Filament Choke Test Results

[ko4nrbs]

As suggested by Roger Singley WQ9E I tested the 2.40-31 Torrid Core filament choke.  I used a carbon 5K ohm potentiometer.  One end of the choke was grounded at the signal generator's chassis the other end fed with a 1.900mHz signal with the Pot in series.  Voltages were measured as specified with my Tektronix 465 oscilloscope.

Made for two 3-400Z Tubes in Parallel

Measured inductance of Torrid Core with 5 turns of 16AWG.
1.850mHz 34uh
1.9mHz 33.93uh
2.000mHz 32.8uh


Results:
Frequency 1.900mHz
Using 1X probe

Scope Vdiv set at 2v
Before Pot 6.02v Scope Vdiv set at 2v
After Pot 3.02v Scope Vdiv set at 2v
Pot measures 670 ohms

Before Pot 5.04 with Vdiv set at 1v
After Pot .608v with Vdiv set at .2v
Pot reads 4.728 kohms


Before Pot 5v with Vdiv set at 1v
After Pot 1.02v with Vdiv set at .5v
Pot reads 2.369 kohms


Vdiv set at 1
Before Pot 6v
After Pot 6v
Pot reads .14 ohms

Comments and suggestions are very welcome.

Bill  KO4NR

[K9MB]

I wonder if you can explain the method you are using?
Why are you using a pot instead of a carbon comp resistor?

Seems like a fixed known value resistor in series with your choke with a fixed rf voltage across it measured at the top, then after the resistor and you will be able to calculate the effective impedance of the choke by ohms law.

If you start with a one volt signal across everything and the resistor is-say- 1k and the voltage across the choke id 500mv, then half the drop is across each element, so the choke is 1k ohm impedance at 1.9mhz.
Wouldn’t that work?
Obviously, there are complex values of resistance and reactance in the choke, but the net effect should be close to the calculated value.
Maybe Inam missing something though….

[ko4nrbs]

I wonder if you can explain the method you are using?
Why are you using a pot instead of a carbon comp resistor?

Seems like a fixed known value resistor in series with your choke with a fixed rf voltage across it measured at the top, then after the resistor and you will be able to calculate the effective impedance of the choke by ohms law.

If you start with a one-volt signal across everything and the resistor is-say- 1k and the voltage across the choke id 500mv, then half the drop is across each element, so the choke is 1k ohm impedance at 1.9mhz.
Wouldn’t that work?
Obviously, there are complex values of resistance and reactance in the choke, but the net effect should be close to the calculated value.
Maybe Inam missing something though….

As per K1JJ recommendations I did that very test on beads slide over a wire.  Here are the results using my Tektronix 465 oscilloscope to measure the voltage.  I just don't have enough room for such a large choke.

Bead Test Results (Looks about 15 beads would have
 produced an even voltage drop across the resistor.

  # of beads on Frequency Generator side of resistor/voltage                 Choke side of resistor Voltage
1- .5v                                                                                                             30mv
2- .5v                                                                                                             60mv
3- .5v                                                                                                            100mv
4- .5v                                                                                                            100mv
5- .5v                                                                                                            150mv
6- .5v                                                                                                            .15v
7- .5v                                                                                                            .2v
8- .5v                                                                                                           .2v
10- .5v                                                                                                         .25v
12- .5v                                                                                                         .3v

I'm far from being an expert on amplifiers.  The potentiometer method as explained to me eliminates calculations as it gives you the reactance when you measure the resistance of the potentiometer.   The input impedance of the two 3-400Z tubes in parallel is approximately 60 ohms.  So, when the resistance across the potentiometer measured 670 ohms at 1.900mHz that indicated about 670 ohms of reactance.   Using the 10X rule for filament choke reactance vs. tube input impedance I should be close enough.
Torrid Core with 5 turns 16AWG
Frequency 1.900mHz
Using 1X probe

Scope Vdiv set at 2v
Before Pot 6.02v Scope Vdiv set at 2v
After Pot 3.02v Scope Vdiv set at 2v
Pot measures 670 ohms

I have lots of guys out there to thank for helping me figure this all out.  Their patience has been well tested with all my questions.

