CushCraft MA5B

When I was looking for an hf mini-beam, I chose the MA5B largely on cost grounds, plus the fact that it offered the Warc bands as well. I had a homebrew 10 metre quad which I had been using. I wanted a beam on some of the other HF bands as well, but I was not expecting the MA5B to equal the quad on 10M. I did some tests, comparisons and measurements which have proved to be illuminating and sometimes surprising.

An initial problem

I have a weekly sked with a friend in Scotland (John, GM4NHI) who purchased a Cushcraft X7. See his X7 review John has been very pleased with the performance of his X7, but he did advise me to look inside the balun box before I put my beam on the mast, as he had not been impressed by some poor electrical work inside his balun box. I duly dismantled the MA5B matching box and had a look; it was very good advice! The build quality was unacceptable, a quantity of braid clippings plus a 1cm length of coax inner wire fell out. The coax shield connecting tag was loose and the quality of soldering was poor. This was all the more surprising because the mechanical construction quality of the beam is excellent but in my beam it was badly let down by this poor electrical assembly work!

I was not happy with the fact that the single tag connecting the feeder coax outer was bolted through the plastic box. This depends on the plastic, which is sandwiched by the nut and bolt, resisting compression with time and temperature variation. This may explain why the tag came loose. I took the opportunity to replace the SO239 socket with another one which had been modified by tapping two of the fixing holes, I could then screw two 4mm bolts through the flange so that the heads of the bolts were tight against the flange. I also soldered the heads to the flange. I fitted locknuts on each of two tags inside the box, so that there was no question of them coming loose. This may seem a bit excessive, but bear in mind that the antenna is rated at 1.2KW, that equates to about 5 amps of rf current into the 50 ohm coax connections.

With hindsight it might have been better to refit the SO239 socket with the flange inside the box. It would then have been possible to solder a connection directly to the flange. It might be better still if it was an aluminium diecast box instead of plastic? Unless the metal would upset the matching function? Although I can't see why?

Assembling the antenna

The first thing to appreciate is that this is NOT a conventional 3 element beam. It is a 2 element trapped and capacity loaded beam on 14, 21 and 28MHz consisting of a driven element and reflector, the centre element is a trapped/capacity loaded dipole for 18 and 24MHz. The matching box construction simply connects the two driven elements via two short lengths of coax which are connected in parallel and joined to a further length of what appears to be about 8 turns of 50 ohm(?) teflon coax wound around three stacked ferrite rings, presumably to act as a choke balun? The antenna needs only one feeder for all five bands.

The manual is excellent I found only one small error in that the EB elements' diameter is wrong in the parts list, it is listed as 1 7/8 inch when it is actually one inch. If you follow the assembly instructions in the manual you should have no problems. I did find it necessary to put a small self tapping screw into the boom to secure the telescoping sections BA and BB at the end nearest to the reflector. I found the reflector could rotate out of line in a strong wind but I did not feel I could tighten the worm drive clamp enough without risking stripping it.

There were no missing parts but there are no extras either! So take precautions against losing small parts, especially the capacity hat fixing nuts and washers. It is very easy to lose them in the grass! I also found the matching box to driven element coax leads were too short to tywrap them to the boom in the manner suggested in the manual. I thought perhaps I had the element spacing wrong, but on checking I found it was correct.

Initial swr measurements........ with the beam set to the dimensions recommended in the manual. I initially opted for the SSB settings for 20 Metres. The antenna was mounted at 33ft above ground. I made some adjustments later to the length of the elements on this antenna to get it exactly as I wanted it, but these were quite small. It is fair to say that you could probably set the measurements as per the manual and get satisfactory swr figures on all five bands without further adjustments.

I have a Microsoft Excel spreadsheet with graphs of all the swr measurements I made with the initial and final settings of the antenna. If you have MicroSoft Excel - click here If you have problems downloading or viewing this spreadsheet, right click and select "Save target as" or "Save link as" to download it to your preferred folder on your disk.

