Another vari-mu: Chiswik Reach Compressor

[skal1]

In fact it's not really 15v from the ac signal: i've read somewhere that 6al5 need a minimum voltage of 45v to begin conduct so it's more 60v from any sides of the output which is 'seen' by 6al5 input. If cv is 30V it means 30db gain reduction... HIGH GR!
I never use comp with such high gr!...  ;D

[/quote]

hummm  so basically what you are saying is the 6al* needs to see 45 ish ac volts at the anode to start conducting and then some more to give you that useful cv so if your output craps out at this point cv is weak, yes, so work on your cv sig.


skal1

pray tell...

? Excuse me english is a foreign language for me i don't understand.

hey ,it just means {tell me what is right kind of tube!}

skal1

 

[KrIVIUM2323]

;D tube: 6BZ7!    I must take a look at datasheet but i think using ***85 need some mod from schematic // the original tube.

hummm  so basically what you are saying is the 6al* needs to see 45 ish ac volts at the anode to start conducting and then some more to give you that useful cv so if your output craps out at this point cv is weak, yes, so work on your cv sig.

Well in fact i'm not sure... i may say something stupid  ??? Don't know have to search where i think i've read that... Because when i look to 6al5 datsheet it don't make sense at all!!!  :

Well from datasheet each section with an input (ac) signal of 15v rms will produce a dc output signal of approximately -16,8/17v for a load resistance of 15000 ohms (aproximately). And as far as i understand there is a 10v drop at 60 milliampere current on plate.

Humm. I need to think about that! And by the way i did understand a sentence i've read telling that 'thump modify the impedance of the gr/cv path'! Great!

PS: Skal1 check your pm please.

 

[mikeyB]

Couple of things - I asked in another thread about trimming out differences in diode halfs ie 6AL5 - if imbalanced will produce ripple on control voltage causing distortion in varimu stage - so can this be trimmed - will try this when i get chance.
More of a design question - tube currents should be matched- right? In the chiswick output stage there is a trimmer in the cathodes to balance current but in the front end there is a trimmer in the anode resistor circuit - could this be changed to a cathode balance trimmer as in the output - or would it cause problems in setup/operation - just thinking of safety point of veiw when dabbling!

 

[gary o]

Kr1v sorry I forgot you wanted me to measure the consumption of my mono chiswik type cpmpressor ? you want me to put milliamp meter in series with the B+ ? I gotta wire the compressor up Im on the case also Miha if yr here il measure the CV voltage for you

 

[rotation]

Yeah, i'm here, please check this out. I measured this myself and my max input to 6al5 cathodes was 8V and it put out -8V max at anodes. Meter and daw shows max 8dB of compression at this conditions.
If i took away 100k FB resistors things get better, IIRC there was max 12dB GR.

Another member tried to add some DC to cathodes (like on many other comps where cathodes are connected also to B+ through divider or whatever that is) of 6al5 and nothing changed, he also got max 8dB GR. Than he tried adding more ac to cathodes of 6al5 and he couldn't get more than -8VDC at anodes of 6al5, no matter how much DC was at cathodes. I hope i'm stating this correctly, maybe person who told me this will jump in because he also tried to add 436's SC..

Miha

p.s.: we better move this to Altec436 thread because discussion is related to other compressor and there are also posts from PRR. It was good discussion there and it would be a shame to divide it in several threads.

 

[KickBack]

Hi

I have re-built several of the Alter 436 compressors. I have worked on gear from the 1920's on up.

I used to service recording studios in LA in the 1970's and 80's. 

A. My solution to the 436 was to take the 6bc8 out and use a 6gh8, 6an8, pentode to split-phase inverter to kill the imbalace problem in the dual triode, unless a special 6bc8 from Altec. Control the grid of the pentode, cap couple to the triode phase inverter.

B. Put another 6al5 in parallel to lower the voltage drop, or feed the Dc into a op-amp to get more DC gain.

C. Use op-amps to make a voltage follower and just forget the 6al5.

6bk7, 6bq7, 6aq8, 6bz8, 12at7, and a few more do the same as the 6bc8, and have different grid control curves.

You can try another dual triode in the front end.

The gain stage works like this.

These tubes were used for TV tuners so they have the ability to "variable Mu" or gain more than standard tubes.

The -DC from the 6al5 lowers the gain of the dual triode. The 6al5 needs a certain amount of AC to "Turn on" as it has a operating curve, and that is the threshold. When the AC on the primary of the out trans, (higest voltage point as the line out trans steps down), is high enough, then -DC comes out of the 6al5 and lowers the gain of the 6bc8.

The problem with this circuit is that the 6bc8 from ALTEC was made special for this bad circuit. If there is any imbalance in the gain of the two triodes, then the drive to the 6cg7 is off, and the compressor "thumps" or worse.

This guy was designed to go into a PA system with no response below 350 Hz so you never heard it at your local high school when the principal was chatting away.


