Hello and thank you for coming by.
After many days and hours of working on a layout for a stereo 2-channel guitar tube amp build, I am left with one issue that is leaving me slightly uncomfortable. While I believe 90% of the layout was able to follow good electrical layout guidelines for which I'm fairly confident off, there is the issue of the input tubes, 12ax7 (or 5751) being within 1 1/2" from the output transformer. Each amp is 20 watts and will be used at fairly low volume, never being over driven to any extent. Its a Marshall Plexi preamp circuit with a basic push-pull pair of 6V6;s at the outputs. Although my 3-4 amp builds have been very successful and quiet, I am not an expert in transformer magnetic field emissions and susceptibility so I thought taking some precautions would be a good idea. After all, it will be my personal gigging amp so, I have no issue with adding precautions even if it doesn't necessarily require it.
Just from initial research on possible ways to address the potential for a problem, I see that there are materials uses for shielding, sheets and foils that are chosen for particular frequency bands and applications. I see mu metal sheets mentioned quite a bit, and for the audio range, materials such as Netic, finemet and Metglas, which appear to be forms of steel that are chosen for shielding in the audio range. I also see mentioned that the thickness improves the performance as well as the geometry of the material. One source said that the application of these materials should be set around the element that you are trying to protect and not the source of the magnetic field emissions.
In my case, the tube (12ax7) is the device I'm attempting to protect, and the source of the magnetic field is an output transformer about 1-2 inches away.
In light of this, is there any approach that occurs to you that you think may provide a solution for protecting the 12ax7 from the output transformer?
Thank you and I appreciate you taking the time for this. I have my heart set on making this stereo amp a good performer to put all the nice stereo time-bases pedals to use that are available now.
Thank you,
Best,
Phil Donovan
After many days and hours of working on a layout for a stereo 2-channel guitar tube amp build, I am left with one issue that is leaving me slightly uncomfortable. While I believe 90% of the layout was able to follow good electrical layout guidelines for which I'm fairly confident off, there is the issue of the input tubes, 12ax7 (or 5751) being within 1 1/2" from the output transformer. Each amp is 20 watts and will be used at fairly low volume, never being over driven to any extent. Its a Marshall Plexi preamp circuit with a basic push-pull pair of 6V6;s at the outputs. Although my 3-4 amp builds have been very successful and quiet, I am not an expert in transformer magnetic field emissions and susceptibility so I thought taking some precautions would be a good idea. After all, it will be my personal gigging amp so, I have no issue with adding precautions even if it doesn't necessarily require it.
Just from initial research on possible ways to address the potential for a problem, I see that there are materials uses for shielding, sheets and foils that are chosen for particular frequency bands and applications. I see mu metal sheets mentioned quite a bit, and for the audio range, materials such as Netic, finemet and Metglas, which appear to be forms of steel that are chosen for shielding in the audio range. I also see mentioned that the thickness improves the performance as well as the geometry of the material. One source said that the application of these materials should be set around the element that you are trying to protect and not the source of the magnetic field emissions.
In my case, the tube (12ax7) is the device I'm attempting to protect, and the source of the magnetic field is an output transformer about 1-2 inches away.
In light of this, is there any approach that occurs to you that you think may provide a solution for protecting the 12ax7 from the output transformer?
Thank you and I appreciate you taking the time for this. I have my heart set on making this stereo amp a good performer to put all the nice stereo time-bases pedals to use that are available now.
Thank you,
Best,
Phil Donovan
Start with a common tube shield that has been used in about a billion old radios. Easy to find.
Thanks stephe.
How it is now, there is room to say, rotate the transformers if I could determine how the flux orients, but there isn't a whole lot of room to move it very far. I could build up the amp as is and do some empirical testing to see if there is an angle of rotation that provides an acceptable result. I have noticed by doing "the headphone trick" that even the smallest amount of rotation, distance and/or tilting can provide the difference between hearing something, and hearing nothing. But I would definitely be into trying some shielding as that would be optimal - don't have to re-structure the layout. That would be a big deal at this point.
Thanks,
Phil D
How it is now, there is room to say, rotate the transformers if I could determine how the flux orients, but there isn't a whole lot of room to move it very far. I could build up the amp as is and do some empirical testing to see if there is an angle of rotation that provides an acceptable result. I have noticed by doing "the headphone trick" that even the smallest amount of rotation, distance and/or tilting can provide the difference between hearing something, and hearing nothing. But I would definitely be into trying some shielding as that would be optimal - don't have to re-structure the layout. That would be a big deal at this point.
Thanks,
Phil D
Hi 20to20.
