AF9Y's 5 Foot prototype. (Just the first half of the 10 foot final design)
IV3WLQ's 4 bay Helix Array based on the New 10 Foot Helix Design
I've attempted to optimize a low cost Helix antenna design suitable for detecting the Mars Observer. Here are the characteristics and modeling results:
Reflector Screen: NONE - A single 10.66" loop reflector is used which
eliminates the cumbersome reflector screen
Boom: 10 Foot length of 4" dia, thin wall PVC (normally used for tile). Weight is less than 5 Lbs and the cost is $3.50 (at Builder's Square)
Helix: 23 Turns of 1/4" refrigerator copper tubing spaced 5.125" per turn. 55 Feet of tubing required at an approximate cost of $20.
This design has a two step taper for the Helix Dia:
Reflector Loop Dia: 10.66" The reflector is positioned 0.375" ahead of the start point of the first helix turn.
First 4 Turns: 10.32"
Next 7 Turns: 8.68"
Last 12 Turns: 8.40"
The above display is the single Helix plot at 14 segments per half wave. (Note that the gain is dBdc, add 2.1 dB to convert to dBic.)
The above display shows the pattern of a stacked pair of these antennas.
The 58 inch spacing appears to be close to optimum. (Note that the gain is dBdc, add 2.1 dB to convert to dBic.) The stacking gain was 3.0 dB.
Based on this model, a four bay array would have 22 dBic gain.
For a 5 Foot array, spacing should be 42" resulting in a gain of 19.3 dBic
I would be happy to model any other helix designs if you email me the dimensions. I'll give any standard reflector screen designs the benefit of a perfect reflector surface. (very time consuming to model a real screen)
For those with AO6-PRO or NEC modeling access, here's the antenna file for the helix:
The first prototype of the new helix design has been completed. A 5' version was built to verify the construction techniques and for verification against the model. The most encouraging result was that the input showed the expected input impedance of 50 ohms. The 5' antenna was tested at Central States in July 96. The predicted gain from AO was 11.2 dBd and the measured gain was 12.9 dBd (after 3dB horz to circular correction).
Pictures are not yet developed for scanning but here are a few tips on building the antenna:
Mark the copper tubing for each helix half turn mounting point as follows:
First 8 marks: 16.375" spacing
Next 14 marks: 13.875" spacing
Final 24 marks: 13.500" spacing
Use a 5/8" hole saw, not a drill bit, for cutting the holes in the 4" PVC tubing. A drill bit tends to chip the tubing producing a ragged hole.
Modify a 4" PVC coupler (about 4.5" dia) to use as a guide for drilling the boom. Cut out the ridge so that the coupler can slide along the boom. Carefully measure and drill two sets of small holes spaced 2.5625" on opposite sides of the coupler. The coupler can then be used as a sliding guide for drilling the boom.
1/2" PVC tubing glued into the 4" boom serves as supports for each half turn of the helix. A 13/64" hole is drilled 11/32" from the top of the tubing. The 1/4" copper tubing is secured to the support with a 3/16" cable tie. A slight crimping of the copper tubing at the mounting point will keep the helix turn in place
Here's a design of a more typical Helix antenna. Note the lower gain:
The screen above is the single Helix plot at 10 segments per wire.
The screen above is the two stack Helix plot at 10 segments per halfwave.
The 58 inch spacing appears to be close to optimum. The stacking gain was 2.9 dB.
Based on this model, a four bay array would have 20.4 dBic gain.
The following text on the new Helix construction is from personal email and the Mars-Net group mailings.
