Co-phasing?

[Marconi]

Once you get both sides of that harness to a 50 ohm match, and as long as you keep to half wavelength (accounting for velocity vactor) multiples of feed line beyond that point, it doesn't matter what the impedance of the feedline is. 50 ohms, 75 ohms, or even 100 ohm feedline should all work equally well...


The DB

DB, does this mean a 1.0:1, 1.5:1, and 2.0:1 SWR at the radio is safe to run?

 

[Shockwave]

DB, does this mean a 1.0:1, 1.5:1, and 2.0:1 SWR at the radio is safe to run?

No, it means that half wavelength cable always reflects the load impedance back to the source and not the cables impedance.

 

[Marconi]

No, it means that half wavelength cable always reflects the load impedance back to the source and not the cables impedance.

Does that happen when the load is not purely resistive too?

 

[Shockwave]

Does that happen when the load is not purely resistive too?

Yes, everything from a short to an open will accurately be reflected through a 1/2 wavelength multiple of cable. Less cable loss of course.

 

[The DB]

Shockwave got in there first, but yes he is right on the nose on what I was referring to Marconi.

DB, does this mean a 1.0:1, 1.5:1, and 2.0:1 SWR at the radio is safe to run?

While this statement is also true it is not what I was talking about.


The DB

 

[ghz24]

Not really Co-phase more out of phase question

I'm considering doing this as the models look so promising.
Doc said/asked

Now, how would you go about making the phasing harness that would change the radiation direction 90 degrees? Make it just a tad easier/harder and only mess with changing one side of that harness so you can make the whole mess 'switchable'!

Use a relay and an extra 1/4 wavelength section of 75 ohm coax so that one antenna is fed through a 1/4 wavelength section and the other through two 1/4 wavelength sections (1/2 total) with the relay controlling which antenna gets the extra 90 degree delay? Oh and cut off a little of both the whips equally to re-tune for best SWR ?

Well Doc (and others) is that right?
Problems I've over looked?
What problems are caused by other non-1/4 wavelength and unequal lengths (like a 135 degree delay instead of 90 )?

 

[W5LZ]

What problems would you run into? Remember that any two antenna phased array as you started with is bi-directional. So changing the phase of those two antennas, one on a 1/4 wave line, the other on a 2 x 1/4 wave line would change the radiation pattern by 90 degrees. That radiation pattern is NOT 'sharp' by any means, so, making that 90 degree shift means that you would cover the whole 360 degrees, right?
How about some miscellaneous phase shift? It would sure change the radiation pattern to something not very predictable. Or if you can make that phase shift using some 'device' variable, you could sort of point the pattern in particular directions.
One problem you will have with a 'fixed' phase shift set up is making those phasing line an electrical 1/4 wave length. The published velocity factors for all coax are 'characteristic', not exact. That means that the direction of the pattern lobes are also going to be 'un-exact', unless you measure the VF of the feed line used. Almost all phased arrays are larger than a single antenna, so, hope you have the room for it. The separate antennas used for this array should be made as 'closely' tuned, or 'alike' as possible. There will still be some variations in the resulting radiation pattern, but they will tend to be predictable. And who knows what other 'catches' there are?
No 'real-life' antenna system is going to exactly match a modeling program's results. There are just too many variables that aren't accounted for. If it get's 'close', figure you got a winner.
- 'Doc

 

[dennis4040]

Thanks for all the input everyone.
It sounds like I would have to spend some extra time checking the tune if I want to use other than 1/2 WL of 50 ohm to make up the difference. I have read that there is less loss using 50 ohm to make up the difference rather than using all 75 ohm 3/4WL because of less mismatched coax. I do not have experience at this but have spent countless hours researching the topic on dozens of forums. This is the first forum I have posted a question on, there is a lot of great help on here. There is an LMR 240 75 Ohm and it is .240 dia. with a .032 inner conductor which should fit nicely into the PL 259 after folding.
The tip regarding "Mutual Coupling" is good to know, I will keep that in mind.
Thanks again, I will be back when I have more detailed questions as I get into this project. I might take me a bit.
dennis

The Co-Phase harness you need to stack the antennas at 34 feet apart is no different than what you would use at closer spacing. All you have to do is add equal 50 ohm lines off the 75 ohm harness to reach the pair of antennas. Once you pass through the 1/4 wave of 75 ohm coax lines, you're back to 50 ohms. It doesn't matter if the antennas are plugged directly into the harness here or pass through any random equal lengths of 50 ohm coax first.