I did infact make sure I ordered the belton 9-pin sockets that will take a twist on tube shield. it would be nice if there were tube shields of various metal types that would provide some deflection of the output transformers field in the audio frequency range. I don't need to worry about any 60Hz or 120Hz buzz, just any potential coupling that "may" occur from the output transformer. Of course, this may all go down as "low frequency" shielding material. I'm right at the beginning of finding out if those things exist. I am not yet sure what the usual tube covers are meant to block or deflect. But I will very shortly!
Thank you,
Phil D
I did infact make sure I ordered the belton 9-pin sockets that will take a twist on tube shield. it would be nice if there were tube shields of various metal types that would provide some deflection of the output transformers field in the audio frequency range. I don't need to worry about any 60Hz or 120Hz buzz, just any potential coupling that "may" occur from the output transformer. Of course, this may all go down as "low frequency" shielding material. I'm right at the beginning of finding out if those things exist. I am not yet sure what the usual tube covers are meant to block or deflect. But I will very shortly!
Thank you,
Phil D
An output transformer is not likely to be a source of trouble unless there is a major imbalance of current and that would create a major hum issue to resolve first. The signal radiation from an output tranny is the same signal that the driver sees and would mask or blend with it so you wouldn't be able to tell if there was any drver ''pickup'' of the output signal. It's the power transformer mechanical buzz or hum that normally finds its way into a driver to create hum.
That said, I could imagine a very worse case scenario of a driver with a high input impedance on the grid picking up tranny radiation of the signal that could become a positive feedback signal and it going into oscillation. Can't say I've ever heard of a case like that though. Never say never in this hobby.
That said, I could imagine a very worse case scenario of a driver with a high input impedance on the grid picking up tranny radiation of the signal that could become a positive feedback signal and it going into oscillation. Can't say I've ever heard of a case like that though. Never say never in this hobby.
Last edited:
And I certainly believe (and hope) that that could very well be the case. Possibilities meander through my head that if there were coupling to the 12ax7 would it be "in-phase interference"? "Out of phase interference"? But this is merely groundless rumination on my part - entertaining worst case scenario possibilities.
Thank you,
Phil D
Thank you,
Phil D
Orienting the output transformer correctly (coils not pointing at the 12AX7) will be fine. Forget shielding. Valve screening cans are for the electrostatic field, not magnetic. Magnetic shielding is invariably expensive, difficult, and not as effective as you'd hoped. Mumetal is required. Did I say it was expensive? It's expensive. Distance and orientation are your friends.
Well, thanks for coming by, but there hasn't been any report of the OT actually causing anything. I stated that I'm all for preventive measurements should there be any chance of magnetic fields coupling into the input 12ax7. After many iterations of layout possibilities the best case scenario I could come up with ended up with the output transformers 1 1/2" from the OT. It would actually be interesting to try some say, mu metal plate sections moved between the OT and tube to hear if there are any interactions. If there weren't any, and the amp sounded great, then I might feel better that there wasn't even any problem to resolve. For now, I'm in the "inquisitive phase" where I see a potential for an issue, and perhaps worth finding a method to show if it is or not.
Thank you,
Phil D
Thank you,
Phil D
Are you using just the one driver for both channels? If so, then feed just the one channel that goes to the OPT in question and see if it radiates to the driver and creates signal that goes to the other output tranny. Put your scope on the unfed OPT to look for signal, or you can feed a frequency that a DMM can read OK and use that if no scope is available. That test would also serve as a run-of-the-mill X-talk check.
Last edited:
I believe you are way overthinking this. If it was a power transformer, maybe? I don't honestly think EMI from an OT will impact a driver tube like it can on a beam pentode. Just don't put it in the flux zone and you will be fine.
Hi 20to20.
Each amp has its own driver. This is actually two completely independent amps in one chassis/cabinet with nothing in common except a little channel switch to select channels (or both). It is both symmetrical and mirror image. You could literally band saw the amp down the center and have two complete soveirgn amplifiers (might want to cover the big holes you sawed out!) that are mirror images of one another. I just thought that layout out with panel would be really nice and so far, it is. The two boards are complete, the front panel has been made, I just started prepping the back panel for switches, fuse, trickle switch (its diode rectification), speaker jacks and FX send/returns. So at this point I'm a bit commited to the layout, which I believe with be fine. Its just that damn OT/V1 that is taunting me a bit. I really think I addressed all other classic potential issues using classic known non-problematic layouts. My 2 other gigging amps I built stayed fairly true to classic Marshall and Fender layouts so I stayed safe with those regarding magnetic and AC coupling. I did however use the headphone trick to finely tweak OT and PT orientation/distance and that works pretty good. I used it for this stereo amp also. It helped confirm that the OT's and related PT's had adequate distance between them for acceptable 60 cycle hum induction to OT. Funny thing though - at full distance there was an extreme reduction in hum, enough to satisfy me. But, if I tilted the OT on its far side 1/4" up, (not 1/8, not 3/8), then any semblance of hum totally disappeared into black silence. That was a little aggravating and caused me to consider an angled aluminum platform to achieve that precise angle. But I decided I'm probably OK the way it is. Ha, for now. That showed me that there isn't always an obvious model to guide transformer orientation down to total silence, that some aspects are a bit invisible, and it takes some detailed physical manipulation to achieve perfection. I guess I digress here.
