EQP1 Dual Pultec *finished* with *Purusha Case and PCBs*

[hotbaby]

Good Evening all. Today I finished my Pultec after 3 Month working and clarify my uncertainties here I have to thank
this forum here and especially Gustav, Frank (layouted the filter pcbs), Volker (XS1100adaptor) and Jakob for the great self-explanatory PCBs and the Case, Sredna and Harpo who explained the tranny with very much patient 
And all works perfectly! Oh i have to thank PRR too who explained it with the LED very well with a funny picture .
Heres my making of:

...at the beginning...

nice helping tool. i tested every thing for getting sure it works. i got this helper on conrad.de for around 20 EUR (Article Number: 121434).

dont forget the cooling gel.

...so much to solder...

...needed to tilt over the con a litle bit that the cooling body can be placed.

this will help saving my life if i need to touch the condensators again:

I used just one power tranny and heres my wiring:

i dont like to touch 230v, so i srink all cables i could touch inadvertently...

METAL WORK:

XS1100-to-LL5402 ADAPTOR

As i have no own DIY Page i hope you have fun seeing my pics here.
_____
AaR0N 

 

[amishsixstringer]

Beautiful!

 

[Silvas]

Looks gorgeous man, congratulations !

 

[keefaz]

Nice job  congrats!
I dig your component tester, nice piece

 

[matthias]

congratulaions !

that's a very nice unit!!  and well documentated...

 

[Sredna]

I say like Borat: "Very nice!"  ;D

Congrats!!!

Now I have to finish my D-Pultec 

Take care/Frohe Ostern,

Anders

 

[hotbaby]

@all, thank you very much 
This is crazy i have a litle bit hum problems on -76db. I thought the whole night about a solution and read matthias, you had a similar problem. in fact the hum comes from the toroid 250v cables and around the filter pcb there are some groundloops i think (sounds like toroid and groundloop noise). i want try to kill the hum with a new constructed balanced working wiring as i think that i did a mistake in my grounding construction.

The powerboards has a balanced input +,0,- an balanced output and a 3pole "in,ground,out" to the filter pcb (which isnt balanced... ???).

I thought about paralleling the grounds of the two power pcbs and the filters and tild one filter ground to the chassi ground ( is it one case.. perhaps it is better paralleling the grounds of the two mono parts).
Another thinking is experimenting with the signal ground "-" path which i usually wanted to connect with the chassi ground anywhere... but i dont know where on the pcb or cables yet (the hum was a litle bit lower after connecting the - path with one pcb ground).
I think it is right connecting all xlr 1 to the chassi ground... external ground noise from other instruments will be grounded this way. but how to eleminate the internal hum with the signal path... urggghhh its so crazy difficulty. anyone who can help with grounding dual pultecs?

the right mono part on the pic has a noise floor of -68db and the left one which is far away frmo the toroid has -76db... :-\ not good for mastering  :'(

 

[babyhead]

Yeah. Your toroid is HUGE. If you were to take a transistor radio and wave it around you would hear what it is surly putting off. IMHO putting the toroid into an external box will most likely solve your hum without major deconstruction. It looks like you have enough slack if you disconnect the IEC to move the transformer to the other side of the right wall and see if the hum lowers. If that's the problem, you may be able to shield the transformer with the top of a hammond box and old monitor parts, but external mounting is much better. Or a smaller transformer...

Very nice build. Sexy.

Good luck with the hum.

 

[hotbaby]

hey babyhead, thank you for posting. Perhaps replacing the toroid will be my only chance killing the hum... I thought about solving the most hum with a better grounding cheme. it is not easy seeing an expedient grounding on the dual pultec pcbs. The power pcbs where produced for mono systems too that i would suggest it would be better paralleling the signal ground. My idea wiring up a balanced signal ground on the left and right pultec section with paralleling the signal ground that the hum can erase itself. If the
hum is louder on the left or right pcb or the grounding cables have different resistances the effect will be impended. I played a litle bit around with my grounding and could influence the toroid hum very interesting... thats the reason why i think perhaps it is possible to eleminate the hum with a good grounding. I took a graphic of the ungrounded pultec system. I read some interesting articles like the "RaneNote 151" http://img211.imageshack.us/img211/3218/internete.gif and it can be that the pcb design ( if you plug it as descripted ) does groundloops in stereo mode ( One chassi and two ground systems).

(* my graphic is free for experimental skizzes...)

So i tryed to think about a parallel signal path to avoid hum and balance/ erase the toroid hum...
SkizzyA

Maybee i think totaly wrong but i think a correct wiring can erase the hum...  : ???
what do you think about it?

 

[keefaz]

For my part, I would use only one GND connection to chassis
From your build photos, I would try to disconnect the GND chassis connection near the torroid and connect it to the GND connection near the XLRs, without 100R resistor and 10nf capacitor
On second thought, -76dB hum does not seem super noisy to me, so maybe just arrange a little the wires

 

[radiance]

Nice build...however...it's a small box to fit all that stuff into.
For starters, remove the PSU from the box...that is, an external PSU. This will definitely help with your hum problems...

