I'm just about to dive in on an EL34 push-pull amplifier build and am curious to know if the left and right output transformers (OTs) can be TOO close together. I found a schematic and notes in an MJ magazine I picked up in Japan and it specified a chassis size that is unavailable. I've got something close, but in order to get the OTs far enough away from the power transformer (PT) I've drawn them up as being about 5mm apart. Does the forum see any issue with this? Cross talk etc.?
Also, in this position do they seem far enough away from the PT?
I've attached a drawing of what I'm planning with relevant dimensions.
Thanks in advance!
Also, in this position do they seem far enough away from the PT?
I've attached a drawing of what I'm planning with relevant dimensions.
Thanks in advance!
Do the output transformers have metal covers over the windings? If yes, then there should not be a problem. If not you may still be okay. But I'm thinking that the power transformer windings may be facing the same as the outputs. If so, rotate the PT 90º. You may have to move the choke a little further toward the edge to accomplish this.
Do not use a magnetic steel chassis.
It is a wonderful conductor of magnetic fields to every point on the chassis. Ouch!
With 4 EL34 tubes, and 2 output transformers, you are either making 2 push pull channels, or 2 parallel single ended channels.
Push pull output transformers are less likely to 'talk' to each other.
Single ended air gapped output transformers are more likely to talk to each other.
If your B+ uses a choke input filter, the choke has a greater magnetic field than the power transformer has.
If your B+ uses a cap input filter, the choke magnetic field is less than a choke input filter.
Now, you have to learn about the causes and preventions of hum-causing ground loops.
Happy amp designing and building.
It is a wonderful conductor of magnetic fields to every point on the chassis. Ouch!
With 4 EL34 tubes, and 2 output transformers, you are either making 2 push pull channels, or 2 parallel single ended channels.
Push pull output transformers are less likely to 'talk' to each other.
Single ended air gapped output transformers are more likely to talk to each other.
If your B+ uses a choke input filter, the choke has a greater magnetic field than the power transformer has.
If your B+ uses a cap input filter, the choke magnetic field is less than a choke input filter.
Now, you have to learn about the causes and preventions of hum-causing ground loops.
Happy amp designing and building.
Steel chassis are OK.
But please read on . . .
I start planning my amplifiers with the power transformer and choke, to be at right angles to the output transformer.
I have 4 aluminum chassis. For most circuit topologies, I get less than 100uV hum, occasionally as high as 400uV.
I have 1 steel chassis. I use it to test out the circuit topology ideas that I get. I often get 500uV to 2,000uV of hum on the steel chassis.
Then I build that same circuit onto one of the aluminum chassis, and I usually get less than 100uV of hum.
The steel chassis is larger. So I can space the power transformer, choke, and output transformer further apart; versus the smaller aluminum chassis.
Perhaps the magnetic fields are not coupling between the power transformer, choke, and output transformer.
I guess that means the magnetic field from the power transformer and choke are coupled by the steel chassis to the vacuum tubes.
The vacuum tubes are sensitive to magnetic fields, Right?
Just one man's experience.
But please read on . . .
I start planning my amplifiers with the power transformer and choke, to be at right angles to the output transformer.
I have 4 aluminum chassis. For most circuit topologies, I get less than 100uV hum, occasionally as high as 400uV.
I have 1 steel chassis. I use it to test out the circuit topology ideas that I get. I often get 500uV to 2,000uV of hum on the steel chassis.
Then I build that same circuit onto one of the aluminum chassis, and I usually get less than 100uV of hum.
The steel chassis is larger. So I can space the power transformer, choke, and output transformer further apart; versus the smaller aluminum chassis.
Perhaps the magnetic fields are not coupling between the power transformer, choke, and output transformer.
I guess that means the magnetic field from the power transformer and choke are coupled by the steel chassis to the vacuum tubes.
The vacuum tubes are sensitive to magnetic fields, Right?
Just one man's experience.
Last edited:
If you really need to put them close, you can cancell stray fields using a thin copper shorted turn around the transformers. In any case, taking magnetic axes at 90° between them, usually sufices.
They have the standard "bell" covers on the sides and nothing exposed on the top or sides. Pretty typical, nothing fancy and rectangular like James, Hashimoto, etc. Should I be concerned about the big ole 100uF 450V x2 capacitor's proximity to the output transformers as well? It doesn't seem to present a problem, but I might be able to move it forward near the choke. Not sure if there is any benefit. See below.
Thanks for your insight. Regarding the design, I should have elaborated... it is push-pull and I've attached a schematic below. This is my first point-to-point, non-kit build so there's a lot to consider!
I'm going with an aluminum chassis for ease of fabrication... might have to add some internal bracing, etc. if it's too flexy with the transformers.
Could you please clarify? I'm not quite sure I understand how to implement and cancel stray fields. I haven't heard of this - among many things I haven't heard of. The output transformers (OTs) will be perpendicular to the power transformer (PT). I was planning to put the choke perpendicular to the PT as well and put it in the far corner opposite the OTs; however, this would mean the choke and OTs are parallel to each other - albeit as far apart as possible. See the sketch below.
Again, thanks in advance for all of your insight. It's an education for sure.
The axis of the power transformer winding is vertical, not horizontal in your drawing. Rotate it 90 degrees. Choke won't have much magetic field, but good to keep it further away.
