PULTEC - Input Options -Buffers- Imp - TX - JLM Hybrid opamp

[drpat]

OK, so now I'm really curious about this mysterious setting that presents a 75Ω impedance. I've done a little bit of screwing around with it, and can't seem to make that SSM2142 misbehave no matter what I do... It would be nice to know what those settings are, so that I won't be chasing my tail looking for distortion. So far, the only difference that I've found between the "SSM2142 before the Pultec", and the standard loop (without the SSM2142) is...

1. The low frequencies are much flatter "with" the driver in front of the Pultec.

2. Greater maximum boosts and cuts "with" the SSM2142 driving the Pultec.

3. No audible distortion with any of the extreme settings that I tried.

Kinda strange. It's as if the Pultec "likes" the chip in front of it...

Jacob? Are you out there somewhere?

 

[drpat]

I guess an explanation of my test setup is in order.

I ran the generator into a console channel, and fed an unbalanced buss to the SSM2142. From there, I multed the signal to the "source" input of the analyzer as well as the input to the Pultec. The Pultec output was then connected to the "response" input of the analyzer before running a transfer function between the two signals.

I did it this way because I wanted to see how the Pultec was reacting to the device (SSM2142) in front of it, and I wanted to take the frequency response of the SSM2142 and the mixer out of the equation. I carefully matched the levels between the raw generator output and generator signal running through the mixer and SSM2142 to be confident that what I was seeing didn't have anything to do with a different level hitting the Pultec.

 

[khstudio]

Well, I just ordered some of the 2142's to try out... so well see.

Also, I started wiring up a JLM Hybrid to try... this will replace the input TX... I'm assuming this would be on the lines of the 990 Roger is referring to & should have no problems with driving it.

I've gotten some feedback on how to hook it up... I'll try to post what I'm doing.

 

[khstudio]

Just sharing Joe's (From JLM) response to my email.

Also... Thanks GREG, the drawing you sent me was exactly like Joe's.:thumb: But the 620 ohm resistor question I had is now answered.

I'd like to try & set it up like Matt did... replacing the 600:600 input
Transformer with the JLM Hybrid.
[Joe] ok see attached gif. It is as simple as that.

Should I add the 620 ohm resistor to GND - between the JLM Hybrid &
Pultec's Hi Boost Pot? (Like the original)
[Joe] NO the hybrid gives a zero ohm output so no load resistor is needed.

Are any other caps needed ???
[Joe] No if you have DC noise on any of the passive EQ pots either fit 2 x
22uF input caps or 1 x 470uF output cap or turn on the Hybrid DC servo. You
only need to do one of these things but none should work just fine.

Last question - Can I join pins 1 & 3 when hooking UN-Balanced gear to
the JLM Hybrid's input ???
[Joe] Yes

Thanks Again,
Kevin

 

[emrr]

I don't have a Pultec to mess with, but theoretically one might wonder about use of series resistance between a 0 ohm driver and the circuit. It may not act exactly like a Pultec in a classic setting without a set minimum input, since it was designed as a matching circuit. We've swung from bridging the input, all the way past matching, to the other extreme in this discussion.

 

[clintrubber]

[quote author="emrr"]I don't have a Pultec to mess with, but theoretically one might wonder about use of series resistance between a 0 ohm driver and the circuit. It may not act exactly like a Pultec in a classic setting without a set minimum input, since it was designed as a matching circuit. We've swung from bridging the input, all the way past matching, to the other extreme in this discussion.[/quote]
Hmm you're right... what does the original manual say w.r.t. intended source impedance ? 600 ?
Might need to add something, anything from a fixed build out resistor (mitigating the need for a potent driver) to a variable front panel control (which will still require muscle in the driver).

Bye,

Peter

 

[khstudio]

[quote author="emrr"]I don't have a Pultec to mess with, but theoretically one might wonder about use of series resistance between a 0 ohm driver and the circuit. It may not act exactly like a Pultec in a classic setting without a set minimum input, since it was designed as a matching circuit. We've swung from bridging the input, all the way past matching, to the other extreme in this discussion.[/quote]

I wondered about the 600 ohm load too.. This is why I asked several times about it. It's easy enough to test it with & without it to see which I prefer or what measures well... who knows, they both might be cool & I can put a switch on it.

