Marconi does his version of the FireStik (FS)

[Marconi]

The PDF file below is my idea for the 72" inch FS that I titled as "Marconi 5 x 102" Helix FS a." This is a model set in Free Space. This model required 443 individual short wires to build the Helix.

I have requested more accurate dimensions for the wire on this antenna, so this model is subject to some change. This version turns out to be longer than the actual antenna, but during my several iterations for this model I was able to get good results at just about any overall lengths, both longer and shorter than 72" inches.

In the images below I also compared this antenna to my Marconi 5x which uses 5 x 102" ss whips, looks similar to the Starduster, and uses 5 whips, 1 as the radiator and 4 as slanted down radials. I also show several different antenna views to help show the antennas detail for the Helix coil used.

The models Average Gain Test shows to be near perfect. The pattern shown here includes both patterns as an overlay for comparison. It shows these two different designed antennas showing very similar results and patterns in Free Space.

The antenna shows a very good SWR curve indicating a 2.50 MHz <2.00:1 SWR bandwidth. I'm not sure about the real antennas actual working BW, but this model surprised me on all counts...and it is very easy to tune using the tip element length if necessary. I think this good tune ability is confirmed by operators.

 

[The DB]

When I look at actual pictures of the firestik, there are three different levels of winding. The windings at the bottom are spread out much further than the windings in the middle, which are further apart again than the windings at the top of the antenna, which are adjacent. The antennas in this image have five different separations of windings, however, these aren't the seven foot version, however they are the six foot version so they are likely close...

An interesting thing to note, the blue one has a different winding pattern than the others...

Something else that I find interesting is how many windings they actually use on these antennas. Maybe it is just me, but it looks like more than I would have expected they would need at first glance.

Anyway, this makes it harder to make a model of this type of antenna as you need to have at least three different coil layouts throughout the antenna length, and possibly as many as five.

Firestik claims these are top loaded antennas, but the reality is they are actually somewhere between your continuous loaded model and a top loaded antenna. That is just one more thing on the Firestik site that isn't technically correct.

Another thing to consider, at least with 4Nec2, a model like this would require more wires than I can make out of a given element (copper) as I can only put said type of conductor on 30 something wires (I think it was). I would be completely unable to put a rubber covering over said wires, much less be able to factor in the fiberglass in the middle of the coil. These little effects add up, just a thought.

Another concern is the segment size of the individual wires that make up the coil. On an antenna like this these would be much smaller than what we would normally expect to see with most antennas. The more wires used to make up each coil will only make this potential problem worse. How many wires did you use per coil? It is possible you may have to add more segments to the radials you used to compensate.

Not being able to factor in these details, as well as potentially others I have yet to consider (as I have not yet made the model), I fear any gain reports from this type of model will be somewhat optimistic at best.

Eddie, have you by chance checked to see how much of the gain that is shown is horizontally vs how much is vertically polarized with your model? That might be interesting to consider. The coil of wires will act like an inductor, but the wires in your model are also not exactly vertical.

I am looking at making such an antenna model and putting it on my Ford Explorer item I made a while back, but I am concerned about the segment count of the overall model. The ford Explorer item has over 2000 segments to begin with, and takes much longer to run than most models we are used to. This type of antenna will only make that problem worse.

That being said, what I am considering doing is dividing the antenna up into five equal parts, and to start with not using coils, but instead inductor obhects that are also evenly spread out as well. These would be tuned to match the antenna as well as I can on the object. I could then replace each individual inductance section with a coil. It may not match the Firestik design exactly, but it will give an idea of what such an antenna design is capable of comparatively speaking when mounted on a vehicle.

Unfortunately this will take a lot of work, and I am not going to have the time to do anything like this for quite a while.


The DB

 

[Marconi]

Thanks DB. I posted somewhere in this thread this model is not perse like the FS 7' footer.

You can see on the Eznec Main Screen view I posted first that I used 443 wires and 458 segments of my 500 limit. Of course if this model is made per the specs...I would not be able to use Eznec with a 500 segment limit.

I might argue with some of your other comments, because I was surprised to see the model even come close, much less be almost identical to my standard 1/4 wave Marconi using 102" elements throughout. When I built this model I have no idea what to expect of my model, but I knew better than to even suggest the Top Loaded FireStik 7' footer...was just window dressing.

Without really knowing for sure, this is what I would expect from such a model, and in my real world experiences this antenna performed very well. This is why I decided to do the model...per the thread above.

Some posters were tending toward making the usual naysayer type claims, and I got curious as to what Eznec might show us.

Here is a small sample of the first few wire of my coil from wire #10 - 457 for the total overall length at 84.5" inches. You will see the current distribution is very stead and evenly distributed...so I think Eznec is working pretty well at this point and the AGT shows = 1.006 = 0.3 db, and if I fiddled with it a bit more I could probably see 1 = 0.00 db.



Here we see a very steady decrease in currents as we get near the tip we see nearly a little less than <zero.

Each coil consists of 3 wires with 1 segment each, and I used most of the rest of my limit on the radials and 2.5" inch wire at the base. I didn't use all the segments, because the best AGT shows less that 500 is most desirable, and I still may be a bit short or over getting to a perfect 1.00 = 0.00 db.

On the issue of the model being arguably in error...I would make that conclusion too, if the model produced results too low or too high. But, that is not what we see here.

So IMO, if one tried to build this antenna using the dimensions for this model...and within reason...I would expect the model to preform about like what we see here.

I also recall that Roy told us in the manual to watch carefully for the current and phase distributions to be irregular, and then question that Eznec was having problems.