73,
Bill KO4NR

[KL7OF]

Thanks Bill....what mix are the beads you used?

[K9MB]

I wonder if you can explain the method you are using?
Why are you using a pot instead of a carbon comp resistor?

Seems like a fixed known value resistor in series with your choke with a fixed rf voltage across it measured at the top, then after the resistor and you will be able to calculate the effective impedance of the choke by ohms law.

If you start with a one-volt signal across everything and the resistor is-say- 1k and the voltage across the choke id 500mv, then half the drop is across each element, so the choke is 1k ohm impedance at 1.9mhz.
Wouldn’t that work?
Obviously, there are complex values of resistance and reactance in the choke, but the net effect should be close to the calculated value.
Maybe Inam missing something though….

As per K1JJ recommendations I did that very test on beads slide over a wire.  Here are the results using my Tektronix 465 oscilloscope to measure the voltage.  I just don't have enough room for such a large choke.

Bead Test Results (Looks about 15 beads would have
 produced an even voltage drop across the resistor.

  # of beads on Frequency Generator side of resistor/voltage                 Choke side of resistor Voltage
1- .5v                                                                                                             30mv
2- .5v                                                                                                             60mv
3- .5v                                                                                                            100mv
4- .5v                                                                                                            100mv
5- .5v                                                                                                            150mv
6- .5v                                                                                                            .15v
7- .5v                                                                                                            .2v
8- .5v                                                                                                           .2v
10- .5v                                                                                                         .25v
12- .5v                                                                                                         .3v

I'm far from being an expert on amplifiers.  The potentiometer method as explained to me eliminates calculations as it gives you the reactance when you measure the resistance of the potentiometer.   The input impedance of the two 3-400Z tubes in parallel is approximately 60 ohms.  So, when the resistance across the potentiometer measured 670 ohms at 1.900mHz that indicated about 670 ohms of reactance.   Using the 10X rule for filament choke reactance vs. tube input impedance I should be close enough.
Torrid Core with 5 turns 16AWG
Frequency 1.900mHz
Using 1X probe

Scope Vdiv set at 2v
Before Pot 6.02v Scope Vdiv set at 2v
After Pot 3.02v Scope Vdiv set at 2v
Pot measures 670 ohms

I have lots of guys out there to thank for helping me figure this all out.  Their patience has been well tested with all my questions.

73,
Bill KO4NR

Hi Bill,
If you mist use a toroid for the choke, then this data from G3TXQ from balun research mogjt be useful.
As you can see, 5 turns may give you 500 ihms, but 9 turns can give more like 2k ohms across the entire spectrum up to 30 mHz using a single FT-240-31. Note the black line showing the resistive component being dominant . It is considered a good thing in chokes for feedlines and may be also for your purposes because it does no go capacitive and provide a series resonant short  73, MB

PS: note that you must have the wire paired to prevent inducitve flux from saturating the core. That just requires that tje wire be laid side by side. It can be taped together before winding to assure this. You can also wind the two windings of different color insulation one at a time and interlace them.MB

[ko4nrbs]

I wonder if you can explain the method you are using?
Why are you using a pot instead of a carbon comp resistor?

Seems like a fixed known value resistor in series with your choke with a fixed rf voltage across it measured at the top, then after the resistor and you will be able to calculate the effective impedance of the choke by ohms law.

If you start with a one-volt signal across everything and the resistor is-say- 1k and the voltage across the choke id 500mv, then half the drop is across each element, so the choke is 1k ohm impedance at 1.9mhz.
Wouldn’t that work?
Obviously, there are complex values of resistance and reactance in the choke, but the net effect should be close to the calculated value.
Maybe Inam missing something though….

As per K1JJ recommendations I did that very test on beads slide over a wire.  Here are the results using my Tektronix 465 oscilloscope to measure the voltage.  I just don't have enough room for such a large choke.

