need help with passive Pultec low/high boost only EQ

[BradM]

I'm trying to make a simple passive EQ that basically is just the low frequency and high frequency boost sections of the Pultec. However, for the high boost it's just going to be a shelving filter so that I can just get away with using capacitors. I want to be able to send a low impedance signal from a D/A converter into the EQ and then connect the output of the circuit to the high impedance DI input of my outboard mic preamps for make-up gain.

I wired up a little prototype with alligator clips to test it out, but it seemed like it need a lot of make-up gain, especially when I subbed in a 100k pot for the treble boost.

Here's a pic of the basic circuit:

http://www.littleredwagonstudios.com/passiveEQ.jpg

C1 will be different capacitors on a rotary switch for the high boost and C2 will be capacitors on switch for the low boost. The 10k lin pot is for the high boost and the 10k log pot is for the low boost.

I have the following questions:

1. Am I using the right ratio of pots and resistors considering I want to interface this with the DI input of a mic pre for make-up gain?
2. Lin or Log taper for the treble boost?
3. I was using the following formula to calculate the corner frequencies for the high shelf:

|Eout/Ein| =

square root of {[R2^2 + (2pi f R1 R2 C)^2]/[(R1+R2)^2 + (2pi f R1 R2 C)^2]}

where R2=(10k + 1k)=low boost pot value and R1=10k=high boost pot value.

Is this correct? I definitely noticed that when I switched the 10k pot to a 100k pot for the high boost that the corner frequency seemed to shift a bunch. Not only did I get much less boost from the 100k but I needed much more make-up gain on my mic preamp. I almost had to turn the thing all the way up.

4. I'm assuming that if I use a 10k pot for the low boost then the Gyraf Pultec capacitor values will give me the frequecies on the Gyraf schematic. Does changing the high boost pot value affect the low boost corner frequency?

Sorry if any of these questions are kind of remedial. I'm still trying to get my head around the basics.

thanks,
Brad

 

[BradM]

Can anyone help me out here?

Brad

 

[BradM]

bump

 

[CJ]

Which Pultec circuit are you using?
I am having a hard time finding any similarities.

 

[BradM]

I should have clarified...I was referencing the Gyraf version. I was also referencing a thread linked in the Meta:

http://www.groupdiy.com/index.php?topic=11342

If you just focus on the filter circuit, what I want to do is remove the high and low cut portions and turn the high boost into a simple high shelf so that I don't need inductors. For make-up gain I would ideally just run it into the high-Z DI input of a mic preamp.

Brad

 

[CJ]

I have not simulated yet, but take out that 1 k resistor and make that pot a 1 K instead of 10 K.

Then remove the shorting wire on the top of the pot.

Just make the wiper go to a 75 ohm resistor, then the other end of the 75 goes to the cap.

 

[BradM]

Are you talking about the bass part of the circuit? Sounds simple enough. Would the capacitor values from the Gyraf circuit still give the intended corner frequencies though? My intuition says no...

thanks!
Brad

 

[cwatkins]

Brad, I had a very similar project, it worked so well many people said I should market it as a after daw "sweetner".
Not to be specific or anything, but I basicly ran a bunch of combinations through the simulator; I was looking for more of a vinyl sound. I eventually ended up building the gyraf pultec and fixing the power supply problem in the US. at that point I liked that better than the passive solution.

If I remember correctly, It depends how much you boost/cut for your makeup gain, so if you want to add a 6db shelf from freq X, you have to boost your mic-pre 6db, or 12 if it goes unbalanced, etc. I(becase if you look at the simulator, basiclly you are subtracting every frequency except the filter and that's how the boost is made. Much like alot of people say, always use subtractive eq first, But if you subtract say highend and low end eq, you end up with a mid "boost". So, then you just have to move everything back up to "level', say with your mic pre. Impendance output was a big PITA for me, so the project developed to using those balanced IC chips, but a transformer would work fine. So you can do either to bring you to 300ohms. it works anyway, without anything, like a unbalanced signal without as much gain. as far as corner/etc, I found the simulator much easier to use because it's only two filters and then you can mess with it a bunch, I think you'll find you may not want exactly the gyraff pultec freqs, which are different than the original. as far as low to high corners, usually it's a flat line in the middle, but again, the simulators give a much better idea.

anyway, when it works, when I put it on the other side of the lavry, it was great, I had the correct low end bump and the correct missing "air". Even with chips and small boosts it makes D/A's sound way better, it feels a little wider too, it's just odd how well it works. And then of course you loose a little perfection if you a/d it again.

 

[BradM]

Thanks for your input. What simulator are you referring to? I'm not sure why I would need to shoot for 300 ohms output impedance if I am justing using high impedance DI inputs on my mic preamps.

If you don't mind me asking, what did end up settling upon for your passive design? What you are describing sounds very intriguing. Part of my intention would be to use it on a drum submix or even my mix bus. If you would prefer not to post to the public forum I can understand. Feel free to PM me any detailed information you wish to share.

thanks,
Brad

 

[Brian Roth]

Indeed, with a passive EQ, you have to start out with a net LOSS through the circuit when it is set for "flat" in order to be able to add any amount of boost. In the circuit you posted, with the pots both set for "no boost", there will be approx 21 dB signal loss.

Bri

 

[BradM]

Hi Brian,

Are you calculating 21 dB of loss for the 10k/10k pot combo or the 100k/10k pot combo? Perhaps I was doing my math wrong...

Brad

 

[Brian Roth]

I was going with the "lower" 10k Log pot set so that C2 was shorted out of the circuit (ie wiper grounded) so that C2 had no effect. In that case you have the 1K R to ground with the "upper" 10K lin pot in series on the voltage divider.

Hence 20-ish dB loss.

BTW, in future schemos, add in "R" numbers for the pots and R's..makes it easier to discuss <g>.

Bri

 

[BradM]

Yeah I agree--I should have done that for ease of communication.

thanks,
Brad