Messing with a pultec filter...

[Neeno]

Hi guys...
I passed the last week messin' with a pultec filter, it is based on the gyraf version but with some more frequencies (ala Tubetech)

Since I'm going to keep it simple (and cheap) but still good sounding I'm using a solid state gain stage.

The gain stage is based on NE5532, i slightly modified an existing design feeding the signal into the non-inverting inputs.
The gain stage is working properly (actually i even tried to use the JLM Dingo) but It didn't solved the problem...

I had a manley stereo pultec lying around for a couple mastering session, and after using it for a couple weeks i decided to build one (had to send back the unit to the rental service)

The problem description.

It looks like my pultec has a problem in the low-boost/cut section. The low-cut pot works properly, but the low-boost acts like a volume control (adding bass frequencies of course) but it sound like it has a very large Q, boosting even frequencies over 1kHz).

In fact i can NOT turn the low-boost potentiometer more than 9 o'clock.

And now... the questions... (sorry)
- Considering that I'm using a IC gain stage I'm loading the filter input with a 10k resistor (do I have to use a 620R instead ?

- I tried different configurations for the low filters summing resistors. Actually I'm using 1k for the low-boost and 10k for the low-cut. Is this right ?

- Anyone had this kinda trouble with his pultec build ?

- Could it be something wrong in the gain stage ????

Thank you for your help.

 

[sintech]

Hey Neeno.. I've had.. I think exactly this problem before, also Purusha.. I'm also guessing it's the same original layout maybe??

In the end we never worked out the problem and rewired the whole thing.

We pulled the gain stage in the testing and used a powered speaker as the makeup and it was exactly the same!

We both used your inductors also (I'm sure this is not the cause) but as a point of reference.

 

[Neeno]

Hi Andy !
If I recall correctly he had a different problem (the same that i encountered when I tried the existing layout) the low-cut didn't worked right. I found that the problem were the 2x1k resistors summing the low-cut/boost filter.

After changing the low-cut summing resistor to 10k solved the low-cut problem, but now the low-boost don't work properly.

I don't think the inductor is the problem, the hi-boost section is working properly (I did some sweep and the curves are right).

 

[syn]

I played with pultec filter section for a months and one thing i know for sure is: if the part of the filter network does not work at all or properly the problem is right there in that section. Filter section will work ok for testing purposes fed directly by a sound card and monitored on a small speakers (PC type), that is the way i experimented with mine. Wrong loading of the f. section will result in a reduced amount of gain available at hi boost, but will still show ~right frequencies. My passive eq can be seen here, and I'm using 100K (10K works as well) from Lo Boost taper to output. No resistance from Lo Cut taper to output. It is fully discrete + transformers in & out.
IMHO you have either a short or no connection or wrong connection, the problem is somewhere within LoBoost section...
Test the filter section only for now, so you can be sure that it has nothing to do with the gain make up stage...But then agin I'm no expert :wink:

good luck

P.S. Mine eq input is loaded with 620r as that gave me better THD for some reason eventhough the opamp has to cope with the heavier load...
10k gave me a bit more thd,and if i was not tripin', it sounded a bit "thiner"...

 

[syn]

http://www.groupdiy.com/index.php?topic=6850&postdays=0&postorder=asc&highlight=pultec&start=15
if you have not already see i think last pages of this thrad regarding
resistors from BBoost and BCut......

 

[ioaudio]

grüezi neeno,

check your wiring - especially the bypass - it´s in the filter section for sure -as syn already mentioned.
620R is correct for the input termination, but it has nothing to do with the type of make up gain. 10k also works, but it seems to be depending on the input transformer´s secondary (just a guess from what i´ve read)

NE5532 - does the input have a dc reference? did you couple it via cap?

-max

 

[radiance]

[quote author="Neeno"]
The gain stage is based on NE5532, i slightly modified an existing design feeding the signal into the non-inverting inputs.
.[/quote]

So how much dB are you able to squeeze out of that NE5532?

 

[Neeno]

Hi guys !
Thank you for your help !

Syn: thank you for the link, that's really interesting, and well, I missed it.
Tomorrow i'll try some different configurations for the resistors and report back.


Ioaudio: gruetzi (that's almost all my german skills... btw I live in the southern swiss, we talk italian here... so... buongiorno sounds better to me... :grin

Regarding your question, yes, I'm coupling the filter trough a cap, 470uF.
Check out the JLM Dingo, that's almost like the configuration I'm using now.

http://www.jlmaudio.com/JLM dINgO complete schematic.pdf