Bead Test Results (Looks about 15 beads would have
 produced an even voltage drop across the resistor.

  # of beads on Frequency Generator side of resistor/voltage                 Choke side of resistor Voltage
1- .5v                                                                                                             30mv
2- .5v                                                                                                             60mv
3- .5v                                                                                                            100mv
4- .5v                                                                                                            100mv
5- .5v                                                                                                            150mv
6- .5v                                                                                                            .15v
7- .5v                                                                                                            .2v
8- .5v                                                                                                           .2v
10- .5v                                                                                                         .25v
12- .5v                                                                                                         .3v

I'm far from being an expert on amplifiers.  The potentiometer method as explained to me eliminates calculations as it gives you the reactance when you measure the resistance of the potentiometer.   The input impedance of the two 3-400Z tubes in parallel is approximately 60 ohms.  So, when the resistance across the potentiometer measured 670 ohms at 1.900mHz that indicated about 670 ohms of reactance.   Using the 10X rule for filament choke reactance vs. tube input impedance I should be close enough.
Torrid Core with 5 turns 16AWG
Frequency 1.900mHz
Using 1X probe

Scope Vdiv set at 2v
Before Pot 6.02v Scope Vdiv set at 2v
After Pot 3.02v Scope Vdiv set at 2v
Pot measures 670 ohms

I have lots of guys out there to thank for helping me figure this all out.  Their patience has been well tested with all my questions.

73,
Bill KO4NR

Hi Bill,
If you mist use a toroid for the choke, then this data from G3TXQ from balun research mogjt be useful.
As you can see, 5 turns may give you 500 ihms, but 9 turns can give more like 2k ohms across the entire spectrum up to 30 mHz using a single FT-240-31. Note the black line showing the resistive component being dominant . It is considered a good thing in chokes for feedlines and may be also for your purposes because it does no go capacitive and provide a series resonant short  73, MB

PS: note that you must have the wire paired to prevent inductive flux from saturating the core. That just requires that tje wire be laid side by side. It can be taped together before winding to assure this. You can also wind the two windings of different color insulation one at a time and interlace them.MB

This is for an old Swan Mark 1 linear.  The filament choke is for the Center Tap or 3rd winding.  The other two windings are connected to the Ameritron filament choke.
https://www.rfparts.com/chokes-inductors/chokes-highvoltagerf/chokes-groundedgrid/filrfc30.html

I just tried the 5 turns on thew Torrid it it doesn't work.  Will try the 9 turns.

73,
Bill KO4NR

[ko4nrbs]

Thanks Bill....what mix are the beads you used?

Type 31 mix.
Bill

[K9MB]

Wow! Sorry, I just looked at the Swan schematic…..
I am not sure that you can just add an outboard separate winding for the center tap of that 10 volt winding….
Makes me suspect that the single winding will saturate the 31 core and tender it useless as soon as current flows. #31 is useless in magnetic flux windings… you need a bifilar winding to cancel the flux in the core…
I assume that the original trifilar wound choke burned up or opened?

I wonder if you have considered just adding a third winding to the Ameritron choke?
You would need to cut the shrink off and examine the bifilar winding and the carefully wing the third winding in the dent between the two windings already there so the third one is close and exactly the same direction and number of turns and then tape or shrink wrap it again….
Frankly, a better choice is to either buy or wind a 5vac center tapped winding  transformer outboard and then one bifilar choke would do as long as it can handle 30 amps.

Another solution is to stack ferrite beads on a fiberglass rod and wind a nrw trifilar winding on it that uses 10gauge magnet wire…

Hairy problem caused by that “clever” decision to use 10vac and series the tube filaments….😬


https://palomar-engineers.com/ferrite-products/ferrite-cores/ferrite-mix-selection

https://youtu.be/5iOkFIJcj2s

[ko4nrbs]

Wow! Sorry, I just looked at the Swan schematic…..
I am not sure that you can just add an outboard separate winding for the center tap of that 10 volt winding….
Makes me suspect that the single winding will saturate the 31 core and tender it useless as soon as current flows. #31 is useless in magnetic flux windings… you need a bifilar winding to cancel the flux in the core…
I assume that the original trifilar wound choke burned up or opened?

