the Poor Man 660 support thread

[mich]

Maybe its time for the schematic of the poorman to go "on air".

maybe the more expirienced tube guys here will have a say... 8)

 

[[silent:arts]]

Maybe its time for the schematic of the poorman to go "on air".

it will only go "on air" if analag decides it.

it is sad to know not everybody likes this project, but that is the way it is (like with any project we have around).
no doubt, it isn't easy to build (this doesn't mean anybody disliking it build it wrong).

the "Fairchild" sign might have kept the Expectation a bit too high,
and I don't know how many of us used a real fairchild up to know ;D

 

[lolo-m]

the "Fairchild" sign might have kept the Expectation a bit too high,
and I don't know how many of us used a real fairchild up to know ;D

You got it right... Most of the builders wanted a piece of myth in their setup and don't have it... Maybe they would be disappointed by the real thing too... Who knows? Most of us have only plugin's to compare so...
PM670 is a good comp IMO.
It was a lot of things to learn about varimu technology and Fairchild topology. I'm not disappointed because I didn't expect too much from the PM670 except what I'm here for : learning. And I learn a lot  .

 

[mich]

it will only go "on air" if analag decides it.

No doubt about that.but since its been a long time now and a lot of people are just selling their units - maybe he doesnt mind..

Maybe they would be disappointed by the real thing too... Who knows?

- Very True.

 

[Kingston]

I will list my gripes, none of them based on any fairchild as I've never heard one.

1. the Edcor transformers simply aren't cut out to balanced push/pull use. They are too asymmeric, and generate bad sounding "lofi grain" distortion with odd phase problems. At least it's somewhat subtle.

2. this thing doesn't handle modern levels, not with clean non-distorted compression anyway. This is something only a true brute force approach can solve in this particular architecture. It would mean a complete remake of the sidechain, a different vari-mu tube (and more of them in parallel) and possibly more juice needed from the PSU.

3. The time constant network is far too simple. At least this one can be easily upgraded.

When I started building this project I didn't know much about vari-mu compression at all (especially related to the 2nd point above). Had I known, I would have avoided this project. can't really blame anyone but myself for being disappointed. I should have done my homework, instead of blindly buying into this project in a feeding frenzy.

 

[mich]

I forgot to mention that i like my unit....

here is some music that has been "mastered" with it :

http://www.myspace.com/digitalprimitives

check out "walkabout" or "people"

will be nice to hear what do you think about it.

 

[elektrovolt]

wow, that is some cool music
well, I can only say that it sounds good, warm and cosy, but muffled. drums sound great!
as far as myspace sound quality goes.

I disagree with Analag - my experience is that this circuit is extremely sensitive to tube mismatch, potmeters / switches balancing and I really think that better transformers would help a lot.
Edcor quality inspection is just not of a consistent level (you want a center tap ? ok, we'll put it somewhere in the middle then) but they are priced accordingly.
If you want some high quality Sowters you are talking 8 - 10 times the price of an Edcor.

I expected more of it indeed, but not much more - what was it called again?
I had the luck to play with a real beaten-up fairchild 660 once. it was humming and rumbling but the compression action was obvious - thick and massive but still natural.

As I am inbetween jobs now and just moved to a new home - a true poorman  I just cannot afford this project right now.

 

[emrr]

Owning a lot of vintage tube limiters, and owning a still uncompleted PM set, I can't see any particular faults with the circuit design on paper.  It seems there are probably QC issues in the build; exact tube parameters, layout (user, that is), etc etc.  For one, I'm skeptical of the use of non-soldered connections for tube plate to transformer connection.  Seems like an easy intermittent failure point.   If there's a single most obvious weak spot, it would appear to be the control amp output transformer rating.  We still don't have a report on it's max output capability, or it's acceptable conditions with regard to current flow.  Looking at the active stage, the OT appears under-rated; sometimes we do that on purpose.    

the extreme basic info I still haven't seen, or missed, would be:

Max input of each amp
Max output of each amp
subsequent flow chart of gain-staging

Gain-staging with vintage limiters is probably the #1 thing people screw up.  We seem to have a 50/50 split between happy users with no problems and folks who can't get theirs functioning correctly.   No one has developed the necessary test procedure to diagnose the problems.    I can't see why we would discount the happy reports in favor of the negative ones.  

 

[lewilson]

I use it for mixdown only.  taking 6 to 10 db off.  + or - a few db. The meter stays about  half way up, around -6 db  , going up a little and  down in quiet parts.

 

[Moby]

it is sad to know not everybody likes this project, but that is the way it is (like with any project we have around).

From the very beginning I had a problem with few facts and I still have a problem with them. It's not a matter of likeness or taste, those are facts.
1. I remember the words of author that matching tubes are not needed because of  his design. Sorry that's not true.
2. Using term "Vu meter" simply don't work because it's not same if use sifam or some other "vu" . I Still don't understand why it's not determined so many people will skip that problem.
3.Edcors are definitely not good enough for this kind of circuit , for mastering or serious pro recording, so info about replacement is really needed. ( I have some ideas but don't want to spend money on that)
4. For stereo link both pairs of tubes must be matched to each other because GR tracking will be terrible. Did somebody ever mentioned that?
5.I understand that unit "works" with all missing aspects but I have a feeling like driving a car with broken muffler. It goes but  :-\

I probably missed something , sorry , I'm typing in hurry, so please fill the missing facts.
At last I typed more than author  , spent an enormous amount of hours thinking about his design (just to help myself and other people) so I don't have a problem if somebody find my comment too personal.
Cheers  ;D

 

[radiance]

Analag, this is the time to comment on all this. Have the gutz...tell us about your poorman unit, how you like it etc. etc.