Date sent: Wed, 2 Oct 1996 23:01:00 -0700 To: email@example.com From: "L. J. Shepherd"
Subject: RE: Mixed results on Helix Builds Forwarded to Mars-Net this evening as you suggested.....JIM. At 03:34 AM 10/3/96 +0000, you wrote: >This is an excellent thread of information which I would like to >post in a section of my webpage. Do either of you object? Also, I >believe it would be beneficial for the Mars-Net group to see all of >this dialogue instead of just the pieces posted so far. Would one of >you consider forwarding it to the Mars-Net group? > >Thanks > > > > >> Date: Tue, 1 Oct 1996 21:41:28 -0700 >> To: "Ronald E. Lile" >> From: "L. J. Shepherd" >> Subject: RE: Mixed results on Helix Builds >> Cc: firstname.lastname@example.org > >> Hi Ron: >> Thanks very much for your quick reply.........I think I understand >> the details of your matching adjustment. It appears to be similar in effect >> to the "tab matching" used by others with solid plane reflectors in that the >> change in impedance comes about by adjusting the proximity of the beginning >> of the helix to the reflector itself. As you may have seen, in the latter >> version, the helix is wound right down to a connector mounted on the solid >> metal reflector and then a small rectangular metal tab is soldered to the >> helical element close to the connector. The impedance is adjusted by bending >> the small rectangular tab toward or away from the reflector until the >> desired match is obtained. >> In effect, it sounds like you're doing essentially the same >> thing.....adjusting the spacing between the helix (near to where it attaches >> to the connector) and the reflector. I did try orienting the helix in a >> somewhat similar fashion, but not with the close spacing that was successful >> for you. As I may have mentioned, I did try lightly touching both the helix >> and reflector simultaneously, with my hand, in the same area and noted a >> significant drop in reflected power. Your success suggests that I should >> have paid more attention to that observation ! >> In any event, we will try your technique as soon as time permits and >> will report the results back to you. By the way, we also intend to use the >> attenuator subsitution method you referenced for our relative gain >> comparisons.........we have a number of circularly polarized satellite >> antennas to use for comparison. >> Thanks again for sharing the details.........JIM. >> >> >> At 07:53 PM 10/1/96 -0500, you wrote: >> >hi Jim, (one of these days I'm going to change the formal signature to Ron, >> which I like better!) >> > >> >Anyway, Visualize an N connector mounted so that the flange is just to the >> left of the support for the reflector at say the bottom of the antenna. I >> used two supports for the reflector opposed by 180 degrees. The N connector >> is mounted so that the pin points to the boom (towards the top of the >> antenna as positioned), i.e., parallel to the support. The pin will now be >> about 5/8 inch from the support and just inside the edge of the support. >> > >> >The reflector plane is the same as the plane of the supports for the >> reflector. >> > >> >The fed element circles around from the top from a spacing of 1/2 turn. >> This forming the beginning of the helix. The element goes through the >> reflector loop passes and is fastened to the reflector support. The element >> is bent to pass the support and meet the connector pin. The end of the >> element is flattened and soldered to the connector pin. >> > >> >The element at the connector will be approximately 3/8 inch center to >> center with the reflector element at this point. I then positioned the >> element parallel to the reflector element keeping the spacing as the element >> passed through the plane of the reflector and at about 6.5 inches began to >> increase the spacing rather quickly away from the reflector so that the true >> pitch of helix could begin. I will need to wait till tomorrow to get on the >> roof to gather more detailed dimensions for you. I did find that adjusting >> the spacing where the true pitch began was critical to acheiving a low VSWR. >> Once I got the hang of what needed to be done it was quick to do and I >> can't tell you now if all are exactly the same. By eye they are very close, >> however. >> > >> >To your second question -- I did not move the reflector at all but kept it >> perpendicular to the boom. All adjustments where completely done with the >> driven element. >> > >> >As to gain, relative to an 18C -- I have never used a Yagi on 435 except a >> link antenna of 7 elements. The two antennas are at about the same height >> and the helix appears to have more gain by about 8-9 dB. This measured with >> the attenuator insertion method as I don't trust "S" meters. It has alot >> more gain than the Lindenblad I had been using for the PACSATs (relative >> value not measured yet). I am laying plans for better gain measurements. >> Experience from other helixes with the traditional ground plane would place >> these with in acouple of dB of calculations, but I must point out that a >> couple of dB less than calculations will mean I won't hear the Mars spacecraft. >> > >> >Don't know if drawing helps, but when related to text, maybe you can see >> what I did. >> > >> >Additional information about the Helix: >> > >> >1) did not use the 4 inch pipe as Mike, but instead used 1.5 inch line. >> >2) element supports are 1/2 inch line (7/8 OD) >> >3) used the copper refrig tubing and prebent to dimensions >> >4) placed "T" in the balance point of the antenna and connected to another >> piece of 1 1/2 inch PVC pipe >> >5) frame is PVC of the same 1 1/2 inch pipe with aluminum tubing to >> strengthen the middle of the frame around the rotor. >> >6) N connectors soldered and pop riveted to the reflector. >> >7) supports are slotted and element is tie wrapped plus glued to supports. >> > >> > >> > >> >---------- >> >From: L. J. Shepherd[SMTP:email@example.com] >> >Sent: Tuesday, October 01, 1996 1:35 AM >> >To: Ronald E. Lile >> >Cc: firstname.lastname@example.org >> >Subject: Re: Mixed results on Helix Builds >> > >> > >> > >> > >> >Hello Ronald: >> > Thanks for sending your comments to the net........I'm one of the >> >builders with high reflected power on the 10 ft helix. I found your >> >description of how you achieved a good match interesting.......like >> >yourself, my version has an swr of about 3.5 or so. >> > I wonder if you would be kind enough to expand a bit please on the >> >exact details of how you achieved a 50 ohm match.........from your >> >description, it sounded like you ran the first 6.5 inches of the active >> >helix parallel (at a distance of 3/8 inch center to center) to the reflector >> >element. I'm curious to know first if my understanding is correct........and >> >secondly, if you found it necessary to move the reflector forward or aft of >> >Mike's design location in order to achieve the parallel coupling ?? >> > I had suggested to Mike the possibility of substituting a solid >> >plane reflector (something like a pizza pan) in place of the loop reflector >> >and using the small rectangular tab matching device that has been described >> >in the literature to reduce the impedance to 50 ohms, but if your technique >> >was successful on 4 antennas, it certainly sounds like something to try first. >> > Did all 4 of your antennas exhibit the same high swr after following >> >Mikes dimensions ??.........we (my co-experimenter N6EGY and I) were very >> >careful in the layout and construction and believe our mechanical >> >construction closely matches what Mike described. Our antenna exhibited an >> >swr of 3.6 at 430 and 3.8 at 439.........dropping to 2.33 at 437. >> > Would you care to offer any relative statement regarding the gain of >> >a single antenna ? I was planning on a comparison with a KLM 18C to get some >> >beginning feel for the gain developed, but haven't tried it yet. >> > Some further details on your matching technique would be much >> >appreciated.......thanks again for responding to Mike's request for comments. >> > >> > Sincerely, JIM SHEPHERD K6OYY >> > >> > >> > >> > >> > >> > >> >At 10:40 PM 9/30/96 -0500, you wrote: >> >>Mike and the group >> >> >> >>I completed 4 of the antennas and have the following results. >> >> >> >>of the 4 all needed adjustment to obtain 1:1 match. But all where within >> >4:1 at the start of the matching process. I did find that the last inch or >> >so of the spacing formed by the reflector and the first turn was more >> >critical as to spacing than expected. Also found that with the 1/4 inch >> >tubing used, the spacing needed for the first 6.5 inches was 3/8 inch center >> >to center. Matching measurements made with a slotted line, HP SWR meter and >> >signal generator. >> >> >> >>The connection to the center pin of the coax connector was such that a >> >quick "S" turn was needed to get the beginning of the first turn to line up >> >with the support and be able to run along the reflector for the required >> >matching distance. >> >> >> >>All supports where made from the schedule 40 PVC (not CPVC) water pipe. All >> >elements have been attached to the supports with different kinds of >> >tie-wraps and then coated with the glue used for joining the PVC components >> >together. >> >> >> >>Testing was done by pointing the antennas straight up with the reflector >> >about 2 feet off the ground. >> >> >> >>The elements where formed prior to installation on the supports by bending >> >around wood circles cut to give the proper dimension when the coil