MK7 - tube mic project

[nashkato]

  It sounds like the shield isn't working properly. Check for conituity between all outer metal parts. Also make sure you put the mic back together before testing audio quality, or that hum will be the result.

tht really seems to be the problem .
Strange thing is , with PS off continuity is a 0 Ω , if i turn on the PS the resitance goes up to app. 30 Ω ??

does someone have an explanation for this ?

 

[HellfireStudios]

  It sounds like the shield isn't working properly. Check for conituity between all outer metal parts. Also make sure you put the mic back together before testing audio quality, or that hum will be the result.

tht really seems to be the problem .
Strange thing is , with PS off continuity is a 0 Ω , if i turn on the PS the resitance goes up to app. 30 Ω ??

does someone have an explanation for this ?

Where exactly are you testing to get these readings?

-James-

 

[nashkato]

between PS case/star ground  and mic case /or rails/metal parts of the mic.

 

[Kingston]

between PS case/star ground  and mic case /or rails/metal parts of the mic.

Again note that you can't make any kind of hum measurements with the mic body open and you need to make sure the body and the grill are actually adequate shields. I've seen some DIY models with no good grill or even grill electrically detached from the body with paint in between etc. carelessness. 

Also, 100mVpp is a lot. Maybe time to go through ALL connections once more and look for shorts where there should be connections.

And how much is your heater VAC while loaded?

 

[nashkato]

Again note that you can't make any kind of hum measurements with the mic body open and you need to make sure the body and the grill are actually adequate shields. I've seen some DIY models with no good grill or even grill electrically detached from the body with paint in between etc. carelessness. 

yeah i´m aware of this . and yes it´s a DIY body , but this is not the first mic i build with these bodies (several G7 and MK47 ) and never had had any problems like this .
so the body is a adequate shield.

i went throuhg the whole thing several times now . always with the same result .
actually the 100mV i measured first are a result of interference at my workbench , not the actual hum .
i tried to measure once more , but interference is to high to measure anything useful .

my heater VAC is 12,5 V , VDC in PS is 7,6V  and in mic at the tube socket it´s 5,7 VDC with EF86 .
actually all voltages look fine and are throughout spot on .

strangely though i measure between 20 and 100 Ω  and about 20mV at the mic body ( mic closed) to Starground in PS when the PS is on

edit : here are some pics of the actual mic
http://www.flickr.com/photos/82843796@N06/sets/72157633834728247/with/8903751944/

 

[nashkato]

finally i got rid of the hum  ;D
unfortunately i cannot say what i did exactly , cause i changed several things at a time , but i definetly was a grounding issue.
thanks to all who tried to help me .

 

[matta]

matt,

check that the tube socket has continuity to ground. the metal part!


or maybe bad coupling cap like skylar suggests on page 35...


and this, max's post about the self biasing:

"for self bias the gold/super caps can be ommited, for a ef86 additionally the 2,2R and the 3,9 resistor should be left out"

Hi EEM01,

Like you mentioned in one of your previous posts, this project has kicked me in the nuts! I've started to doubt my own abilities, and so like you I put it on the shelf and finally have some time to pick it up again.

Thanks for your suggestions, I've checked the grounding on the tube socket, it's all good, ohms out perfectly!

I'll look into the bad output caps as a last resort, firstly want to clear this up.

"for self bias the gold/super caps can be ommited, for a ef86 additionally the 2,2R and the 3,9 resistor should be left out"

I'm using and EF86, self-biased, can't seem to find where Max suggested leaving those resistors out.

EDIT: Ok, I just found it:

as already noted, a different approach changes the bias method , adding a cathode resistor for self bias and disconnecting the heater rail completely from the audio path - this is the easy way and works with the unmodded stock psu, except for the pattern switch assembly. for self bias the gold/super caps can be ommited, for a ef86 additionally the 2,2R and the 3,9 resistor should be left out.