 

[elektrovolt]

Moby, second that. well, i bought my 8BZ8's as a matched quad, but I do not know how well they are matched. no testing report or tube tester here.
every combination of 6BZ8's gives slightly different results. the design and transformer ratings are just not optimal. it is a good study - redesigning the circuit instead of hacking and modding the existing boards. don't get me wrong here, it is not bad at all! but it just has some quirks.

I am not selling my poorman set because I am disappointed, just because I really am a poorman now 
all the trouble I had with my poorman did cost me lots of money and time, and now that I found out that I do not have a good use for it. well, put it up for sale then.

 

[Moby]

Analag, this is the time to comment on all this. Have the gutz...

I don't think it's a matter of gutz, guy is simply busy . I have a plenty of time in the pause of current 6 recording projects what I produce. Well, nobody's perfect 

 

[lolo-m]

For "rich" poeple who want to go in a transfo exchange (T1 + T2), I would recommend those two from Sowter :
SOWTER 9063 as T1 and SOWTER 9900 as T3 (about 80£ each). But I bet it won't change the overall characteristics of the comp as it is mainly due to the 6BC8 character and to the TC network + sidechain amp... EDCORs are ok for the sidechain IMO. Their unbalance isn't really problematic there.
If poeple are interrested I've got an idea of an easy balance checker for the 6BC8...

But all this should belong to the PM MOD Thread...

Analag, this is the time to comment on all this. Have the gutz...

I don't think it's a matter of gutz, guy is simply busy . I have a plenty of time in the pause of current 6 recording projects what I produce. Well, nobody's perfect 

 

[Kingston]

But all this should belong to the PM MOD Thread...

not really if it's about making a somewhat flaky project into a decent one. the original design is clearly flawed (or at best very poorly documented) and needs more work. Not modding, but getting it right.

 

[[silent:arts]]

... the original design is clearly flawed (or at best very poorly documented) and needs more work ...

excuse me, but this is wrong.
or do you want to say the successful stock-builders are bollocks

 

[lolo-m]

... the original design is clearly flawed (or at best very poorly documented) and needs more work ...

excuse me, but this is wrong.
or do you want to say the successful stock-builders are bollocks

Please don't feel any real offence in what's said. Try to understand what's happening.
I'm a successfull builder. I think the comp is Ok but it is a poorman compressor, nothing more. Its character can be liked or not because it does have a character.
Fact is : there's a mythic name on it (Fairchild), the schematic isn't open source, the PCB designer (you) isn't a tube eagle, the project is one of the most expensive, there are many problems to build this compressor and to get it working and the designer doesn't post anything on his own design. Most of the trouble shouting have been done by poeple like me, Moby, EMRR, Kingston (...) but none (or so) by the designer himself... Evidence is that you'll have disapointed customers. Man, don't forget you sold boards at a price so poeple want to have something working...

 

[bernbrue]

It´s fully understandable that someone sells his poorman for personal/financial reasons. There are no further reasons to be mentioned. Bringing this fantastic project in miscredit in order to justify the decision to sell the project is not fair at all. The project has been discussed and documented in various threads in the Drawing board. Prototypes were built, soundsamples and lots of explanatory graphics were posted. Lots of people helped other people troubeshouting their mistakes and errors. Some experts suggested mods and improvements. There is no reason to complain about anything.
regards
Bernd  

 

[Kingston]

or do you want to say the successful stock-builders are bollocks

no one is accusing you of poor documentation. you've covered the build-by-the-numbers aspect really well. Just that the actual design choices, parts of the architecture are left in the dark, with no one to comment on why.

lolo-m kindly covered with much the same as I was about to say, but there is an aspect with the build-by-numbers folk that is often ignored: they might not know a flaw even if they see or hear one. They're just happy that the unit didn't explode and that they managed to stay alive.

I had a kind of extreme surprise with G9 when it came to that. This is project that had been going for like 5-10 years, and it has severe impedance issues and the B+ part of the PCB is basically broken. This means a standard G9 has some -50dBu 50hz hum at normal gains and no bass (due to impedance mismatch). Some people even say "it's part of the sound"! I'm talking hundreds of happy builders. I find this kind of collective blindness baffling.

Keep that in mind next time someone says poorman sounds "awesome"...

PS. I would also advise on bringing the schemo out in the open, so the more enlightened tube guru's could have their say.

 

[radiance]

I had a kind of extreme surprise with G9 when it came to that. This is project that had been going for like 5-10 years, and it has severe impedance issues and the B+ part of the PCB is basically broken. This means a standard G9 has some -50dBu 50hz hum at normal gains and no bass (due to impedance mismatch). Some people even say "it's part of the sound"! I'm talking hundreds of happy builders. I find this kind of collective blindness baffling.

Well, kind of baffling. I searched in the official G9 help thread and I was the first to comment on the flawed bass response http://www.groupdiy.com/index.php?topic=17980.100
At that time I knew nothing about tube circuits (still don't  ) so I thought it was something I had messed up. Therefore I don't think you can say "hundreds of happy builders" or "collective blindness" since most of those builders were newbees at the time.
Beeing a newbee you are not really in a position to question a supposedly "proven" design AND, for a "supposedly proven" designer it's very hard to admit a design flaw. So far I've seen NON doing this.