was >> >placed on the forms. Same for the reflectors and the reflectors were >> >soldered together using lead free water line solder. >> >> >> >>Mike, if there are variations in the diameter, particularly close to the >> >feed point, what Za variations does your modeling show? >> >> >> >>Tests by using each antenna in turn to track satellites does not show >> >significant gain variations. >> >> >> >>Using CPVC pipe (made for higher pressure than that for drains) may have >> >different filler materials causing different loss factors? >> >> >> >>Metal supports cutting the plane of the loops, used for support of the >> >helix in the middle? >> >> >> >> >> >> >> >>---------- >> >>From: AF9Y Mike[SMTP:mwcook@CONCENTRIC.NET] >> >>Sent: Monday, September 30, 1996 5:54 PM >> >>To: Multiple recipients of list MARS-NET >> >>Subject: Mixed results on Helix Builds >> >> >> >> Mixed results on Helix Builds >> >> >> >> >> >> I've received mixed results from others attempting to reproduce the >> >>low cost Helix antenna published on my webpage. So far, two models show a >> >>near 1:1 SWR while two others show aprox 3:1 SWR. After extended >> >>dialog with the builders, I'm at a loss to understand why some appear >> >>to match the model while others do not. >> >> >> >> If anyone else has constructed one of these antennas with either >> >>good or bad results, I would appreciate the feedback. I am especially >> >>interested in areas where my construction technique was unclear. >> >> >> >> Some ideas offered so far: >> >> >> >> 1) PVC material used in some builds was lossy at 437 Mhz. >> >> (Those stubs are close to a half wave length.) >> >> >> >> 2) High sensitivity to the reflector circumference. >> >> (But I slid mine around and observed little change in SWR) >> >> >> >> >> >> Anyone else have ideas? >> >> >> >> >> >>Thanks, >> >> >> >> >> >> >> >> > Date: Sun, 29 Sep 1996 17:30:01 -0700 > To: email@example.com > From: "L. J. Shepherd" > Subject: 10 ft helix impedance tests > Cc: firstname.lastname@example.org > Hi Mike: > John (N6EGY) and myself have completed our first round of testing > with our 10 ft helix, based on your design. Our initial testing was only > concerned with the measurement of reflected power, we have made no attempt > at gain evaluation as yet. > Our test procedure involved moving the helix to a remote location on > John's property, mounting the antenna on a non conductive fixture with an > upward angle of approximately 30 degrees, feeding the type N connector with > an adjustable rf source through a Bird model 43 wattmeter located in close > proximity, but out of the near field of the antenna (15 ft off the back). > > The SWR measurements are summarized as follows: > > 439.0 mHz 3.80 > 437.1 mHz 2.33 > 435.8 mHz 2.44 > 430.0 mHz 3.59 > > Our feedpoint attachment was as you described, with the first turn > actually connecting slightly aft of the plane of the reflector element. > After the first round of measurements, we tried reorienting the connector > such that the first turn was in front of the plane of the reflector, with no > significant > change in measured reflected power. > > Conclusions: > > 1. The antenna exhibits a somewhat broadband resonance near our desired > frequency of operation at 437 mHz. > 2. The antenna, as constructed, does not exhibit an input impedance of 50 ohms. > > Comments: > > We took great care in the layout and construction of the helix, the > supports are very precisely oriented and all spacing dimensions are > referenced to a single point on the end of the array. We marked the tubing > at the intervals you specified and were careful to precisely align the > interval marks on the element supports at a point corresponding to the > specified design diameter. We followed your suggestion and used black nylon > cable ties for the attachment. > During the measurement session, I noted that SWR did drop somewhat > if I touched the radiating element (about 6 inches from the feedpont) and > the reflector element simultaneously. We made no attempt to quantify this > effect, but it suggests to me that some additional design modifications may > be needed in the 10 ft design in the area of the feedpoint. Our N connector > was attached to the reflector with two 4-40 machine screws and oriented such > that the coax was arrayed to the side of the array, slightly to the rear. I > tried reorienting the first turn of the helix in an attempt to increase and > decrease coupling to the reflector with no significant reduction in > reflected power. All our measurements were made by first adjusting to a > reference forward power level at each test frequency, and then measuring the > reflected power using the Bird 43 (we checked our feedline with a 50 ohm > resistive UHF dummy load to rule out any coax