I wonder if you have considered just adding a third winding to the Ameritron choke?
You would need to cut the shrink off and examine the bifilar winding and the carefully wing the third winding in the dent between the two windings already there so the third one is close and exactly the same direction and number of turns and then tape or shrink wrap it again….
Frankly, a better choice is to either buy or wind a 5vac center tapped winding  transformer outboard and then one bifilar choke would do as long as it can handle 30 amps.

Another solution is to stack ferrite beads on a fiberglass rod and wind a nrw trifilar winding on it that uses 10gauge magnet wire…

Hairy problem caused by that “clever” decision to use 10vac and series the tube filaments….😬


https://palomar-engineers.com/ferrite-products/ferrite-cores/ferrite-mix-selection

https://youtu.be/5iOkFIJcj2s

I have been trying to add 160 meters to the Swan Mark 1.  Have had no luck at all trying different chokes on the 3rd winding.  I will try to wind the 3rd winding on the Ameritron choke and see how it goes.  That was how the original filament choke was wound.  The 3rd winding was 16AWG magnet wire.

Bill

[K9MB]

Wow! Sorry, I just looked at the Swan schematic…..
I am not sure that you can just add an outboard separate winding for the center tap of that 10 volt winding….
Makes me suspect that the single winding will saturate the 31 core and tender it useless as soon as current flows. #31 is useless in magnetic flux windings… you need a bifilar winding to cancel the flux in the core…
I assume that the original trifilar wound choke burned up or opened?

I wonder if you have considered just adding a third winding to the Ameritron choke?
You would need to cut the shrink off and examine the bifilar winding and the carefully wing the third winding in the dent between the two windings already there so the third one is close and exactly the same direction and number of turns and then tape or shrink wrap it again….
Frankly, a better choice is to either buy or wind a 5vac center tapped winding  transformer outboard and then one bifilar choke would do as long as it can handle 30 amps.

Another solution is to stack ferrite beads on a fiberglass rod and wind a nrw trifilar winding on it that uses 10gauge magnet wire…

Hairy problem caused by that “clever” decision to use 10vac and series the tube filaments….😬


https://palomar-engineers.com/ferrite-products/ferrite-cores/ferrite-mix-selection

https://youtu.be/5iOkFIJcj2s

I have been trying to add 160 meters to the Swan Mark 1.  Have had no luck at all trying different chokes on the 3rd winding.  I will try to wind the 3rd winding on the Ameritron choke and see how it goes.  That was how the original filament choke was wound.  The 3rd winding was 16AWG magnet wire.

Bill

Bill,
I think adding the winding to your Ameritron choke makes sense.
As I said, you may need to open the wrapper to see how the windings go and how many. If you can identify the pairs, you can simply wind your third winding in the notch between the pairs, so it looks like it was wound with the third winding between the two “hot” legs. After you wind it, some scotch #33 or shrink tube can restore it to glory. MB

[ko4nrbs]

Wow! Sorry, I just looked at the Swan schematic…..
I am not sure that you can just add an outboard separate winding for the center tap of that 10 volt winding….
Makes me suspect that the single winding will saturate the 31 core and tender it useless as soon as current flows. #31 is useless in magnetic flux windings… you need a bifilar winding to cancel the flux in the core…
I assume that the original trifilar wound choke burned up or opened?

I wonder if you have considered just adding a third winding to the Ameritron choke?
You would need to cut the shrink off and examine the bifilar winding and the carefully wing the third winding in the dent between the two windings already there so the third one is close and exactly the same direction and number of turns and then tape or shrink wrap it again….
Frankly, a better choice is to either buy or wind a 5vac center tapped winding  transformer outboard and then one bifilar choke would do as long as it can handle 30 amps.

