160M Inverted L Matching Question

[k3msb]

I constructing a box to hold the matching "network" for a new 160M Inverted L.

I'm running RG-213 to the box which will contain a commercial multi-tap UNUN (I'm using this as I have one on hand but not a suitable air variable cap.....).   The UNUN has SO-239 connectors.

I was going to put an SO-239 feed through on the box.  My question is this:  From the interior end of the feed through (SO-239) to the SO-239 on the UNUN, does this short (6 inches....) piece need to be a full RG-213 jumper, or can I just use the center conductor and be done with it?

Thanks

[WQ9E]

Is your unun going to be happy with the voltage it sees at that point?  A very easy way to match an inverted L is to cut it a bit long so it is inductively reactive and then use series capacitance which does NOT need to be in the form of a variable.  I feed my Hy Gain HT as a modified inverted L on 160 and use capacitors made from RG-11U switched by a BC-375 tap switch.  6 positions provide good coverage of the entire band and these caps stand up to contest duty with 1,500 watts output without complaint.  The far end should have the shield cut back about 1 inch and then taped.  My only planned mod is to use a motorized switch in the future.

If you go with the unun this short length can just be an unshielded jumper.

[W3GMS]

The short wire should be fine.  No issue with that.  You did not say if your inverted L is voltage or current fed based on its length.  I think I would use just a L matching network.  1 cap and 1 Inductor and that's it.  Many years ago I had a 1/4 inverted L up and worked lots of DX on Top Band with it.  Of coarse a current fed wire needs a good low loss radial system.  My next one will be 3/8 wave in length.

Joe, W3GMS

[k3msb]

Gentleman

Thank you for your comments. As I said I am using the UNUN because I do not have the other parts available (although I obviously could wind a coil.)

The only part that I have available is a 200 pf air variable but from what I have read I need something in the range of 300-400 pf. 

So, I'm making do with what I have on hand.

[K1JJ]

Hi Mark -

A few additional comments:

Whenever I need to run an unshielded 50 ohm wire for RF, I try to space it 1/4" away and parallel to the chassis. This simulates the impedance of a low impedance transmission line so there will be less impedance bump.  A short, isolated wire will make no difference on 160M, but on the higher bands it may mean a 1: 1.2 swr improvement.

As for matching, low to high impedance - in case you come up with some suitable caps - .... put up the L and then go out there with some variable capacitors, a coil and clip leads.  Using an MFJ-259, (or whatever) experimentally try an L network, a series L/C and a parallel config to match the L and you will surely find a good combination that gives a 1:1 match.  Try tapping the inv L off the coil with various cap configs, as well as tapping the coax off the coil bottom at low impedance.      Once configured with clip leads, build it in the box.

This way if the inverted L changes in the environment over time, you can tweak it. When the MFJ-259 is matched 1:1 at the input feedpoint junction   of the inv L, you know the match back to the shack will also be 1:1 and will show up that way at the transmitter.

T

[Jim, W5JO]

Mark if you can make the entire length as much as 165 ft. then do as Pete suggests.  Put the variable in series with the feed from the center conductor to the antenna.  If you lack enough capacitance with the variable, add some fixed to make the 300 or 400 pf needed, then adjust the fixed as necessary to reach a good match.  Attach the shield to your ground/radial system and have a ball.

I am using a quarter length piece of wire here for 1.885 and don't have any capacitance in series.  It reflects some power but works fine fed direct.  The proof in the pudding will be the ground system/ radials.

[k3msb]

Well I need just under 128 feet for the length and I have 140 feet on hand.   

I'll give the 200 pf series cap a try with the antenna about 10 feet too long and see how it goes.

[W4EWH]

My question is this:  From the interior end of the feed through (SO-239) to the SO-239 on the UNUN, does this short (6 inches....) piece need to be a full RG-213 jumper, or can I just use the center conductor and be done with it?

I'd use some RG-213. I've never been comfortable with purely mechanical connections for RF, and that's what running a straight wire would do: force the "ground" connection to go through two (or more) mechanical shield connetions.

If I had a choice, I'd substitute a moisture barrier for the feed-through, and run the input cable directly to the UNUN.

My $0.02. YMMV.

Bill, W1AC

[sndtubes]

Just FYI I have a similar antenna fed eith an unun.  My unun has nutiple taps.  As I added radials, I had to change the taps a couple of times.  Bandwidth will narrow as you add radials.  Mine is at 1.2:1 at resonance using the UnUn.  This is a better DX antenna thsn a local NVIS antenna.

[W0BTU]

This might give you an idea or two: http://www.w0btu.com/160_meters.html#inv-l_antenna

[k3msb]

Well the Inverted L is up.  The vertical part is around 50 feet; that's the best I'm gonna get with the trees I have. 

I decided to try the series capacitor as the antenna is a bit long - it's 139 feet and it should be 138 feet.