I have a full bench with scope so can do some experiments to see imposed ac signal waveforms. The OT"s are rear middle about an inch apart. I really don't see a problem of say, channel A OT coupling with channel B 12ax7 as there is considerably more distance between opposite channel OT's and drivers. But you have given me and idea I can try out that may very well show any deviation due to the proximity. Also, there may be a slight opportunity to unfasten the OT to give just enough "service loop" to see if manipulating the OT in space is causing any deviations in the output of the driver tube. I will keep that in mind for sure, sounds like a workable test.
Thanks for your ideas, this is why I came here! Good input here.
Best,
Phil D
Each amp has its own driver. This is actually two completely independent amps in one chassis/cabinet with nothing in common except a little channel switch to select channels (or both). It is both symmetrical and mirror image. You could literally band saw the amp down the center and have two complete soveirgn amplifiers (might want to cover the big holes you sawed out!) that are mirror images of one another. I just thought that layout out with panel would be really nice and so far, it is. The two boards are complete, the front panel has been made, I just started prepping the back panel for switches, fuse, trickle switch (its diode rectification), speaker jacks and FX send/returns. So at this point I'm a bit commited to the layout, which I believe with be fine. Its just that damn OT/V1 that is taunting me a bit. I really think I addressed all other classic potential issues using classic known non-problematic layouts. My 2 other gigging amps I built stayed fairly true to classic Marshall and Fender layouts so I stayed safe with those regarding magnetic and AC coupling. I did however use the headphone trick to finely tweak OT and PT orientation/distance and that works pretty good. I used it for this stereo amp also. It helped confirm that the OT's and related PT's had adequate distance between them for acceptable 60 cycle hum induction to OT. Funny thing though - at full distance there was an extreme reduction in hum, enough to satisfy me. But, if I tilted the OT on its far side 1/4" up, (not 1/8, not 3/8), then any semblance of hum totally disappeared into black silence. That was a little aggravating and caused me to consider an angled aluminum platform to achieve that precise angle. But I decided I'm probably OK the way it is. Ha, for now. That showed me that there isn't always an obvious model to guide transformer orientation down to total silence, that some aspects are a bit invisible, and it takes some detailed physical manipulation to achieve perfection. I guess I digress here.
I have a full bench with scope so can do some experiments to see imposed ac signal waveforms. The OT"s are rear middle about an inch apart. I really don't see a problem of say, channel A OT coupling with channel B 12ax7 as there is considerably more distance between opposite channel OT's and drivers. But you have given me and idea I can try out that may very well show any deviation due to the proximity. Also, there may be a slight opportunity to unfasten the OT to give just enough "service loop" to see if manipulating the OT in space is causing any deviations in the output of the driver tube. I will keep that in mind for sure, sounds like a workable test.
Thanks for your ideas, this is why I came here! Good input here.
Best,
Phil D
Yeah, I know. I overthink many things a lot. But it's a safeguard against appreciating what I don't know. I need to remain humble to that and not be too assuming on things I don't have a lot of knowledge about. Safe than sorry kind of thing. And above all, I hope you are right and that I am overthinking this!
Then there won't be anything to fix or put a Band-Aid onto.
Best,
Phil D
Then there won't be anything to fix or put a Band-Aid onto.
Best,
Phil D
If you look at old high gainers like the Soldano SLO100, or at new compact high gainers in the market, alot of them have the first gain stage close to the PT with a shield, but not to the OT.
Being a guitar amp, supposed to have at least 4 gain stages in the preamp plus the phase inverter, I would stay away from the OT not that much for the magnetic fields above the chassis, but for the connections below the chassis: you will have cables and grid resistors with few mVpp close to cables with 1kVpp swing on the primary of the OT.