 

[hotbaby]

@keefaz
yes i need to update this, at moment i have disconnected all grounds and search/wait for a flash of genius  ...
on the pics i did the two "cable shields" of "filter PCBs in/outs" and all "XLR 1" to the starpoint near the "XLRs"... i read a lot that i should use
just one star point for signal ground or chassi ground. i did two easy chassi grounds and the shield cables seemes to work like an antenna. if i touch the cable rubber i can hear the hum... Two Starground Connections can produce groundloops or louder hum... i think this got true. Dont you think seperating chassi ground and signal ground and connect both with an condensator could be better (like on my skizzy)? I read a lot good things about it. they sey good grounded balanced systems normaly has an noise floor but it erases itself as it´s well wired. I dont want to displace the toroid but i think i must...

@radiance
do you mean both PSU pcbs or just the Tranny? This would suck if this is the only solution  :-\
what do you think about the grounding? I have problems seeing what need to be grounded to the chassis and working with signal grounds on the chassis...
perhaps you have an idea.


The noise is not much loud but the right channel has a -65 noise floor which is distracting if i want to use it for mastering or on stereo files.
thank you very much for helping.

 

[keefaz]

For my part, I would connect the grounds like this:


(The in and out from filter board to line amp board using the same ground wire)
[edit]
(Picture edited)

 

[Stagefright13]

That IS Beautiful! And I love seeing pics. Thanks for sharing. Usually a hum problem can be tracked down "Fairly" easily. Don't despair. I have seen tons of stuff crammed into a chassis including the power supply with no hum. When there is a will there is a way. Congrats and hope ya get it sorted out soon!

John

 

[hotbaby]

Thank you very much Stagefright. Yeah, i did it for us guys who love seeing pics here I hope i can inspirate someone with it. I like the tube tech style i modded the purusha case a litle in this way...

@keef
Thanks for your draw, you would tild left and right direct to the xlrs... I think that could cause an pin 1 problem that the shield/chassi noise goes direct to the input signal ground "0". What do you think about this mix i did with your draw:
SkizzyC
it is based on the pin1 problem pdf where the xlrs shouldnt go direct to the signal ground:
http://www.audiosystemsgroup.com/Pin_1_Revisited_Part_2.pdf
the only difference is that the signal ground is attached to the chassi on an other place/other site and is connected to the chassi ground.
I read in some articles, they said that the xlr ground should be connected direct to chassi and not to the signal ground if i understand it right...
what do you think? it would be great if Jakob could say anything to the dual pultec grounding too...

another thinking was wiring a complete internal grounding that all noise which hit the pcbs will "balanced out"/erased and the noise of the extern cable shields/chassi should go direct to the ground:
SkizzyD can it work?
(the last two green/blue signal ground strips which are going to the star point are shielded cables that they cant attach other chassi noise...)

 

[keefaz]

On your first drawing, by connecting filter pcbs gnds togeither, there is a risk of crosstalk between channels
Your second drawing should work although I don't follow the drawn shield things, not my personnal preference though,
I prefer connect signal ground and chassis ground close to XLRs inputs

 

[hotbaby]

Ah yes you are right, on the first there are two connections to the chassi. this is not good.
the second "SkizzyD" i wire the two signal grounds to a groundstar and then to the chassi ground. the two signal ground wires could be shielded wires i thought that they do not eat other hum on the way to the star point and the shield of the cables would go direct to the chassi (the two blue wires). that was the idea.
i dont know what is better but "SkizzyD" or your Skizzy. The last one "Skizzy D" would be the way like rane 110 and the other grounding articles suggested if i understood it right.
perhaps on the star point i could add an 100ohm Resistor and a 10nf capacitors to the chassi to avoid extern hum in the signal ground... I asked Jakob to look on our skizzes, a third view wouldnt be wrong. But you helped me out, i think i am on the right way now and can stop this noise with one of this grounding ways.

 

[keefaz]

Be aware that your signal ground holds high voltage, it is a tube circuit remember...
Applying ohm formula with your 100 ohm resistor: 300 * 300 / 100 = 900 watts!
Rane is good reference, but I think it is for solid state circuit (I could be wrong though)
Forget the 100 ohm and 10nf capacitor to connect signal ground to chassis ground, just use direct connection, like everyone do with this circuit
[edit]
Maybe I am wrong with the signal ground holding high voltage, after all it should be a 0V ref..
I think I need to do some electric homework  Better wait some expert advice...

 

[hotbaby]

uh yes the tube thing is really dangerous, i tested all knobs and chassi for current before i touched it the first time... building my mp2 was quet holiday for me against the tube things  thanks for the advice with the 100ohm... seems it would be a fast death if there would be 300v and the resistor on the chassi  :-\ ok direct connection.

 

[keefaz]

I think I was wrong, it seems not possible that there would be a 300V voltage between signal GND and chassis GND (hearth),
maybe if a short happens with B+ and signal gnd but even in this case maybe the fuse would blow before the 100 ohm resistor
Anyway I don't know and never see a DIY pultec build with isolation resistor between signal GND and chassis, so I can't say.. Better be safe than worry