The 200 pf matching cap is fully meshed for an SWR of 2.5:1 at 2.2 MHz.  The analyzer shows a resistance of 90 ohms.   No radials yet just a ground rod - I have the wire but ran out if daylight!

Next steps:  Add a 200 pf fixed doorknob cap to give me between 250 and 400 pf capacitance and of course add radials.

[k3msb]

That should read  "it's 139 feet and it should be 128 feet".

[k3msb]

Hi guys. 

Well,I  have the gearbox torn off the SX-28A (whew…..) and decided to switch over to play with the inverted L a bit.   I was looking for answers, but found more questions!

I typically use my MJF-259 analyzer (SWR and R), but I borrowed a RixExpert AA-170 from a friend of mine last night.  Sweet device!  I wonder if Santa stocks them?  Heck the AA-30 would be FB OM for me!

I'm running 150 feet of RG-213 form the Inverted L to my shack.   In the shack this morning I'm seeing an R of 90.4 ohms and an X of 212 ohms.

At the antenna matching box last weekend I saw about 90 ohms (was using an my MFJ-257 so had no way to see the X).

So, for an Inductive X of 212 ohms, that translates into a needed C in the box (at 1830 KHz) of

212= 1/(2*pi*f*c) =>  c =  1/(2*pi*212) => 410 pF in series.

My concern is that I already have 200 pF in series, so I need a total of 600 pF. 

That seems excessive expecially since my antenna is only about 10 feet over the required 234/f value. 

Is it excessive?

As a further test (just for fun) I added a 2 foot RG-213 jumper between my feedline and the AA-170.  Now I get an R of 62 ohms and an X of 178 ohms.

I don't understand why the R value differs so much with just a 2 foot jumper.

I'll lay out the radials this weekend.  From studying the books, I expect that to change R, but not change X very much.  Correct?

Thanks for your help guys!

[k3msb]

Made a mistake in my post;  that 2 foot jumper is RG-58, not RG-213.

[WQ9E]

Mark,

You need to make your measurement at the feedpoint of the L which is the point where the series capacitance will be added.   It sounds like you made your measurement this morning through a long run of coax which will not provide the measurement you need.  With the series capacitance it should provide a near perfect match at the point where the RG-213 attaches to your series capacitor.

Keep in mind that if you have extra coax and you need a high power series capacitor you can use the coax.   The center conductor is one side of the cap, the shield is the other.  Start with excess and trim to what is needed.  Provide some linear space between the shield and center conductor at both ends to avoid arcing at high power.

[KA0HCP]

I don't understand why the R value differs so much with just a 2 foot jumper.

Probably body capacitance effect.  With no radials and only a ground rod (which is really a poor RF ground) your body proximity is probably having significant effect on the readings. 

Try clipping on even one or two wires as counterpoises temporarily and see how the readings change.

[k3msb]

Keep in mind that if you have extra coax and you need a high power series capacitor you can use the coax.   The center conductor is one side of the cap, the shield is the other.  Start with excess and trim to what is needed.  Provide some linear space between the shield and center conductor at both ends to avoid arcing at high power.

Hi Roger;

I'm not sure what you mean by that.  I don't have very much extra RG-213 in my 150 foot run.

When I enter the box, the center conductor goes to one end of the cap, and the antenna to the other;  the coax shield doesn't touch the cap (because the cap is in series....).    I'm obviously not understanding something that you meant.

[WQ9E]

Mark,

Where is the physical point where you are making the measurement of R and X?  You should do this at the feed point and not at the station end of your RG-213 coax run.  You want to use enough series C at the feed point to create a match before attaching the feeedline back to your station.

My reference to the coax shield is if you need an additional series capacitor then you can easily make a high power capacitor out of scrap pieces of coax.  For example RG-213 has a capacitance of around 30 pf per foot between the center conductor and shield so a 7 foot section could be used as a 210 pf capacitor by connecting the shield as one plate of the capacitor and the center conductor as the other "plate".

[k3msb]

Gotcha Roger!

This morning I was measuring inside the shack.   Last Saturday I was measuring at the matching box that contained the series C -- which I read 90 ohms R and an off the scale SWR with 200 pF in series.

[WU2D]

I have a 135 ft inverted L fed from my bulkhead where I have a motorized cap L match tuner with a tapped switch for the band. It works well from 80M to 10M. I could not get it to tune 160 with the maximum coil and I lacked enough C as well. Now I could have added turns and C but I tried something simple. I grounded the bottom of the coil so I formed a parallel tuned circuit and grid dipped it connected to the antenna. It resonated on 160M with about 200 pF. Then I tapped up the coil till I hit a happy match to the coax. Finally I tapped down the coil on the antenna side and bingo - perfect match. This much like an UNUN in effect and it gives some harmonic rejection and noise reduction and selectivity on receive as well. 

[k3msb]

Hello Lads.

