ART M-Four microphone parts upgrade

[The Sugar Lion]

Very quick question, after I remove the caps do I simply bridge the connection with a small piece of wire to close the circuit?

 

[Khron]

Since when is UCC / NCC a "chinese fake"? SMS is "just" one of their general-purpose 85C rated series.

Khron - why you are using "*only* if they're from one of the handful of reputable Japanese brands"?
What gives you that? A sticker or better parameters or what?
If better parameters, then  typical cheap foil cap usual have most parameters (at least on the paper) better than any "reputable manufacturer" elctrolytic

It gives me (way) more trust in their reliability / lifetime, that's all I'd trust one of the japanese caps i've recovered from whatever gear i've torn down a lot more than a brand new chinese no-name or a crap brand ;D

I'm not claiming they "sound better", only that i trust them not to get bloated and crapped out in a year or two

I've seen CapXon caps bloated in the signal path a couple years ago, in a dbx active crossover. Not to mention the countless cases they crapped out in computer & tv power supplies. 

If this is about that 1uF from the capsule bias supply, you definitely do NOT want to connect that voltage directly to ground
Ever heard of a "short circuit"?

Very quick question, after I remove the caps do I simply bridge the connection with a small piece of wire to close the circuit?

 

[ln76d]

Very quick question, after I remove the caps do I simply bridge the connection with a small piece of wire to close the circuit?

No these three caps are only for remove, if there something would be for jumpering, then i will let you know

 

[ln76d]

Since when is UCC / NCC a "chinese fake"? SMS is "just" one of their general-purpose 85C rated series.

Since it is chinese fake which you can buy from on ebay or aliexpress.
Look at the poor print on the foil and pattern on the top. It's not genuine Nippon SMS capacitor.
Rubycon probably same fake, but there's no pattern on the top, so it's hard to guess.
It's cap like capxon which you mentioned.

It gives me (way) more trust in their reliability / lifetime, that's all I'd trust one of the japanese caps i've recovered from whatever gear i've torn down a lot more than a brand new chinese no-name or a crap brand ;D

I'm not claiming they "sound better", only that i trust them not to get bloated and crapped out in a year or two

With foil cap you don't have to trust "in their reliability / lifetime", if the voltage rating is proper, then you soldering it and forget about it.
Filtering caps at capsule connection path, where btw. is high impedance, have significance for quality of your microphone.

Now i have question for you
Why you would like to use 1uF electrolytic capacitor at this position?

 

[Khron]

If anything, i think i'd shoehorn a bigger value cap in there, depending on what i have in my "cap bin"

I'm not so hung up on the chemistry, when it comes to (linear) power supply filtering

PS: "Even" with electrolytics (good ones, anyway), if it's rated for, say, 2000 hours @ 105C (and that's at its max ripple current rating), i believe the rule of thumb is the lifetime is doubled for every 10C less than the rated temperature, so 4k @ 95c, 8k @ 85c, 16k @ 75c etc.

 

[ln76d]

In common supply node in PSU i partially agree, but not without a reason there's no ultra big caps in B+ supply in most of tube mikes. Especialy in not regulated PSU. There's small current, high voltage, big caps charging longer, and also there's no need to use oversized values.
In my opinion at polarisation voltage node, cutoff frequency below 20Hz is what is needed. Especially if another filtering cap at this node is set that way. Rest is in the outside box.
Still don't see any point of using 1uF electrolytic here.
I think that Georg Neumann, Gerhardt Bore, Bernhard Weingartner also had no point to use eletrolytics at these positions

 

[Khron]

High-voltage caps in high values have large sizes - that's possibly one of the bigger reason they're not used that often And likely not all that cheap either.

You have to keep in mind, if you wanna make a profit in mass-production, you have to (also) optimize cost Although, of course, the quality-to-cost ratio (or level of compromise) varies from company to company...

 

[ln76d]

Try few 1000uF at B+ and you will see Today it wouldn't be so hard to use bigger caps, sizes are reduced, prices also not so high.
Especially if manufacturer using fake caps.
Look at U47 schematic, Even in 50's, when there was much different pricing, 1uF caps were paper not electrolytic.
Saving money in mass production? Yes - and that's why is worth to reverse that process modding microphones.
And now we are producing posts related to one capacitor - this is not cost effective

 

[The Sugar Lion]

Great thanks! I removed the three caps last night. Managed to do it with out destroying anything I will turn it on and do some recording  tonight and let you know how it goes.

No these three caps are only for remove, if there something would be for jumpering, then i will let you know

 

[ln76d]

I reccomend you to test it by simple voice test.
You should be able to catch difference without any problem.
This is only beginning
Now microphone should be brighter (not necessary pleasant) but with less distortion.
If it would be that way sounding then it's good.
It's a first step to clean the mess
Then we go to another step

 

[The Sugar Lion]

Yay! I know this is a bit late but great to have for verification.

 

[ln76d]

Great!

we don't have to figure out pad connection
Do you need hi-pass filter?
Maybe you want low-pass instead?

 

[The Sugar Lion]

It already has a high pass filter and a - 10 db pad. I think we should focus on more advanced next step.

Great!

we don't have to figure out pad connection
Do you need hi-pass filter?
Maybe you want low-pass instead?

 

[ln76d]

I would suggest to change both as a mod, that's why i propesed lpf instead hpf.
Also lpf in k67 is really useful feature, you can have microphone with two different responses.
More modern more oldschool.
In both cases this would be also better for input and output topologies.

 

[ln76d]

Here's how it could look with most of the modification and lpf instead hpf.

 

[The Sugar Lion]

Thanks again for all your help!

Ok for one I turned the thing on and it did not blow up. This is very encouraging to me

I have looked at the new schematic that you drew up (thank you very much for doing this by the way!) and it makes my head lock up. Trying to go from that drawing to the microphone in my hand is not easy to say the least. I will stare at it a little while longer and see how it goes.

Could you please explain what the goal of the first more advanced mod you suggested? It looks like changing C4 from 1000pF 600v to 470pF/250V and moving one connection from the capsule from the ground R6/R9. How would doing this affect the sound?

Oh yes as for the listening, I could hear a difference. The best way I could describe it is less nasal. It seems that there are some frequencies in the upper mid range that are now being represented more naturally. The 8k + range is harder for me to hear a difference in. This impression was made by recording acoustic guitar.

I would suggest to change both as a mod, that's why i propesed lpf instead hpf.
Also lpf in k67 is really useful feature, you can have microphone with two different responses.
More modern more oldschool.
In both cases this would be also better for input and output topologies.

 

[The Sugar Lion]

Thanks for pointing this out btw

If this is about that 1uF from the capsule bias supply, you definitely do NOT want to connect that voltage directly to ground
Ever heard of a "short circuit"?

 

[Gus]

ln76d in reply 54
Removing the 10k can cause issues
Ask yourself why it might be there it is another part that costs money and assembly time
Look up stability of followers

 

[ln76d]

Gus,

if most ccda microphones doesn't have resistor between plate of first staage and grid1 of another and works stable -  i don't see any problem.

 

[Khron]

What's the (added) benefit of removing it, then?

Gus,

if most ccda microphones doesn't have resistor between plate of first staage and grid1 of another and works stable -  i don't see any problem.