Another problem people usually face when phasing antennas together is called "Mutual Coupling". The effects of mutual coupling cause the resonant frequency of antenna to lower when combined together. For example, on the FM broadcast band, a dipole tuned to 98.1 Mhz. as a single antenna will usually drop its resonant frequency down to 97.9 Mhz when combined with second dipole that was tuned at 98.1 Mhz. by itself.

To compensate for this we simply tune the individual antennas to be resonant 200 Khz. higher than the desired frequency. The offset in frequency will probably be a little different on 11 meters with two yagi's but the effect and direction of frequency shift will be the same. For maximum performance it's important that we correct this problem by tuning the antennas and not mistake this as needing to compensate by adjusting the length of the phasing harness.

 

[W5LZ]

"All you have to do is add equal 50 ohm lines off the 75 ohm harness to reach the pair of antennas. Once you pass through the 1/4 wave of 75 ohm coax lines, you're back to 50 ohms."
Sorry, not hardly. There are conditions that were left out of that.
- 'Doc

 

[ghz24]

The model I was looking at is 2 vertical dipoles separated by 7 ft. I wouldn't call them bidirectional with gain of 6.75 dBi and f/b ratio of ~5 (beam width depends on phase delay).
This model doesn't perform well at all co-phased (SWR >3)
Had to trim the elements to ~101 inches to get SWR ~1.4

That radiation pattern is NOT 'sharp' by any means, so, making that 90 degree shift means that you would cover the whole 360 degrees, right?

No not really 360 coverage. Best gain has a beam width of 120 degrees, which antenna receives the delayed signal changes which end the array end-fires from.
So I get 120 degree beam north or south but very little east and west without physically turning the array.
Changing the phase delay by + or - 5 degrees changes the pattern very little.
So I'm thinking errors in VF will be tolerable.

Or if you can make that phase shift using some 'device' variable

Could you elaborate or point to an example

 

[W5LZ]

ghz24,
None of this is 'new' by any means, it's been published quite a few times. One source is the ARRL HandBook, William Orr's HandBook is another good source. I would suggest you take a look at either (Orr's is better in my opinion).
There is no 'front to back' for most simple phased arrays ('dual antennas'). They are bi-directional with little difference between 'front' and 'back'.
- 'Doc

 

[dennis4040]

Tuning CP harness with MFJ 259B

Marconi and others,
I have read and understand the MFJ 259B manual regarding tuning stubs and transmission lines. I do not understand ho to apply this to tune the CPH using 2 dummy loads.
How do I tune each side?
Do I have to solder Pl259s on each end then cut them off and re solder after each measurement?
Can't get the picture in my head of how this is done.
Thanks, dennis

[Marconi, The open ends of this harness were also stubed-out and attached to the antennas and tested for resonance. These ends were made a bit long and then trimmed down to center frequency for CB. Each end ended up being about 6' feet long per side and would easily reach across the back end of a P/U truck.[/QUOTE]

 

[Marconi]

Marconi and others,
I have read and understand the MFJ 259B manual regarding tuning stubs and transmission lines. I do not understand ho to apply this to tune the CPH using 2 dummy loads.
How do I tune each side?
Do I have to solder Pl259s on each end then cut them off and re solder after each measurement?
Can't get the picture in my head of how this is done.
Thanks, dennis

The open ends of this harness were also stubbed-out and attached to the antennas and tested for resonance. These ends were made a bit long and then trimmed down to center frequency for CB. Each end ended up being about 6' feet long per side and would easily reach across the back end of a P/U truck.

