DIY fan dipole vs ST-2

[nanZor]

Huh, I never looked at the canonical ST-2 until today. Some great threads about it here. Gave me a brainstorm.

To put it simply, it is a very nice fan-dipole with a 4:1 (300-75 ohm) balun/transformer. Maybe that's too simple, but that balun caught my eye.

What other simple broadband diy antenna uses a 4:1 transformer? Yeah, the OCFD. But this isn't about that - it has been discussed forever.

But where does the OCFD go wrong where the ST-2 fan dipole goes right?

1) The elements are fed in the center, rather than offset, thus the balun can actually do a bit of common-mode rejection. The total imbalance of the ocfd means that the balun is overwhelmed, and cannot do much common-mode filtering. In fact, that is what makes it work on low-band - using the coax braid as much of the antenna.

2) On a related note, the fan-dipole being balanced, also means that the 4:1 balun can do a more accurate job of impedance step-down, whereas the massively offset ocfd means a LOT of reactance that really skews transformer impedance efficiency.

3) Mostly important is the fact that the fan dipole has varying element lengths, which means that you have a much better chance of having a decent radiation / reception angle, vs the ocfd's "vertical long wire" effect that basically points the lobes straight up above vhf.

4) Broad-band is helped for pure non-resonance with the 4:1 transformer, where impedances can go very high. Both the fan-dipole and ocfd share this feature, BUT again, the varying element lengths of the fan-dipole make it a winner.

So how about testing a simple fan dipole with 3 elements (6 total for dipole configuration) of varying lengths and a 4:1 transformer balun? Could it be this simple?

I'll have to think it over for what element lengths to use. I'd probably shoot for simple "halving" of lengths, like 36 / 18 / 9 inches to start and play with that. Hmmmm.. it all depends. Note that I'm NOT trying to shoot for a 50-75 resonant impedance at the center of the bands I'm interested in, as the 4:1 transformer would take that down to 12.5 ohms for example. I'm purposely cutting things off-tuned for a higher overall impedance, and hoping the lobes don't get too wacky.

So instead of trying to duplicate an ST-2, how about a *custom* fan dipole? As we've seen with the ocfd, mere low impedance matching does not a good antenna make when the radiation angle is straight up.

For an rx-ony setup, perhaps a slightly lossy imperfect 4;1 tv-type transformer might be the easiest practical solution, rather than trying to make it all perfect as if I'm trying to make it transmit-worthy -- ie radiation angle is more important than trying to achieve a 1:1 swr with lobes pointing straight up!

Hmm - have to think before I cut. Opinions please - good or bad!

 

[chief21]

Assuming that your target is VHF/UHF, would your fan dipole be vertically polarized?

John

 

[KC1UA]

This is a neat idea and something I had not considered before. I may try to do this myself. It would indeed be vertically polarized. I recently built a 40/20/10 meter fan dipole for amateur radio and it works quite nicely. Building one for receive purposes to cover VHF/UHF/800 shouldn't be difficult or take up a great deal of space.

 

[KC1UA]

Based on the way I built my HF fan dipole, why wouldn't this work? It's obviously not drawn to scale, but this would conceivably be a 4 band "scanner fan dipole". You could cut for as many or as few bands as you wish to listen to. The idea is based on the 1/2 wave dipole formula of 468/freq = overall length. Then half of that is each dipole leg. For example:

468/155.000 = 3.019 feet (rounded) overall length, half on each side would be just a bit above 1.5 feet per leg.

The antenna could be hung vertically as shown with some non-conductive material. Simple copper antenna wire and ceramic or plastic insulators would be used. The center piece of PVC would have copper plumbing strapping (thin strip with holes in it) run along each side of the INSIDE of the PVC, directly across from each other. Short eye hooks with nuts would go through the PVC and be secured on the inside with nuts and lock washers. The transformer would be secured in a similar manner but probably just locked down with nuts and bolts. Antenna wires connected to the eye hooks at one end and the insulators on the other, and the insulators connected to thinner PVC pieces at top and bottom by some decent antenna rope.

This is nothing new for antenna design, for all I know it's been posted here before, but I believe it would create several half wave resonant dipoles (which of course require no ground plane to work) and feed them with one common 75 ohm coax cable. Just hang it and see what happens. I'm going to give it a try at some point in the near future. Should be a cheap build with parts that I by and large already own.

 

[n9mxq]

Posting to hear results. But can't find any reason this wouldn't work.. you have piqued my curiosity..