I should mention that I'm not "Stuck" to any one of these ideas & that I'm building a few of these Dual Pultecs... One will definitely be set up with the 600:600 input TX, which was probably my main concern when starting this thread. But the more I read & got feedback, I realized there's more than one way to skin a cat. :green:

I'm still waiting for my 600:600 input TX's to get here... Until then, I'm working on the Hybrid Test.

UPS just dropped off the 2142's too... but I still have to figure out how to properly get rid of the extra 6dB.
Correct or not... I'm going to try it like Patrick did, just to see.

I'm not hard set on an "Exact" Pultec clone... just some good sounding Passive EQ's that can interface with my studio.

 

[clintrubber]

[quote author="khstudio"]UPS just dropped off the 2142's too... but I still have to figure out how to properly get rid of the extra 6dB.[/quote]
C'mon, be a man :wink: Now what's then exactly the difficulty with 'google', resistive' & 'divider' ? :roll:

I guess this is the third time it's being mentioned, it's pretty frustrating to see all those questions being asked again & again while answers are already in between...

Have a good weekend :evil:

 

[khstudio]

[quote author="clintrubber"][quote author="khstudio"]UPS just dropped off the 2142's too... but I still have to figure out how to properly get rid of the extra 6dB.[/quote]
C'mon, be a man :wink: Now what's then exactly the difficulty with 'google', resistive' & 'divider' ? :roll:

I guess this is the third time it's being mentioned, it's pretty frustrating to see all those questions being asked again & again while answers are already in between...

Have a good weekend :evil:[/quote]

I DID Google the L -PAD right after I posted:

10k shunt & a 4.9k series = 6dB attenuation


You've got to relax on me dude... the way you talk is pretty downgrading. I guess you know it ALL & don't ever ask dumb questions.

YOU have a nice weekend. :evil: :evil:

 

[khstudio]

As long as I'm on the PAD (for the 2142)

I was wondering (because I don't know it all) if the PAD would be better being AFTER the ssm2142 for any reason? Since the 2142 is what's boosting it anyway.

If so... should it be a T or U PAD?

 

[clintrubber]

[quote author="khstudio"]You've got to relax on me dude... the way you talk is pretty downgrading. I guess you know it ALL & don't ever ask dumb questions.

YOU have a nice weekend. :evil: :evil:[/quote]
Oops, the dude-saying has started...

With that L-pad under your belt, why then still asking about that 6 dB attenuation ? Didn't mean to downgrade you, it's just that I didn't get that.

I can't tell for sure how you meant it, but to me it sounded like 'we' hadn't helped you enough yet with the 6 dB attenuation.

[Please take this in jest]... but at least in my possibly twisted perception you have a way to give people the feeling that your problems are neglected by people. That the world isn't fair to you by not spoonfeeding you the solutions you have in mind, while you're not really apt to hear about other alternatives. Things like that can start to make people feel a bit guilty (for no reason, but still, it does) You sure can become a successful manager because of that capability ! :wink: [/Please take this in jest]

I guess you know it ALL & don't ever ask dumb questions.

While dumb questions don't exist I certainly do ask those, but I hope I don't keep repeating them after they've been answered.

Enough/cheers/let's both take that Chill Pill from the Brewery now,

Peter

 

[khstudio]

I like you dude... I do... But you read WAY too much into things.

Also, this is my first time dealing with a Pultec EQ & these loading issues... & it takes a little time to learn & absorb ALL the options. I share a lot back to the forums but in this case, for now, I have to be the taker. :wink:

I've got soldering to do... I'll post back what I find, Hear & Like.

Thank to everyone for their patients & help. :thumb:

 

[clintrubber]

All fine :sam:

 

[emrr]

Just make sure to have a look at the way build-out resistors get used to set an output impedance for the next stage. Quad Eight stuff for instance had direct output connections along with secondary 560 ohm series resistor output connections for systems requiring 'true 600' interfacing. Series resistance is determined by actual output impedance of driving amp. Many products now have build-out resistances sheerly for output short-circuit protection.

There's a nice Altec Tech Note that can be downloaded somewhere concerning Altec-specific build-out resistances. It's useful for grasping the concept. May be the same one discusses loading of passive equalizer networks of varying types. It's been linked here before.

Whether or not a Pultec needs any input series resistance when driven with a 0 ohm source is beyond my immediate comprehension. Thought it was worth considering.