You have to really spend a lot of time studying all the tubes made to get some ideas as to how they were used.

i can post some circuits, let me know.

 

[gary o]

Hi Kickback thanks for your knowledge I wpild love to see your circuits Im sure other will want to see them too I would be very interested in see a schematic of how you would wire up a penrode to split the phase and kill the inbalance......Am I right in saying some other older vari mu limiters used this like the RCA BA6A ?

Thanks

 

[skal1]

Me too  ;D

skal1

 

[KrIVIUM2323]

Hi everybody.

Gary, I think Kickback is refering to pentode because some of them are remote cutoff and do well in comp job (when triode connected). For the phase splitter take a look at RCA's BA21A. The first stage of this preamp run single ended and the 6072 second half is used as a phase splitter. This is the way to go: first tube GR and second a phase splitter (or first triode and second one in case of a dual triode tube). This is convenient for mono unit for stereo you still need matching for two channels (and from the rca Ba21A manual you need a selected tube for this task, but it deal with very low voltage as preamp so need of hi spec, in comp i think i'ts not that crucial)  .  

Kr1v sorry I forgot you wanted me to measure the consumption of my mono chiswik type cpmpressor ? you want me to put milliamp meter in series with the B+ ?

Yeah that would be nice.

unless a special 6bc8 from Altec.

Kickback you have already stated this before. In what are 6bc8 from Altec special? Pairing of some spec (transconductance and paired triodes i suppose) or something else?


PS:Gary could you backup and clean your PM box as it's full and cannot be accessed anymore.

 

[Tubefreak]

I have heard from a very reputable source that this build/layout
(and he made 3 of these comps) gave only about 2dB of compression...

I was seriously thinking of doing one of those Purusha special boards but instead, I think
I'll (like Gary, cheers for the pic!) build a pair of 436s in a box, make them
stereo linkable and tweak them out a bit. Or maybe a modified PRR?

Unless, that is, the VK1 gets fixed with confirmation.

I've just finished mine and it rocks 100%. There are no faults or flaws and it sure does more then 2dB of compression! Check http://www.diy-racked.com/diy-talk/viewtopic.php?f=12&t=98 and www.tubefreak.com/varimu/

Kind regards,
Maarten

 

[ChrioN]

Nice build! And scratched up panel. But I wouldn't expect anything less from our friend tat 

 

[Tubefreak]

Nice build! And scratched up panel. But I wouldn't expect anything less from our friend tat  

Scratching is all my fault. Tat delivered a flawless case/pcb and with the docs it was not so hard to build. Finished version is cleaner inside, with all cables tied together.

Maarten

 

[SIXTYNINER]

nice unit !
any picture about the "jewel" ?
6t9r

Nice build! And scratched up panel. But I wouldn't expect anything less from our friend tat  

Scratching is all my fault. Tat delivered a flawless case/pcb and with the docs it was not so hard to build. Finished version is cleaner inside, with all cables tied together.

Maarten

 

[Tubefreak]

Something like this? Found it at a local store, uses a small push in light bulp and (this one) runs on 6V from the heaters. On start-up the light slowly fades in, which looks really nice.

 

[SIXTYNINER]

...yeah , cool jewel light..

how many compression db , from the vk1?
20?

6t9r

 

[Tubefreak]

No, no 20dB GR with my version. But maybe if you take out the negative feedback circuit you'll get there. But if you REALLY want 20dB of GR, maybe a varimu is not the type of GR you want. It does amazing compression between 1-5dB, nothing I've heard/used comes close to this type of compression. Between 5 and 10dB GR it really depends on the audio and the settings, how it will suit the audio. Check the samples on my site www.tubefreak.com/varimu. The drms3 files have fast, slow and the modded extra slow attack, which I like a lot. There's also a file with 10dB of GR in the 'drums' files.

I'm gonna mod the attack & release times some more. For mastering I like some longer attack times and some shorter release times. Don't care much for how the time values compare to other units (=way off!), the mods are done by taste and needs. With a few additional mods this thing is gonna be even more killer, especially the longer attack times are a real bonus. And the high pass filter is so sweet, really opens up the music and an absolute must for the way I use it. Really can't live without a HP filter in a compressor. Don't know why and how others encountered troubles with the VK1 project, mine works flawless and sounds killer.

Kind regards,
Maarten

 

[Tubefreak]

I've added a specific page regarding this project on my website www.tubefreak.com  It's filed under Effects & Rackgear.

Kind regards,
Maarten

 

[SKJGProject]

are there any options for the sowter 8650?

 

[SKJGProject]

....and is it possible to use a 1/2 mA meter with the clone?

 

[Tubefreak]

The 8650 is not a problem, I've made mine with it. All the docs are here that explain it. http://purusha.smokinggunrecording.com/XVXDY/

As for the meter, 1/2 mA ain't a problem either, it's just a matter of using a different resistor for R106, VR100 or Rmeter. Don't know exactly which one, but there is enough info in the lab regarding this. And otherwise it's just trial and error.

Maarten