The HF AC bias current added to tape heads for recording was to help overcome the coercive magnetic forces in the tape material. Essentially acting to improve the permeability (temporarily during the field imprinting for later storage).
It would not be too surprising if the audio signal acts similarly with a small DC imbalance in an OT. Like a table of sand being shaken with a slight tilt, most of the sand will eventually shift downhill. So I can see where a slight DC imbalance could damage the sonics of an OT over time. It is certainly well known that increasing the DC imbalance worsens the quality of operation of an OT.
However, OT core material is magnetically much "softer" than recording tape material. (lower coercive force to overcome) Then again, there is a distribution of coercive level among the many magnetic domains, so some will be "harder" in nature.
The question in my mind would be whether just correcting the DC imbalance would eventually correct the situation, after "shaking" the magnetics with some strong audio for a while. Or whether some stronger demagnetization specific signal would be required to unlock some "hard" magnetization domains. In either case the demagnetization effort should help the sound some, just a question of whether it is -necessary- for getting there. Probably the "harder" magnetic domains don't contribute much to the normal operation, so might have minimal additional effect.
Obviously advantageous if the amplifier repairer/restorer can produce an immediate improvement in the sound, using a demagnetization process. Rather than the user seeing a slow improvement over time afterwards (DC now balanced), maybe just ascribed to tube "burn-in". (but actually from audio signal demag. effects)
..
It would not be too surprising if the audio signal acts similarly with a small DC imbalance in an OT. Like a table of sand being shaken with a slight tilt, most of the sand will eventually shift downhill. So I can see where a slight DC imbalance could damage the sonics of an OT over time. It is certainly well known that increasing the DC imbalance worsens the quality of operation of an OT.
However, OT core material is magnetically much "softer" than recording tape material. (lower coercive force to overcome) Then again, there is a distribution of coercive level among the many magnetic domains, so some will be "harder" in nature.
The question in my mind would be whether just correcting the DC imbalance would eventually correct the situation, after "shaking" the magnetics with some strong audio for a while. Or whether some stronger demagnetization specific signal would be required to unlock some "hard" magnetization domains. In either case the demagnetization effort should help the sound some, just a question of whether it is -necessary- for getting there. Probably the "harder" magnetic domains don't contribute much to the normal operation, so might have minimal additional effect.
Obviously advantageous if the amplifier repairer/restorer can produce an immediate improvement in the sound, using a demagnetization process. Rather than the user seeing a slow improvement over time afterwards (DC now balanced), maybe just ascribed to tube "burn-in". (but actually from audio signal demag. effects)
..
Last edited:
IMHO:
If core magnetization is a problem, then for a:
1. Push Pull Output Transformer:
If a large un-symmetrical music signal causes the core to be magnetized, then a nice large sine wave signal should be able to de-magnetize the core (but not if the amp has large 2nd harmonic distortion . . . that is a form of an un-symmetrical signal).
If unequal DC quiescent current in the output tubes causes the core to be magnetized, then a nice large sine wave signal should be able to de-magnetize the core (but not if the amp has large 2nd harmonic distortion . . . that is a form of an un-symmetrical signal).
But . . . as soon as the signal goes away, the core is re-magnetized because the DC imbalance is still there.
Get the DC currents balanced first, and then de-gauss using a sine wave.
2. Single Ended Output Transformer:
The core will be magnetized due to the DC current.
3. Parafeed Output Transformer:
There is no DC (unless the coupling cap is leaky), but perhaps un-symetrical signals cause magnetization. If so, a large sine wave should be able to de-gauss the core (but again . . . it will not if there is significant 2nd harmonic distortion of the amp . . . that again is un-symetrical).
If core magnetization is a problem, then for a:
1. Push Pull Output Transformer:
If a large un-symmetrical music signal causes the core to be magnetized, then a nice large sine wave signal should be able to de-magnetize the core (but not if the amp has large 2nd harmonic distortion . . . that is a form of an un-symmetrical signal).
If unequal DC quiescent current in the output tubes causes the core to be magnetized, then a nice large sine wave signal should be able to de-magnetize the core (but not if the amp has large 2nd harmonic distortion . . . that is a form of an un-symmetrical signal).
But . . . as soon as the signal goes away, the core is re-magnetized because the DC imbalance is still there.
Get the DC currents balanced first, and then de-gauss using a sine wave.
2. Single Ended Output Transformer:
The core will be magnetized due to the DC current.
3. Parafeed Output Transformer:
There is no DC (unless the coupling cap is leaky), but perhaps un-symetrical signals cause magnetization. If so, a large sine wave should be able to de-gauss the core (but again . . . it will not if there is significant 2nd harmonic distortion of the amp . . . that again is un-symetrical).
The only problem with PP OPT is the hysteresis, that give a big problem that not happen with the SE, the SE are semisaturated all the time and the little gap of work that remain usable not has Hysteresis. Maiybe with high quality of ferromagnetic material is possible to improve the hysteresis behaviour. My question is: Is possible to get better linearity of the hysteresis curve using High current of class A?
Best Regards
Best Regards
Hysteresis is a behavior of all magnetic materials independent of type of operation. It is easily visible by the open loop of a BH curve. Applying a DC bias will change the shape of the loop from on that has all odd harmonics to one that has both even and odd harmonics but DC bias does not somehow eliminate this core characteristic of magnetic materials. The airgap used to combat the DC biad does have a pronounced effect on the hysteresis loops but that is simply because you are effectively averaging the characteristics of air which has no hysteresis (for this practical discussion) with that of the material of choice.
I see a crude proof of this goes something like this...
Since the area inside the hysteresis loop represents a major aspect of core loss, if applying a DC bias eliminates hysteresis, it would also have to eliminate hysteresis loss.
dave
The problem with OT's is magnetizing current, of which hysteresis is a minor component. Neither of these directly affects voltage transfer through the OT anyway, just current loading on the output tube(s).
Magnetizing current is reduced by a high primary inductance, and its affects on tube distortion is reduced by a low output Z from the tube(s). P-P OTs are an excellent solution with high primary inductance, while SE OTs are the worst possible compromise, with low inductance and typically high Z drive (no N Fdbk). The high SE distortion produced "fortunately" ends up as syrupy 2nd harmonic distortion. Everyone to their own taste.
Magnetizing current is reduced by a high primary inductance, and its affects on tube distortion is reduced by a low output Z from the tube(s). P-P OTs are an excellent solution with high primary inductance, while SE OTs are the worst possible compromise, with low inductance and typically high Z drive (no N Fdbk). The high SE distortion produced "fortunately" ends up as syrupy 2nd harmonic distortion. Everyone to their own taste.
Hi Smoking member, Thanks for your comments now, this are related to Triodes, more Z better linearity? I was agreeing with you but recently I read an article that discuss about the difference between SE and PP, I guess until now, Further the euphonic character of even harmonics, the best at all always is, no extra harmonic. I didn't compare yet the sonic difference btwn both kind of amps, All my amps were PP except the FWJ2 I do listen every day but not with real soul. Now I have not the article to show you and the rest of gentlemen.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.