Thats a good point that I probably havent put full thought into - how the connecting wires underneath are oriented, especially the B+ and wires feeding the plates of the 6V6's. There is a pretty large voltage swiing going on there even with my 6V6 clean pedal platform amp. I may be able to reposition the OT to move those wires substantially further away from the input section. Its looking like what I need to do is have at hand a few well thought out test tactics that can be done to show if, and how much interplay is or isn't happening between the V1's and OT.s And surely all of this has me thinking.
Thank you for the good insights
Phil D
Thank you for the good insights
Phil D
"Thats a good point that I probably havent put full thought into - how the connecting wires underneath are oriented, especially the B+ and wires feeding the plates of the 6V6's. "
I agree, wiring layout or "lead dress" is much more of a concern to get right than the proximity to magnetic fields. Cross at 90 degrees and use the 3D space inside the amp to put some distance between them, especially the grid lead to the input tube and the plate lead to the output tube. You should be able to route them away from each other. When building amps, I spend a lot of time working this part out!
I agree, wiring layout or "lead dress" is much more of a concern to get right than the proximity to magnetic fields. Cross at 90 degrees and use the 3D space inside the amp to put some distance between them, especially the grid lead to the input tube and the plate lead to the output tube. You should be able to route them away from each other. When building amps, I spend a lot of time working this part out!
Magnetic fields pretty much can't be sheilded against - orientation and distance are your only real choices. And, of course, don't use a steel chassis if you're concerned about magnetic coupling. But mag fields are low on the priority list - good compact layout, good short grounding, signal paths twisted with signal returns, power supply considered as part of the signal path - conductors are antennae, so treat them as such.
Maybe the most general rule of good design is to remember to consider all signal as appearing across two conductors. "Ground" is not a magic hole of infinite electrons; it's just a fiction. Signal appears across two conductors and both are important. Twisted pairs or coax can cancel a lot of antenna-bility.
Regular vacuum valves have enough slop in their short beam path to be fairly immune to small stray mag fields. Some exceptions might be the deflected beam dual anode valves and CRTs and such long path valves. Local E-fields are your real foe.
All good fortune,
Chris
Maybe the most general rule of good design is to remember to consider all signal as appearing across two conductors. "Ground" is not a magic hole of infinite electrons; it's just a fiction. Signal appears across two conductors and both are important. Twisted pairs or coax can cancel a lot of antenna-bility.
Regular vacuum valves have enough slop in their short beam path to be fairly immune to small stray mag fields. Some exceptions might be the deflected beam dual anode valves and CRTs and such long path valves. Local E-fields are your real foe.
All good fortune,
Chris
Magnetic fields need close coupling to a coil and iron core to create a difference of potential that is a voltage/signal. There just isn't the type of structure in a tube to create that voltage and current. The materials should not be magnetic to the point of vibrating in a magnetic field. If the grid or plate or cathode vibrated with a magnetic field then that would create the same effect as microphonics, which does create voltage flux and signal noise, but the PT field would be the main source of that and it would be obvious hum. That field would affect every tube on the chassis because the PT field is very strong. I've had a PT radiate over 2 ft. away to a phono cartridge.That was an eye opener to discover as I repositioned a TT and it started sending hum to the amp. So, if a driver is microphonic it may pick up vibration from an output tranny, but that is a stretch. They do vibrate to the signal through them. You can hear music from an OPT when it's loaded with a dummy load. But about the only radiation signal that may get into an amp is a strong local radio station via RFI/RMI.
Hi 20o20. If that is the case, then I likely have nothing to worry about. In my mind, I've carried a working/loose idea that a magnetic field crossing a metal path in a closed loop would at least theoretically induce some voltage across the exposed part of the circuit even if extremely small. Of course, the type metal in that closed loop/path and certainly its geometry would come into play but again, to what extent is beyond my ability at this point to ascertain. Again, this is only a loose picture from my limited understanding of how AC magnetic field will induce a voltage, how, and how much.
It sounds like I should just pack away any reservations I"ve had for now, and just build the amps with the layout I have. I know I'm able to recognize the big no-no's of keeping power away from signal, keep good distance between PT and OT, route signal and power wires away from each other, and that has worked for me very well so far.
Thanks again for sticking with me here, and I'll just get on with it, and report back with some results.
Best,
PJD3 - PHIL D
It sounds like I should just pack away any reservations I"ve had for now, and just build the amps with the layout I have. I know I'm able to recognize the big no-no's of keeping power away from signal, keep good distance between PT and OT, route signal and power wires away from each other, and that has worked for me very well so far.
Thanks again for sticking with me here, and I'll just get on with it, and report back with some results.
Best,
PJD3 - PHIL D