Dennis the dummy load idea was suggested as a learning tool. You have to have two good non-reactive dummy loads that produce very similar results at your frequency of choice for this idea to work. All the dummy load does is replace the antennas with a near perfect match. This is so you can test the harness with a known good load, and see the positive results as you trim it to the proper length. Using a dummy load allows you to get the harness close to the tuned length.

However, your antennas individually will not show a perfect match. So, your harness will still need to be trimmed and tuned for the antennas you'll use. The term "trim" used here assumes the harness was originally cut a little too long of course. We used stub ends on the coax so we could trim. This was done so we didn't have to deal with the connectors as we tuned the harness. We used SWR meters, but you can use an analyzer too and it may be more accurate.

These stub type connections works just as good or better than 259/239 connectors...as long as the stub ends or kept as short as possible.

Since these joints are exposed, you also need to make these 3 connections water tight.

Personally I think a well mounted single antenna can perform just as well.

 

[jazzsinger]

I don't mean to butt in here with something s small bit off subject, but I didn't want to start a new thread just for this, and this may be for co-phasing of some sort idk. but what is it for?'

View attachment 5809

LMAO, going by the spacing it can be for only one thing, Booty Monster cophasing microwave ovens for Brontoburgers at cookouts.
op:

 

[Shockwave]

SW may be right, but I've never seen it done that way, with both 75 ohm and 50 ohm mixed to make up the added length for the harness itself. From the harness back to the radio with 50 ohms at any reasonable length is OK however.

I would think that if what SW says is possible and you can get a good tune...then the 50 ohm line extension used to make the 30'+ gap would have to also be a tuned length 1/2 wave line, and not just any old length. Here is a YouTube video I did after talking to SW about the subject of co-phasing two antennas some time back. In this video I cophase two of my base antennas, using their feed lines and a T connector. This setup showed >2.0:1 SWR on my radio and the radio's cut-back was working, so I switched on my auto-tuner to take care of the mismatch. So, we see that SW is right, we can use a random length for our harness, and we can even use 50 ohm coax, without a phasing harness, but we won't get a workable tune.

Marconi Testing Co-Phased Base Antennas Video 1 - YouTube

I've never extended a real CP harness myself, but I know adding random feed line to any antenna setup can skew the results in many ways...if the feed point match (load) is not very close to perfect.

All one has to do to check these ideas out, is make the setups and check the results out...at the termination point on the radio end of the harness connected to the antennas, and to two accurate 50 ohm resistive dummy loads.

If the dummy loads are accurate I would think they should permit a good match with a random length 50 ohm extension to the harness.

Since the antenna loads are often not perfect, such an addition of a random length 50 ohm line to the harness, could skew the results badly, so any such setup as a CP harness needs testing.

SW, I just didn't think that Dennis was up to checking out such a big setup without first getting familiar with a mobile CP harness, but you are right, I think tuned 1/2 wave 50 ohm jumpers could be added to the harness to extend the length, so it would reach in his case.

I was suggesting he use a tuned 1/2 wave length of 75 ohm instead to make up the space, and if he did this setup needs checking out.

So Dennis, take your pick.

Once you get to the end of the pair of 75 ohm cables, the impedance match is done. At this point the only important thing is to keep the phase matched. The only requirement for this is that the 50 ohm extension cables must be equal length. The actual length matters none as long as they are equal and in phase.

Another option has already been pointed out and that is to extend the 75 ohm cable to 3/4 wavelength. Eliminating two connections in the harness. In this case the actual length of cable used is very important. This is because if you want the extended 75 ohm cable to reflect the 50 ohm impedance, you must add an exact 1/2 wavelength of 75 ohm cable to the existing harness.

The tradeoff here is that to extend the harness with 75 ohm cable requires you get two more lengths of cable precisely tuned at an electrical 1/2 wavelength. Errors here will reduce performance just as easily as adding two more splices in the cable to extend with 50 ohm cable of equal but random lengths.