Crack as amp

[chaslieb]

Has anyone gone to the next level and tried to use this circuit in parallel as a low power amp?  Given the speed of otl amps it seems a natural for a small tweeter amp, particularly for a high efficiency tweeter, with a slower tube amp on the mid/woof.  I have a 2a3 amp on my homebrew TL speaks and wouldn't mind a tiny tweeter amp to get the crossover out of the circuit and speed up the highs.  Just thinking about what would be necessary to lower the impedence handling to speaker proportions, without an autoformer, and bring up the power to a watt or two.

Charles

[Paul Joppa]

Power calculations are straightforward.

First, the bad news. To get one watt into an 8 ohm load, you need 350mA rms, or 500mA peak current capability. For a perfect Class A amp, that means the quiescent current must be 500mA, so the output can swing from 0 to 1000mA on peak outputs. Crack runs about 30mA quiescent per channel, so you need seventeen triodes per channel in parallel - if they could be paralleled perfectly. Looked at another way, the peak crack output current is 30mA, which is 21mA RMS and would provide 3.6mW into an 8 ohm load.

Now the better calculation. Pushing the capability to the max, you could double the standing current without exceeding the PT-3 power transformer's capability. Converting to a monoblock by paralleling the two halves would double it again, so you could get 120mA peak. That would get you up to 55mW or so, which could be enough for a high efficiency horn with compression driver.

You'd still need a lot of negative feedback to get the output impedance down to a reasonable level. A tube version of the very simple Nelson Pass MOSFET amps might be a good approach.

[Doc B.]

From a more subjective standpoint I believe there will be problems as well, as using feedback and using a cathode follower to get low impedance would put two different sonic signatures into the tweeter that would probably not be considered as positive attributes. Since tweeters consume so little power relative to larger midrange and bass drivers it might be more rewarding to consider taking advantage of the resolution and tonal attributes of a small transformer output DHT amp.

[chaslieb]

Thanks for the responses.  I better read up on the circuit topologies that some big name OTL amps, using similar tubes, put out so much power into low impedence loads with minimal or no feedback.  My mistake to think that the Crack headphone amp would be easily scalable up into speaker amp form.

Thanks again,

Charles

[Paul Joppa]

You do have to read between the lines when comparing such different topologies.

The fact is, there are hardly any OTL tube power amps that do not operate in push-pull and with substantial negative feedback. Tubes just can't handle the steady-state current. I've tried, from time to time, to come up with a way to make this work, and failed every time. Such amps are usually current-limited more than voltage-limited, therefor they are most practical when many tubes are paralleled and the amp is designed to deliver high power.

I'm not saying that they can't, or don't, sound good with the compromises that are necessary. Engineering is all about making things work as well as possible within the necessary compromises. Certainly us DHT and SET enthusiasts have to make compromises in the real world, they are merely different compromises.

I do think a true zero-feedback Class-A SET OTL could be made with a single 6C33 for each channel. It would consume about 100 watts per channel, and produce no more than 1 watt per channel. I'd expect it to sound awfully good, in fact.

[xcortes]

I do think a true zero-feedback Class-A SET OTL could be made with a single 6C33 for each channel. It would consume about 100 watts per channel, and produce no more than 1 watt per channel. I'd expect it to sound awfully good, in fact.

Are you thinking a cathode follower like the Crack?

I know this one is a very long shot but, given Crack's success, is there any chance of having a BH kit on these lines? Modifications to a current one?

Seems like the ultimate amp for very high efficiency horns!

[Doc B.]

The most straightforward approach would be to try some old 600 ohm Jensen drivers.