Output transformers do not usually suffer seriously from crosstalk even when almost touching. What little does exist is generally masked by the crosstalk already inherent within the audio source. I think you'll be fine.
It's fine, the can of the capacitor is cathode and therefore connected to ground, the transformers have shrouds too. No electrostatic coupling problems there.
You should add a diode from grid to cathode of the 12AU7 phase inverter to protect against internal arcing at startup.
https://i.ibb.co/CnvdCJk/120347359-991784128003409-5031309580399918735-o.jpg
Last edited:
Thanks Tom. For some reason, the laydown PT is confusing to me in terms of orientation. Since it is on its "side" isn't it already at 90 deg. to the OTs? That said, I moved things around - see attached - which creates a bit more space between the PT and the OTs. Thoughts?
Thanks for the heads up re: the diode!
If possible, assemble the circuit / "amp" without drilling some of the holes and experiment with placement. You could even forego a chassis altogether and see if there's any noticeable noise with the OPTs on a 2x4 right next to each other, then slightly apart, and so forth. There really is no substitution for that sort of experimentation in situ.
I vote for moving the 100uF capacitor to where you have indicated as an alternative. Or maybe locate it under the chassis instead of on top, unless you have a strong reason not to do that.
I vote for moving the 100uF capacitor to where you have indicated as an alternative. Or maybe locate it under the chassis instead of on top, unless you have a strong reason not to do that.
If you not use a magnetic steel chassis there is still a magnetic field around some components and they can influance other components.
What would be the best solution?
For sure. I will try to come up with a strategy to mock things up in such a way as to not have to do a ton of 'undoing' after I arrive at an acceptable solution.
It's a bit tight in that arrangement, but not impossible. Beyond keeping the power supply away from the signal, do you see this as being overly problematic? I mocked up a couple of layouts and aesthetically prefer rotating the PT 90 degrees, keeping the 100uF near the back and pulling the choke forward a bit. It solves one problem by getting the PT another 10cm away from the OTs. See attached.
Still struggling with the laydown PT and its orientation. The OTs and choke are vertical mount and after some thought and feedback, I feel that having the PT oriented "front to back" - as in the first image - is the best option.
Am not terribly experienced with custom layouts, but what others have said here indicates that you are likely okay with your original plan--having the OPT very close together.
I've found that agonizing over this level of placement choice is less important than the way things are eventually wired up. That power switch is potentially an issue--on my first amp, I had a power switch on the front. The AC wiring went right behind the choke. Turned out that I could hear nasty hash from my LED dimmer lights when turned on. Moving the power switch (and associated wiring) helped with that. It's probably possible to wire it up without issue, but moving that may un-cramp the top plate a bit?
Personally, I like that capacitor between the PT and the choke. The current loop between the rectifier, choke, and C1/2 looks to me like it would be minimal there compared to putting it next to the OPTs.
I've solicited plenty of layout and build advice online and, like I said, it's really hard for anyone to say what's going to happen or best. You really just have to do it!
I've found that agonizing over this level of placement choice is less important than the way things are eventually wired up. That power switch is potentially an issue--on my first amp, I had a power switch on the front. The AC wiring went right behind the choke. Turned out that I could hear nasty hash from my LED dimmer lights when turned on. Moving the power switch (and associated wiring) helped with that. It's probably possible to wire it up without issue, but moving that may un-cramp the top plate a bit?
Personally, I like that capacitor between the PT and the choke. The current loop between the rectifier, choke, and C1/2 looks to me like it would be minimal there compared to putting it next to the OPTs.
I've solicited plenty of layout and build advice online and, like I said, it's really hard for anyone to say what's going to happen or best. You really just have to do it!
tubes4all,
You said: "If you not use a magnetic steel chassis there is still a magnetic field around some components and they can influance other components.
What would be the best solution?"
This thread has already given solutions to the main factors related to magnetic coupling:
Relative orientation between the various coils (90 degrees, not 0 degrees).
Spacing between the devices that have coils (larger spacing couples less).
You said: "If you not use a magnetic steel chassis there is still a magnetic field around some components and they can influance other components.
What would be the best solution?"
This thread has already given solutions to the main factors related to magnetic coupling:
Relative orientation between the various coils (90 degrees, not 0 degrees).
Spacing between the devices that have coils (larger spacing couples less).
Zung,
That is a nice amplifier, but you do need Ultra Linear taps on the output transformer.
Not all push pull output transformers have ultra linear taps.
And, pentode / beam power tubes with global negative feedback around the output transformer is generally harder to make stable, versus triode wired pentode / beam power tubes.
That is a nice amplifier, but you do need Ultra Linear taps on the output transformer.
Not all push pull output transformers have ultra linear taps.
And, pentode / beam power tubes with global negative feedback around the output transformer is generally harder to make stable, versus triode wired pentode / beam power tubes.
Hum induction by the power transformer can be checked as follows : connect earphones with the primaries of the output transformer.. Switch on the PT. Now move the OT in the direction of the PT. Some hum will be heard. Now move the OT away from the PT and\or turn it around until hum is hardly noticeable. Sometimes you have to lift it somewhat. If there is a slight hum you must decide for yourself if you can live with that, if not move the OT's a little further away.
I always do it this way and it works perfectly.
I always do it this way and it works perfectly.