the pimp / mod the Poor Man 660 thread

[SaMpLeGoD]

Listen to your comp ! ;D ;D ;D

I will Lolo, I will
It's my 5th DIY comp and I like a lot the 1176 REV D and the GSSL! so I' looking for a fast and clean comp...
I'll use the PM670 on a mix next week, so I'll post the results 

Thanksssss!

Eddie ;D

 

[lolo-m]

I made a board layout for my second mod. I will share it. It is single-sided so I would like to know if it's worth to make some professional boards. The thread is here :
http://www.groupdiy.com/index.php?topic=32406.0

 

[mikeyB]

It's getting late and I can't find what I thought I had seen - maybe another thread -but didn't someone mention using a hybrid poweramp chip to replace the valve sidechain? Will this affect the sound drastically?
nite nite - need sleep!!

 

[lolo-m]

It's getting late and I can't find what I thought I had seen - maybe another thread -but didn't someone mention using a hybrid poweramp chip to replace the valve sidechain? Will this affect the sound drastically?
nite nite - need sleep!!

It's maybe here :
http://www.groupdiy.com/index.php?topic=31088.0

 

[lolo-m]

A few days ago, looking the amazing results Rotheu had on his PM660, I realized that there was maybe an "all tube" pimp that can be done on the Analag's PM660...
The first stage of the real Fairchild SCamp is a 12AX7 in class B mode. On Analag's design, this stage isn't present. I didn't scope (yes, I know, I should have  ;D) the SCamp, but I'm quite sure the PM670 SCamp is a classic push-pull amp. So I added this missing ( ??? ) stage... And the result is: it seams to work !!!
I will have a lot of shows in the next 15 days so I won't be able to check this idea correctly untill next month. If someone wants to test it, it is quite simple. Look the Fairchild schemo and copy the first stage, then connect the output to the "threshold connector". Use the Threshold pot output to feed the 12AX7 grids.

Analag, am I right to think the 5687 are in pure AB mode ? I don't have anything to simulate the 5687 so...

 

[lolo-m]

My last investigations on this mod aren't really revelant  :-\... I only managed to add a DC threshold with this mod so I give up...
Anyway, for poeple who want to implement the "missing" DC threshold (not sure it is really a missing point but more Analag design wish), there's a way to do it with the SCAB (SCAmp Booster)... I will fix this Tuesday or Wednesday and post results here :
http://www.groupdiy.com/index.php?topic=32965.0

Thank you again Analag and Silent Arts for this wonderfull comp !

 

[lewilson]

Hello all,
  I have found the Russian 6n5p tube to be a nice sub for the 6bc8 tube, Much smoother fuller compresion and overall a better quality tube than the american tubes. Its the same pinout as the 6bc8  but uses more filiment current. 

 

[Moby]

Hello all,
   I have found the Russian 6n5p tube to be a nice sub for the 6bc8 tube, Much smoother fuller compresion and overall a better quality tube than the american tubes. Its the same pinout as the 6bc8  but uses more filiment current. 

It will be nice if you can post some sound samples, 6N5P against 6BC8 

 

[lolo-m]

Hello all,
   I have found the Russian 6n5p tube to be a nice sub for the 6bc8 tube, Much smoother fuller compresion and overall a better quality tube than the american tubes. Its the same pinout as the 6bc8  but uses more filiment current. 

Is it a direct substitute ? I mean did you just plug them with the same settings or did you change the 2 bias settings ? I bought some to have different compression flavours on my 2 PM670...

 

[khstudio]

I'm gettin close. I'm also lucky that my brother can make me Vinyl stickers.

Pos 1 thru 5 will be some of Blue Bird's & Pos 6 will be stock... should be cool & give enough options.

 

[khstudio]

See Below

 

[khstudio]

I posted a Blank - VOLTAGE CHART for everyone to download & hopefully fill out & post.
these measurements will help others troubleshoot none working units & narrow down problem areas.
Also confirm if your unit is working properly.

I made a voltage chart in EXCEL & a TEXT doc.

EXCEL VERSION = http://www.khstudio.us/DIY/PM670/PM670-VOLTAGE-CHART.xls

PL1 - Control Voltage (Threshold Off, Then 1st ON Position, Med & Full on) This one is tricky but important!
...Must have audio going thru unit, say 1.2v Test tone @ Input (Please measure AFTER connecting to input! due to loading)
& then set gain to have 1.2v at the output (with Threshold fully off)
Obviously everyone's measurements will be slightly off due the differences of Threshold controls, Taper & values chosen...
but if you KNOW your switch & values, posting the value along with the Voltage reading will be more accurate!
((((Sorry... This Test should have been added to the Test Sheet.))))