I first raised the antenna with the settings as per the manual and the 14MHz "chart A" option = 8". All measurements were made using a 33ft RG8 coax feeder. The swr's were as follows.

12M:-...... On 12 Metres the swr was 1.35:1 across the whole band 24.890 - 24.990MHz.

The 10M swr bandwidth is excellent but note that 21MHZ and especially 14MHz do not produce a "modern transceiver friendly" swr over the whole band. Most transceivers will start to reduce power at 2:1 or even less. This narrower swr bandwidth is the inevitable consequence of the element loading and all minibeams suffer from this. The 20M elements are the heaviest loaded so 20M will have the narrowest swr curve. However, note that the swr bandwidth is considerably better than the Cushcraft specifications on all bands.

Although these readings were acceptable I felt that I would like the 21MHz swr minimum frequency lower down towards the cw end, it looked like it might be possible to get the swr acceptable across the whole 15M band. This would mean that the astu would only be necessary on 14MHz. It seemed to me that small adjustments to the 18/24MHz dipole section should not affect the 14/21/28MHz 2ele beam section. My computer model seemed to confirm this and in practice this turned out to be true. The 18MHz band swr seemed to suggest that the swr minimum frequency was a bit high as well.

So what to adjust? Well I reasoned that on the 2 element, lengthening the centre section of the reflector and director would affect 10M,15M and 20M. Adjusting the"A" and "B" dimensions of the reflector and director should only affect 20M. This proved to be correct.

I made the following adjustments to the reflector and director of the 2ele beam:

1. lengthened the centre section by 0.5 inches each side (reduce the insertion depth from 4 inches to 3.5 inches.

2. Increased the 14MHz "A" and "B" dimension to 9.3 inches.

The centre section of the warc band dipole was lengthened by 0.5 inches each side, i.e. a total of 1 inch overall.

Final measurements......These alterations produced the following figures:-

Results of the changes: The approximate swr minimum frequency by band is now:-

I say approximate because the swr curve is quite flat near the minimum on most bands, this can make it difficult to be exact, it becomes a bit academic anyway!

Note that it has been possible to get the 15M swr bandwidth close to an acceptable level across the whole band. This means that the only band requiring the use of an antenna matching unit is 20M and even here the 195 KHz 2:1 bandwidth is considerably wider than the 90KHz CushCraft specification.

Since I have had the antenna up and down several times I checked the change in swr from the raised to lowered position. I tilted my mast over until the reflector was 3 metres from the ground with the beam pointing almost vertically upwards. I then measured the change in the swr-minimum value and frequency. I thought this might be useful to others installing the MA5B

The reduction in the swr minimum frequencies, with the mast lowered were as follows:-

10M - 45KHz, 12M - 37KHz, 15M - 23KHz, 17M - 18KHz, 20M - 12KHz.

The difference was less than I expected, the swr curve became broader presumably because the Q was lowered due to the proximity of the ground. In my case it was also about 2 metres from some vegetation when lowered.

Comparisons between the MA5B and my homebrew 10M quad........The 10M quad which I had been using was a very good performer, however note that the quad was mounted at a centre height of 24ft and the MA5B was at 33ft, so it has a 33% height advantage. Measurement of antenna gain is notoriously difficult, however I had a yardstick for comparison with the MA5B in that I also have a 145ft doublet at 45 ft which happens to have a very good lobe to South America on 10M.

Tests on a number of stations over a period of time had established that the quad did have a small but consistent gain over the doublet in the direction of South America. Computer modelling using EzNec had suggested the doublet had a gain of 11dBi and the quad had 12.2dBi at a take-off angle of 19 deg. The MA5B computer model showed an advantage of about 1dB over the quad, 13.2dBi at 14 deg take-off angle. The lower take off angle is due to the height advantage.