Well, the Lord provided some decent antenna weather on Saturday morning so my wife helped me run radials for the Inverted-L.  I now have 7 radials each over 100 feet long, unfortunately not equally spaced around the antenna, but at least they're out there.

Before I adjusted the series C at the antenna base I did some measurements in the shack.  These were at the end of 150 feet of RG-213 and without any C/L (ie, direct connection):

1.820 MHz:

SWR   3.4:1.  RL = 5.3 dB     R = 26.3 ohms.  X = 40.2 ohms.

The antenna is resonant at 1.680 Mhz

SWR  1.05:1   RL = 32.5 dB,  R = 48.8 ohms   X = -2.0  ohms.

I was alarmed at the Return Loss (RL) value being high at resonance, but after doing some reading learned that's what it's supposed to do.

So, on Wednesday I'm going to lower the antenna and chop off about 8 feet or so (incrementally) to bring it in line with the 234/f value.

At the antenna base (with no matching, all measurements at 1.820 Mhz):

SWR 3.7  RL = 4.9 dB  R = 63.3 X = 77.4

To counteract 77.4 of inductive reactance I need to add the same in series capacitance.  At 1.820 Mhz that would take 1135 pF.  I managed to scrounge up 600 pF.  After adding the 600 pF I get

SWR 4.6  RL = 3.8 dB  R = 45.8 X = -80.2

Huh?   I expected to still see a positive X as I have not yet counteracted all the inductance!!

My Heathkit SA-2060 handles it well, but I don't understand why X has gone negative.

Inside the shack (with 600 pF series C at the antenna base and after my 150 feet of RG-213):

SWR = 3.8  RL = 4.7 dB   R = 167.3   X -56.5

Why is R now 167 ohms?   

The antenna seems to work well.  I worked into the Caribbean and South America without too much problems on 100W.  But, the math doesn't work out and I'd like to know why.

[AB2EZ]

Mark

Capacitive reactance is inversely proportional to C.

At any frequency, if C is lower (e.g. 600pF instead of 1135pf), then the capactitive reactance will be higher (e.g. -j146 ohms instead of -j77 ohms)

At any given frequency, a specific length of coaxial cable will transform the impedance looking into the load at one end (in this case your inverted L plus the added capacitor in series) into a different impedance looking into the cable at the other end.

The formulas that are used to calculate the transformation are somewhat complicated (look up "Smith chart") ... and you should expect that both R and X, looking into the input end of the coax, will change... if you change the reactance, looking into the load, at the output end of the coax.

Stu

[k3msb]

Hi Stu

I am using 600 pF (where about 1130 pF is needed to counteract +j77.4 @ 1.820 MHz).

600 pF at 1.820 Mhz will be a capacitive reactance of –j146

+j77.4 – j146 = -j68 (which is ballpark to what I am seeing).

So, for inserting series capacitance, I need to start HIGHER than what I need.   Half isn't going to work due to the inverse relationship.   I knew about the inverse relationship, but didn't realize the impact on this physical application. 

When I prune my antenna on Wednesday, if I want to make it a bit longer so I just need a series C to compensate,  I should ensure that the reactive component is between (about)  +j150 and +j1619   so that I can tune it out with my available capacitance:

600 pf       -j145   200  doorknob + 200 doorknob + air var
400           -j218    200  doorknob + air var
200            -j437    Air var fully meshed
54           -j1619   Air var fully unmeshed

However, since I'm now at +j77, shortening my antenna will cause me to approach j0, and that will require more capacitance than I have available!   Thinking about this, as the antenna shortens to the correct length, the inductive reactance approaches 0, so the required compensating capacitance approaches infinity.  Is that right?

In summary, I can't do a series match with the caps I have on hand!   My best option is to prune close to 234/f as planned (and recommended by a few others).

[AB2EZ]

Mark

Zero capacitive reactance corresponds to a straight through wire (no series capacitor)

Stu

[WQ9E]

Mark,

I went back through the notes I made when I put up my inverted L, it is part of my HyGain 18HT vertical and the L transitions from vertical to horizontal at the 45' level via an added 40 meter trap (to avoid detuning the basic 18HT system) and a metal stub for additional vertical height.  The horizontal portion is an additional 115' out to the tower that supports my 4 element quad.

The maximum series capacitance I needed at any point to get down to 1.5 to 1 across the 160 meter band is around 320 pf.   As I noted earlier I am using capacitors made from lengths of RG-11U which easily handles 1500 watts all modes in contest duty used in this fashion.

Ground system consists of 60 1/4 wave buried radials.

I don't think I took any close-up photos of the matching system and since it is 29 degrees with a 45 MPH wind and blowing snow I won't be heading out to take one anytime soon.  The attached photo shows the 18HT in the foreground which is where the L originates and it runs out to the 55G tower that holds my quad and terminates at a sidearm at the rotor level.  The antennas sit in the pasture near our second garage and is fed via 150 feet of coax from the contest station which I built into a finished section of the barn that sits just north of the pasture.