EQP-1 With Cinemag on output...need help

Hi ,
I am a building eqp-1 , not a first time , but this time with Cinemags on interstage and output.
Because they are little different from the original specs , Daren was so kind to send me changed schematics.So , first problem....the very first thing that I notice is that whole circuit is very sensitive, and without 620 pf anode cap (originally 20 pF) I get nasty oscillation and pulsing on the output.Second , filter start to act strange , and I was able to get only 10 db of low boost.The worst thing are voltages....until those 200k 1W resistors they are like they  suppose to be , but when I connect two interstage termination resistors(47 k) to ground , anode voltage drops from 140v to 60-70V. Cathode voltage is also very low , only 0.3v.
So, on the end , I change the two cathode res from 470 r to 2K2 , and that increased anode v to 120 and cathode to 1.3 V , but  filter network response was the same , only 10 db of low boost.I increase those two 47k res on the CMLI 10/600 secondaries to 470 k and  only than I start to get near the expected Pultec eq curve...something like 13-14.5 low freq boost.
I really don't know enough on the tube stuff , so if someone can look at the schematics and give me some sugestions how to finish circuit , I will be really grateful.

> I connect two interstage termination resistors(47 k) to ground

The grids HAVE to have a DC path to ground. If the transformer CT is not grounded (a common approach, though not shown in your schematic), then you need <1Meg resistors grid to ground (47K is fine).

> anode voltage drops from 140v to 60-70V

I don't think 60V is wrong. We only need 9V of swing into the 12AU7. A 12AX7 with these parts can easily do that with Vp as low as 50V.

> voltages....until those 200k 1W resistors they are like they  suppose to be ... Cathode voltage ... 0.3v. ... change the two cathode res from 470 r to 2K2

Does not add up.

0.3V in 470 is 0.64mA. 250V-60V is 190V, 190V in 200K is 0.95V. Unless your meter-loading is very severe, there's a third of a mA missing.

> increase those two 47k res on the CMLI 10/600 secondaries to 470 k and  only than I start to get near the expected Pultec eq curve

Looks like something is not wired they way you think it is. The specified 2*47K on secondary of a 1:4 transformer will throw 5,875 ohms to the primary. This is a "no-load" condition for what appears(?) to be a 600 ohm termination. Increasing the secondary load to 2*470K should make only a tiny difference. But it does. Something isn't what you think it is.

Although the EQP1A is sort of a 600 ohm design, the low cut is not. As this is what feeds the interstage transformer, it needs to be loaded with no less than 10K for the lo cut to work as advertised.

Cheers

ian

PRR said:

> Something isn't what you think it is.

Click to expand...

Yes, this is my first though , and probably most expected scenario....I learn it ...to many times.
But ....I checked everything again and again , and at the end ask my friend to check everything again , every wire , every res is on right place.
Finally I was forced to exchange Cinemag CM-9600-T output transformer, because I suspect that tertiary winding , which btw don't have CT (same for interstage ),make all this strange cathode voltage, with Carnhill vtb 2425 (12k5:600R:24.5R) , and all instabilities gone.Strange thing with 630pF cap on anode with Cinemag on output , was also boost from 5Khz to 20Khz which was more than 4db!!!! , With Carnhill I put 20pF  , and response was totaly flat.
I wire Carnhill , like it was on original  eqp1-a schematic , with 360 r res , 1K on CT to ground and 820r on cathode, but honestly I liked sound of the Cinemag more. So my question is , how to configure resistors with tertiary winding so whole ecc83 start to work more stable?

ruffrecords said:

Although the EQP1A is sort of a 600 ohm design, the low cut is not. As this is what feeds the interstage transformer, it needs to be loaded with no less than 10K for the lo cut to work as advertised.

Cheers

ian

Click to expand...

Tnx jan for response ,
I am glad that somebody that really understand passive circuits wants to help to novice like me.
I constantly looking at the curve of lots Pultec clones, and on many I can notice almost 17 db of low boost in some extreme settings(100Hz on 10) .
I stacked with 13-14 db max , so I am wondering what part most influence bass boost , never have problem with low cut and highs. That few db mean that in relationship with low cut attenuation , curve will not be quite eqp1-ish.

sasarist said:

ruffrecords said:

Although the EQP1A is sort of a 600 ohm design, the low cut is not. As this is what feeds the interstage transformer, it needs to be loaded with no less than 10K for the lo cut to work as advertised.

Cheers

ian

Click to expand...

Tnx jan for response ,
I am glad that somebody that really understand passive circuits wants to help to novice like me.
I constantly looking at the curve of lots Pultec clones, and on many I can notice almost 17 db of low boost in some extreme settings(100Hz on 10) .
I stacked with 13-14 db max , so I am wondering what part most influence bass boost , never have problem with low cut and highs. That few db mean that in relationship with low cut attenuation , curve will not be quite eqp1-ish.

Click to expand...

This document describes how many passive EQs work, including the EQP1A. If you think of VR4 as the hi boost,  VR2 as the hi cut,  VR3 as the lo boost and VR1 as the lo cut then you have an EQP1A.

http://www.ianbell.ukfsn.org/EzTubeMixer/docs/EzTubeMixer/UniversalEQ/PCB/UNIEQConfiguration.pdf

Cheers

Ian

check Cinemag OPT with scope, look for phase shift on teedback windings compared to pri signal.

use Zobel network on OPT sec as per EQP schemo
maybe try 1 K resistor on OPT output to load circuit,

CJ said:

check Cinemag OPT with scope, look for phase shift on teedback windings compared to pri signal.

use Zobel network on OPT sec as per EQP schemo
maybe try 1 K resistor on OPT output to load circuit,

Click to expand...

Tnx CJ,

Yes I already try 620r 0.003 on the output and it just make more shift of high freq (this was in scenario with 630 pf on Anode).
But what concern me most is why Cinemag tertiary with 150r on each side and 430r from cathode to ground was dropping cathode v to 0.3v , and Carnhill with 360 r on each side 1k on CT with 820r  on cathode produce 1.2v!!!???

Ok , I did little research and it seams that Pultec gain circuit is extremely sensitive to output transformer inter-winding capacitance, which can cause nasty phase-shift between feedback and primary.Is there anybody who actually have success with CM9600-T on Pultec circuit?

Does the Cinemag have an inter-winding screen? If you ground this it will significantly reduce inter-winding capacitance.

Cheers

Ian

The CM-9600T's tertiary winding is bifilar with the secondary windings, realy dont know about screen........probably not.
You mention that low cut topology is actually 10k....my interstage is 600 ohm on primary....can that make trouble with low cut curve?

Cheers
Sasha

hi

do you sorted out what was the problem with your circuit? I'm building the same build as you with the same cinemag transformer. Do you try reversing wire on the interstage and opt? I saw on many build and even on the schematic that the interstage is reverse wired.

also, do you try to talk with David at Cinemag? He could help you for sure.

Deepdark said:

also, do you try to talk with David at Cinemag? He could help you for sure.

Click to expand...

David...that guy is probably the best support you can get in whole audio industry.
I am simply amazed with attention he give to me and with effort which did to help me out.
Cinemag...David....10+

totally agree. the guy is awesome. This is one of the reasons I try to buy as much from cinemag as possible.

Deepdark said:

totally agree. the guy is awesome. This is one of the reasons I try to buy as much from cinemag as possible.

Click to expand...

+1