Torroids for cmc

[redbeast]

Hooked up new qt-80 in the truck, and have been fighting noise problems. First Alt whine, which I figure will go away when I install filter, and second, receive noise on almost all channels which I figure is CMC. So to start, I inspected and cleaned feed line and antenna connections, all were good, if nothing else more a maintenance check. Swr and ability to tune is not a problem. Have been doing my research, especially KOBG, and got a ferrite torroids coil and managed to get like 6 turns through with my existing rg8 feed line; to get any more is gonna require cutting off the pl259 from existing feed and redoing it. The recommendations I've been seeing have been 7-9 turns. Rx has definitely cleaned up with the six turns, so my question is, how critical is this. Should I cut it out and make it for eight? And also, should it be close to antenna point, or is a convenient spot inside the truck going to work? Not trying to be lazy, just trying to save my old bones and back!! Thanks guys!

 

[redbeast]

Hooked up new qt-80 in the truck, and have been fighting noise problems. First Alt whine, which I figure will go away when I install filter, and second, receive noise on almost all channels which I figure is CMC. So to start, I inspected and cleaned feed line and antenna connections, all were good, if nothing else more a maintenance check. Swr and ability to tune is not a problem. Have been doing my research, especially KOBG, and got a ferrite torroids coil and managed to get like 6 turns through with my existing rg8 feed line; to get any more is gonna require cutting off the pl259 from existing feed and redoing it. The recommendations I've been seeing have been 7-9 turns. Rx has definitely cleaned up with the six turns, so my question is, how critical is this. Should I cut it out and make it for eight? And also, should it be close to antenna point, or is a convenient spot inside the truck going to work? Not trying to be lazy, just trying to save my old bones and back!! Thanks guys!

Yeah, never mind. I think senility is setting in.

 

[Crawdad]

I'm curious as to what size toroid you wound with RG-8??

 

[redbeast]

Hooked up new qt-80 in the truck, and have been fighting noise problems. First Alt whine, which I figure will go away when I install filter, and second, receive noise on almost all channels which I figure is CMC. So to start, I inspected and cleaned feed line and antenna connections, all were good, if nothing else more a maintenance check. Swr and ability to tune is not a problem. Have been doing my research, especially KOBG, and got a ferrite torroids coil and managed to get like 6 turns through with my existing rg8 feed line; to get any more is gonna require cutting off the pl259 from existing feed and redoing it. The recommendations I've been seeing have been 7-9 turns. Rx has definitely cleaned up with the six turns, so my question is, how critical is this. Should I cut it out and make it for eight? And also, should it be close to antenna point, or is a convenient spot inside the truck going to work? Not trying to be lazy, just trying to save my old bones and back!! Thanks guys!

Yeah, never mind. I think senility is setting in.

I'm curious as to what size toroid you wound with RG-8??

It was a 240-31, 1 1/4 " id. I have ordered a bigger one, trying to get 8 turns with least amount of trouble. I'll usually smaller one for different application

 

[loosecannon]

save yourself a world of headaches and get a set of these:

Ferrite Snap On RF Choke/Isolator - 1/4"/6mm cable, RG-58/59/06/LMR240 - 3-300 M | eBay

However, it is strongly suggested that they be covered to protect the plastic covers from weathering – use electrical tape for best results or heat shrink to ensure longer life. – use electrical tape for best results or heat shrink to ensure longer life.

www.ebay.com

no need to remove any connections or coil any cable.

LC

 

[Dr_DX]

Here is something I put together a while back for my HL2 Discord server:

 

[brandon7861]

My 10 cents on the topic...

Torroids and snap-on beads can do two things, they can increase the inductive reactance of the common-mode path, or they can increase the resistance of it (often both).

Depending on the length of the common-mode path, it can be capacitive or inductive (just like an antenna). If it is capacitive, adding inductance can increase common-mode current since the added inductance can cancel the already-existing capacitive reactance. One should ensure the path is not capacitive if adding inductance, or at least be aware of that implication. It don't always work against you, but it can.

Adding resistance (a core that is lossy at RF) works better since resistance does not cancel any of the reactance of the common-mode path. A reflection from a resistance is just as good as a reflection from inductance, so its best to choose ferrites that are lossy at the frequency you are using.

Ugly baluns, the coiled up coax, is probably the best route if made properly. The reason is that the inductance of the turns is in parallel with the capacitance between them, and that coil of coax can create a parallel LC blocking your TX from seeing the rest of the coax, so whether the rest of the CM path is inductive or capacitive doesn't matter.

Probably the best solution is to use a VNA or dip meter to wind the coil so you know what the resonance point of the balun is. Charts get you close, but they require the coax be the same as what the chart was made with. A thicker jacket, smaller diameter braid or different materials, for example, all change the inter-winding capacitance and therefore changes resonance point.

If you have a nanoVNA, make a fixture that connects the grounds of port 0 and port 1, and then run the centers of both to alligator clips that go to the shield so you can measure the S21 of the common-mode path and watch the log mag as you add turns looking for the resonance dip.

From my experience, the more ferrites you add, the less the effect of each additional one, so IMO, there is a point where adding more isn't justified. I could be wrong, your mileage may vary.

 

[redbeast]

My 10 cents on the topic...

Torroids and snap-on beads can do two things, they can increase the inductive reactance of the common-mode path, or they can increase the resistance of it (often both).

Depending on the length of the common-mode path, it can be capacitive or inductive (just like an antenna). If it is capacitive, adding inductance can increase common-mode current since the added inductance can cancel the already-existing capacitive reactance. One should ensure the path is not capacitive if adding inductance, or at least be aware of that implication. It don't always work against you, but it can.

