Hi all, first poster here. I'm launching into my first non-kit build based on this design. I'm going to drop the DC heater circuit in favor of AC for simplicity. This leads me to two basic transformer questions which I've had a hard time sourcing a simple answer for. Please forgive any lack of proper terminology or misunderstood concepts, I'm just learning. 1) I plan to use a center-tapped, 6.3V, 6 amp transformer (Edcor) but am uncertain how wiring of the heater circuit differs from a non-center tapped transformer. I understand that each side of the wiring is 3.15VAC out of phase, and along with the grounded center-tap this helps reduce hum. What I don't get is "flow" for lack of a better term. How do two positive voltages, wired into a filament, cause it to heat up? The voltages add up to 6.3, do they create a pathway back through the transformer to ground? Do i just pick a pin for one of the 3.15V taps and stick with it as I daisy chain the filaments? Question 2) the reason I chose this project is that I have two 3.5K output transformers from an Elekit project that I didn't use. There is little info online about these as they are Japanese (Etine or something similar, don't have them with me). They have five wires on the primary side, 3 that i plan to use (B+, UL, and plate) but also a ground and negative feedback. Since this amp has no NFB, i'll just tape that one off. What about the ground? Want to make best use but if that is just taping it off, than so be it. Thanks for your input.
I think I can help with the first part of your question. The two 3.15 volt windings do not generate two positive (or negative) voltages at each end at the same time.
Wired in series you will measure 6.3 volts end to end and 3.15 volt from the centre to either end.
If you were to reverse one of the secondaries then you would still measure 3.15 from centre to each end but crucially zero volts between the two ends. This is because the two windings are now in phase and relative to the centre point do now both generate the same voltage and with the same phase relationship. The net voltage between the ends is therefore zero (so cold heaters 😉)
Wired in series you will measure 6.3 volts end to end and 3.15 volt from the centre to either end.
If you were to reverse one of the secondaries then you would still measure 3.15 from centre to each end but crucially zero volts between the two ends. This is because the two windings are now in phase and relative to the centre point do now both generate the same voltage and with the same phase relationship. The net voltage between the ends is therefore zero (so cold heaters 😉)
Thanks! Right, I see that if I were to combine the two leads then measure at a ground point I'd get 6.3V but I guess my real question is how does one wire this to the filaments? I'm looking at something like the first "Correct" (funny it's spelled wrong) here but confused over how to wire the two 3.15V ends to the tubes. Something is not connecting in my brain.
To make sense of this you have to understand that voltages are always measured 'with respect to somewhere else'.
So a valve with 6.3v filaments has to end up with 6.3 volts across the heater pins.
It doesn't matter what the voltage is from the heater to 'somewhere else', it has to have the voltage across the heater winding.
The crucial bit with the transformer winding is that the winding itself is floating, it is not in itself referenced to anything (not yet anyway) and so you can not measure voltage 'to the winding' from ground unless the winding actually connects to ground in some way.
So you connect the end points of our correctly connected series 3.15 volt windings across the heater and the heater lights.
As that stands there is no voltage measurable from 'ground' to the heaters because there is no continuity or connection to ground.
Any point of that winding can if we want be connected to ground. If we connect an end pin to ground we would then see 6.3 volt on one heater pin and zero on the other (because it connects to ground that we are reading from).
If we connect the centre point to ground and then measure from ground to each valve heater pin we would see 3.15v. The heater still has 6.3 volts across the filament though.
Does that make sense 🙂
So a valve with 6.3v filaments has to end up with 6.3 volts across the heater pins.
It doesn't matter what the voltage is from the heater to 'somewhere else', it has to have the voltage across the heater winding.
The crucial bit with the transformer winding is that the winding itself is floating, it is not in itself referenced to anything (not yet anyway) and so you can not measure voltage 'to the winding' from ground unless the winding actually connects to ground in some way.
So you connect the end points of our correctly connected series 3.15 volt windings across the heater and the heater lights.
As that stands there is no voltage measurable from 'ground' to the heaters because there is no continuity or connection to ground.
Any point of that winding can if we want be connected to ground. If we connect an end pin to ground we would then see 6.3 volt on one heater pin and zero on the other (because it connects to ground that we are reading from).
If we connect the centre point to ground and then measure from ground to each valve heater pin we would see 3.15v. The heater still has 6.3 volts across the filament though.
