Pultec filter section question

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

JPrisus

Well-known member
Joined
Sep 9, 2004
Messages
145
i'm thinking about adding the mid cut filter from an MEQ to the Gyraf Pultec EQP1 clone... is it possible to just 'drop in' the mid cut circuit somewhere in the Pultec schematic? If so, what makes the most sense? anyone try this before? It's pretty standard amongst the modern pultec-inspired EQs like the DW Fearn VT4 etc.

I'll probably be doing the entire filter section point-to-point and using the PCB for the power and makeup gain sections.


BTW, got my GSSL PCB stuffed and soldered in about 4 hours today... waiting on a voltage regular which was backordered, can't wait!!!
 
I always wanted to try that, so I will get on the simulator and see if it can be done.
The mid cut would sit underneath the treble boost section, so interaction might need to be addressed with some isolation resistors.
MEQ is 10db cut, EQP is 16, so I will try and tweak the MEQ for 16db, a matter of changing the pot and therfore inductor/cap impedances.
cj
 
glad you're interested CJ, as it's way over my head to figure this out! I was hoping someone else shared the idea. Should make for one helluva box!
 
Yeah, the mid cut can give you some really neat vocal effects, like the nasal sound ala Rob Thompson on the Santana song smooth. In the very beginning of the song that is.
 
ok, i think I got it!
works on the slimeulator at least, (which means it ain't crap til we build and test)
Sorry for the oversized pic, but it won't reduce without chopping up the image.
You have to change the midcut pot from 10k to 2 k, add a 10k resistor underneath the midcut pot, otherwise the eqp bass boost curve gets tweaked for the worst, and add the 510 ohm resistor. I squeezed as much mid cut as I could out of the circuit without sacrificing the eqp curves. I think I got 12 or 14 db instead of 10, which ain't too shabby.
The 100k on the output is just a dummy load for the transformer:
z
meq+peq_combo.jpg
 
As I read that schematic, there will be little variation (from a lot of cut to a little less cut) when turning the mid-cut pot. It won't be "off" at either end of it's travel...

Jakob E.
 
[quote author="gyraf"]As I read that schematic, there will be little variation (from a lot of cut to a little less cut) when turning the mid-cut pot. It won't be "off" at either end of it's travel...

Jakob E.[/quote]

any way to solve this?
 
R6 is controlling the Q though, no? In the original MEQ shematic it's shown as a 270k variable resistor in the same spot.
 
ah i see, so the Q and cut are kind of dependent on each other... thus experimenting with different values for R6 would let me tweak the Q to my taste? (or make it variable) If so, that's pretty damn cool.
 
I'm a little concerned about the mid-cut never being 'off' as stated above. you're implying that with the cut level as zero, i'll still be getting some degree of cut?? Can't have that.

The other thing mentioned, cut not being as deep as possible due to the Q, that's ok with me, i'll mainly be using this for 3-4dB gentle cuts on vocals and i'll probably just find a Q that works well for my tastes, taking out some of the midrange 'room', helping the vox to fit in the track a little better etc. I'll stick to a band-pass filter for the Rob Thomas effect heh. :thumb: :sam:
 
> concerned about the mid-cut never being 'off'

With R7 shorted, the mid-dip is out of circuit at lowest pot setting.

R6 affects Q, but so does the rest of the network. I see much more than 270Ω hanging around, so the value of R6 may be meaningless.
 
OK so R6 won't be much of a factor... what if the Q of the midcut doesn't sound good, am i stuck with it that way?


BTW this is great stuff guys, very much appreciated. Can't wait to get my SSL wrapt up and start on this guy.
 
also, how do these circuit tester programs like SPICE work? I hear you guys talking about them, but i have no clue how they work or what to even look for within them. any guides available?
 
another thought... what would it take to cut the mid-cut circuit in half, instead of having 11 Hz options, maybe doing 2 mid cuts with 6 options each... a lo-mid and hi-mid? i suppose we'd need a seperate inductor for each, not a problem, i'll make as many as i need to. Is it as simple as having both circuits in series?
 
Yes, good idea. I used a bigger core for the lower half.
This improves the Q of the inductor and reduces stray capacitance.

I threw that circuit together in about a half an hour, so don't expect any miricles, but it does work on the simulator.
I will try some other tweaks and layouts.
The 270 ohm resistor has quite an effect durring simulation, but not in real life. I guess the ear responds to peak values, like the eye responds to peak values of light. I put a 1k pot inplace of the 270 ohm for a vari-Q thingy, but ripped it out. Only when the 270 ohm approaches zero will you get any super noticable effects, at least on my project.
Note that the MEQ uses a reverse log pot.
I had about 2 db of unwanted cut in the off position , but I had do do this to keep the bass boost circuit from dropping out too much. You can experiment with different values of pots and the 10k resistor under the pot, take it out alltogether if you like, but then the bass boost drops form 16 db to about 10db. It's a compromise, which is probably why you don't see a mid cut on the EQP-1A. But remember that these Pultecs where meant to eq phone lines, not audio, so they wanted cheap and simple.
You should be able to get some pretty whacked out curves with that mic-cut added on. A lot of combos to simulate, especially if you check all frequencies and combos also.
Rumor has it that a certain company has a proto ready, kind of a mini-massive, so keep an eye out, might make a cool diy eq.

cj
 
you have to think to build the NewYorkDave EQ, it got the same make up stage of the G pultec but different filter arrangement it got also 3 band with possibiliy to add a 4th. Bluebird (hellofaguy ) hase build six.

the ideal solution for me is to have 4 band and cut at the cornered.
I'm planing, to build one of this ( NYD Eq, with G-Pultec SRRP stage card )
but maby the passive topology is not ideal for this purpouse.
EAR EQ is not a passive one, MEQ Fairmann isn't passive too...
so if you a need tube Eq for his flow G-pultec it's ok, if you need a Tube mastering Eq, other topology are prefered, maby.
 
for me it's got more to do with the passive inductor-based filters and the transformers than even the tube thing... my fav pultecs were the opamp guys anyway, and the Lang PEQ. But an extra tube or two, well implemented, never hurt anyone. No mastering going on, just gentle sculpting of vocals, bass, etc.

the DWFearn VT4 gave me the idea for this guy. i can live with 10dB of bass boost in order to get the mid cut filters happening. Another idea i'm having is losing the bass cut and replacing it with a smooth HPF with 3 or 4 selectable frequencies, say 30, 60, 120 or thereabouts. That way you can cut at a certain frequency and boost at another, the overlapping effect is quite awesome sometimes. And also possibly losing the HF cut too, since the hi-mid cut will reach up to 7k or 8k.

so let's sum up the ideas:

trafo input
HPF 30, 60, 120, 180
Bass boost 40, 60, 80, 100, 120, 150
low-mid cut 200, 300, 500, 700, 1k, 1k5
hi-mid cut 2k, 3k, 4k, 5k, 7k, 8k
Hi shelf boost 10k, 12k, 16k
SRPP tube gain stage
trafo out

sounding cool? This will make it more of a unique unit, set it apart from a typical Pultec clone. Lots of control over lows and mids, and that sweet top we all know and love.
 
Back
Top