Mk II ... Mk III

Is it feasible to upgrade a MkII to a MkIII? What would it take to do this?

The reason I'm asking is that I may have an opportunity to purchase a pair of MkIIs. They would be used in a bi-amped system with MkIIIs. It would be very important for the amps to be closely matched.

It is not difficult. I've done it for customers who asked.

You need to replace the 5U4 rectifier tube with a 5AR4/GZ34; and the 50Ω 10W resistor in the B+ supply with a filter choke. The one sold by Dynakit Parts for the Stereo 70 is exactly correct and you will need to drill the chassis to mount it.

Then you would replace the EL34 output tubes with 6550 tubes, and reset the bias. Nominal idle current for the EL34 in a Mk II is 65mA per tube; for the Mark III with 6550, it's 70mA per tube. The specified 1.56 Biaset voltage will no longer apply; you would set the Biaset measurement to 1.68 volts instead. If you want to get a nominal 1.56 Biaset voltage on a Mark II with 6550s, you would replace the 12Ω cathode resistor with an 11.2Ω resistor.

Nowadays many users set the output stage bias lower than the original Dynaco specs, since modern tubes appear to be less durable than the old ones. If you do, you'll have to calculate what the Biaset voltage measurement will be for your desired bias current.

You still will not have a 4Ω tap on the output transformer, unless you have one of the very few Mark II that was special-ordered with 4Ω and 8Ω taps instead of 8Ω and 16Ω. Some claim to have seen Mark II with Mark III output transformer, and the 4Ω wire hanging loose and taped off. This is conceivable but I've never seen it myself.

Thank you. Those were both great answers, just the information I was looking for. Bob, I'm not as much concerned about the gain as I am about the "character," as you put it. Here's why; I'm using Vandersteens which are ideally set up for bi-amplification, but the Vandersteen manual advises caution when considering this. "Even if the levels of the amplifiers are matched for one volume level, the amplifiers will still exhibit different dynamic characteristics, different imaging characteristics, different degrees of detail and instrument texture and different tonal balances. With the crossover between the woofer and midrange occuring in a range where the ear is very sensitive to any sonic changes, these differences will create considerable confusion through the important midrange." The differences between the MkII and MkIII may be so slight that they are not significant, but the Vandersteen advisory has at least made me consider if adding the MkIIs could be a step down rather than a step up. For this reason I think I would be more confident in upgrading the MkIIs to MkIIIs.

I think it would be sufficient, actually, to change the output tubes and rebias. All else is power supply modification. and if the Mark II are the tweeter amps, the demands will not be as great on them.

There is a difference in sound between EL34 and 6550 which would be greater than the ten-watt difference in output power. In fact, while you are swapping tubes (don't forget readjusting the bias each time), you may find you prefer the sound of the EL34 over the 6550, using them in the Mark III (instructions on setting the bias for EL34s in Mark III are in the instruction manual). For all the fuss people make about how one brand of EL34 sounds different from another (and they do!), the different tube types are far MORE different; and since the two types are effectively interchangeable with the appropriate bias adjustments, I'd think the time taken to do listening tests would be worth it.

There is no problem with doing this; the EL34 handles the higher plate voltage of the Mark III just fine. In fact, original Mullard EL34s are rated at 800 volts on the plates, far above what a 6550 can handle.