PLEASE DOWNLOAD & POST YOUR VOLTAGES... Thank You!

Below is the TEXT version & Attachment Version Below that.
------------------------------------------------------------------  
PM670 Voltage Chart USER NAME =
"Inital setup"   input gain & threshold control at 0

Measure voltages referenced to ground - Set RV3 wiper to -2.4VDC - Set RV6 wiper to -4.5VDC

Power Transformer & High Voltage output =
HR* = Heater resistor Value =
Volts - Before & After HR* =
VU METER "Type" =

POWER SUPPLY NOTES
KL1 AC-IN HIGH VOLTAGE AC IN TO DIODES
R1 Before 47r - 2w
R1 After 47r - 2w
R2 Before 47r - 2w
R2 After 47r - 2w

CHANNEL - 6BC8 - Section
PL1 PIN 2 CONTROL VOLTAGE - THRESHOLD OFF
Q1 Emitter "Common node voltage of R1, 2, 3, 4 = 330r"
R2 330r Cathode = V1a
R1 330r Cathode = V1b
R4 330r Cathode = V2a
R3 330r Cathode = V2b
PL2 Pin 2 Center Tap  of T2 - 10k:600 - 136v from R6 / VU Meter
PL2 Pin 1 Outer Tap 1 of T2 - 10k:600 - Voltage to Plates - One side of R6
PL2 Pin 3 Outer Tap 4 of T2 - 10k:600 - Voltage to Plates - One side of R6
PL3 Pin 2 Should be same as PL2-Pin 2 voltage
PL3 Pin 1 136v INPUT - to R6 & VU Meter

CHANNEL - 5687 - Section
R10 330r "Common node voltage of R11, 12, 13, 14"
R11 47r Cathode = V3a
R12 47r Cathode = V3b
R13 47r Cathode = V4a
R14 47r Cathode = V4b
PL8 Pin 2 Center Tap  of T4 - 10k:600 - 245v INPUT
PL8 Pin 3 Outer Tap 1 of T4 - 10k:600 - Voltage to Plates - One side of R15
PL8 Pin 1 Outer Tap 4 of T4 - 10k:600 - Voltage to Plates - One side of R15

 

[buschfsu]

kev,
i have a 2 deck 6 position switch for my TC that works with my purusha case.  Can you share how you wired yours?  looks like you jumpered the 10uf on the board? is that right?


Also my purusha case has a hpf knob silk screened on the back.  did i miss a thread on this?  anyone have one built this way?

thanks

 

[khstudio]

I will try to make a drawing as soon as I have time... it's not too hard.
I think if you read back I mentioned what values i used & I like them a lot!

 

[khstudio]

This is part of an email I sent to Rowan concerning using only ONE 5687 in the sidechain:

I ran some tests, using 1x 5687 in the left ch & 2x 5687's for the right.
All of the voltages into & out of the sidechain section & the control voltages matched at various settings!
So then I did some "Burst" tests to check & see the "Attack Time" in the computer... they also were no different & appeared to be pretty fast (aprox 2-5ms)
Then I listened & could not tell a difference between the 2 channels...

I'd like to run the 670 with just One 5678 instead of 2 but I need to ask, what is the downside of this if any
& does the cathode resistor need to be adjusted & to what?
The Voltage at the 330r & 47r node is:
2 tubes = 12-13v
1 tube =  10-11v
The main reason for me wanting to run it like this is the HEAT is drastically reduced everywhere except the heater resistor.
I know the sidechain amp has a lot to do with the "Attack Time" but how is it that the One tube does the same job as having 2?
*** The Philips & newer Tung-Sol both reacted the same when doing the "single tube" test. ***

Rowan's TC network doesn't need a lot of power to work... It charges a really small cap on attack...

Thanks for the info...
Do you see any downside to the one tube side chain at all?
What about charging the 10uF Cap?

This comp really pulls a lot of current & if it reacts the same with 1 or 2 tubes I wonder why it even has 2 to begin with.

My next question should be about the possible need to adjust the cathode resistor for the single tube operation.

I must say I'm really surprised that no one else is having issues with a HOT Edcor Power transformer.

Here are the readings:
Total Heater Current draw = 5.27 amps

15 min = 98 deg F
30 min = 110 deg F
45 min = 115 deg F
60 min = 120 deg F
90 min = 128 deg F
120 min = 135 deg F

At this point the PT is too hot to keep your hand on for more than 5 seconds... if that!