problems). > While initially disappointed in not achieving a 50 ohm input > impedance, we recognize that the helix may develop very useful gain and > directivity. Before proceeding further however, I wanted to submit this > report for your review and comments, as we recognize the obvious need to > optimize the feed impedance to the calculated design value. > It would be interesting to hear of the results obtained by others > attempting to replicate your design. > 73, JIM K6OYY > > Jim, Thanks for the report. I wish I could offer some ideas but I have none. As for others building the antenna, I've had the one station from Italy that got the 1:1 SWR with his replication of the design (10 ft). You know already that Christine also had aprox 3:1 SWR with her 7 ft and after a cut back to 5 ft. It appears that two attempts matched the model and two did not. Not a good track record. I used a WELZ meter which showed 1:1 with a UHF 50 ohm load and moved just a little above 1:1 when attached to the antenna. Maybe there is a common problem with my construction instructions. Here's some checking info I sent to Christine: > Date: Sat, 10 Aug 1996 12:47:25 -0700 > From: Christine Hindman > Organization: SHARC (Starlink HAm Radio Club) > To: email@example.com > Subject: helix test > mike, > > > 1) i need to know how you measured things and how precise i need to > be for this to work. we tried measuring very precisely and still no > luck. i.e. did you stretch the copper out straight to measure? are > you measuring from the center of the copper tubing? the outside > diameter? the inside diameter? > I measured the tubing length with a tape measure and was probably only about 1/16 accurate at best. In fact, that error can be considered to accumlate since I just marked the first 16.375" length and then used that mark to measure the next 16.375" length. I marked all the segments in one sitting: First 8 marks: 16.375" spacing (but 1" added for first segment) Next 14 marks: 13.875" spacing Final 24 marks: 13.500" spacing The stubs were cut to the length necessary for a center to center measurment for the diameter of the three different helix steps. > 2) what is the distance between the bottom of the support stub (at > attachment point to the top 4 inch pvc) and the top of the copper > tubing (tubing attachment point on the support) for each series of > turn diameters? for top of 4" surface to top of copper tubing: reflector loop = 3 3/8" 1st four turns = 3 3/16" last seven turns = 2 3/8" what are the distances from the 4 inch pvc pipe to the top of > the copper tubing at the 90 degree marks on each turn in the series? I don't understand this question. The 180 degree stubs are the same distance as the 0 degree stubs if that is what you are asking. > > 3) what is the length of copper tubing behind the reflector? in > other words, from the feed point to the place where the turn crosses > the reflector is how long? About 2.5" for the length to the point where the helix turn intersects the plane of the reflector (crossing point considered at the center of the copper tubing). At that point the surface to surface distance between the reflector copper tube and the helix copper tube is aprox 1/4" > > 4) is the feed point on the end of the tubing or one inch down from > the end of the tubing? (you said the last turn was an inch longer > than the turn previous.) The feed point (connector pin) is 1/2 inch below the end of the copper tubing. A #12 wire is used to bridge the distance. > > 5) what is the outer diameter of the reflector? what is the length > of the copper tubing you cut to make the circle? reflector loop dia (center to center) is 10.66"; (outer to outer) is 11.16" length of copper tubing cut was 33.5" > > 6) did you measure from center to center on the copper tubing pitch > measurements? did you measure 5.125 inches with calipers around > each turn or what? I used just the sliding sleeve to achieve the 2.5625 spacing between the stubs. A good check here is to measure the total distance between the center of the reflector stub and the last helix stub. My measured distance is 56.375". That checks with the calculation of 11 helix turns x 2 stubs per turn x 2.525" = 56.375" > > 7) is the feed point of the connector inside the circle of the > reflector? does the center conductor of the connector make a line > parallel to the plane of the reflector and 3/8th of an inch away > from the plane of the reflector? does the connector face away from > the center of the reflector? There is a tilt of the coax connector which places the connector center pin at an angle which is behind the plane of the reflector. The angle is such that it meets the plane of the start point of the helix copper tubing but is apox 1/2" below it. A #12 wire is used to bridge the distance.