Does that make sense 🙂
And just to help clarify this. The diagram shows the two 3.15v winding are in series and so producing 6.3 volts between the ends, and those ends go to the valve filament.
That is all the valve needs to work... but how you connect it for minimum hum rejection is then up to you.
If you take the image with a preset resistor and you were to measure from ground to each heater pin then you would only see 3.15v on each heater pin if the preset were centred.
Move the preset toward one end and the ratio changes such as seeing 5v on one pin and 1.3 on the other (as measured from ground) but the 6.3v across the heater is a constant.
Ok, now this is starting to make sense! In series we see 3.15 +3.15V across the heater for a total of 6.3V. In the case of a 6L6, pins 2 and 7 each get one of the 3.15V legs from the center tapped transformer. I guess I was mentally thinking about a "return" but disregarding the fact that I'm working with a transformer. Thanks....now I just need some positive encouragement on #2.
OK, I think your getting the idea 🙂
I'll have to leave it for now but I'm sure someone more experienced in valve amp design will be better placed to answer your other question 😉
I'll have to leave it for now but I'm sure someone more experienced in valve amp design will be better placed to answer your other question 😉
To find out what wire is what on an unknown tfmr start with "buzzing" it out using a DMM diode/continuity check, EG which wire connects to another. Primary wires usually have a red wire as the centre tap,but not always.
You then need to find the tfmr winding ratio by putting a sinewave into one set of windings, either 1v or so from a sig gen or 6v ish from a heater winding,then measure what comes out the other winding, EG if you put 10v intothe pri and get 1v out the secondary you have a winding ration of 10:1, to find the impedance ratio you square the pri, so 10x10=100 so Z pri = 100:1 meaning with an 8r secondary load you'll have 800r on the pri. See here for more info - Output Transformer Impedance
Andy.
You then need to find the tfmr winding ratio by putting a sinewave into one set of windings, either 1v or so from a sig gen or 6v ish from a heater winding,then measure what comes out the other winding, EG if you put 10v intothe pri and get 1v out the secondary you have a winding ration of 10:1, to find the impedance ratio you square the pri, so 10x10=100 so Z pri = 100:1 meaning with an 8r secondary load you'll have 800r on the pri. See here for more info - Output Transformer Impedance
Andy.
That is the primary.
That belongs in the secondary.
Hint:
* primary wires will be at +V potential, hundreds of volts.
* Ground and Nfb will respectively be at Ground/0V and at or near 0V (it *might* be a cathode feedback winding, I mention this just for sake of completeness)
In a nutshell: you can not call all of them "primary wires".
Last edited:
I see what you're saying, again, my terminology may not be the best. These additional wires come out of the transformer at the same location as the "primary" 3. I was trying to indicate that they don't come out the "other side" of the transformer (speaker connections). In the end, I'm trying to figure out if I need to 1) use the ground wire as a ground and 2) if i should use the "NFB" wire or i should just supply the NFB as outlined in the schematic (this is what I'm leaning towards). I was wrong in my earlier statement that the design didn't incorporate NFB.
The 0v and Feedback are the the speaker connection - ie the secondary.
if you don't want any nfb, just omit Rf.
Have you made sure your output transformer is designed for an SE amp?
I'm also building my first tube amp, from the same website.
6SL7 SRPP / KT77 Push-Pull Class-A Ultra-Linear DIY Tube Amplifier
I am also using AC heater supply and not fitting the switch on delay. My mains transformer has a separate 40v ac winding to use for elevating the heater supply (once rectified)
Output transformer is a Hammond 1650H and tubes are 6L6GC.
if you don't want any nfb, just omit Rf.
Have you made sure your output transformer is designed for an SE amp?
I'm also building my first tube amp, from the same website.
6SL7 SRPP / KT77 Push-Pull Class-A Ultra-Linear DIY Tube Amplifier
I am also using AC heater supply and not fitting the switch on delay. My mains transformer has a separate 40v ac winding to use for elevating the heater supply (once rectified)
Output transformer is a Hammond 1650H and tubes are 6L6GC.
No, these are additional wires which emerge from the transformer at the same location as the primary. The "secondary" 8, 5 ohm and ground emerge from the other side of the transformer. Again, will have to look at the elekit schematic to see how the go back into the design.