 

[khstudio]

Hey Doug,
I'm making good progress now that I actually started soldering... but the load resistor I was referring to is the 620 ohm that's normally on the secondary of the Pultecs input TX... The series resistance is something else i don't know about or even considered.

 

[emrr]

I doubt much of anyone has bothered matching their sources for true 600 ohm drive to a Pultec, and I hadn't remembered the Pultec had a 620 ohm load resistor already implemented. Just thinking once you implement a buffer amp in front to gain absolute Z control, it's only logical to wonder exactly what Z is it really supposed to be. Someone who knows the circuit far better than I may easily come in and shut me down. I haven't seen any evidence of detailed tests regarding very low Z input drive with this network.

 

[JLM Audio]

Driving with zero ohms is fine as it's allows you get a slightly higher Q at the sharpest Q pot setting but since the Q pot is a series resistor with the LC turning the Q pot slightly back returns you to the normal maximum Q. The biggest thing is that the Q range with the input buffer stays constant no matter what is driving the input.

When using the transformer input the maximum Q changes depending on the output impedance of the gear driving the input transformer.

The rest of the Pultec high cut, low cut and low boost are run using the high boost pots 10k resistance as the series feed resistor to them so changing the input impedance will do very little to them.

Remember the pot tolerances of +/-10% between pultecs would make a bigger difference to max Q than any of the above changes.

But the biggest problem is the normal Pultec input transformer changes its tone to be harsher to my ears when it is driven by modern solid state low output impedance gear as the transformer has a softer nicer top end when driven by 600ohms. The 600ohm feed also drops the input level 6dB at the same time so that probably also helps. You will find with a lot of transformers driven by low impedance will bring on ringing in a transformer which will have a smooth response when driven by 600ohm but I have not tested if this is the case with the Pultec input transformer.

Also if you are driving your Pultec with modern +18dBm to +22dBm levels don't think driving the symmetrical valve make up circuit in the Pultec this hard is going to warm up the signal as it cancels out the fat 2nd THD very well and only makes a lot of odd bad harmonics the harder it is driven.

Try a Male to Female XLR tube with 2 x 300R resistors in series with the pins 2 and 3 and have a listen for yourself as it helps with both the above issues if you have a original or clone Pultec EQP-1A or 1A3 or 1R.

 

[Jonkan]

Interesting thread for me.
Im thinking of building some passive eq:s with the pultec passive filters as some kind of starting point since i got some inductors from Ioaudio when he had his big sale.

Havent decided what kind of makeup gain im going to use, or what to put on the input.

Would you guys suggest it would be better to use a 10k:600r transformer on the input, or to go with something like the hybrid opamp schematic posted in this thread? I will use the eq connected to the input/outputs of my motu soundcard most of the time.


I hope im not too far off topic, i dont want to "steal" your thread.

thanks!
/J

 

[rodabod]

[quote author="khstudio"]I guess you know it ALL & don't ever ask dumb questions.[/quote]

I know evrything's cool, and we are not "getting on" at each other, but I think Peter's genuinely been trying to help you here (as have I) by trying you to get to think for yourself.

You will not learn much by repeatedly asking for working solutions. We're all mainly in the same boat - nobody is 100% correct about everything. If you just follow exactly what others say then I think you're missing out on part of the fun of DIY. It's good to work on solutions yourself and implement them, then you can apply this new knowledge to other projects/problems which you meet later.

I think you'll understand the issues better here if you first get your head completely around matching (eg. 600 -> 600) and bridging (eg. 50 -> 10K) impedances. Why did we ever use matching impedances? What were the implications of doing so? What are the implications of using 600 Ohm gear with equipment designed for bridging impedances? How do the two different methods differ as far as voltage and current transfer are concerned?

With this in mind, how might this affect you using your Pultec with the other gear might want to use?

The series resistance is something else i don't know about or even considered.

I believe we are talking about setting the output impedance of the de-balancing driver at the input to 600 Ohms to mimic the conditions of plugging a 600 Ohm source into the input. I think I mentioned this in one of my posts earlier.

I'd be inclined to pad before any de-balancing stages. This would let you drive the input 6dB hotter before clipping. Noise shouldn't really be an issue since you are at line-level.

 

[rodabod]

[quote author="Jonkan"]
Would you guys suggest it would be better to use a 10k:600r transformer on the input[/quote]

This is what I would probably choose, but there are many ways to skin a cat.