Racking two cascaded v72's to make a poor man's v76. Is it possible?

[SamuelPepys]

Hi! I'm buying a pair of Siemsns v72 units, and I remember Oliver explaining how the v76 was more or less two v72 units with a few control additions. I'm wondering if it is possible to rack these units in a way that makes them useful as a set of two seperate preamps, and also combine them in a way that with the push of a button, you can use a third set of controls where they are working one after the other to produce a v76-ish result (hopefully)?
Are there anyone out there that would know how to do this? Or know if it's even possible to do this?

Thanks!

 

[Winston OBoogie]

With appropriate circuitry you can certainly cascade two V72 amps. But doing so doesn't really equate to a V76.

The second half of the V76 amp exists, first and foremost, to act as an output buffer for the first half, so that the changing of gain (via negative feedback) doesn't also result in a change in output impedance of the complete amplifier.
Back when these amps where designed, adhering to a defined impedance within the confines of a desk was important. These days, not so much.
Anyway, the V72 and its kin were designed around a 200 ohms working impedance


A V72 input transformer therefore wants to see a source impedance of 200 ohms. Deviating much from this results in frequency response aberrations.
However, the output impedance of a stock (unmodified 34dB gain) V72 is circa 30 ohms. So running one amplifier straight into another isn't the best.

If you "build out" the output impedance of the first V72 with a resistor of circa 170 ohms, the total output impedance is then 200 ohms: 30 ohms of the V72's internal output impedance + 170 ohms "build out" R = 200 ohms. Having done that, it works just fine to run one V72 into another.

When the cascading of amps is done within a floating environment, the "build out" can be a single resistor from the "hot" (positive) side of the first V72's output. If you need to adjust the level of V72 no. 1 into V72 no. 2, this is best done with a constant impedance attenuator, normally a "bridged T" type.

However, when you have other stuff in the path such as phantom power feed resistors this is no longer a floating environment so build out R's and attenuators then need to be balanced and "H" type.

Another issue/difference between a V72 and a V76 is headroom. A V72 has output headroom of around 4dB lower than a V76. There's nothing you can do about this without completely changing the circuitry of the V72. This headroom difference may not be a practical issue, but it's worth mentioning. A V72 is good for output peaks up to around +18 dBu while a V76 is OK up to circa +22dBu

* Edited for clarity, although I'm a bit suspicious that the edit made it even less clear than it was already! Hopefully not though.

 

[MaxDM]

V72's are a better circuit, considering the gain, as they aren't riddled with electrolytics and they use a bit of positive feedback along with the negative feedback, which makes things sound faster, from my experiments.

I had a V76s and thought it was too soft and tubey sounding for my (picky) tastes. A lot of the sound comes from the input transformer which has a very high ratio, compared to the V72, which is already high. It tends to have a softer top as a result. I changed all the worn-out Electrolytics and retubed with sealed NOS tubes, and though it was very useful in terms of variable gain, very low noise etc. I thought it wasn't forward sounding.

To make the V72's closer to a V76, you would need to connect a pot to the anode cap of the output tube, maybe 10K, and feed the signal to the input of the first tube of the second V72, bypassing the cap. I think it might sound better than a V76.

By the way, you can get up to 55 dB with a V72, provided you are going into a high impedance (10K) line input, by varying the negative feedback. There will be more overall distortion, but it's useable, and sounds a bit more relaxed, from memory.