Another solution is to stack ferrite beads on a fiberglass rod and wind a nrw trifilar winding on it that uses 10gauge magnet wire…

Hairy problem caused by that “clever” decision to use 10vac and series the tube filaments….😬


https://palomar-engineers.com/ferrite-products/ferrite-cores/ferrite-mix-selection

https://youtu.be/5iOkFIJcj2s

I have been trying to add 160 meters to the Swan Mark 1.  Have had no luck at all trying different chokes on the 3rd winding.  I will try to wind the 3rd winding on the Ameritron choke and see how it goes.  That was how the original filament choke was wound.  The 3rd winding was 16AWG magnet wire.

Bill

Bill,
I think adding the winding to your Ameritron choke makes sense.
As I said, you may need to open the wrapper to see how the windings go and how many. If you can identify the pairs, you can simply wind your third winding in the notch between the pairs, so it looks like it was wound with the third winding between the two “hot” legs. After you wind it, some scotch #33 or shrink tube can restore it to glory. MB

I measured the Ameritron filament choke while installed in the amp and found it is almost 19uh.  Accounting for strays that’s close to the advertised 15uh.
I took the heat shrink off and counted the bifilar turns and came up with 24.  If I wound the 3rd winding the entire length of the choke it would be 47 turns which is too much inductance since the bifilar turns are only 24.  I will wind a few more than 24 and measure the inductance adjusting turns as need.

On 160 I measured about 22uh of inductance in the tank circuit.  The G3SEK spreadsheet calls for 15uh.  I tried the recommended padding capacitors for the Load capacitor and found it was too much capacitance.  Using the VNA and a 1600-ohm resistor it tuned 160 pretty good with no padding for the Load capacitor and 150pf padder for the Tune capacitor.  Both capacitors are meshing with the Load capacitor a little over 25% or so.

Thanks for the recommendations.  Much appreciated.

73,
Bill KO4NR

[K9MB]

Wow! Sorry, I just looked at the Swan schematic…..
I am not sure that you can just add an outboard separate winding for the center tap of that 10 volt winding….
Makes me suspect that the single winding will saturate the 31 core and tender it useless as soon as current flows. #31 is useless in magnetic flux windings… you need a bifilar winding to cancel the flux in the core…
I assume that the original trifilar wound choke burned up or opened?

I wonder if you have considered just adding a third winding to the Ameritron choke?
You would need to cut the shrink off and examine the bifilar winding and the carefully wing the third winding in the dent between the two windings already there so the third one is close and exactly the same direction and number of turns and then tape or shrink wrap it again….
Frankly, a better choice is to either buy or wind a 5vac center tapped winding  transformer outboard and then one bifilar choke would do as long as it can handle 30 amps.

Another solution is to stack ferrite beads on a fiberglass rod and wind a nrw trifilar winding on it that uses 10gauge magnet wire…

Hairy problem caused by that “clever” decision to use 10vac and series the tube filaments….😬


https://palomar-engineers.com/ferrite-products/ferrite-cores/ferrite-mix-selection

https://youtu.be/5iOkFIJcj2s

I have been trying to add 160 meters to the Swan Mark 1.  Have had no luck at all trying different chokes on the 3rd winding.  I will try to wind the 3rd winding on the Ameritron choke and see how it goes.  That was how the original filament choke was wound.  The 3rd winding was 16AWG magnet wire.

Bill

Bill,
I think adding the winding to your Ameritron choke makes sense.
As I said, you may need to open the wrapper to see how the windings go and how many. If you can identify the pairs, you can simply wind your third winding in the notch between the pairs, so it looks like it was wound with the third winding between the two “hot” legs. After you wind it, some scotch #33 or shrink tube can restore it to glory. MB

I measured the Ameritron filament choke while installed in the amp and found it is almost 19uh.  Accounting for strays that’s close to the advertised 15uh.
I took the heat shrink off and counted the bifilar turns and came up with 24.  If I wound the 3rd winding the entire length of the choke it would be 47 turns which is too much inductance since the bifilar turns are only 24.  I will wind a few more than 24 and measure the inductance adjusting turns as need.

