the pimp / mod the Poor Man 660 thread

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
I wired two 6BA6's  to a 9 pin tube socket with buss wire kind of like the Manley T bar to eliminate some oscillation I was getting earlier when using  4 inches of wire from the 7 pin tube sockets to the 9 pin tube socket. Both adjustment points are at 7 volts.
Bluebird, did you connected 6BA6's dirrectly in the triode mode or U used some resitors?
 
Has anyone ever implemented a temperature sensor in their builds?.. how about doing it in this one?

To have a fan spinning all the time in the studio is a no-no here. But if the PM670 gets hot like an own, then it could be ok to have an automatic switchable fan.  Sound like a fun thing to build. Hope that the switching of the relay won´t affect the sound in any way....I am afraid that it might produce clicks or level-changes due to power consumption.

I was thinking to maybe build a thermistor sensor -based switchboard (with a LM339 comparator), which switches a relay controlling one or more small fans, say reacting around 60-70 Celsius. The set point for the temperature could be adjusted by a pot on the back of the unit, for later fine-tuning.


 
Guys, what's the difference between 6BA6 and 5749. Or is it 5749 good replacement for 6BA6 Vari mu ? Also there is 6660.... similar but how good?
 
Bump for Bluebird  ;) are you around? How did you connected 6BA6's in triode mode? Directly or U used some resistors? Also , do I see well on pic? U used 5749's versions of 6BA6?
 
Moby said:
Guys, what's the difference between 6BA6 and 5749. Or is it 5749 good replacement for 6BA6 Vari mu ? Also there is 6660.... similar but how good?
5749 is more shock resistant... A kind of military specs 6BA6... But I can't say more for the sound... I bet it will sound the same. ::)
If you or someone else don't know this url, it's the best tube data sheets site I found :
http://frank.pocnet.net/
 
5749 is more shock resistant... A kind of military specs 6BA6... But I can't say more for the sound... I bet it will sound the same.
Yes, I compared the sheets, looks similar (but not exact the same). I just wonder about sound.  ??? Must be that it's less microphonics since it's industrial-mil tube .
 
Hi to everybody trying to mod the PM670 with the Fairchild's time constants... I was kind of suspicious about the fact that cloning the network will give the same results. It was impossible to check this at that time for me because my PM670 wasn't finished... Now it is. :)

The fairchild's time constants network is working in the PM670 with better results IMO than the stock  ;D
but it doesn't give the same results  :'( :'( :'(.
The first position give about 300ms of attack and release times... The answer why is there :
http://www.groupdiy.com/index.php?topic=31345.0
The PM670 has got a SC-amp of 5 Watts or so, the Fairchild has got about 25W... The current out of the transformer that fills the RC network cannot fill the same caps at the same speed... That's certainly one of the reasons Analag didn't copy it ...

I'm trying to find the good cap to reach the 0,2ms of the position 1,2 and 6... But there's another problem I suspect to occur : thump will maybe show its ugly face !  :mad:
I'll keep you informed  ;)

PS: The Analag's time constants network is based on the position 5 and 6, a kind of automatic function program compression with quicker release (I didn't check really if it's quicker but it seams to...).
 
At the moment I've got on channel modified and the other stock for a quick possibility to compare.
I wil certainly stop my research now for today (I've got children to pimp ... sorry, to wash  ;D)
So, about 8ms attack time and 300ms release time : remove R17, shunt (or not it's not important here) R18, shunt C4 and change the pot (RV7) for a fixed 1M resistor.  :D

Listen, compare, it's really different !!! A lot quicker than stock, and maybe clearer, but I'm not sure I must perfect my settings between the two... I want to be sure they sound close without compression but I won't have time enought today to do this...

See ya later  ;)
 
Hi, i'm back with some new results...
The real difficulty is to know how fast attack we can get with a PM670...

I found two solutions to have about 0,1ms attack time and 300ms release time.  ;D

One is a change in Analag's network but have the same compression signature, with a more program dependant maximum compression...
R17 : 221K
C3 : 100nF
R18 + RV7 : Fixed 10K
C4 : 1uF


The other is certainly more different, more "a la Fairchild position 1 compression", but I would like to have the advice of some "tube eagles". It seam to be stable in my PM670 but my knowledge in tube technology don't allow me to say do it ! I don't know if this second mod is secure, so wait some time before trying this !!!
I simply did :
C3 :100nF
R17 : 3M. Just unplug everything else...