For my self-bias of the EF86 I'm using the diagram in the first post, replacing the 3R9 resistor as screened with the 2K2, I assume that's the 3R9 Max is talking about?

RE the 2R2, mine is fitted! Am I correct in thinking it should be removed and jumpered? Is that what you've done with your own mic? I also have the 2 x gold super caps fitted as well? Just looking for some clarification?

Really hoping to get to the bottom of this all not that I've taken some time away from it.

I spotted this in an earlier comment from Max.

for tubes with a 6,3v heater (ef86 etc) , the psu has to be modded to supply ~9vcd and two resistors (2,2R* & 3,9R*) have to be

I'm using an EF86 and haven't changed anything on the power supply, I built it as spec, is this the issue, were those changes and resistors documented for those who used EF86 tubes?

@shabtek, the rails are grounded to the PCB via wire from the PCB ground plane to the rails via a soldered wire and lug.

Thanks in advance!

Matt

 

[EEMO1]

hey,

by checking the ground, I meant the eyelets on the screw's where the 68m on one side and 2.2k/10nf/pin2+4 connect.


my tube socket was concaved just right, so there wasn't a connection to ground until I put in a washer in between...

 

[matta]

hey,

by checking the ground, I meant the eyelets on the screw's where the 68m on one side and 2.2k/10nf/pin2+4 connect.


my tube socket was concaved just right, so there wasn't a connection to ground until I put in a washer in between...

Hey EEMO1,

All good there, ohms out as it should. So that's not the issue, sadly. Are you using an EF86 tube? Above Max's talks about changes that need to be made, as you shared in an earlier post.

"for self bias the gold/super caps can be omitted, for a ef86 additionally the 2,2R and the 3,9 resistor should be left out"

Obviously in the case of the self-biased option the 3.9R is replaced by the 2.2K, but in the case of the 2.2R, did you bridge it? I can't see how you can omit it as there is no way for the heater to reach the tube if you don't.

Either way I did remove it and bridged it, but it also doesn't seem to make a difference.

It would be great if you could share what you did with your mic RE those values, changes or omissions.

I went back and looked at my grounding again and lifted a wire here or there and just got hum, so the current grounding seems to be correct as per Emperor Ketchup's diagram, just with no sound  :-[

I just want to make sure that I've made any necessary changes that need to be made running and EF86 before I move on to replacing the output cap that you and Skylar mentioned.

Thanks in advance!

Matt

 

[ioaudio]

as i was re-reading your posts, i found this quote:

The moment I add the EF86 the Heater voltage drops from the 8.1V to 4.2V, I've tried it with a few EF86 tubes and it does the same thing each time.

the heater voltage must be dialed in with the tube in place (or a load resistor with the equivalent resisance)!
Hope that's the solution!

 

[matta]

as i was re-reading your posts, i found this quote:

The moment I add the EF86 the Heater voltage drops from the 8.1V to 4.2V, I've tried it with a few EF86 tubes and it does the same thing each time.

the heater voltage must be dialed in with the tube in place (or a load resistor with the equivalent resisance)!
Hope that's the solution!

Hi Max,

No, sadly not, as I went on to share in another post, the issue there was that the power lamp was wired into the heater output which kept dropping the voltage, the moment I removed it the voltages corrected themselves and I was able to dial in the correct heater voltages.

What worries me is as I shared above, you share that the 2.2R resistor and 3.9R should be omitted with an EF86 and the self-bias, my 2.2R resistor is fitted, though I did remove it and bridged the trace with a piece of wire, as I can't see how if omitted the heater voltage can move from the heater pad on the PCB to the mic amp/tube socket. Still no improvement.

As you requested I've just mailed you and shared that if my understanding is correct to measure the DCR reading I need to ohm out between Pin 2 on the XLR and the transformer primary, after the coupling cap and Pin 3 on the XLR and the same junction, in both cases I have nothing, the meter reads open.

