Hillsborough
Sheffield
S6 2LL
All Rights reserved
This site is optimised for a screen resolution of 96 DPI.
Main Page List
Aerial test site on the moors above Sheffield, at an altitude of around 345m (1140ft).
Now that`s what I call “line of sight”, you can see for miles......
Over the years we have tested dozens different models of antenna on Belmont
(A / Wideband), Waltham (CD / Wideband), Bilsdale (A / K), Crosspool/Sheffield
(A / Wideband), Emley Moor (B), Stocksbridge (CD) Oughtibridge/Wharncliffe (CD).
The results are interesting to put it mildly, we rely on our own data now and more or less ignore manufacturers figures. This is particularly the case where they only give a (peak)
gain reading. Even if the latter is accurate, this figure can be misleading, especially with wideband aerials. What you really want is the gain graph, but many manufacturers won`t release this information, arrogant gits.
We tested the aerials against each other at the same time, in the same location,
and in actual reception conditions. We have also retested the same antennas and find
the signal reception can vary for no apparent reason (possibly the temperature, the
humidity or even differing -
In summer 2007 we decided to institute a major new test programme and identified a site
on the hills above Sheffield which would enable us to do this as it has good signals off four main transmitters and three repeaters. Between them they cover virtually all the TV band frequencies. Some of the most significant data concerned the response of grouped
antennas above and below the designed for frequency band, this is particularly relevant for some Digital (i.e. Freeview) transmissions. As an example it proves that a B group aerial
(certainly the ones we sell ! ) will work perfectly well on CH55, which is MUX6 off Sutton Coldfield transmitter, see relevant graph.
At least two readings were taken -
every aerial (including many not published on this site), so some idea of the work involved can be imagined. I reckon I took around 8,000 individual readings, and that was just the 2007 tests ! Whilst it was very interesting to start with, it could became a hell of a grind....
The results are arranged in tables / graphs as follows :
Aerial Groups (the same model of aerial but all the different groups)
The C/D group aerials we stock
It should be noted that as antennas leave their designed for band their performance can become erratic and the further out of band you go the more pronounced this becomes.
This can give spurious readings, particularly in a spot where multiple TXs can be received, because the aerials directivity can become very poor and inconsistent. This is the main reason why we have not published some readings (e.g. C/D off an A group) as they are meaningless.
I thought long and hard about how to make the results relevant and comparable with
each other, and have decided to show all readings (in dBuV) relative to a DM Log Periodic aerial. The latter has a pretty flat gain curve and thus is an ideal control aerial.
Quite apart from anything else, wherever possible (i.e. in reasonable signal areas), Logs
should be used as a matter of course, so the table(s) will show how much gain you have to make your decision on antenna choice. Note how the Log performs pretty well (in terms of gain) against small/medium wideband Yagis at the bottom of the band(s) but falls behind
as the frequency rises.
For those who are interested I reckon that, conservatively speaking, the gain of a DM log would be about 6dB at the bottom of the band, then rising by about one dB in the middle, before dropping again at the top end. These are the figures I used when calculating the absolute gain for the graphs. The absolute gain figures are conservative estimates,
but I bet they`re not that far out.
One things for sure, they`re all accurate relative to each other !
Note this is dBd, gain compared to a half wave dipole, NOT the (dishonest ? ) dBi....
Such are the signals available at 1140ft that even the Log was giving the signals shown in the table below. Bear in mind that the Yagi18s were giving up to 7 dBμV more than that
and the XB16s up to 9 more, that`s NINE ! Just to put it in perspective, one should be aiming for 60 to 80 dBμV (on analogue), and not more than that or cross modulation interference can result.
Differences in the test results of one dBμV on any individual frequency should be disregarded, manufacturing tolerances or a truck going over a hill somewhere between the test site and the TX could easily account for those ! Furthermore the response of any aerial is not linear, it`s gain curve will have bumps and dips in it. That said, a 1dB difference
across the whole band is significant, and an increase (or decrease) of two or three dB is
very significant. It should be remembered that 3dB is a 50% higher signal level, and such
an increase at the aerial (as opposed to through an amplifier) is very difficult to achieve.
Obviously if you live in a decent reception area these amounts of signal variation are not important, but if that`s the case, you should just use a Log Periodic anyway !
In order to give some idea of the gain of the aerials we also included a Tri Boom yagi,
e.g. DAT 45, on the tables. We do not stock Tri Booms, but for those who know of them, the figures could be illuminating, over rated.........
I didn`t test the DAT 75 because it was too unwieldy, but Televes report it as being about
the same as a DAT 45 at CH21, rising to +1dB at CH45 and peaking at +2dB by CH62.
