carpenter said:I'm going to build a tube amp someday, Choky. I just have to get up the nerve to handle such high voltage--no room for careless fingers in that venue!
Having done the tube thing and had the pee-widdle-dookie shocked out of me more times than I can count, I will certify that it is unpleasant. Not enough to keep me from coming back for more, though.
The reason I hang out here rather than in the Tube Department is not because I got tired of getting shocked by electricity, but because I got tired of sticker shock every time I went to buy new tubes. They do sound awfully good, though,
One neat thing about tubes: All the common dual triodes are matched to a degree incomprehensible to the average solid state maven. And they're a lot easier to find than these silly Japanese JFETs.
Grey
flg said:
Carpenter,
That is exactly why you would get better coupling of your + and - inductors if wound together, rather than 2 separate units side by side
Bifilar makes for good coupling, but does the "every other" wire thing make for poor inductance in the individual coils? Each coil has half the density when wound bifilar--or so it seems.
BTW, it's nice to see so many familiar avatars in this little thread.
I should come up with an avatar, shouldn't I...
Yes I beleive your individual inductance value would be less than ideal. But, this thing doesn't work optimally with one at a time. Beyond that, I once tried to describe that the actuall L value is 4 times the individual when coupled (maybe like a ZV7-T). I expected a more fluent inductor person to confirm or explain correctly... That did not happen??? Oh well...
I beleive the calculation/Formula for one single inductor in the ZV7-T is not the correct X subL to use or expect, for the AC load it actually presents the lower stage of the circuit My thinking is, the total inductance the differential signal works into can be considered as 1 large inductor with the + and - halfs coupled. That would be 4 X 1 side of the inductor. However, the + and - side of the diff only see 1/2 the total each as a load.
Therfore, a 100 mH center tapped choke (50mH-50mH) in this type of differntial mode would load each side of the diff with an X sub L value of 31.4 ohms @ 50Hz. Or, the inductor dips to 8 ohms (per side) at 12.7Hz...
I suppose a look at #of turns and the L formula would explain this better than I??? And, an experiment vs 2 8 ohm Rs would prove it???
Hopefully, I wasn't opening my big mouth to much and I will have to prove this to myself??? I'm, still working on a SE, L loaded circuit without the benifits of the cancelled DC flux and coupling. And right now I'm sitting here next to all my high voltage BOSOZ/BZLS Power Supply components in their new chasis, waiting for holes and screws and more buss wire and soldering iron and connectors, and oh yea, can't forget the blue LED. I don't need to be spending time on another amp...
Yes I beleive your individual inductance value would be less than ideal. But, this thing doesn't work optimally with one at a time. Beyond that, I once tried to describe that the actuall L value is 4 times the individual when coupled (maybe like a ZV7-T). I expected a more fluent inductor person to confirm or explain correctly... That did not happen??? Oh well...
I beleive the calculation/Formula for one single inductor in the ZV7-T is not the correct X subL to use or expect, for the AC load it actually presents the lower stage of the circuit
My thinking is, the total inductance the differential signal works into can be considered as 1 large inductor with the + and - halfs coupled. That would be 4 X 1 side of the inductor. However, the + and - side of the diff only see 1/2 the total each as a load.Therfore, a 100 mH center tapped choke (50mH-50mH) in this type of differntial mode would load each side of the diff with an X sub L value of 31.4 ohms @ 50Hz. Or, the inductor dips to 8 ohms (per side) at 12.7Hz...
I suppose a look at #of turns and the L formula would explain this better than I??? And, an experiment vs 2 8 ohm Rs would prove it???
Hopefully, I wasn't opening my big mouth to much and I will have to prove this to myself??? I'm, still working on a SE, L loaded circuit without the benifits of the cancelled DC flux and coupling. And right now I'm sitting here next to all my high voltage BOSOZ/BZLS Power Supply components in their new chasis, waiting for holes and screws and more buss wire and soldering iron and connectors, and oh yea, can't forget the blue LED. I don't need to be spending time on another amp...
I don't mean to be disagreeable, or, contadictory, or whatever, But, In my bis we always speak of bi filar meaning 2 wires wound together. We are usually considering paralleling them for lower leakage, Less DCR and high current... The same L value results because they are parraleled windings, not a total number like series wound would be...
This would be the secondary of a flyback transformer in the case you are wondering but, the same rules apply...
One point I was trying to make previously was that the bifilar wound choke with one end connected as a center tap, and a differential signal is like a series connection (with a center tap). And it would be 4X the L, not double the L.
O.K. that does it, where's that formula... ... ...
This would be the secondary of a flyback transformer in the case you are wondering but, the same rules apply...
One point I was trying to make previously was that the bifilar wound choke with one end connected as a center tap, and a differential signal is like a series connection (with a center tap). And it would be 4X the L, not double the L.
O.K. that does it, where's that formula... ... ...
O.K., I just went to the first inductor calc bookmark I found and tested my theory on it! Yea Grey? Simulation however simple, excuse me. Yea, 990 turns on a 3" dia form with 12 Ga wire, Bla, Bla, was 50mH. (it would wiegh 20lbs US, 9 kg) 1696 turns on the same form would be 200mH. 4X the L but 2X the turns.