On 160 I measured about 22uh of inductance in the tank circuit.  The G3SEK spreadsheet calls for 15uh.  I tried the recommended padding capacitors for the Load capacitor and found it was too much capacitance.  Using the VNA and a 1600-ohm resistor it tuned 160 pretty good with no padding for the Load capacitor and 150pf padder for the Tune capacitor.  Both capacitors are meshing with the Load capacitor a little over 25% or so.

Thanks for the recommendations.  Much appreciated.

73,
Bill KO4NR

Bill,
Not sure why you need 47 turns.
You simply wind the third winding over the existing windings- so it will tske 24 windings to accomplish it. Between the original windings is an indentation,where the windings touch. Put your third winding in that indentation so it mirrors the other two windings turn for turn. (24 turns- just like the other two)
MB

[ko4nrbs]

Wow! Sorry, I just looked at the Swan schematic…..
I am not sure that you can just add an outboard separate winding for the center tap of that 10 volt winding….
Makes me suspect that the single winding will saturate the 31 core and tender it useless as soon as current flows. #31 is useless in magnetic flux windings… you need a bifilar winding to cancel the flux in the core…
I assume that the original trifilar wound choke burned up or opened?

I wonder if you have considered just adding a third winding to the Ameritron choke?
You would need to cut the shrink off and examine the bifilar winding and the carefully wing the third winding in the dent between the two windings already there so the third one is close and exactly the same direction and number of turns and then tape or shrink wrap it again….
Frankly, a better choice is to either buy or wind a 5vac center tapped winding  transformer outboard and then one bifilar choke would do as long as it can handle 30 amps.

Another solution is to stack ferrite beads on a fiberglass rod and wind a nrw trifilar winding on it that uses 10gauge magnet wire…

Hairy problem caused by that “clever” decision to use 10vac and series the tube filaments….😬


https://palomar-engineers.com/ferrite-products/ferrite-cores/ferrite-mix-selection

https://youtu.be/5iOkFIJcj2s

I have been trying to add 160 meters to the Swan Mark 1.  Have had no luck at all trying different chokes on the 3rd winding.  I will try to wind the 3rd winding on the Ameritron choke and see how it goes.  That was how the original filament choke was wound.  The 3rd winding was 16AWG magnet wire.

Bill

Bill,
I think adding the winding to your Ameritron choke makes sense.
As I said, you may need to open the wrapper to see how the windings go and how many. If you can identify the pairs, you can simply wind your third winding in the notch between the pairs, so it looks like it was wound with the third winding between the two “hot” legs. After you wind it, some scotch #33 or shrink tube can restore it to glory. MB

I measured the Ameritron filament choke while installed in the amp and found it is almost 19uh.  Accounting for strays that’s close to the advertised 15uh.
I took the heat shrink off and counted the bifilar turns and came up with 24.  If I wound the 3rd winding the entire length of the choke it would be 47 turns which is too much inductance since the bifilar turns are only 24.  I will wind a few more than 24 and measure the inductance adjusting turns as need.

On 160 I measured about 22uh of inductance in the tank circuit.  The G3SEK spreadsheet calls for 15uh.  I tried the recommended padding capacitors for the Load capacitor and found it was too much capacitance.  Using the VNA and a 1600-ohm resistor it tuned 160 pretty good with no padding for the Load capacitor and 150pf padder for the Tune capacitor.  Both capacitors are meshing with the Load capacitor a little over 25% or so.

Thanks for the recommendations.  Much appreciated.

73,
Bill KO4NR

Bill,
Not sure why you need 47 turns.
You simply wind the third winding over the existing windings- so it will tske 24 windings to accomplish it. Between the original windings is an indentation,where the windings touch. Put your third winding in that indentation so it mirrors the other two windings turn for turn. (24 turns- just like the other two)
MB

24 turns was my plan all along.  I just noticed that if I wound along the entire length of the existing Ameritron choke it would be 47 turns.  I knew that was way too much and always intended to use 24 turns.