Incidentally this trend for extra directors to add gain at the top end of a widebands gain
curve (and less so at the bottom) was reflected in our own tests of other antennas.
It should be borne in mind that gain is not everything, which is why Logs are still one
of the best aerials despite being having relatively low gain figures.
Impulse noise rejection, directivity, out of band rejection, cross polar rejection and
front to back ratio and can all have significant effects on the quality of the received signal.
Having said that, gain is a function of the others (apart from impulse noise rejection) as it demonstrates that the antenna is performing efficiently, indeed gain must be a function of the others. Why must ? Well an aerial is a passive device, the only way it can increase its gain (in a particular direction) is at the "expense" of gain in the other directions. As an aerial`s gain increases, its acceptance angle (and its Cross Polar Rejection come to that) decrease. They must do, there is no other way for the aerial to increase its gain.
All of this is before you even start talking about wind loading considerations and/or
build quality. Fortunately, we definitely know what we`re talking about when it comes
to the latter, there are no “Bacofoil” aerials here......
This table compares six Yagi 18s of the same model, but of all five groups plus
Wideband. Note, we don`t actually sell the Yagi 18 wideband (because our preferred
model of XB10WB outperformed it) but the gain plot is included because it directly
compares aerials of the same model. Interestingly these grouped Yagi 18s performed
just as well as grouped XB10s, and since they are cheaper, have lower wind loading
and are (arguably) more robust in construction, we do stock those.
The test methodology, and points to bear in mind when interpreting the results, are explained above. The figures are obviously for this particular model of antenna.
Other models varied in performance and this was particularly the case for gain outside the designed for band. As an example, the point above which significant gain fall off occurred,
for A group aerials this varied from CH40 to CH42, and for B groups from CH56 to CH59.
For gain within band, other tested 18 element A groups (which we don`t stock) gave
around 3dB (av) more than a DM Log, the other B groups gave 4 to 4.5dB and other
C/D groups gave 4 to 4.8dB.
I must confess to some surprise that the Yagi18 K performs as well as it does up to CH62,
but this was consistent over three sets of tests on three different days. It will be noted that the B group “beats” the K (on average) over the K band. However this is primarily due to its excellent performance in the B group, note how the B trounces the Tri Boom (e.g. DAT45), which is a larger and more expensive aerial don`t forget. The K is superior to everything bar an A group at the bottom of the band, and this is the area that is the weakest by far for all wideband (or semi wideband) aerials.
Note that the A group antenna is a huge 3.8dB up on the wideband (of the same size) for
the A group frequencies, it`s also a massive 3.2dB up on the (larger) Tri Boom antenna.
In fact such is the widebands inferiority that there is no such thing as a “High Gain” wideband aerial for the A group frequencies, or if there is I`ve yet to come across it.
The C/D`s performance was inconsistent, it was sometimes only on a par with the E and
the W/B, but sometimes it was in front ! It would have been easy to “massage” the results
to say what they should, but being supremely honest, I decided not to........
On the left is Ferrybridge power station and on the right we have West Burton.
Belmont is situated just to the left of the latter, except it`s a further 26 miles away.....
From the area of the site you can see an incredible 7 (seven ! ) power stations and these are marked with white pins on our TX map. From the NE to the SE they are :
Ferrybridge (24m to the NNE, opened in 1966, 2GW)
Eggborough (26m to the NE, opened in 1968, 2GW)
Drax (32m to the NE, opened in phases 1974 & 1986, 4GW, largest in Europe)
Thorpe Marsh (22m to the ENE, open 1963 to 1994, 1GW)
West Burton (32m to the ESE, opened in 1967, 2GW)
Cottam (34m to the ESE,opened in 1969, 2GW)
High Marnham (36m to the ESE, open 1962 to 2003, 1GW)
Incidentally I`ve been on trips round West Burton and Drax and they were mind blowing !
Also compare to :
The above table is a summary of the results, for the full results click here.
Incorrectly Fitted Cradle
In this test we deliberately installed the aerials cradle incorrectly, that is to say parallel
with the dipole/directors as opposed to perpendicular to them, see the pictures below.
As mentioned on Aerial Arrays, many years ago we deliberately mis-
if you think about it the whole tuning of the antenna is changed by such a bodge.
So the purpose of this experiment was to put some sort of figure on just how much signal one can lose by placing the cradle in the wrong position. The actual loss (in dBs) varied according to the frequencies involved, which is what one would expect. It must be stressed that the level of loss would probably vary depending on the model of aerial but it`s a fairly safe bet that it would always be significant. That said, if you`re in a strong signal area anyway then your picture would still be fine. The only problem you`d have is any passing aerial installers laughing at your antenna.....
