If you have 3 phase power and delta connected xfmrs, the 3rd harmonic just circulates around the ring, making a loud buzz and heat. With a tuned inductive pick-up, you might be able to extract the 180 Hz power for free. Just heats up the xfmrs otherwise. The utility will find -some- way to charge for it though.
Maybe with 3 OTs and a similar set-up to sum the circulating 3rd H distortions, one could absorb the OT distortion. 9X the price $$$$$$.
(there is a way to convert 3 phase into 2 phase. I don't think you can get it down to single phase though, unless you throw out half the power. You'll just have to put up with a rotating sound pattern with 4 speakers. Merry-Go-Round music! )
Maybe with 3 OTs and a similar set-up to sum the circulating 3rd H distortions, one could absorb the OT distortion. 9X the price $$$$$$.
(there is a way to convert 3 phase into 2 phase. I don't think you can get it down to single phase though, unless you throw out half the power. You'll just have to put up with a rotating sound pattern with 4 speakers. Merry-Go-Round music! )
Last edited:
All I know about the special case push pull that uses 2 SE air gapped transformers, it "sounded" good, and not just to me.
And there did not seem to be any reduced power, reduced gain, unwanted cancellation, etc.
Of course, the primary inductance was a low frequency limiting factor, use real good SE output transformers.
It is really easy to try that topology if you have a triode push pull amplifer that does not use negative feedback.
I never heard of the French objecting to the performance.
Co$t is probably the only reason that topology is not built more often.
$0.03
And there did not seem to be any reduced power, reduced gain, unwanted cancellation, etc.
Of course, the primary inductance was a low frequency limiting factor, use real good SE output transformers.
It is really easy to try that topology if you have a triode push pull amplifer that does not use negative feedback.
I never heard of the French objecting to the performance.
Co$t is probably the only reason that topology is not built more often.
$0.03
OK, so no inversion from K to P. The degenerative FB signal found on a K is in phase with the grid and since it does not get inverted vs. the plate it attenuates the plate signal. Lowers gain. So NFB on the K must be out of phase with the plate signal, (and same phase as the grid.) I see the light! And the scope sits next to me on a shelf.... The lesson is that NFB is not simply a signal that is inverse of the source signal. Thanks!
All of my amps are GNFB types and I've mistakenly assumed that the OPT was also a point of inversion. Clearly, it must be wired to not invert on Mullard topology. Driver, OP's, invert, and send the in phase (to the driver grid, source) OPT signal back to the K.
In my designs, I've always included active regulation for PA screens. In these cases, the active regulators were hollow state, but there's no reason why it couldn't be done with solid state. A "capacitance multiplier" woud be better than nothing (and considerably more than what commercial designs back in "the day" included). The main problem with the 6V6 is that it was designed to operate with the same voltages on the plate and screen. That wasn't a consideration with the 807 or 6BQ6 as these types call for screen voltages considerably lower than the plate voltages.
Hello Miles Prower
One question please if possible - in the Vixen amp where does the -300 Vdc regulated goes ?
Looking at the schematic of the amp i can not see where it goes.
The -300V rail connects to the cathode tails of the 6SN7 grid drivers and the grid bias voltage dividers. On the schemo, it's labeled "VKK".
Do you all think the 6AU6 in this circuit could provide enough voltage swing to drive a 6HJ5 or 6HD5?
(I'm guessing around 50v, control grid to cathode, on the 6HJ5 or 6HD5.)
(I'm guessing around 50v, control grid to cathode, on the 6HJ5 or 6HD5.)
As I mentioned before, I'm tempted to rebuild a pair of PL519 monoblocks into something similar to the RCA amp in this thread (but bigger, obviously).
It occurs to me that I could save both space and money by using PCF80 (9A8, the 300mA series heater Euro version of 6BL8) in the input stage. Each bulb contains a triode quite similar to 1/2 12AU7 and a pentode that isn't too far off from a 6AU6.