This is a 3.5k SE transformer that came from an Elekit Tu8800 amplifier, I upgraded to the lundahl transformers for that build so these are the original units.
My bad. Just tape them both up.
I think they will be for a separate feedback winding so won't be needed. (you can confirm by checking for continuity to the main primary winding.
Oddball OT with extra secondary winding
I think they will be for a separate feedback winding so won't be needed. (you can confirm by checking for continuity to the main primary winding.
Oddball OT with extra secondary winding
Last edited:
My thinking too, specific to the Elekit design. I did the continuity test/resistance check and nothing, so I do think they're outside the primary.
You could experiment by connecting the FB to the cathode and the 0v to the cathode resistor (lower value?). i.e. in series with the resistor.
One thread I read suggested that this increases the output power, which you can speculate is because the cathode current is flowing through the winding on its way to the anode and otp primary?
i found some comments re cathode feedback windings
[FONT=MS Sans Serif, Microsoft Sans Serif]These will help generate very high quality bass response unheard from conventional SE amp since the controlled motion of woofer is greatly increased. Minuet, therefore, redefines superb bass response (thus enhances full body) along with its excellent midrange and high frequency response. [/FONT]
[FONT=MS Sans Serif, Microsoft Sans Serif]Additional benefits of using cathode feedback circuit are:
1) Lowering output impedance
2) Increasing Damping factor
3) Extending frequency response at both ends of audio spectrum[/FONT]
[FONT=MS Sans Serif, Microsoft Sans Serif]
[/FONT]
[FONT=MS Sans Serif, Microsoft Sans Serif]http://www.sacthailand.com/AmpMinuetEL34.html
[/FONT]
One thread I read suggested that this increases the output power, which you can speculate is because the cathode current is flowing through the winding on its way to the anode and otp primary?
i found some comments re cathode feedback windings
[FONT=MS Sans Serif, Microsoft Sans Serif]These will help generate very high quality bass response unheard from conventional SE amp since the controlled motion of woofer is greatly increased. Minuet, therefore, redefines superb bass response (thus enhances full body) along with its excellent midrange and high frequency response. [/FONT]
[FONT=MS Sans Serif, Microsoft Sans Serif]Additional benefits of using cathode feedback circuit are:
1) Lowering output impedance
2) Increasing Damping factor
3) Extending frequency response at both ends of audio spectrum[/FONT]
[FONT=MS Sans Serif, Microsoft Sans Serif]
[/FONT]
[FONT=MS Sans Serif, Microsoft Sans Serif]http://www.sacthailand.com/AmpMinuetEL34.html
[/FONT]
Last edited:
I have a, perhaps tangential, question that may be pertinent here since phase has been mentioned (and so not just for me).
I'm building a power supply with dual secondaries, each one feeding a bridge rectifier, common mode choke and capacitor as a raw supply to the regulators.
On the schematic there are "dots" associated with the windings, both primary (dual) and secondary. I can see that the primary windings must be wired "dot to dot" for 120V.
And I guess that on the other side of each bridge the "dot" orientation probably doesn't matter... or does it?
The grounds from the signal board will eventually connect together and, perhaps, connect to the neutral of the AC (perhaps through a bridge a la the Pearl 2). Maybe the dot sides of the primary should connect to the "hot" line of the AC and the dot sides of the secondaries connect to similar pins of the two bridges and so have a similar phase relationship to the "+" and "-" outputs of the bridges and the raw supply?
I don't know if this matters. Just askin'.
I'm building a power supply with dual secondaries, each one feeding a bridge rectifier, common mode choke and capacitor as a raw supply to the regulators.
On the schematic there are "dots" associated with the windings, both primary (dual) and secondary. I can see that the primary windings must be wired "dot to dot" for 120V.
And I guess that on the other side of each bridge the "dot" orientation probably doesn't matter... or does it?
The grounds from the signal board will eventually connect together and, perhaps, connect to the neutral of the AC (perhaps through a bridge a la the Pearl 2). Maybe the dot sides of the primary should connect to the "hot" line of the AC and the dot sides of the secondaries connect to similar pins of the two bridges and so have a similar phase relationship to the "+" and "-" outputs of the bridges and the raw supply?
I don't know if this matters. Just askin'.
It doesn't matter when the windings are kept separate, since you have DC at the smoothing caps.
It would matter if you were to connect the secondaries together.
It would matter if you were to connect the secondaries together.