OK, is there any consensus on the difference in sound between SE & PP OPTs? Or are there too many variables in the mix to isolate out the sound characteristic of the transformer?
Hmm, it seems like ideally the OPT should be avoided altogether but this introduces other issues. OTL, ZOTL (Berning), Don, did your "Switched Capacitor Impedance Converter" ever get built? Any others worthy of note?
Hmm, it seems like ideally the OPT should be avoided altogether but this introduces other issues. OTL, ZOTL (Berning), Don, did your "Switched Capacitor Impedance Converter" ever get built? Any others worthy of note?
rdf said:
It's called "bridging" and used to be done a lot in PA systems.
A repeat coil (1:1 line transformer) can be used to invert one
input if the amps don't provide balanced inputs. You do need
low impedance speaker taps to do it with tube amps though.
This should not be too hard to do. Does the SE magic disappear?
Is the SE magic due to the even distortion or the low level detail?
Does SE magic even exist?
Cheers,
Michael
jkeny:
"any consensus on the difference in sound between SE & PP OPTs? "
Would be difficult to separate the SE xfmr effects from the SE amplifier effects. Rdf's suggestion sounds interesting. I would also try two SE amps (one inverted input vs the other again) but connect the speaker up so as to have a nulled output. Whats left would then be what the SE is adding to the sound.
"Hmm, it seems like ideally the OPT should be avoided altogether but this introduces other issues. OTL, ZOTL (Berning), Don, did your "Switched Capacitor Impedance Converter" ever get built? Any others worthy of note?"
I have all the parts together for the switched capacitor converter and a Berning like (but patent avoiding) converter (can also operate the Berning configuration too). Both use the same digital controller for switching. I haven't had time to work on them though. Too many ideas, stuck in a little bitty room I can hardly move around in. I need a corporation to carry out all the projects sitting on the shelf here. I also have to decide where to live, just renting a little apartment at the moment.
The simplest way to avoid the OT though is to just use a transistor in Darlington or CFP (Sziklai) configuration. Drain or collector connected to the tube plate for Darlington (or to cathode for CFP) so that triode operation is preserved. (these act as current multipliers to reduce loading on the tube) There have been at least three threads on this idea in various forms. Another variant is using a Mosfet buffer off a pentode plate with feedback back to the screen grid (giving triode like results too). The success of these approaches depends on a high quality, low capacitance Mosfet or a constant Beta bipolar and avoiding class B operation crossover problems.
If SE sound is what you want, but using a P-P xfmr, take a look at the current thread "Spud-assist: Totem-Pole Current-Mirror PP Hybrid" by Michael Koster. This doubles the power output of SE too.
If price is no object, a full permalloy core should solve most problems nicely. For a bit less, there are the pin-striped cores (nickel lams periodically spaced among the the M6 lams) which should come close. This may be the best practical solution for building an amplifier with no feedback. Would be REAL NICE to have a toroid with a thin nickel strip co-wound with the M6 strip. I'm surprised Menno hasn't come up with this product yet. It will mandate a DC control servo of course.
I would like to see some cheap winder come out with toroid OTs too. After all, once you have a CNC toroid winder, which any of the toroid makers already have, there is nothing hard about making an audio version. It just requires progressive winding technique which is easily programmed. Probably some of the power ones use it already, but most seem to be stuck in the bad practice of co-winding two 120V windings at once, which ruins HF performance.
Don
"any consensus on the difference in sound between SE & PP OPTs? "
Would be difficult to separate the SE xfmr effects from the SE amplifier effects. Rdf's suggestion sounds interesting. I would also try two SE amps (one inverted input vs the other again) but connect the speaker up so as to have a nulled output. Whats left would then be what the SE is adding to the sound.
"Hmm, it seems like ideally the OPT should be avoided altogether but this introduces other issues. OTL, ZOTL (Berning), Don, did your "Switched Capacitor Impedance Converter" ever get built? Any others worthy of note?"
I have all the parts together for the switched capacitor converter and a Berning like (but patent avoiding) converter (can also operate the Berning configuration too). Both use the same digital controller for switching. I haven't had time to work on them though. Too many ideas, stuck in a little bitty room I can hardly move around in. I need a corporation to carry out all the projects sitting on the shelf here. I also have to decide where to live, just renting a little apartment at the moment.
The simplest way to avoid the OT though is to just use a transistor in Darlington or CFP (Sziklai) configuration. Drain or collector connected to the tube plate for Darlington (or to cathode for CFP) so that triode operation is preserved. (these act as current multipliers to reduce loading on the tube) There have been at least three threads on this idea in various forms. Another variant is using a Mosfet buffer off a pentode plate with feedback back to the screen grid (giving triode like results too). The success of these approaches depends on a high quality, low capacitance Mosfet or a constant Beta bipolar and avoiding class B operation crossover problems.
If SE sound is what you want, but using a P-P xfmr, take a look at the current thread "Spud-assist: Totem-Pole Current-Mirror PP Hybrid" by Michael Koster. This doubles the power output of SE too.
If price is no object, a full permalloy core should solve most problems nicely. For a bit less, there are the pin-striped cores (nickel lams periodically spaced among the the M6 lams) which should come close. This may be the best practical solution for building an amplifier with no feedback. Would be REAL NICE to have a toroid with a thin nickel strip co-wound with the M6 strip. I'm surprised Menno hasn't come up with this product yet. It will mandate a DC control servo of course.
I would like to see some cheap winder come out with toroid OTs too. After all, once you have a CNC toroid winder, which any of the toroid makers already have, there is nothing hard about making an audio version. It just requires progressive winding technique which is easily programmed. Probably some of the power ones use it already, but most seem to be stuck in the bad practice of co-winding two 120V windings at once, which ruins HF performance.
Don
Don,
You mentioned some Avel-Lindberg PS toroids with progressive winding back in 2002 http://www.diyaudio.com/forums/showthread.php?postid=14336#post14336
- do you remember the model no?
You mentioned some Avel-Lindberg PS toroids with progressive winding back in 2002 http://www.diyaudio.com/forums/showthread.php?postid=14336#post14336
- do you remember the model no?
Barkhausen noise occurs as the core is magnetized along the steepest part of the magnetization curve, but the minor hysteresis loops occur along a less steeply sloped line (representing incremental permeability) that is likely free of noise caused by "stiction". In a single-ended OPT, the Barkhausen noise theoretically occurs during the initial magnetization of the core, namely when the amplifier is turned on. I believe this can only really be proved through experimentation, which will be somewhat costly and time consuming. I say this because the sources I've seen state only that Barkhausen noise "occurs during magnetization when the slope of the B-H curve is steepest" without giving any value for permeability that would be considered "steep".
John
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.