It occurs to me that I could save both space and money by using PCF80 (9A8, the 300mA series heater Euro version of 6BL8) in the input stage. Each bulb contains a triode quite similar to 1/2 12AU7 and a pentode that isn't too far off from a 6AU6.
If you did that, you could experiment with making the input stage a twin-triode LTP, instead of the cathodyne/concertina in the RCA amp.
Who knows if that would sound better, though. It would be different, most likely.
Who knows if that would sound better, though. It would be different, most likely.
I've thought about that too. An LTP would probably be a better phase splitter, at the cost of some gain and a negative rail for the CCS.
Open loop gain is something I wouldn't want to sacrifice, my intention is to use the amp(s) to drive a pair or 18" woofers in 340L BR boxes from from 25 up to 60-70Hz, I'll probably need all the negative feedback I can get for woofer control.
Open loop gain is something I wouldn't want to sacrifice, my intention is to use the amp(s) to drive a pair or 18" woofers in 340L BR boxes from from 25 up to 60-70Hz, I'll probably need all the negative feedback I can get for woofer control.
Given the ease of driving a pentode, something like a 5751( or 6SL7/6SU7) or a 12AY7 would make a fine input stage for this beast. It is only driving a pair of 6AU6 grids... 🙂
Douglas
Yes, there is probably no shortage of suitable tubes to use in this topology and I can't imagine it would be particulary picky either. This amp is pretty much the opposite of those posh zero NFB SETs that only accepts directly heated triodes that has been burned in for exactly 215,45 hours under the full moon during the solstice, supervised by naked virgins covered in snake oil... 😎
The question is, how much gain do you actually need?
Even with a 12AU7 or 6SN7 as the input tube, the gain is quite high. Upping the gain by using a 12AX7, 6SL7 or 12AY7 would roughly double the gain. Maybe that's an excuse to add 6dB of global NFB from the OPT secondary to the cathode of the input triode? Otherwise, I think a 12AU7 or 6SN7 would do the job with gain to spare.
Good question, I haven't done any real math yet. The amplifier won't need to have high gain per se but I think any extra dB of feedback, local and global, will be welcome when it comes to hold those 18" woofers in a tight grip. Nothing has been decided yet but I have eight or ten NOS Philips PCF80 and it would be convenient to use them if possible.
Tubes in a subwoofer amp, wouldn't that be just stupid? Maybe, but I know from before that the amps I'm planning to dismantle and rebuild are capable of producing some earth-shattering bass when asked to.
Tubes in a subwoofer amp, wouldn't that be just stupid? Maybe, but I know from before that the amps I'm planning to dismantle and rebuild are capable of producing some earth-shattering bass when asked to.
I'm thinking of doing a variant with a 6AU6 front end, MOSFET (with heavy gate protection) "cathodyne" splitter, and 6V6 outputs. I have some push - pull transformers that are purported to be from Onetics for output iron - I will throw them on the bench and characterize them - that, or some Hammonds I have hanging around. And yes, the amp will get some global feedback.
A pentode input stage would probably do well in this amp, especially if you want a little extra gain for GNFB.
I'm also keeping an eye on the 50-watter with additional plate to plate (grid...) local feedback loops. The transmitter tube SET I'm listening to as I type uses a little bit of both plate to plate and plate to cathode feedback, plus some CFB. Works like a charm.
I'm also keeping an eye on the 50-watter with additional plate to plate (grid...) local feedback loops. The transmitter tube SET I'm listening to as I type uses a little bit of both plate to plate and plate to cathode feedback, plus some CFB. Works like a charm.
Yep (attached). This is the design I favour, and included two versions. The one that uses the 6KE8 is the design from Le Renard. The alternative makes use of a rather hard to use type that was originally an oscillator/mixer for TV front ends. Not much audio usage for a triode/pentode with a shared cathode, but just what you need for the asymmetrical LTP error amp. Have a bunch of 6X8s since people just give them away.
Attachments
Yes, I've posted that amp by Graeme Gohen of Australia here on DIY several times.
Why don't you store it this time?