Actually I'm off a little, 10% or something, or the calculator?, but we are close enough
Thank You
however simple, excuse me. Yea, 990 turns on a 3" dia form with 12 Ga wire, Bla, Bla, was 50mH. (it would wiegh 20lbs US, 9 kg) 1696 turns on the same form would be 200mH. 4X the L but 2X the turns.Actually I'm off a little, 10% or something, or the calculator?, but we are close enough
Thank You
Very interesting, Lee.
I keep seeing two wires in parallel, not series. I also see more current flowing, which allow greater electromagnetic strength per winding--in a DC mode. With AC, I haven't a clue.
Magura wrote that his test demonstrated the bifilar choke operated far below the expected cutoff of 20hz.
I'm going to be happy with what both of you have suggested--no matter what, there will be a gain in inductance.
I'll just remember to keep the profile of the windings symmetrical. I appreciate everyone's involvement.
I keep seeing two wires in parallel, not series. I also see more current flowing, which allow greater electromagnetic strength per winding--in a DC mode. With AC, I haven't a clue.
Magura wrote that his test demonstrated the bifilar choke operated far below the expected cutoff of 20hz.
I'm going to be happy with what both of you have suggested--no matter what, there will be a gain in inductance.
I'll just remember to keep the profile of the windings symmetrical. I appreciate everyone's involvement.
flg said:O.K., I just went to the first inductor calc bookmark I found and tested my theory on it! Yea Grey? Simulation
Actually I'm off a little, 10% or something, or the calculator?, but we are close enough
OK, now how can I actually measure this? I have as I vrote earlier such an inductor laying around (actually got a pair), so it could be interesting to get a real life measurement.
Magura
I remember measuring an 800VA toriod Pri some time ago. A dual Pri (110/220). It measured something like 270mH on 1/2 and almost 1.2H on the 2 in series.
All this means is that, when calculating the low frequency gain or cut-off freq, each winding will be producing double the reactance of the L in my SE version due to the differential coupling bis.
Carpenter, Think of the current through the inductor in differential operation. The current through one inductor is the opposite the current in the other. Imagine the inductor in a strait line, from drain to drain, you effectively have one long inductor. You can than consider the whole thing as twice as many turns with the same current.
You have it a little easer than some, me, since you are driving the upper freq range and your inductor size should be more manageable. I have a 13lb 120mH choke in mine.
Magura, Actually measuring it? Well, I guess you don't have an intrument ha? Yep, gotta think about it to be accurate, but here goes. You get a DC power supply and connect a big FET to it. On the other end of the FET you connect your inductor. On the other end of your inductor you connect it to a small value (.1ohm) sense resistor and then back to the other side of the supply. You are free to select the Voltage, Current and gender of FET to make this thing operate in a range you can use... Then you connect a signal generator to the gate of the FET and drive it good and hard (maybe 10V) with a square wave. You need a scope to look at the Voltage on the sence resistor. It will be a sawtooth wave ramping up with the square wave. With the measurements from the sawtooth wave you can calculate the L.
I know that was a little incomplete but if you really want to try I can come up with better details...
All this means is that, when calculating the low frequency gain or cut-off freq, each winding will be producing double the reactance of the L in my SE version due to the differential coupling bis.
Carpenter, Think of the current through the inductor in differential operation. The current through one inductor is the opposite the current in the other. Imagine the inductor in a strait line, from drain to drain, you effectively have one long inductor. You can than consider the whole thing as twice as many turns with the same current.
You have it a little easer than some, me, since you are driving the upper freq range and your inductor size should be more manageable. I have a 13lb 120mH choke in mine.
Magura, Actually measuring it? Well, I guess you don't have an intrument ha? Yep, gotta think about it to be accurate, but here goes. You get a DC power supply and connect a big FET to it. On the other end of the FET you connect your inductor. On the other end of your inductor you connect it to a small value (.1ohm) sense resistor and then back to the other side of the supply. You are free to select the Voltage, Current and gender of FET to make this thing operate in a range you can use... Then you connect a signal generator to the gate of the FET and drive it good and hard (maybe 10V) with a square wave. You need a scope to look at the Voltage on the sence resistor. It will be a sawtooth wave ramping up with the square wave. With the measurements from the sawtooth wave you can calculate the L.
I know that was a little incomplete but if you really want to try I can come up with better details...
Attachments
flg said:
Carpenter, Think of the current through the inductor in differential operation. The current through one inductor is the opposite the current in the other. Imagine the inductor in a strait line, from drain to drain, you effectively have one long inductor. You can than consider the whole thing as twice as many turns with the same current.
You have it a little easier than some, me, since you are driving the upper freq range and your inductor size should be more manageable. I have a 13lb 120mH choke in mine.
I really can grab hold of your explanation, Lee. Thanks. Previously, I was focused on the DC portion of the choke. This morning, I tried thinking of the center-tapped choke as you just described and "bingo"--it made sense.
flg said:
Carpenter, Think of the current through the inductor in differential operation. The current through one inductor is the opposite the current in the other. Imagine the inductor in a strait line, from drain to drain, you effectively have one long inductor. You can than consider the whole thing as twice as many turns with the same current.
You have it a little easier than some, me, since you are driving the upper freq range and your inductor size should be more manageable. I have a 13lb 120mH choke in mine.
I really can grab hold of your explanation, Lee. Thanks. Previously, I was focused on the DC portion of the choke. This morning, I tried thinking of the center-tapped choke as you just described and "bingo"--it made sense.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.