What was Freecel telling us about cancelling reactance?

[Marconi]

Freecell posted a comment in the following thread on WWRF a while back.

http://forum.worldwidedx.com/viewtopic.php?t=9596&start=0

He said:

now the load impedance at the transmitter is exactly 50 ohms. we do however have an inductive value of reactance on the order of approximately +j18.935 ohms, preventing a more normalized input resistance but 46.276 ohms ain't that bad. the remaining inductive reactance can be cancelled easily using an open length of the same feedline in conjunction with a t-connector, the feedline input and the transmitter output to exactly cancel the remaining inductive reactance and restoring a purely resistive, non-reactive match between the input of the feedline and the transmitter. approximately 9 inches of RG-8X from T-plug to open end almost perfectly cancels the inductive reactance present at the transmitter.

I have been unable to catch up with him to ask him to describe this function a little better, indicating to us how it is applied to a mobile antenna in order for cancel the remaining reactance like he suggests.

Read it over and if anyone else has a clue what he is talking about, then please explain. I have tried what I believe is the t-connector scheme, but I cannot seem to get to first base.

 

[Guest]

Sounds to me like he's using a coaxial stub as a capacitor.

Radio into the "T", coaxial stub on one end (open) , and feedline to the antenna to the opposite side of the "T".

 

[Marconi]

Well 181, that is the way I see it too. By adding a little cap to the inductive reactance you get rid of the reactance.

He says by using a 9" open piece of coax. I assume he means that 9" is what it takes to fix that particular condition at the feed point that he reported on first testing the antenna. And that he found that condition on his analyzer with nothing in the third open stubing port. Then he tested a length of coax in that port and started triming until the analyzer showed to be 50 ohms resistive and 0 reactance, right.

Then when he replaces the analyzer with the Transmitter he has a perfect match. Is that the way you see this too?

I am not getting any success doing that, but I will try again.

 

[W5LZ]

...Not sure about the 9 inches, but haven't tried figuring the capacitance for that length either. Capacitance per length is usually one of the things listed in a coax table/chart. If you know what capacitance (ball park'ish) is needed, you can get an idea of how long the stub has to be. "Check and chop" is always a good way of finding it though...
- 'Doc

 

[Marconi]

Well guys do you think this thing will work, if one finds the right length? Will it fix the reactance or will it in the process of tuning the stub mess up the resistance?

When I looked at a line the other day that was considerably longer than 9", as I chopped, the X value did get better. But at the same time when I looked back at the resistance it started to change also. So even if I could trim the stub to show the X=0, the value for R would also be changing on me. So I ended up with X=0, OK, but the resistance goes to pot.

Now I know that the resistance in the load did not really change, but we are transforming something here, are we not? So no telling what is really going on. Maybe as the stub can transform the reactance to 0, but maybe it also transforms the resistance to another value that will be no good for us. If so, then this thing can't do what we want, unless I am going about it wrong.

What do you guys think?

 

[Guest]

I haven't tried it, but I tend to believe it works. At work we have pretty large "cans" to notch out adjacent channel on our low-band fire bases. We tune them to notch what we want, and then fine tune them the same way. A coax stub. Although they're a lot longer than 9 inches. Just about 3 feet as a matter of fact.

He didn't say what type of coax he's using, the Capacitance/ft or the impedence. Not all coax has a PERFECT 50 ohm impedence. I would tend to think those specs would be vital.

Although I would tend to think 52 ohms, 70% VF, and 29 pF/ft is a good generic starting point.

 

[Marconi]

Yes I was hopeful that we could attach this to a reactive load with an analyzer and trim the stub, and maybe affect both cap and inductance by triming length to see X=0 on the analyzer, then hook that puppy up to a transmitter and she would see a near perfect match, but as I trimmed for reactance, the resistive value went away. I tried shorting the stub if it was open and open it if it was shorted, and all it did was go back the other way with the reading on the meter.

Maybe cap will be handled with a shorted stub and maybe inductance will be handled with an open line, not sure though. That is why I have been trying to get a hold of FC and see what he could add. But no luck. Hope he is OK.

Thanks for you help Steve.

 

[W5LZ]

...shorted or open stub, can be either capacitive or inductive, depends on length (also capacitive ~and~ inductive). Just to make it a little bit more complicated - lol.

Yes, stubs do work. Too many of them being used in lots of different applications for it not to work, right? (and not all of them are coax)
- 'Doc