I did triple check and I have continuity from the between the XLR Pin 2 and transformer secondary at the PCB where marked A+ and XLR Pin 3 and the transformer secondary at the PCB where marked A-

Hope that helps!

Cheers

Matt

 

[matta]

Right, just so we are on the same page I decided to take all my voltage readings and put them against Max's, I'm using an EF86 and have it set up for self-bias, all the voltages are loaded voltages.

I do still have the 2.2R fitted and the 3.9R is replaced by the 2.2K as shown in the above self-bias image and am still unclear about what to do with them, as Max shared, don't stuff them, but then I assume one needs to jumper the 2.2R.

My heater is a little low by comparison, but seems to regulates just fine for the required voltages.

Any thoughts?

EEMO1, on your recommendation I pulled the output cap, but it tests fine and have since put it back in circuit.

Thanks in advance!

Cheers

Matt

 

[matta]

Success!

I can't tell you how this project has kicked me in the guts, I've stripped it at least 3-4 times, rechecked, recalibrated, tried various tubes, even desoldered individual components to see where the problem lay.

The voltages look good, components were good, wiring was good, I kept coming back to the Max's transformer. We exchanged a few e-mails and as I shared in previous responses I had an open circuit on the secondary, that shouldn't happen!

Max gave me some DCR measurement and eventually I desoldered the transformer to find that Max sent it to me without soldering any of the output pins! He assures me it's the first time it's ever happened, it must have passed his intial test, but the windings came loose.

A quick solder around all the taps, ohmed out, perfectly, placed the transformer, powered it up, pure audio bliss, clean, no hum.

Spend a week or two over the past year on this, so relieved that it's finally over, Ha Ha!

Thank you ALL for your suggestions, tips and advise, I'm sorry we couldn't get to the bottom if it sooner and grateful for Max's help in troubleshooting it.

Cheers

Matt

 

[ioaudio]

Thank you, Matt!

 

[EEMO1]

nice!!!

 

[duantro]

I just finished up my mk7, that I recently gathered parts for. Sounds great on my bench. I'll try to test it at the studio ASAP. I used an equinox u47 body and headbasket, a Beesneez k7 capsule, and telefunken pf86 tube. I used Dan's at Collective Cases psu enclosure, with Max's passive psu bread boarded with the suggested triad choke. Quiet and clean of hum.
Thanks Max for making this possible!

 

[duantro]

Ok, so after checking it out at my studio, I'm noticing that the front portion of the mic amp sounds a little quieter and dipped down in the upper mids. The back side sounds fantastic in omni! I originally accidentally had the front capsule connection bridged to pin 8 on the tube instead of pin 9 where it was supposed to be. I know, stupid mistake.
I've replaced the 68m and styroflex cap that is accross the capsule. I've also checked the capsule by hooking up the opposite sides to make sure that It wasn't the capsule that was damaged. It sounds the same regardless of capsule orientation, so I don't think it is the smoking gun.
I guess I'll try to change the tube and double check for damaged parts elsewhere.
If anyone knows what I should zero in on, feel free to chime in.

 

[shabtek]

probably not the tube--maybe body/grill:you exchanged connections front to back--was anything else changed?--can you easily flip capsule on mount to rest this theory?

 

[duantro]

probably not the tube--maybe body/grill:you exchanged connections front to back--was anything else changed?--can you easily flip capsule on mount to rest this theory?

Yes, I've exchanged the front and rear capsule connections and the issues stays with the amp connection and not with the capsule, nor the body/grill orientation.
I've just changed 68m and the 1nf styroflex. It looks like it was all in line that I may have damaged.
I must note, the deficiency is pretty subtle. Maybe 2db down?

 

[0dbfs]

You can also remove the capsule and replace with a pair of about 60pf caps in order to remove any "acoustic" interactions. Inject signal at the front/rear nodes, change patterns, and sweep to test just the amp.

The backplate biasing section has some filter caps too and patterns change the capsule biasing so could be in the rear/backplate biasing sections too.

-jb