Sent from my SM-G950U using Tapatalk

 

[popnokick]

A few questions that I couldn't answer from the diagram -
- How long is the PVC pipe containing the copper strap that appears to be the common "bus" connector for each dipole?
- Won't the length of the copper bus from each of the vertical dipoles need to considered in the resonant lengths?
- Won't the proximity of the two copper straps inside the PVC have a capacitive reactance and velocity factor... similar to what you have with ladder line? If so these factors may serve to detune (or enhance) the antenna elements.
- Similarly, the length of proximity of the elements to each other may produce interactions much as you'd have with a beam antenna and driven / reflector elements.
- Why not bring each dipole element's center to a common point right at the 300 to 75 ohm transformer, rather than have to deal with all of the above resulting from the copper straps inside the PVC? (And yes, this would make it sort of like the ST-2, but also much more like a regular fan dipole).
- Have you tried modeling this first with 4NEC2 or EZNEC?

 

[prcguy]

In this case EZNEC is your friend and will find answers to all your questions.
prcguy

Huh, I never looked at the canonical ST-2 until today. Some great threads about it here. Gave me a brainstorm.

To put it simply, it is a very nice fan-dipole with a 4:1 (300-75 ohm) balun/transformer. Maybe that's too simple, but that balun caught my eye.

What other simple broadband diy antenna uses a 4:1 transformer? Yeah, the OCFD. But this isn't about that - it has been discussed forever.

But where does the OCFD go wrong where the ST-2 fan dipole goes right?

1) The elements are fed in the center, rather than offset, thus the balun can actually do a bit of common-mode rejection. The total imbalance of the ocfd means that the balun is overwhelmed, and cannot do much common-mode filtering. In fact, that is what makes it work on low-band - using the coax braid as much of the antenna.

2) On a related note, the fan-dipole being balanced, also means that the 4:1 balun can do a more accurate job of impedance step-down, whereas the massively offset ocfd means a LOT of reactance that really skews transformer impedance efficiency.

3) Mostly important is the fact that the fan dipole has varying element lengths, which means that you have a much better chance of having a decent radiation / reception angle, vs the ocfd's "vertical long wire" effect that basically points the lobes straight up above vhf.

4) Broad-band is helped for pure non-resonance with the 4:1 transformer, where impedances can go very high. Both the fan-dipole and ocfd share this feature, BUT again, the varying element lengths of the fan-dipole make it a winner.

So how about testing a simple fan dipole with 3 elements (6 total for dipole configuration) of varying lengths and a 4:1 transformer balun? Could it be this simple?

I'll have to think it over for what element lengths to use. I'd probably shoot for simple "halving" of lengths, like 36 / 18 / 9 inches to start and play with that. Hmmmm.. it all depends. Note that I'm NOT trying to shoot for a 50-75 resonant impedance at the center of the bands I'm interested in, as the 4:1 transformer would take that down to 12.5 ohms for example. I'm purposely cutting things off-tuned for a higher overall impedance, and hoping the lobes don't get too wacky.

So instead of trying to duplicate an ST-2, how about a *custom* fan dipole? As we've seen with the ocfd, mere low impedance matching does not a good antenna make when the radiation angle is straight up.

For an rx-ony setup, perhaps a slightly lossy imperfect 4;1 tv-type transformer might be the easiest practical solution, rather than trying to make it all perfect as if I'm trying to make it transmit-worthy -- ie radiation angle is more important than trying to achieve a 1:1 swr with lobes pointing straight up!

Hmm - have to think before I cut. Opinions please - good or bad!

 

[KC1UA]

Good questions all.

On the HF fan dipole I built, I used a 4" interior diameter piece of schedule 40 PVC. A bit heavy but the way I have it mounted is not an issue. This was per the specs of many designs I found, many of which I combined to create what I built.

Regarding the copper bus, you're right, and with the HF fan it was necessary to shorten each dipole a bit so I suppose it would be ideal to have a MFJ-269 or equivalent antenna analyzer. But, given that you're not transmitting I'd say it's less critical and if properly thought out the antenna could be wrapped back on itself, shortening the length if need be, and the antenna rope could be lengthened, with proper prior planning.

The proximity of the copper strapping on my HF dipole is to my knowledge not an issue as you suggest.

As far as the spacing between elements? Well, look at the ST-2, which is the original poster's reason for an alternative. They're pretty close together and I have yet to find a scanner antenna that outperforms it. The Omni-X is close but I feel the ST2 is superior. That's a debate for another thread, perhaps.

Bringing the dipole elements to a common point is a dandy idea, and maybe far better than mine. I'm just going on what worked for me in the ham bands, as well as a more solid method of support for the elements and the feed point themselves.

I have not tried modeling this with any software, I drew up that diagram on a whim because I like the idea. Many that are far more schooled than I am at this stuff will no doubt come up with better ideas.

 

[lmrtek]

The Channel Master monitenna 5094 was a better design