Using two Tube Program Equalizers on a stereo buss – can I match two units?

Today I want to talk about some questions I’ve been asked several times lately: 

I want to the use two Tube Program Equalizers on stereo busses – can I do a matched pair? How can I match them? Does it make sense at all?

Before we jump right into it, let me tell you this:

You probably might already know that back in the days the tolerance of electronic components for the most part was much higher than what’s possible today. More specifically – the tolerance of most components in vintage Pultecs was about 20%!

That’s also why no two vintage Pultecs sound exactly the same.  

Besides – if we’re talking using a pair of vintage Pultecs nowadays – depending on the conditions under which they were used over the years (humidity, heat, etc) components may have aged and values may have shifted very differently.

But still it’s custom, in many studios around the globe that are lucky enough to have a pair of vintage Pultecs in their repertoire, to use them on stereo sources or busses during recording and mixing.

So what’s the deal?

Alright now – first of all in regards to our Tube Program Equalizer project to build an authentic recreation of a vintage Pultec EQP-1A we have a definite advantage: some components nowadays have tolerances of 1% and most of the components nowadays have tolerances between 5-10%. Only a few of the components we need have a higher tolerance. But I’ll talk about that a little later.

That means two units we build are most likely a lot closer to each other than two vintage units.

But can we match them – or should we match them to use them as a pair? The answer depends on the application we want to use them for.

  1. recording & mixing – or
  2. mastering. 

Obviously the latter makes it a little more complicated.

Let’s focus on 1.

If you read my „3 bands of glory“ e-paper (part of the second Build Your Legend Series) you know that the Pultec EQP-1A consists of two separate circuits so to speak: the passive filter circuit and the amp circuit.

There are several ways to do that but one of the easiest ways to tweak the output level of the amp is to adjust the value of two resistors in the feedback path of the amp circuit. How exactly it works I’ll explain in the building guide, ok? Just so you know it’s not a big deal.

The more sophisticated task would be to match the filter circuits, because that means you’d have to get a bunch of each value of the filter caps and test every single one of them to find matching pairs. It can be done but it’s tedious.

Now let me remind you – the tolerances of the components nowadays are rather small – in case of our Tube Program EQ complete kit – the custom made Mundorf filter caps have a tolerance between 3-5%.

So I guess your units will be rather close in this regard without matching the caps 😉

Plus, some of the components with the most tolerance of all are the pots. In some cases 10% can also be 20%! That means even if you match all caps and resistors in the circuit, the knob settings on two units to achieve the same result might vary quite a bit.

I’ll leave that open for discussion, but imho for recording & mixing matching the output level of two units and going by ear from there when dialing in the knobs works great and I wouldn’t go as far as matching every single component. What’s your opinion?

One could argue that there are also high precision pots on the market that go as low as 1-5% – but they are like 75-150 EUR ea. AND here’s the thing:

The log pots used in the Pultec EQs were made by Allen & Bradley or Ohmite in the USA and their logarithmic taper (curve) was quite different from modern log tapers. It was called a „modified log“ curve. And there are no high precision pots with a modified log taper available.

The diagram below shows the tapers of a linear (blue), a standard logarithmic (grey), a modified log (yellow) and a modern Alpha log pot (red).

As you can see they are completely different and I even tested the PEC pots used in the modern reissue Pultecs and they don’t match the tapers of the vintage Allen & Bradleys either – they are somewhere between the Alphas and the A&Bs.

The pots used in the complete kit are custom made with the original „modified log“ taper.

For mastering that’s a different chapter. 

I think it couldn’t be close enough. Then you’d have to use stepped attenuators (fine stepped switches) instead of pots and use matched resistors for every step of the way. But then again it most likely won’t behave like an original Pultec in terms of boost and cut settings.(remember the modified log taper)

I learned it the hard way building myself a custom 2-channel Neve 1084 a couple of years back and I matched every single part and component within 1%. Instead of pots I soldered my own step attenuators – which was a huge PITA to do… 

…and in the end I swapped them for custom pots because the unit just didn’t behave like the Neve EQ I expected in terms of boost and cut settings #fail

Now with the pots it’s behaves as epexcted and it still works great for stereo applications.

Anyway, I hope I could shed some light on that subject and answer some of your questions. I’d still like to know what your opinions or experiences are – please let me know in the comments.

Martin

PS: Pre-orders are about to close very soon!

Days
Hours
Minutes
Seconds

Let others know about our analog passion:

Leave a Reply