Help with DIY mic noise!!!!

Well, I just etched my first circuit boards with a homemade UV box and they are a psu and an ELA M 251 board.
Both came out very well ( for my first go!! ) The psu uses the same
ht circuit as the original with a simple LM317 heater supply all on the same
board with a pcb mount 9v to 120v transformer with 240v to 15v feeding
it and the heater circuit.
I tried using an AC701 I had lying around and.......no hum at all but LOTS of hiss. Also, I tried using an EC1000 but I get the same hiss mixed in with the signal. Tried replacing the 8m resistor at the grid to no avail.
Output transformer is a 1:7 Stevens-Billington.
Any clues?
Coud it be the output cap - a really old Sprague red top 2.2uf?
Heeeeeeeelllllllppppppp!

Thanks

Hi Buttercup,
How is your output level? Is it very low?
I don't know this circuit but try checking all the Hi-Z connections to make sure they are clean & free of flux.

Cool!!!

I will try that.

Thanks.

One thing that helps is to make the high-Z connections in the air, i.e. off the circuit board. Have a look at this

Does anybody know if there are any other possibilities though as I would
like to try a few things before I pull all my hair out!!

Thanks.

Are you sure your tubes are working right, and not rejects?

Put the ac701 away for now.

What 251 circuit the 6072a or the ac701 one? Plate R and Cathode R are different IIRC.

What B+ voltage 110VDC?

What capsule type a cardiod only or a multi pattern?

If multipattern 4 wire or 3 wire capsule. For the 251 you want a 4 wire for the stock pattern switching

No more than 4V to the AC701. I have not used one: however people I trust tell me they are very sensitive to heater voltage over 4VDC will hurt the tube fast.

A picture of the built board will help.

I hope you don't have the grid wire to the PCB. Look close at zebras picture above or pictures in the meta of some of my builds

I have been told a fingerprint can be as low as 40megs. I tried to test this but my meter went to over range. It was higher than 40 meg maybe 40 meg with a very dirty or oily finger.

FWIW One of my first microphone builds had alot of noise. It turnd out that I handled the input cap in a way that made it somewhat conductive from lead to lead. That very high R low leakage path will upset the grid charge at the tube side. I cleaned the cap and the noise went away. I then replaced the cap.

I try to use tools to handle the parts by the capsule grid/gate interface sometime I will form the lead by hand. You need to be careful not to make a leakage path.

You can't underheat the ec1000 it will go noisy. The ac701 was made for microphones so 4V is its best.

Hmmmmm!

LOTS of useful stuff there.
The grid lead is wired to a pcb pin soldered to the track connection
of the 8m resistor.

The mic I have put it into is an MXL992 - but it was just to TEST
the circuit as I have some bits on the way to build a mic from scratch
completely.

The AC701 was measured at the mic end with a heater voltage of 4v
and the ht is 120 ( 127 sometimes with local variations ).
I realise I REALLY need to think about where to put that 8m resistor
suspended somewhere.

Cool challenge - I love this stuff!!!!!!!

The circuit is the standard AC701 ( cardioid only centre terminated capsule - the finished mic will be 2x backplate as originally specified )circuit though I may try a 6072 as well
with a 30m resistor instead.

I read the thread over at Klaus's siste about underheating so when I
tried the EC1000 it was at 6.3v.

It may well be that BOTH tubes ( AC701 was £50!!!!!!!! ) are duds too!

Would love to post pics but nowhere to host and not much time right
now so will post later though I have the EAGLE layout which I can
email if anyone would be kind enough to cast a second pair of eyes
over it.

No hum though!!!!!!!!


Thanks hugely guys.

If Eagle will spit out a jpeg or gif then email me the pic and I'll stick it up for discussion.
Email
stewart at zebraindustries dot com

Cool Stewart.

I shall do that right away.

voila!



:thumb:

Oh dear.

Did did I do that?
No wonder the bloody thing doesn't work!!
But seriously, lots of other mics manage to put the grid
resistor on the board ( G7? ) so if I tinned the traces and used
rubber gloves, would it help?

You can leave the one end that is grounded connected to the PCB, Just lift the node like In Zebra's build.

Back at TT I had an Idea for a 250/1 type circuit. The extra was the use of 1 gig in the grid to ground as well as the 8 or 30 meg. The 1 gig would be shorted for the high pass and open for more bass.

Alot of sub minis should work fine have you noticed how close all cathode bias tube circuits are?

I have not heard a ela m250/1 but from looking at the circuit and doing experiments. I think the use of a REAL CK12 capsule and the grill and 110v b+, are what make it what it is.

The stock transformers were 10:1 to 11:1 this is from reading posts at other webs site over the years

How long did you leave the tubes on? If it was a short time leave them on for an hour or two and see if the noise goes down or away.

Thanks Gus!!

I have figured a neat way to do just that!!!!!

Will also try soldering the node dirctly to the pin on the tube
socket when I try the 6072.

I think you are right about the combination of the CK12, b+ and
the grille so I shall try and re-create perhaps a little bit of that
magic and post pics when I am a bit further down the line.

I like your idea for the switchable grid/ground res.
Did you ever put it into practical use?

Stevens-Billington were kind enough to 'custom' wind one of their
stock transformers so getting them to make a 1:10/1:11 should
be easy.
Very helpful people and VERY reasonably priced but can't comment
completely on sound quality until I get the circuit tweaked!
Will post results when I get a mo in the next 2 days.

Huge thanks

Giles

Just googled the 992.

That is a microphone design I was thinking about sometime ago specs like a 1006 and V67 mix. I have a number of mxl1006s and v67s. The 1006 circuit can sound OK and I kind of like what the V67 grill does with the china 32mm capsules.

Is the 992 circuit the standard S one? 1 fet 170, two PNP EFs, and one transistor DC to DC converter?

80% Mu lams?

Well, I'm not the shiniest pan in the kitchen as far as these
MXL circuits go but I do know that it is a transformerless
design, cardioid capsule only but I do rather like the grille - which is
why I got it.
I shall have a look at the circuit though and see if I can learn a little
bit about it.
Let me know if you want some pictures of it and I can email them.

Aaaaaaannnnnnddddddddddd I found it!!!!!

What may be the bloody problem.
Having looked again and again at my first little pcb
attempt, the diaphragm is earthed mistakenly when it should
be receiving a polarising voltage so I have rectified the mistake
in the design.
But look at all the cool stuff I learnt while I was waiting to discover
my mistake.
I love this place!!!!!!!!!!

Thanks

Ok.

Just got one of thos photobucket things so here is my slightly
updated board just to test it out.

Is it supposed to do anything or just stay there as a link?

BMP's can't be viewed by standard HTML viewers...

Use .JPG or .GIF in stead..