Here are the readings with "Just ONE Tube in the Side Chains" - Both Ch:
Total Heater Current draw drops from 5.27 amps to 3.58 amps

15 min = 95 deg F
30 min = 98 deg F
45 min = 103 deg F
60 min = 106 deg F
90 min = 107 deg F
120 min = 107 deg F

MUCH Better, no?

 

[noulou]

Right! Pimping time!!  ;D

First of all, many thanks to lewilson for the Russian valves tip. I tried them and they work great as a 6bc8 sub. I have to say I also prefer the sound to the american valves. East-West 1-0.

I am thinking of upping the supply voltage to them, like bluebird did with his 6ba6s.

I think 200v should be cool with them, what do you all think?

http://www.mif.pg.gda.pl/homepages/frank/sheets/049/6/6BC8.pdf
http://www.mif.pg.gda.pl/homepages/frank/sheets/113/6/6N5P.pdf

Now, with the all important time constants...

In my opinion, this is where it's at!

The original design has far too long release times for my taste. I tried to measure the release time, got bored, went for lunch and when I came back the meter returned to 0vu. About 25 minutes i think.  ;D ;D
(joking)

So, I took out c4, jumpered it and substituted it with different capacitors and resistors in parallel.

At the moment I have
1u2 // 50k
3u3 // 100k
4u7 // 150k
5u6 // 1meg
6u8 // nothing
10u // nothing

Mind you, these are just values I had in hand. I installed them in order to start aurally tweaking the beast.

BOOM!!!

I haven't taken it to the studio yet, but listening at home I can easily see the poorman becoming my favourite compressor.

Beautiful compression action, especially with those short release times.

Who cares about 50 watt sidechains and sub-microsecond attack times?

If I wanted to completely trap the transients I'd turn to a digital limiter, which can also look into the future. 

I don't think I have a need for lolos boards after all. I say let some of those sneaky transients through!

I'll report back once I tweak those time constants even more. (hopefully with sound examples).

 

[lolo-m]

This is part of an email I sent to Rowan concerning using only ONE 5687 in the sidechain:

I ran some tests, using 1x 5687 in the left ch & 2x 5687's for the right.
All of the voltages into & out of the sidechain section & the control voltages matched at various settings!
So then I did some "Burst" tests to check & see the "Attack Time" in the computer... they also were no different & appeared to be pretty fast (aprox 2-5ms)
Then I listened & could not tell a difference between the 2 channels...

I'd like to run the 670 with just One 5678 instead of 2 but I need to ask, what is the downside of this if any
& does the cathode resistor need to be adjusted & to what?
The Voltage at the 330r & 47r node is:
2 tubes = 12-13v
1 tube =  10-11v
The main reason for me wanting to run it like this is the HEAT is drastically reduced everywhere except the heater resistor.
I know the sidechain amp has a lot to do with the "Attack Time" but how is it that the One tube does the same job as having 2?
*** The Philips & newer Tung-Sol both reacted the same when doing the "single tube" test. ***

Rowan's TC network doesn't need a lot of power to work... It charges a really small cap on attack...

Thanks for the info...
Do you see any downside to the one tube side chain at all?
What about charging the 10uF Cap?

In stock TC network (as in Fairchild's pos 5 &6) the SCAmp charges the cap paralleled to the grid resistor first. This gives you the attack time ( in Analag's it is a 220nF). Then it charges the rest of the circuit... Less power will certainly mean in Analags' TC network only less release (or a weird release action)...
Paralleled tubes in push pull  technology gives you more power and more linearity....

 

[khstudio]

Do you have any tips you could share about how best to test the release times?
I'd like to do a comparison once I get my unit back up & running.

Thanks for you help. 

less release (or a weird release action)

Do you mean slower, faster or what?

 

[lolo-m]

Certainly a faster release as the cap won't be charged enough but certainly too with a fast peaky reduction and a release time starting at a different gain reduction... I don't know how it can sound...
All this is what I suppose, I can't say this is what happens...

 

[khstudio]

Cool... The reason I'm asking is, before my unit went down I had tried it both ways but settled on the Single side-chain tube because of the heat (or until I got it situated) I had recorded a few different tracks with each setup + I still remember how it felt... especially on this one Bass track I'm working on.
I do think it was a little better sounding with both tubes... something is a little strange with the release I think but can't put my finger on it.

I will say that it still does work very well... just maybe not optimal.

Thanks.