I built this some weeks ago and although it sounds good enough for me I'm getting quite a bit of hum from the power transformer. It was purpose machine wound although i don't think there's any shielding.
This pic shows the location of the transformer.
So what I did is to extend the wires and locate the transformer about 40cm away from the chasis and the hum was gone, so I tried with that copper foil around it and it really didn't improve, not noticeable. When I closed the transformer near the pentodes I got the hum again, so I was considering making a metal box for it. Any suggestions ?
This pic shows the location of the transformer.
An externally hosted image should be here but it was not working when we last tested it.
So what I did is to extend the wires and locate the transformer about 40cm away from the chasis and the hum was gone, so I tried with that copper foil around it and it really didn't improve, not noticeable. When I closed the transformer near the pentodes I got the hum again, so I was considering making a metal box for it. Any suggestions ?
An externally hosted image should be here but it was not working when we last tested it.
Three things straight off:-
1. Your copper foil looks very thin. The usual gauge is 1 mm or thicker.
2. Is that a steel chassis or aluminium? If steel, some transfomer flux may be circulating in the chassis. No ammount of foil will fix that. Try mounting the transformer with more non-magnetic spacers between it and the chassis.
3. How did you join the ends of the foil together? The join must be of very low resistance for the foil to do its job and short out the leakage flux. The foil should be cleaned of all fingerprints at the join with saucepan cleaner powder, overlapped a few mm, and soldered with a high wattage iron, not the usual low power iron for electron use.
1. Your copper foil looks very thin. The usual gauge is 1 mm or thicker.
2. Is that a steel chassis or aluminium? If steel, some transfomer flux may be circulating in the chassis. No ammount of foil will fix that. Try mounting the transformer with more non-magnetic spacers between it and the chassis.
3. How did you join the ends of the foil together? The join must be of very low resistance for the foil to do its job and short out the leakage flux. The foil should be cleaned of all fingerprints at the join with saucepan cleaner powder, overlapped a few mm, and soldered with a high wattage iron, not the usual low power iron for electron use.
Last edited:
it's thin indeed, it's what I use to shield stages inside the chasis like separators.
It's iron, but even holding the transformer with my hands close where it should be is enough to induct noise.
Ups!, I did not solder the ends and perhaps there not even contact between them, so this one should be well joined as opposed to the inner shielding between the winding's, which by the way there isn't any in this one.
The shield should be grounded ?
It won't make any difference to the shielding effect of the foil if it is grounded or not.
However, grounding a metal that is otherwise not electrically connected to anything is good practice - from a safety point of view and for minimising noise not magnetic in origin. From teh photo, it looks like you have a glass fibre board of some kind between the transformer core and the chassis. Assuming it is a thermoset and not a thermoplastic, that is good practice, but teh foil should go on teh core side of it, not the chassis side. It probably won't make a noticable difference though.
If your foil is thin, you can proably get away with a standard soldering iron.
However, grounding a metal that is otherwise not electrically connected to anything is good practice - from a safety point of view and for minimising noise not magnetic in origin. From teh photo, it looks like you have a glass fibre board of some kind between the transformer core and the chassis. Assuming it is a thermoset and not a thermoplastic, that is good practice, but teh foil should go on teh core side of it, not the chassis side. It probably won't make a noticable difference though.
If your foil is thin, you can proably get away with a standard soldering iron.
The copper band around the transformer is not a shield for electric fields, which is what I suspect the OP thinks. Such a shield can be almost as thin as you like, and need not form a complete turn, but must be grounded.
This is a belly band, and it reduces external magnetic fields by acting as a very low resistance shorted turn. So it must be thick enough and have a very good low resistance connection at the join. It need not be grounded.
Basically the problem is that your transformer has far too much external flux. Custom winding does not guarantee a good quality result. Is it possible that the core is saturating? It looks quite small. One might expect a power transformer to be similar in size to the OPTs, not smaller.
This is a belly band, and it reduces external magnetic fields by acting as a very low resistance shorted turn. So it must be thick enough and have a very good low resistance connection at the join. It need not be grounded.
Basically the problem is that your transformer has far too much external flux. Custom winding does not guarantee a good quality result. Is it possible that the core is saturating? It looks quite small. One might expect a power transformer to be similar in size to the OPTs, not smaller.
If the core is saturating, it will certainly leak a lot more flux. But that is not likely to be his problem.
Don't overlook that it is excessive voltage that causes saturation, not the load. Flux density in a transformer is determined by voltage and self inductance, and is independent of load. So, even if the OP chose a transformer too small, so long as he chose the correct voltage, hum won't be his problem. An overheated transformer yes, regulation issues perhaps, but not hum.
It is quite common for the power transformer to be smaller than the output transformers. It appears that there are two output transfomers - hence the amp is two-channel. But only two large tubes, so its single ended - DC in the output transformers. That requires a large gapped core in the ouitput trans to retain sufficient inductance.
Quality output transformers tend to be large because they are designed to be efficient from 20 Hz to 20 kHz or so. That means plenty of copper window (= large core), an eye on core losses in design, and space consuming winding techniques to maximise coupling and minimise capacitance.
A power transformer only has to work at 50 Hz, so even though in a Class A amplifier typically has to carry 5 or 6 times the audio output power, the design emphasis is minimum cost.
But they OP probably grabbed whatever output trans he had available anyway.
Last edited:
The OP said he didn't solder the ends of the foil together, so as to make it indeed a very low resistance shorted turn. That's what he should do first. If it doesn't make sufficient improvement (it most likely will), he should increase the effective foil thickness by either replacing it or adding another layer.
