the Poor Man 660 support thread

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
;D Not the prettiest but it does the job ;)

Thanks Analag and Silent art


varimu.jpg
 
wow...Nice front panel Gachet!!!I especially love the DIY graduations/FP layout!!! ;)

Do these sosckets melt down cause of tube's heat ??
 
Hi

The attack control is the normal time constant on a stereo 10K lin pot.
The "so called " release is the suggested bleubird mod:
I've tied a 500k log pot parallel to C4.It results in faster release settings(to my hears)
I can switch it back to original with the "mod" switch to compare.

I have removed the 22k vumeter resistor with no problem.

I've tried 8 rca 6bc8 to find my best sounding quad.

I've tried both schotky and 1n4148 diodes with the dip 8 socket trick .After some audio testing I kept the 1n4148 but this is very subjective.
After the mixing sessions I changed the treshold steps and they are following (in ohms):
1-500 ,2-500 ,3-100 ,4-200 ,5-300 ,6-400 ,7-500 ,8-500 ,9-500 ,10-500 ,11-1000 + 10000 wich is not selectable.
It means that I have control from 14,9K to 10K attenuation for the  treshold.

The input attenuator is basically log steps.

I add a fan for the european transfo wich is heating a lot to me .

That's all.

Sanstitre2.jpg

 
Hi Cyrille,
whow, thats a funky looking unit. I love the orange colour. Did you tie the time constant controls of both channels together? In my unit it´s the rectifier for heater supply that is getting quite hot. Congratulations. Many greetings to France. Bonne annee.
regards
Bernd 
 
bernbrue said:
Did you tie the time constant controls of both channels together?

All controls(except for input and treshold) are electronically separated but not mechanicaly(I used stereo pots)
Easier for stereo operation but limited in dual mono

I've built it this way to better suit my needs ...(overall mix compression)
 
The attack control is the normal time constant on a stereo 10K lin pot.
The "so called " release is the suggested bleubird mod:
I've tied a 500k log pot parallel to C4.It results in faster release settings(to my hears)
I can switch it back to original with the "mod" switch to compare.
So you left C4 10uf and you just added pot across it to "speed up" a time? 500k/log? Why not linear?
I've tried both schotky and 1n4148 diodes with the dip 8 socket trick .After some audio testing I kept the 1n4148 but this is very subjective.
So there is some sound change  ;) Good to know, I still didn't tried  :-\
After the mixing sessions I changed the treshold steps and they are following (in ohms):
1-500 ,2-500 ,3-100 ,4-200 ,5-300 ,6-400 ,7-500 ,8-500 ,9-500 ,10-500 ,11-1000 + 10000 wich is not selectable.
It means that I have control from 14,9K to 10K attenuation for the  treshold.
Did I understood correctly? You turned the pot into tapper with stop value of 10k max 14,9K?
 
Gachet said:
I add a fan for the european transfo wich is heating a lot to me .
Bonne annee !!!
I'm surprised your transfo can get hot. :eek:
The EMRR test that closed the too long discussion about regulated or unregulated filament PSU makes me think that the filament can eat a lot more current than they should. Getting hotter the resistance of the filament seam to lower... That would explain the little voltage raise during the test... But that would mean extra-current out of the transfo if you didn't put a high enough resistance...  :-\

Try to know how much current goes through the resistor... It should be 5,2A... If it's more, here is your problem !!! If it's not  ??? ??? ???

Let us know, this is weird and interresting...
 
RE: Codered's Avell Lindberg Toroid.

Now that they have arrived I just want to be sure on how to wire it. Here's the colour code.
http://picasaweb.google.com/edcor670/AvelLindbergPM670ToroidalTransformer#

Deciphering this stuff is not my strong point  ::)

Expecialy the primarys I usualy only have 2 wires to worry about.

So far I have worked out.

Orange and White go to the Rectifier
The 2 Black go to AC
Green to Ground
Red and yellow go to AC250 (does it matter which of the two)?
 
Moby said:
So you left C4 10uf and you just added pot across it to "speed up" a time? 500k/log? Why not linear?

Yes.500k log to have some accuracy neer the full short cut of C4.
After that point I prefer the normal mode.

Moby said:
I've tried both schotky and 1n4148 diodes with the dip 8 socket trick .After some audio testing I kept the 1n4148 but this is very subjective.
So there is some sound change  ;) Good to know, I still didn't tried  :-\

This is just my opinion but I found the pm670 less reactive with schotky.
Maybe less crispy when compressing. :-\

Moby said:
After the mixing sessions I changed the treshold steps and they are following (in ohms):
1-500 ,2-500 ,3-100 ,4-200 ,5-300 ,6-400 ,7-500 ,8-500 ,9-500 ,10-500 ,11-1000 + 10000 wich is not selectable.
It means that I have control from 14,9K to 10K attenuation for the  treshold.
Did I understood correctly? You turned the pot into tapper with stop value of 10k max 14,9K?

[/quote]

It means that I can't go into extreme compressing values.
1:
The pm 670 is a feedback compressor (the output is feeding the sidechain) so I use it like kind of 1176 (It's a fixed threshold compressor with in and out pots) .
2:
If you increase the threshold the output level is going down.And we don't have make up gain.
So to keep the output loud I prefer to increase input and fine tuning with my 5k threshold.

I did it to suit my very special needs. :)

lolo-m said:
Gachet said:
I add a fan for the european transfo wich is heating a lot to me .
Bonne annee !!!
I'm surprised your transfo can get hot. :eek:
The EMRR test that closed the too long discussion about regulated or unregulated filament PSU makes me think that the filament can eat a lot more current than they should. Getting hotter the resistance of the filament seam to lower... That would explain the little voltage raise during the test... But that would mean extra-current out of the transfo if you didn't put a high enough resistance...  :-\

Try to know how much current goes through the resistor... It should be 5,2A... If it's more, here is your problem !!! If it's not  ??? ??? ???

Let us know, this is weird and interresting...

I think you're right about current problem but sorry ... I can't measure it right now.
But in fact I had to down the resistor value to raise the voltage from 6.1V to 6.3V.
It result in less current in the heater.I didn't check the heat of the transformer  after that but it must be much safer now with proper voltage.



 


 
emrr said:
The filament is really quite simple.  A given current at a given voltage.   
Just for fun, I measured the resistance of the filament of a 5687WA... About 2,4 ohm each section meaning 1,2 ohm parraleled... So I'm sure filaments doesn't follow the Ohm's law... Getting hotter their resistance grow... It would be interresting to know how !!!
This doesn't mean unregulated PSU is stupid or... It just explain why unregulated PSU is safe !!!
 
lolo-m said:
Just for fun, I measured the resistance of the filament of a 5687WA... About 2,4 ohm each section meaning 1,2 ohm parraleled... So I'm sure filaments doesn't follow the Ohm's law... Getting hotter their resistance grow... It would be interresting to know how !!!

Heaters are very much like light bulbs. So 2,4 ohm is nearly a short circuit when it's cold, but not when it's lit.

Have a look at http://en.wikipedia.org/wiki/Lightbulb section "Electrical characteristics"
 
It's looking like I finally have all the parts to put one PM660 together, I am one toroid away from having two done. Here is the preliminary layout, hopefully I am not making any huge mistakes.

Thanks


Howard
DSC03429.JPG


 
Back
Top