Chiswick VK1 meter problem

[johnheath]

Hi…

I have just built a clone of the Chiswick VK1 vary-mu comp and I have some trouble getting the meter working… at all.

After reading about the Altec 436 and its meter …thing I have tried almost everything that I have read but I can't make it work.

First of all I did not have any 6BZ7 or ECC85 home so I tried a 12AT7 and 12AY7 instead (This might be the problem but I do not know)

Otherwise the compressor is built after the attached schematic.

I designed the B+ and heater supply filtering on my own and the sound is good and working… I have no former experience with the Altec or Chiswick so I cannot say much about the compression but it works even though it is rather … kind but with a transparent touch to it.

Anyway… the meter. I have tried a vast amount of different trimpots to get the meter honing in on zero… tried all different settings to see if there is a change. Tried it with and without a preamp… vocal or guitar.

The meter I have is bought from Don-audio in Germany and it is actually the only one I have found that would fit into a 1U front panel. Link to the meter is also attached for those who want to see.

I also took it apart to see if there was a rectifier inside ( read that it would not work with a internal rectifier in the VU-meter) but I couldn't find anything except the magnet and coil.

Thanks in advance

/John

 

[johnheath]

And here is the link to the meter itself.

https://www.don-audio.com/VU-dB-Meter-D1010SL-Slim-Audio


Thanks

/John

 

[PRR]

What is the DC voltage at V1 cathodes, idle and with BIG signal?

 

[moamps]

Anyway… the meter. I have tried a vast amount of different trimpots to get the meter honing in on zero… tried all different settings to see if there is a change.

Maybe I'm wrong,  but "zero reduction" in a vari-mu compressor shows maximum current/voltage, so you should  revert your meter's scale and adjust zero at nominal (maximum) tubes currents.

 

[johnheath]

PRR: Here are all the voltages
V1 (12AT7) at the moment
p1 - 200V
g1 - 0V
k1 - 3,92V
p2 - 211V
g2 - 0V
k2 - 3,92V

V2 (12AT7)
p1 - 220v
g1 - 0V
k1 - 3,36V
p2 - 220V
g2 - 0V
k2 - 3,36V

V3 (6AL5)
p1 - 0V
p2 - 0V
k1 - 3,1V
k5 - 3,1V

Maybe the difference between P1 and p2 in V1 is the problem, I don't know.

I do not have an oscilloscope but I threw in some BIG signal levels via a microphone and the voltage on the cathodes of V1 maintained the same… 3,9V (good or bad?)



moamps:
I have thought of that too, but after fiddling around with the settings and the meter there is no reaction in the meter at all so I cannot see where the needle want to go =)


Thanks

/John

 

[johnheath]

Anyone?

 

[PRR]

> the voltage on the cathodes of V1 maintained the same... 3,9V (good or bad?)

Then trim the meter so with power-up it lays at the opposite end from where it lays power-down (as moamps says).

Then figure out why your side-chain is NOT making negative voltage to turn-down the bias on V1.

(No wonder it is "transparent", it isn't limiting.)

C4 voltage should be zero at idle (this may be OK) and should go NEGative DC when there is BIG signal at the OUTput (before the attenuator) (this is not happening, or not getting to V1).

You *really* want a reliable AC/Audio volt meter for checking a limiter. You *need* to be able to measure the input and output audio levels. Sound-card "meters" are not usually suitable.

And the 12AT7 will not be a great control tube; there's reasons limiters mostly use odd tubes (and you should be prepared to feed them what they want). However the side-chain should be *trying* to turn-down the 12AT7 (reduce its cathode current), and there will be some turn-down (but going over to distortion too soon).

The 6BZ7/6BQ7 is actually a very fine tube for this purpose, and ought to still be inexpensive (even though I used to burn them up in my headphone amp)? I'm seeing 4 bux, can't kick at that. (BTW, the Z and Q are NOT the same tube; I burned-up enuff to know the difference. But the differences are too slight to matter here.)