Why go looking for a problem that his amplifier symptoms don't fit? If you present to a doctor with influenza, he doesn't order a urine protein test.
Why go looking for a problem that his amplifier symptoms don't fit? If you present to a doctor with influenza, he doesn't order a urine protein test.
Last edited:
I doubt the transformer is hardlimiting enough for a light bulb to light, otherwise excess heat would of been reported. if the foil doesn't help, next up, better isolating the mounting screws with fiber washers top and bottom.
remove all tubes and measure secondary voltage with a variac will give the most useful data, linearity etc.> if the sound changes with removed load then it's probably a buzzing winding, if it changes with input voltage then look at core saturation issues.
remove all tubes and measure secondary voltage with a variac will give the most useful data, linearity etc.> if the sound changes with removed load then it's probably a buzzing winding, if it changes with input voltage then look at core saturation issues.
I get hum even if I hold the transformer with my hands close to the chasis but without touching it.
Not quite, it has 250+250 200mA 5v 3A and 7.5 + 7.5 v (yes, PCL86). Machine wound by a factory for sure not aware this was intended for audio
Forgot to mention at last minute I decided to use a preamp stage so I added the 6j7g's with a separate 6v 1A transformer which is inside the chasis, though that doesn't seem to have any problem, installed a switch to power off that small transformer and didn't change anything.
The power transformer is as big as any 150watts transformer plus a bit more I order it with comfortable margins, however I don't know nothing, but I know even less about transformers.
I will record a video so you can hear the effect of the transformer field.
Last edited:
I said you probably grabbed whatever OUTPUT transformer... not POWER transformer...
Instead of recording a video (you have described what's happened very clearly - more clearly than most in fact), tell us if you have soldered the foil ends together. What effect did that have? Note that overlapping or clamping the ends together is not sufficient.
You said:-
a) moving the power trans off the chassis lowers the hum
b) holding the trannie at about its installed position but clear of the chassis makes little or no difference
c) You did not realise the need to make the foil a closed loop and you did not solder the ends together.
That pretty much proves it as far as I am concerned. The only thing you've changed by moving it off chassis is remove the effective magnetic field from the power transformer away from the output transformers and the tubes.
Solder the ends together and I expect it will improve things. Is there some reason why you can't do this?
Instead of recording a video (you have described what's happened very clearly - more clearly than most in fact), tell us if you have soldered the foil ends together. What effect did that have? Note that overlapping or clamping the ends together is not sufficient.
You said:-
a) moving the power trans off the chassis lowers the hum
b) holding the trannie at about its installed position but clear of the chassis makes little or no difference
c) You did not realise the need to make the foil a closed loop and you did not solder the ends together.
That pretty much proves it as far as I am concerned. The only thing you've changed by moving it off chassis is remove the effective magnetic field from the power transformer away from the output transformers and the tubes.
Solder the ends together and I expect it will improve things. Is there some reason why you can't do this?
Last edited:
I soldered the ends of the copper foil but unfortunately did not improve or at least I didn't notice improvements.
It's interesting how rotating the transformer causes the hum to pan between the channels , also tried with a metal 6j7.
sorry about the mess
before ditching the transformer I'm going to try to get those metallic covers, perhaps they will help.
It's interesting how rotating the transformer causes the hum to pan between the channels , also tried with a metal 6j7.
An externally hosted image should be here but it was not working when we last tested it.
sorry about the mess
An externally hosted image should be here but it was not working when we last tested it.
before ditching the transformer I'm going to try to get those metallic covers, perhaps they will help.
An externally hosted image should be here but it was not working when we last tested it.
It is suprising you noticed no improvemnt at all when you soldered the foil. As we said before the foil is too thin but you should have noticed some improvement.
I doubt that metal cheek plate covers as depicted in your picture will help much.
What is more likely to help is completey enclosing the transfomer in a mild steel box, with the transformer installed inside the box so that there is no metal contact between the transformer parts and the box. To be effective, the box will need to be magnetically continuous all round - 10 mm or more overlap at joins, preferably welded, but rivetting should work well enough. Or braze it.
The box will need to be made of thick sheet and have small holes punched in it for ventilation. It may be cheaper to go for a toriodal transformer.
I smell a rat somewhere. Transformer field hum problems are not usually this difficult - at least not in an amplifier with plenty of room like yours.
Can you measure the hum voltage actually produced at the speaker terminals?
If you leave the amplifier running for 1 hour, how hot does the power transformer get? Too hot to hold? Hot enough to smell? Just warm to the touch?
What is the thickness of the foil?
I doubt that metal cheek plate covers as depicted in your picture will help much.
What is more likely to help is completey enclosing the transfomer in a mild steel box, with the transformer installed inside the box so that there is no metal contact between the transformer parts and the box. To be effective, the box will need to be magnetically continuous all round - 10 mm or more overlap at joins, preferably welded, but rivetting should work well enough. Or braze it.
The box will need to be made of thick sheet and have small holes punched in it for ventilation. It may be cheaper to go for a toriodal transformer.
I smell a rat somewhere. Transformer field hum problems are not usually this difficult - at least not in an amplifier with plenty of room like yours.
Can you measure the hum voltage actually produced at the speaker terminals?
If you leave the amplifier running for 1 hour, how hot does the power transformer get? Too hot to hold? Hot enough to smell? Just warm to the touch?
What is the thickness of the foil?
Last edited:
That was due to the capacity between the 5879 anode and your hand.
From the OP's description, his problem is magnetic coupling - a tube shield won't do much for that. Still, he could try some tempory grounded foil around a tube without much trouble.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.