After it is wound, I will measure the inductance and also test it on the bench.  This is my last resort, so I want to cover all the bases.

73,
Bill KO4NR

[K9MB]

Wow! Sorry, I just looked at the Swan schematic…..
I am not sure that you can just add an outboard separate winding for the center tap of that 10 volt winding….
Makes me suspect that the single winding will saturate the 31 core and tender it useless as soon as current flows. #31 is useless in magnetic flux windings… you need a bifilar winding to cancel the flux in the core…
I assume that the original trifilar wound choke burned up or opened?

I wonder if you have considered just adding a third winding to the Ameritron choke?
You would need to cut the shrink off and examine the bifilar winding and the carefully wing the third winding in the dent between the two windings already there so the third one is close and exactly the same direction and number of turns and then tape or shrink wrap it again….
Frankly, a better choice is to either buy or wind a 5vac center tapped winding  transformer outboard and then one bifilar choke would do as long as it can handle 30 amps.

Another solution is to stack ferrite beads on a fiberglass rod and wind a nrw trifilar winding on it that uses 10gauge magnet wire…

Hairy problem caused by that “clever” decision to use 10vac and series the tube filaments….😬


https://palomar-engineers.com/ferrite-products/ferrite-cores/ferrite-mix-selection

https://youtu.be/5iOkFIJcj2s

I have been trying to add 160 meters to the Swan Mark 1.  Have had no luck at all trying different chokes on the 3rd winding.  I will try to wind the 3rd winding on the Ameritron choke and see how it goes.  That was how the original filament choke was wound.  The 3rd winding was 16AWG magnet wire.

Bill

Bill,
I think adding the winding to your Ameritron choke makes sense.
As I said, you may need to open the wrapper to see how the windings go and how many. If you can identify the pairs, you can simply wind your third winding in the notch between the pairs, so it looks like it was wound with the third winding between the two “hot” legs. After you wind it, some scotch #33 or shrink tube can restore it to glory. MB

I measured the Ameritron filament choke while installed in the amp and found it is almost 19uh.  Accounting for strays that’s close to the advertised 15uh.
I took the heat shrink off and counted the bifilar turns and came up with 24.  If I wound the 3rd winding the entire length of the choke it would be 47 turns which is too much inductance since the bifilar turns are only 24.  I will wind a few more than 24 and measure the inductance adjusting turns as need.

On 160 I measured about 22uh of inductance in the tank circuit.  The G3SEK spreadsheet calls for 15uh.  I tried the recommended padding capacitors for the Load capacitor and found it was too much capacitance.  Using the VNA and a 1600-ohm resistor it tuned 160 pretty good with no padding for the Load capacitor and 150pf padder for the Tune capacitor.  Both capacitors are meshing with the Load capacitor a little over 25% or so.

Thanks for the recommendations.  Much appreciated.

73,
Bill KO4NR

Bill,
Not sure why you need 47 turns.
You simply wind the third winding over the existing windings- so it will tske 24 windings to accomplish it. Between the original windings is an indentation,where the windings touch. Put your third winding in that indentation so it mirrors the other two windings turn for turn. (24 turns- just like the other two)
MB

24 turns was my plan all along.  I just noticed that if I wound along the entire length of the existing Ameritron choke it would be 47 turns.  I knew that was way too much and always intended to use 24 turns.

After it is wound, I will measure the inductance and also test it on the bench.  This is my last resort, so I want to cover all the bases.

73,
Bill KO4NR

Well, you are counting the edges of both of the bifilar windings. That is not what you want. Just put 24 turns into the grove between those two windings and you have the trifilar winding required.

As for measurement, the entire choke design is marginal at best., IMO.
It is likely made on the old 5x rule, which is marginal, but might work.
A proper choke would have at least twice the inductance, IMO, but for this mod, just add the 24 turns between the pairs of 24 turns in the choke. 73, MB

[ko4nrbs]

Wow! Sorry, I just looked at the Swan schematic…..
I am not sure that you can just add an outboard separate winding for the center tap of that 10 volt winding….
Makes me suspect that the single winding will saturate the 31 core and tender it useless as soon as current flows. #31 is useless in magnetic flux windings… you need a bifilar winding to cancel the flux in the core…
I assume that the original trifilar wound choke burned up or opened?