The 185V 200V 211V different V1 plate voltages are non-issues. The two halves are never exactly matched, and the large-signal (DC plate voltage) mismatch won't be the small-signal (thump) mismatch. First get it working (limiting)! Then check that the plate DC voltages are not absurdly different (like 80V and 230V). Do the thump balance. This design does not give a lot of leeway for balance, so the two plates will be near-enough (unless one half is just dead).

 

[gyraf]

6BQ7 are still surprisingly cheap: http://www.wallcoinc.com/Sylvania_6BQ7_Receiving_Tube_p/6bq7.htm

Jakob E.

 

[Nele]

Don't you need a 1mA meter? Not a VU-meter?

 

[johnheath]

Thanks….

PRR:  Ok we are touching a few things now and that is very good

A couple of ECC85 have been ordered already and I am curious to try them… I hope they are near spec as the 6BZ7 / 6BQ7?

I did not expect much from the 12AT7 but as you say there should be at least some reaction… hmm, back to the workshop

The meter itself is nothing that I believe is a very accurate meter but merely something to get an indication of the level of gain reduction.

Nele:  I think the meter in fact is a 1mA meter … it surely is not a true VU-meter though.

The project is awaiting the new tubes and then we will see if there is a difference.


Jakob: The tube is very cheap on its own but they want 165usd for shipping and to Sweden it is yet another 10% in customs fee and then another 25% VAT… the final price is very expensive.

 

[Nele]

Doh! You're right. It is a 1mA meter..

Nele:  I think the meter in fact is a 1mA meter … it surely is not a true VU-meter though.

 

[johnheath]

Ok… this is what I have found out this far:

I have now changed the V1 to a ECC85 and I had to change the heater supply resistors for obvious reasons.

The meter works as you predicted Moamps and it is now working …backwards, but working (I guess the next thing to do is to get a scale that fits??)

A few things that I have noticed and I wonder if it is correct?

The value of R106 alter the sensitivity of the compression??? There is a value around 330R when the meter starts wobbling back and forth depending on settings on level and release pots. I am now up to 1k2 and wobbling is not present.

And… the input level pot is the key to compression… that might have been the issue before actually because I did not turn it all up. Fully open it is possible to adjust the compression in small steps but when the level is rather low the compression starts only at very BIG signal levels and goes all the way back… i.e not useful at all.

So my question here is if it would be preferable to get rid of the input level pot and go for a output attenuation pot and thereby have the useful sensitivity of the compression and alter the overall volume with the "output" pot? If that is a good idea where would you recommend to connect it… before or after the output transformer?

Also: R112 (68K) at the thump switch makes the compression less "all or nothing" as compared to without the 68k… is this wrong or normal?

I am asking because I really have no experience in this vari-mu comp at all… haven't heard it before so nothing to compare.

I guess I have more questions but these are the most tricky ones at the moment.

Thanks in advance

/John

 

[PRR]

> I did not turn it all up.

It sounds like you have not used any compressor much?

You are correct: that meter is mis-marked. When not connected, the needle should sit on "-20", not zero.

You may be able to pop it open and turn the scale upside down, then turn the whole meter upside down, and get it right.

But that is NOT important. One end is "no compression", the other end is "heavy compression". You want to have the needle dancing between "none" and "medium" for many situations. (If the compression is for protection only, you want to sit on "none" but accept slight jerks. If the compressor is for conference speech or smashed AM radio sound, bang the needle as needed.)

 

[johnheath]

Well, for sure I have used compressors before but … but the ones I have used doesn't have to be turned "full throttle" before they start compressing.

I have been thinking about the exact same thing… turn the meter upside down and open it up and flip the scale and I will have a functioning meter. By the looks of it and former experience of opening it it will work… even though I have to resemble the whole front panel.

The thing is that I don't want that all-or-nothing compression but rather an overall smooth compression.

I decoupled the input level pot and replaced it with a 100k pot and coupled that one as a output attenuator pot instead. I didn't have much time to test it much but my theory seems to make sense in this case. I got a nice compression with the needle moving nicely around gain reduction of 2 - 4 on the meter ( I know that the meter really says nothing… just a hint on whats going on). This was tested with an acoustic guitar.

/John

 

[EvLoutonian]

How did you go with this project in the end?

Everything end up satisfactory?