I am thinking of using a center tapped choke inverter for my 832A PP amp with a 6dj8 driver.
There is transformer company here in the Philippines that winds my OPTs and PTX and my chokes with great results. However, he has not wound any center tapped choke for phase inversion.
I have searched for any specs of this kind of phase inversion and the only specs I got are those posted by Shoog using toroidal power tranny as phase inverters.
Any advise on how to wind these?
There is transformer company here in the Philippines that winds my OPTs and PTX and my chokes with great results. However, he has not wound any center tapped choke for phase inversion.
I have searched for any specs of this kind of phase inversion and the only specs I got are those posted by Shoog using toroidal power tranny as phase inverters.
Any advise on how to wind these?
I think a bi-filar winding and any techniques possible to minimize stray capacitance are a couple of things I would be inclined to try. Leakage inductance is important too for high frequency balance, hence the recommendation for a bi-filar winding with the problems that brings. The insulation on the magnet wire must withstand full B+ (if present) and more than 2X the maximum plate swing of the driving stage, some margin say 50% would be a good idea.
It goes without saying that 1/2 the winding must have sufficient inductance by itself to permit flat response down to the lowest frequency of interest.
It goes without saying that 1/2 the winding must have sufficient inductance by itself to permit flat response down to the lowest frequency of interest.
Got all the information i need now:
Center tapped choke, bi-filar winding, no gap. very similar to a plate choke winding except for the center tap, the end of the first wind and the beginning of the second wind, together will form the center tap, the end of the second wind will have a 180 degree signal compared to the beginning of the first wind.
I will just ask my winder to make sure it has the same bandwidth as all the plate chokes he did for me.
Thanks for all your help.
Center tapped choke, bi-filar winding, no gap. very similar to a plate choke winding except for the center tap, the end of the first wind and the beginning of the second wind, together will form the center tap, the end of the second wind will have a 180 degree signal compared to the beginning of the first wind.
I will just ask my winder to make sure it has the same bandwidth as all the plate chokes he did for me.
Thanks for all your help.
hey-Hey!!!,
This is a fairly simple wind. With a dual bobbin, wind 1/4 of the turns in each bay, one reverse direction compared to the other. Continue winding the next 1/2 of each bay's turns in the opposite bay( but in the same direction they were started in ). Finist the last 1/4 of each bay in the same one it started in.
In each bay you'll have half of each half, and half of the other side( wound in opposite direction ). Each of the turns near the CT will be nest to the core, and the signal ends will be at the outside of the bobbin.
Best Regards,
Douglas
This is a fairly simple wind. With a dual bobbin, wind 1/4 of the turns in each bay, one reverse direction compared to the other. Continue winding the next 1/2 of each bay's turns in the opposite bay( but in the same direction they were started in ). Finist the last 1/4 of each bay in the same one it started in.
In each bay you'll have half of each half, and half of the other side( wound in opposite direction ). Each of the turns near the CT will be nest to the core, and the signal ends will be at the outside of the bobbin.
Best Regards,
Douglas
I wasn't thinking about the bi-filar capacitance issue when I made that suggestion so you might well want to consider it as a potential problem.
The tradeoff between capacitance and leakage inductance is a difficult one to make, generally with a low source impedance more capacitance can be tolerated, but less leakage inductance - for example with a triode with relatively low rp, otoh if you were using a pentode the capacitance becomes the dominant issue, and leakage inductance less so due to the greater source impedance.
Depending on budget you may want to try several different approaches and see what gives you the best overall hf balance.
The tradeoff between capacitance and leakage inductance is a difficult one to make, generally with a low source impedance more capacitance can be tolerated, but less leakage inductance - for example with a triode with relatively low rp, otoh if you were using a pentode the capacitance becomes the dominant issue, and leakage inductance less so due to the greater source impedance.
Depending on budget you may want to try several different approaches and see what gives you the best overall hf balance.
Don't these phase spiltter chokes require special consideration due the asymetrical loads? That's the answer I got back from winders (without more detail) when questioning measurements I took using of unloaded transformer windings, etc. in bench tests. One end of the choke is loaded by the output impedance of the driving stage, the other flaps in the electron breeze loaded only by the output tube grid impedance, effectively infinity. In tests the result was severe ultrasonic ringing.
Usually you can damp the ringing with an appropriately sized load resistor on the none driven end. Worst case it should be a few times larger than the driver tube rp, (say 3 - 4 x to avoid excessive distortion) but usually can be much larger. You'll probably have to determine this empirically.
My last (and only) one used AC coupling (yes there was dc on the core) to the grids of the output tubes and the grid resistors provided enough damping with the particular choke out to at least 40kHz or so. Sonically it didn't seem to be a problem, and the idea appears to have some merit.
My last (and only) one used AC coupling (yes there was dc on the core) to the grids of the output tubes and the grid resistors provided enough damping with the particular choke out to at least 40kHz or so. Sonically it didn't seem to be a problem, and the idea appears to have some merit.
Actually no, it forms a series resonant rlc circuit, and I sure wouldn't count on that for damping except at its specific resonance where it should look resistive, depending on q it may ring like crazy below the rlc resonance. Other capacitances may be dominant in which case the resonance will be at a lower frequency and q on the undriven side will be dominated by the esr of the shunt capacitance and the dcr of the winding.
You probably need some shunt resistance to damp the winding hf resonance.
You probably need some shunt resistance to damp the winding hf resonance.
Bandersnatch said:
Wacky iron of course. The 80K CT winding of some vintage Hammond interstages, various values on the other winding attempted with no success. Granted, most of the really old Hammond audio iron I've tested was no great shakes to begin with, usually demonstrating odd behaviour not far above the audio band.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.