Of course in the process of building this one I had Geometry and Segment issues, but they all went away with a little tweaking of the coil diameter and getting the wire length ratios right. I built the first 10 coils by hand, and then used a generator feature in the Eznec Geometry editor to build the rest using the copy coils feature to stack them one upon another.

I have not yet set the model over real Earth, but the FS model shows little horizontal at -59 db of gain. I'll post the model over real Earth later...but I can't be sure how good it will look yet. It may be off the wall with so few segments per wire.

Thanks for adding your comments.

 

[Marconi]

Here is the overllay model over Average Earth at 36' feet compared to my Marconi 5x SD'r type design at 36' feet. There is not much difference.

In the past when I tested my real Marconi with different types of loaded 1/4 wave elements for the radiator and the 4 slanted down 102" whips as radials...I saw little to no difference in on-air performance.



Here is the Helix model with both Vertical and Horizontal turned on and I see just a touch of Horizontal in the pattern.



Here is the Eznec main screen view for the Helix over real Earth at 36' feet.



BTW, I forgot to add a mast to this model over real Earth...that might make a little difference.

 

[Marconi]

An interesting thing to note, the blue one has a different winding pattern than the others...

In the course of doing this model I generated maybe 10 different coil configurations DB, and believe it or not most would produce a perfect AGT with gain, angle, pattern, pretty close to the results shown for the model I posted. The only difference was the overall length I ended up with.

I was shooting for 72" inches, but never quite got there.

So, I'm not surprised that, in this regard, we see different loaded design ideas working about the same. I've heard a 5/8 wave is a 5/8 wave, but IMO a 1/4 wave is a 1/4 wave too. I generated one coil configuration that came out a little more than 60" inches overall, and it too worked just about as good as my 84.5" incher above. I didn't save it though.

Something else that I find interesting is how many windings they actually use on these antennas. Maybe it is just me, but it looks like more than I would have expected they would need at first glance.

I just added coils until I got a good tune, then I added a short straight vertical wire on top to adjust...very similar to what some of these type antennas do with a tuning screw setup.

Anyway, this makes it harder to make a model of this type of antenna as you need to have at least three different coil layouts throughout the antenna length, and possibly as many as five.

It would be more difficult and I can't do it because I'm working with wire lengths of less than 1" now, and my limit of 500 segments just won't stretch. Getting smaller wire spacing will likely produce geometry problems I think.

Another concern is the segment size of the individual wires that make up the coil. On an antenna like this these would be much smaller than what we would normally expect to see with most antennas. The more wires used to make up each coil will only make this potential problem worse. How many wires did you use per coil? It is possible you may have to add more segments to the radials you used to compensate.

As long as Eznec does not give me critical error reports, I don't stop optimizing the old fashion way...one step at a time. I don't get a step by step record like 4Nec2 makes when optimizing, but I have pencil an paper.

I used 443 x 1 segment per wire with .188" spacing between coils = about 82" inches. I think I tried to cut this in half and ran into geometry and segment problems. I want to go back and cut the .188" to maybe .16" inch and see it I can get the model closer to 72" inches overall.

I am looking at making such an antenna model and putting it on my Ford Explorer item I made a while back, but I am concerned about the segment count of the overall model. The ford Explorer item has over 2000 segments to begin with, and takes much longer to run than most models we are used to. This type of antenna will only make that problem worse.

DB, that will be a challenge. I don't have my mobile model on my new computer yet, so I can't fiddle with mobile ideas. However, I've found that when I place a 1/4 wave radiator on top of my car mobile the antenna has to be shorter than normal. Do you find the same to be true. I think I recall M0GVZ, Conner telling me he finds this true in his static mobile work too. Maybe he sees this when he mounts close to the ground vs. on top of his mobile.

Even if this is true and it allows me more segments to work with, my model only has about 100 segments left...so a simple1/4 wave radiator is about all I can do with my version of Eznec. You will have to be the one that 'splains this dilemma.

That being said, what I am considering doing is dividing the antenna up into five equal parts, and to start with not using coils, but instead inductor obhects that are also evenly spread out as well. These would be tuned to match the antenna as well as I can on the object. I could then replace each individual inductance section with a coil. It may not match the Firestik design exactly, but it will give an idea of what such an antenna design is capable of comparatively speaking when mounted on a vehicle.

I don't have a clue about inductor aspects with modeling, but that might work too. Sounds plausible.

Unfortunately this will take a lot of work, and I am not going to have the time to do anything like this for quite a while.

Well, that is a revolting development. If you don't do it quick...ole Grampa forgets stuff.

I have not yet set the model over real Earth, but the FS model shows little horizontal at -59 db of gain. I'll post the model over real Earth later...but I can't be sure how good it will look yet. It may be off the wall with so few segments per wire.

I've made this model over real Earth, but I don't have enough segments left for the wires that should have more segments than 1. This limits the model from responding to is maximum performance, but it is still close enough for government work.

 

[The DB]

Unfortunately this will take a lot of work, and I am not going to have the time to do anything like this for quite a while.

Well, that is a revolting development. If you don't do it quick...ole Grampa forgets stuff.

Can't be helped, leaving today to go backpacking on part of the Appalation Trail in Georgia with a group. Will be back sometime next week.


The DB

 

[Marconi]

Can't be helped, leaving today to go backpacking on part of the Appalation Trail in Georgia with a group. Will be back sometime next week.


The DB

Shoot, I thought you could get right on it, Never Mind!

Be careful and have fun.