GoodSoundClub - Romy the Cat's Audio Site

In the Forum: Melquiades Amplifier
In the Thread: Single-stage Melquiades vs. DHT amps

Post Subject: Operating points and cathode currentPosted by floobydust on: 6/22/2009

Romy,

I don't think we're in any major disagreement. It's clear you have a good operating point and performance indicates this as well. I did a "normalized" model of this with the 2A3 and 45 and it specs well. All I'm suggesting is that all early and final specifications for the 45 showed a maximum recommended cathode current of 36ma in any and all operating modes. As a result, I simply think there is a higher risk of running the tubes down faster. I've seen some older 45 tubes where the getter is used up a fair amount and there is some brown hazing on the top inside of the glass. I would suspect this to be a result of running them too hot which resulted in some internal contamination. As all 45 tubes were made before the mid-1940's, it's uncertain how pure and/or advanced some of the cathode coatings were. For my own reference, I've put more than 1000 hours on a NOS pair of Zenith branded (and sealed boxes) 45 tubes in my current amps and they are still maintaining their initial performance. They are made by Sylvania and of my preferred internal construction.

There is one other 45 tube you might look into... just a wild shot. EML make a 45 solid plate and a mesh plate. I have a pair of the solid plate and they are huge. I don't consider them a real 45, but one thing is for sure... they are built like a tank (butt ugly too) and actually sound good at 40ma cathode current, use less and you notice it. I changed my amps specifically to run these for some listening tests. Ultimately, I find the EML to be too sensitive to mechanical exitation and not quite as neutral sounding as the real 45 and went back to my NOS Zenith/Sylvania pair (and back to 34ma). Still, maybe it would perform better in your circuit.

As for load impedance, most of the operating points I've calculated (via plate curves, a straight edge and a calculator) and tried in practice confirmed that, at least in my application, a 5K load seems quite optimal at the voltage and current I'm using. Increasing to 7K resulted in a loss of signal linearity at higher power before clipping. Dropping the impedance would get some additional power, but only if you run the current up as well. Again, your application is pretty much doing what you want it to. Only comment is that (as you know) you'll loose some power as the load impedance goes up. You can easily go to 10 watts on the plate as you seem to have adequate airflow, but with your currently available plate voltage I would look to limit the cathode current more to spec. If you're happy with the 40ma current... I won't argue. I would be highly interested to see what your view of tube life is if you get to a point where you think they are losing performance. Likewise, if you opt to drop the current down, again... more feedback on your observations.

Regards, KM

Rerurn to Romy the Cat's Site