This was confirmed in a comparison with my doublet, it was just possible to be sure that the MA5B did give marginally stronger signals than the quad; I would estimate it at about 1dB. I was pleasantly surprised at this, as I had fully expected to lose out on 10M compared to my quad. It looks like the gain in my MA5B installation on 10M is about 13dBi at a 14 degree take off angle. This is a good illustration of the fact that, if you can get a smaller antenna higher, then you may actually get a better low angle performance than a bigger, higher gain antenna, which you have to mount lower down. It is well worth remembering that the angle of maximum radiation of a horizontal antenna is almost completely dependent on height!

Conclusions:

It is worth mentioning here, that if you want to understand what can be reasonably be expected from a 2 element yagi, you would do well to look at "What can we expect from a 2 element beam" on W4RNL's web site. He uses a 10M two element as an example. There is a lot of detailed information but generally it shows you will do well to get more than 10dB front to back and 5 or 6dB forward gain (in free space). So the Cushcraft specs look entirely reasonable, especially for a multiband antenna.

As one would expect in a two element beam, the forward lobe is quite wide, my quad was similar. I would estimate it to be about 70 to 80 degrees, the front to back appears to be about 10 to 12dB on 10M and 15M, which would be to specification, however I found the 20M f/b was also about 12dB not the 22dB claimed. However accurate measurements of such things are not easy to make.

The nulls off the side on 10M,15M and 20M appeared to be about the 25dB specified. The review by G3SJX, Peter Hart in the RSGB RadCom mentions that he found almost no nulls off the end of the warc band dipole. I did not find this to be the case, the figure for the warc bands was the same as the others. However Peter did have the antenna at only 20ft, if I remember correctly? Perhaps that had an effect? I could discern no significant front to back on the warc dipole, which of course is to be expected. At times the warc performance on 18MHz in particular surprises me, it quite frequently outperforms my 145ft doublet, it is obviously filling in some nulls on the doublet radiation pattern, which is very useful.

I would agree with Peter's comment about the 10M performance, the MA5B does seem to perform very well on that band. The reasonably wide bandwidth is also quite pleasing as I don't like fiddling about with matching when changing frequency within a band, but this is only necessary on a small section of 20M, or a very small part of 15M if your fussy!

As I stated earlier it is notoriously difficult to make antenna measurements, especially in an amateur back garden setup. However the described checks against my doublet, the antenna modeling and the 10M comparisons with my homebrew quad are quite convincing. Operationally I always feel one of the best tests is how an antenna performs in cracking pile-ups! In that respect on 10M & 15M it is very effective especially as I normally run only 150W. I can't really say yet how good it is on 20M as I have not used the antenna seriously on that band so far.

The fact that my antenna ended up being set to slightly different dimensions from the manual is not all that surprising. Individual specimens of the antenna are bound to have slight differences and more importantly the actual site installation has an effect. So don't assume my settings will suit your installation! If your mast is not easily raised and lowered, then measure your swr settings with the mast tilted over as I have describe, so that the reflector is 3M or 12ft from ground. The figures I found for the frequency change from lowered to raised should be of use for estimating the swr minimum frequencies when raised.

Am I happy with my purchase?...... Well, Yes I am actually. Although I may sound a bit critical at some points in this review, I think it is necessary to remember that this beam is reasonably priced. (I purchased mine in the U.K. from Waters & Stanton) It is not really fair to pay a low price and then criticise using the higher standards that would normally be associated with a beam at twice the price!

For the ham who does not have the financial resources or the space for a full size beam, it is an effective solution which performs well. It is very well constructed mechanically using quality materials, but could do with some quality control in that matching unit!........Alex G3ZBE.

UPDATE: March 2004. I have used the beam for some years now and it has been entirely trouble free. The 20M performance seems to be quite reasonable, it certainly consistently outperforms any of my wire antennas on 20M. It doesn't seem as competitive as on 15M and 10M but then 20M is a much more competitive band generally........Alex