Reading the 6bc8 data sheet, I seen that grids should be grounded though a 500K maximum resistor... Even if in a varimu grids are not grounded, I wonder if it shouldn't be the same in the PM670...
If someone knows the answer, I would be glad to read it  !!!!  ;D ;D ;D

Edit : The second mod is highly unrecommended !!! Read these PRR explainations :
http://www.groupdiy.com/index.php?topic=31088.0


 
lolo-m said:
One is a change in Analag's network but have the same compression signature, with a more program dependant maximum compression...

[snip]

The other is certainly more different, more "a la Fairchild position 1 compression".

You just gave me an idea. When I get to where you are with building the PM670 and tweaking the time constants, I'm going to teach myself "basic PCB making" by creating a tweakable time constant network expansion, with relay selectable networks.

I was already going to do a "lat/vert" board to hide the large amount of let/vert related wires behind relay switch and in a small PCB. There's an opportunity to put feedback/feedforward switch on that same PCB as well.

Guess I have work to do now.
 
At the moment, I do all those tests really easily...
I made a few wires shunts on the Volker's board, and putted out a few components...
Shunt C4 and R18, put out R17 and C3 and connect the new networks to the attack connector (only 2 wires are used there) !!!  ;)
 
Here are some snapshots of the mod :
12K "sine wave" sent into the PM670...
The upper wave is the unrecommended mod.
The lower wave is the sure to work properly mod.
ATTACK snapshots :
ATTACK-N1a.jpg

ATTACK-N1Ba.jpg


RELEASE snapshot :
RELEASE-N1a.jpg


Not that bad isn't it?  ;D

Edit : The second mod is highly unrecommended !!! Read these PRR explainations :
http://www.groupdiy.com/index.php?topic=31088.0
 
Not that bad isn't it?
Not at all  :). But , how it sounds? Are you happy with overall results? Are you tweaking 660 for tracking , mix bus or mastering purpose, or you want to have "all in one"  ;) Also, it will be great if you can do some sound samples. I will try your time constants but it will be great to hear results ;D
One is a change in Analag's network but have the same compression signature, with a more program dependant maximum compression...
R17 : 221K
C3 : 100nF
R18 + RV7 : Fixed 10K
C4 : 1uF
Is that recommended mod?
 
Moby said:
how it sounds? Are you happy with overall results?
It sounds really good if you like ultra fast comps. I do. There's a magic thing with the varimu comps : the compression is always very natural, quite impossible to hear...

Moby said:
Are you tweaking 660 for tracking , mix bus or mastering purpose, or you want to have "all in one"  ;) Also, it will be great if you can do some sound samples. I will try your time constants but it will be great to hear results ;D
I'm trying to have a quick comp... I love quick comps... I will work on differents attack and realease times to have a really versatile compressor... I don't do a lot of mastering but because I'm always disapointed with others results, I would like to have a mix and mastering comp...
I will put a few samples online in a few days if you want...

Moby said:
One is a change in Analag's network but have the same compression signature, with a more program dependant maximum compression...
R17 : 221K
C3 : 100nF
R18 + RV7 : Fixed 10K
C4 : 1uF
Is that recommended mod?
Yes, it is that one !
 
I will put a few samples online in a few days if you want...
Of course  :)
From the very beginning I was bit suspicious about time constants , now I see that it's possible to tweak this beast  ;D
 
Hey Lolo... nice you're working on the PM670 time constants... I almost finished mine, and I'm looking forward to apply this mod!

Cheers,

Eddie  ;D
 
Hello, is anybody have some better solution for the PM670's Time Constant? to get them near to the original Fairchild? I think the 10k pot wired with 2k step resistors is not that good as I expected...  Any conclusions Laurent?
Cheers,

Eddie  ;D
 
Eddie, try my recommended mod and tell us what you think about it  ;) !!! You surely have the components at home...
I will post some audio samples this week end (I'm not at home now)...
 
Back
Top