HifiZine
The enthusiast's audio webzine

The AVA Hifi UltraValve

As one of the (many, I am sure) people who were introduced to tube amplification through the Dynaco ST-70, I was intrigued by the Audio by Van Alstine UltraValve amplifier. Well-received at shows like AKFest 2010 and RMAF 2010, this amplifier could be viewed as the latest step in the evolution of the line of upgrades to the original ST-70 sold as kits for many years by Audio by Van Alstine. I asked Frank Van Alstine if he would be prepared to share some of his insights with us.

JR – Hi Frank, thanks for speaking with us today. Let’s start with the best part first – your Ultravalve amp has been very favorably received – you must be very happy about that?

FVA – Yes, I am really happy about the very favorable recognition of the musical quality and outstanding value of our Ultravalve amplifier.

JR – The UltraValve is the culmination of years of evolution based on the original Dynaco ST-70, is it not? How much of the original ST-70 remains?

FVA – What remains is some of the circuit basics (although drastically improved), the biasing scheme (also improved and regulated), and the basic 5AR4 vacuum tube R/C power supply (again hugely improved and supplemented). We also have pretty much kept the overall physical layout simply because it worked well, is very quiet, compact, and can be assembled at a rational cost. No point re-inventing the wheel.

JR – Are there specific circuit refinements that you can share with us?

FVA – The major refinement is actually making the tubes work the way they were intended to. The original – as do all other vacuum tube amps I am aware of – has hopelessly inadequate power supplies. A simple RC network, no matter how big or how outrageously priced the parts are, no matter what the outlandish claims for their wonderful sounding parts, simply fails in real world use conditions. There is a huge undesirable interaction between the tube elements, both channel to channel, and even between each separate tube section. This occurs because the output impedance of the supply ends up being several hundred ohms at some frequencies, no matter what musical qualities are claimed for the parts. To eliminate this interaction, separate, isolated, high current, high speed, analog, regulated power supplies with very low output impedance are required for the plate of each tube. With the Ultravalve, we have done this. The sonic results are the proof of the pudding.

We did make a bad original design assumption in the first iteration of the design, in the early production Ultimate 70 version. We tapped the raw supply for the regulators ahead of the choke as we could measure no ripple at all after the regulated supplies. Later, thanks to some very useful advice on our forum on AudioCircle, we changed this to after the choke and that made a huge sonic improvement, although there were no measurable differences in the performance, either of the regulated supplies’ purity or at the output. The listening comparisons were instant no-brainer improvements.

JR – What about output transformers? Tube lore has it that the original Dynaco A470 output transformers are among the best ever made. How do modern transformers stack up (pardon the pun)?

FVA – We really do not hear any differences between the various versions of the output transformers that have been used over the years. Ours are made by Dynakitparts to our own specifications and we are very happy with the results.

JR – Could you outline for us your evaluation methodology? How do you decide which of numerous circuit alternatives is the one to go with for production?

FVA – Normally we evaluate by building one new channel in a production unit, keeping the other channel as it was. We then have someone scramble the interconnect and speaker wire connections and then leave giving us a random setup. We then listen in mono and use the balance control to swap from channel to channel. We make notes answering two questions: First, can you hear any differences at all? Second, are they better or worse differences? Sometimes the results are instantly very obvious, sometimes there is nothing at all, sometimes what is obvious is that the changes made things worse. We also test in general on two very different kinds of speakers, the Salk HT3s with ribbon tweeters and a metal cone mid-range, and on a set of highly upgraded B&W 801 speakers. Both are difficult loads, both are highly regarded, and both easily show up design flaws in electronics, but differently.

JR – Were there moments of frustration during the design process of the UltraValve, or was it all smooth sailing?

FVA – The only real frustration is the amount of time it takes to put everything together for production and get it right the first time. We are too small and don’t have the financial resources to make significant mistakes. Fortunately I have a great mechanical design engineer who is turning out perfect chassis designs for us time after time and I am pretty good with circuit board layout and between us and all the other internal advice and design help, we are pretty good at getting it right the first time.

JR – If you had a time machine, which decade would be the one in which you would most want to be designing amplifiers (and why)?

FVA – Actually it would be really fun to jump back in time to when David Hafler was designing the original Dyna ST-70 along with a supply of the necessary regulated power supply parts (not available back then), and make the amp work then the way we can do it now. It was a killer in the marketplace then, I wonder how the Ultravalve would have done.

JR – What about music? What are some of your personal favorites?

FVA – I like many kinds of music from bluegrass to steel drum bands to opera. Just as long as it is a good performance well recorded.  For live music, I only have one rule: no electricity allowed! I want to listen to the music, not some terrible PA system thrashing around.

Thanks much for this opportunity.

Frank Van Alstine