd be eternally grateful!!

[Kai]

P.S., I was told that my microphone malfunctioned again today, in very short order. 

Are you sure it's the tube, not the capsule?
Can you replace the capsule by a little capacitor and still hear the noise, using one of those "broken" tubes?

BTW: you can "refresh" a tube by overheating it for a short time. You don't have something to loose in this case, you can just try how much and how long you need to.

I have advised for cleaning the high impdance path in another tread here - I still vote for that when building a mic (partly from dirty, vintage parts like old tube sockets), but doing it over and over isn't necessary.

Thereare 2 things I dislike in the circuit:
- The grid resistor I would use 60M instead of 1G for a more stable bias.
- The heating arrangement: I would feed 6.3V into the other side of the heater (now on 0V) bypass R2 and let R3 determin the amount of underheating (Voltage on R3=1,2V). It would make a clearly defined 0V rail instead of two.

These mods would come closer to the proven (with another tube of course) U47 circuit.

Regards
Kai