Adding resistance (a core that is lossy at RF) works better since resistance does not cancel any of the reactance of the common-mode path. A reflection from a resistance is just as good as a reflection from inductance, so its best to choose ferrites that are lossy at the frequency you are using.

Ugly baluns, the coiled up coax, is probably the best route if made properly. The reason is that the inductance of the turns is in parallel with the capacitance between them, and that coil of coax can create a parallel LC blocking your TX from seeing the rest of the coax, so whether the rest of the CM path is inductive or capacitive doesn't matter.

Probably the best solution is to use a VNA or dip meter to wind the coil so you know what the resonance point of the balun is. Charts get you close, but they require the coax be the same as what the chart was made with. A thicker jacket, smaller diameter braid or different materials, for example, all change the inter-winding capacitance and therefore changes resonance point.

If you have a nanoVNA, make a fixture that connects the grounds of port 0 and port 1, and then run the centers of both to alligator clips that go to the shield so you can measure the S21 of the common-mode path and watch the log mag as you add turns looking for the resonance dip.

From my experience, the more ferrites you add, the less the effect of each additional one, so IMO, there is a point where adding more isn't justified. I could be wrong, your mileage may vary.

Funny, being an old guy, and having run old cobras for years, kinda finding out that new radios are a lot more sensitive to CMC, etc. the snl and noise blanket works ok, but never seemed to have this problem with old board cbs......price for modern technology?? Anyway, got another coil, and reworked for eight wraps, seems to be much better, just waiting for the power line filter to show up. Almost have to say that the rigexpert is showing a little bit wider swr and resonance....so I'll update.

 

[loosecannon]

There is a very famous chart that became popular quite a few years ago that people like to use for making CMC's by winding coax.

its the one with all the green, red, and yellow. the guy who made it had a foreign call sign, something like GTX... or something like that.

the thing about this chart that everyone misses is that he only made it in order to show that air core chokes were inferior to those that used ferrite.

I haven't been able to find any of his notes in quite a while, but he made posts telling people that they were misusing his chart and that his point was to show people that ferrite is the way to go when dealing with CM currents.

Im certainly no expert in this but i have personally cured both mobile and base station RF problems by using five or six snap on beads of the right mix at the feedpoint of the antenna.

LC

 

[Crawdad]

There is a very famous chart that became popular quite a few years ago that people like to use for making CMC's by winding coax.

its the one with all the green, red, and yellow. the guy who made it had a foreign call sign, something like GTX... or something like that.

the thing about this chart that everyone misses is that he only made it in order to show that air core chokes were inferior to those that used ferrite.

I haven't been able to find any of his notes in quite a while, but he made posts telling people that they were misusing his chart and that his point was to show people that ferrite is the way to go when dealing with CM currents.

Im certainly no expert in this but i have personally cured both mobile and base station RF problems by using five or six snap on beads of the right mix at the feedpoint of the antenna.

LC

Choke charts:
https://web.archive.org/web/20180422195858/http://www.karinya.net/g3txq/chokes/

G3TXQ home page:

https://web.archive.org/web/20180424075003/http://karinya.net/g3txq/

7 3

 

[redbeast]

Choke charts:
https://web.archive.org/web/20180422195858/http://www.karinya.net/g3txq/chokes/

G3TXQ home page:

https://web.archive.org/web/20180424075003/http://karinya.net/g3txq/

7 3

I read K0BG's info on CMC, basically uses the same chart, so pretty much used that as a frame of reference.....

 

[19wrc333]

There is a very famous chart that became popular quite a few years ago that people like to use for making CMC's by winding coax.

its the one with all the green, red, and yellow. the guy who made it had a foreign call sign, something like GTX... or something like that.

the thing about this chart that everyone misses is that he only made it in order to show that air core chokes were inferior to those that used ferrite.

I haven't been able to find any of his notes in quite a while, but he made posts telling people that they were misusing his chart and that his point was to show people that ferrite is the way to go when dealing with CM currents.

Im certainly no expert in this but i have personally cured both mobile and base station RF problems by using five or six snap on beads of the right mix at the feedpoint of the antenna.

LC

Hi,

That made me do a bit of "google dorking" and search for Steve's (G3TXQ) original quotes. Google groups appears to have archived some of it.

They were interesting, let me share a few of those original statements by linking to the forum threads.

https://groups.io/g/hexbeam/topic/common_mode_chokes/41433972

https://groups.io/g/hexbeam/topic/41434806

and a few direct quotes on air-based chokes vs. ferrite toroid:

https://groups.io/g/hexbeam/message/9605

https://groups.io/g/hexbeam/message/8519


--
Skyline

 

[Crawdad]

Hi,

That made me do a bit of "google dorking" and search for Steve's (G3TXQ) original quotes. Google groups appears to have archived some of it.

They were interesting, let me share a few of those original statements by linking to the forum threads.

https://groups.io/g/hexbeam/topic/common_mode_chokes/41433972

https://groups.io/g/hexbeam/topic/41434806

and a few direct quotes on air-based chokes vs. ferrite toroid:

https://groups.io/g/hexbeam/message/9605

https://groups.io/g/hexbeam/message/8519


--
Skyline

Thanks for those links.

Here's an important quote from G3TXQ (SK):
"The difference between a coiled-coax choke and a well-designed ferrite-cored choke is that the former acts like a high-Q inductor: it will be narrowband; it will be influenced by metalwork around it; and it may actually increase braid current if the impedance of the common-mode path is capacitive. On the other hand a ferrite-cored choke acts more like a resistor: it will always reduce braid current and it is relatively immune to its environment."

Seems plain that Ugly Baluns could, under the correct circumstances, cause more problems than they cure. Torroid or bead chokes are more likely to help prevent CMC without being reactive, while being inexpensive and simple to make. Looks like a no-brainer.

7 3