I wonder if you have considered just adding a third winding to the Ameritron choke?
You would need to cut the shrink off and examine the bifilar winding and the carefully wing the third winding in the dent between the two windings already there so the third one is close and exactly the same direction and number of turns and then tape or shrink wrap it again….
Frankly, a better choice is to either buy or wind a 5vac center tapped winding  transformer outboard and then one bifilar choke would do as long as it can handle 30 amps.

Another solution is to stack ferrite beads on a fiberglass rod and wind a nrw trifilar winding on it that uses 10gauge magnet wire…

Hairy problem caused by that “clever” decision to use 10vac and series the tube filaments….😬


https://palomar-engineers.com/ferrite-products/ferrite-cores/ferrite-mix-selection

https://youtu.be/5iOkFIJcj2s

I have been trying to add 160 meters to the Swan Mark 1.  Have had no luck at all trying different chokes on the 3rd winding.  I will try to wind the 3rd winding on the Ameritron choke and see how it goes.  That was how the original filament choke was wound.  The 3rd winding was 16AWG magnet wire.

Bill

Bill,
I think adding the winding to your Ameritron choke makes sense.
As I said, you may need to open the wrapper to see how the windings go and how many. If you can identify the pairs, you can simply wind your third winding in the notch between the pairs, so it looks like it was wound with the third winding between the two “hot” legs. After you wind it, some scotch #33 or shrink tube can restore it to glory. MB

I measured the Ameritron filament choke while installed in the amp and found it is almost 19uh.  Accounting for strays that’s close to the advertised 15uh.
I took the heat shrink off and counted the bifilar turns and came up with 24.  If I wound the 3rd winding the entire length of the choke it would be 47 turns which is too much inductance since the bifilar turns are only 24.  I will wind a few more than 24 and measure the inductance adjusting turns as need.

On 160 I measured about 22uh of inductance in the tank circuit.  The G3SEK spreadsheet calls for 15uh.  I tried the recommended padding capacitors for the Load capacitor and found it was too much capacitance.  Using the VNA and a 1600-ohm resistor it tuned 160 pretty good with no padding for the Load capacitor and 150pf padder for the Tune capacitor.  Both capacitors are meshing with the Load capacitor a little over 25% or so.

Thanks for the recommendations.  Much appreciated.

73,
Bill KO4NR

Bill,
Not sure why you need 47 turns.
You simply wind the third winding over the existing windings- so it will tske 24 windings to accomplish it. Between the original windings is an indentation,where the windings touch. Put your third winding in that indentation so it mirrors the other two windings turn for turn. (24 turns- just like the other two)
MB

24 turns was my plan all along.  I just noticed that if I wound along the entire length of the existing Ameritron choke it would be 47 turns.  I knew that was way too much and always intended to use 24 turns.

After it is wound, I will measure the inductance and also test it on the bench.  This is my last resort, so I want to cover all the bases.

73,
Bill KO4NR

Well, you are counting the edges of both of the bifilar windings. That is not what you want. Just put 24 turns into the grove between those two windings and you have the trifilar winding required.

As for measurement, the entire choke design is marginal at best., IMO.
It is likely made on the old 5x rule, which is marginal, but might work.
A proper choke would have at least twice the inductance, IMO, but for this mod, just add the 24 turns between the pairs of 24 turns in the choke. 73, MB

I chose this choke because it has been used successfully in many SB220 160-meter modifications.  I counted the space between the two turns to find out how many turns Ameritron used to construct the choke.  My plan was to use an equal number of turns for the 3rd winding and then measure inductance plus test it.  Disconnected from the filament circuit and laying on my test bench the Ameritron choke has little over 17uh of inductance.  It is advertised as having 15uh of inductance.

Thanks for the recommendations!!

73,
Bill KO4NR