An average signal loss of 3.7dB is very significant. In fact to achieve a increase in gain of that amount from an aerial takes a
lot of effort. As an example an XB10WB only averages + 3.7dB gain over a DM Log.
Just out of interest we then positioned the aerial on the other side of the mast
(picture C) and there was an additional signal loss of about 2dB (on average), just for
moving an aerial about 14” to one side ! ?
CH41 (which is analogue C4 off Emley) was particularly badly affected, a total of 9dB
down on the meter, and the picture was very much worse.
I have no explanation for this other than the aerial was then 30” from the chimney as opposed to 36”. But the fall off didn`t occur to anything like that extent when the aerial was mounted correctly (as in picture D).....
To be honest your guess is as good as
mine, but one things for sure this is
a classic case of RF being a black art.....
A = Cradle installed correctly,
right hand side (RHS) of the mast.
C = Cradle installed incorrectly,
left hand side (LHS) of the mast.
B = Cradle installed incorrectly,
right hand side (RHS) of the mast.
D = Cradle installed correctly,
left hand side (LHS) of the mast.
Just how much difference does a longer aerial pole make ?
In these experiments (and Aerial Installation Ridge Tests) we set out to discover the answer, although it must again be emphasised that these specific results are only applicable to this particular installation. Remember RF is a black art, particularly where an aerial doesn`t have “line of sight” to the transmitter.
We erected two DM log Periodic aerials 3ft 3in apart, with the lower one (when aligned on Emley or Belmont) clearing the roof line / ridge by 2ft 8in. There was no ridge in the other direction, towards Sheffield /Crosspool TX. These apparently arbitrary dimensions were governed by accessibility to our mast, it wasn`t easy (or safe) to go any higher without having to dismantle our aerial array ! It should be remembered that the most commonly adopted method of raising an aerials height is to swap from a 6ft pole to a 10ft. Bearing in mind that one then needs a bigger bracket (thus losing some of the additional height) this would normally give a nett increase in height of between 3ft 3in and 3ft 9in, i.e. we`re not actually far out on that one !
For Emley Moor we`re in a medium signal level area (thus we utilise a Log 40 on that
transmitter) but for both Belmont and Sheffield / Crosspool the reception is weak.
None of the three transmitters gives direct “line of sight” but in this respect Emley is the best, Belmont is in the middle, and Crosspool is the worst.
We have not published all of the frequencies for Belmont and Crosspool because some of the readings were too low to be reliable.
As can be seen in the results table, off Emley the additional 3ft 3in height gave an average increase
of 1.7 dB. An increase of this magnitude from the aerial, as opposed to from an amplifier, is significant, particularly from an increase in height of only 3ft 3in.
I was surprised.
To put 1.7dB into context, that is the increase you`d get upgrading to an
However, as the ridge tests revealed we had, by sheer chance, put the upper aerial near the “sweet spot”, much higher and the signal actually falls slightly !
Similarly the bottom aerial was just at the height where
the ridge on the roof begins to affect the reception ! Obviously the differences in gain vary with
the frequency / wavelength of the signal(s) but all channels off Emley Moor were up, between
1.1 and 2.0 dB.
The Belmont results were less consistent the Emley`s, unsurprisingly as “line of sight” is more problematic, i.e. the signal has more chance of being reflected, refracted or even cross polarised. In fact one of the five measured frequencies off Belmont showed no increase at all. Having said that, the average increase was 1.2 dB, which is still worthwhile if you`re really struggling for signal.
As mentioned in the article on the Topography Map there is a hill between us and Crosspool / Sheffield and this was certainly instrumental in the results off this transmitter being the most inconsistent. The thing is that being hidden from the transmitter by high ground is one of the main reasons that people fit taller poles.... Five of the ten measured frequencies off Crosspool were increased and one showed no difference, but four were actually decreased !
The average increase was 0.3dB, which is so small as to be unreliable.
I think that it`s reasonable to conclude that generally speaking mounting an aerial higher
will usually give more signal, but it`s not a certainty by any means and moving it laterally may be just as effective. The latter is particularly true in the case of ghosting problems.
However, unless the additional height is required to clear a roofline (or other obstruction)
by at least three feet, then don`t expect miracles.
Many people know that raising an aerial in height can affect the signal level, but an aerials lateral position can also have just as much bearing on reception. I have suspected the latter for years but wanted to know just how much the signal can be affected so we set up a five foot long horizontal pole, moved the aerial in one foot increments to the side and measured the results.
A length of five foot was chosen because we sell a 6ft “Supercrank” pole with an offset of about
2ft 6in, i.e use of the latter can move an aerial up to two and a half foot sideways.
Two times 2.5ft = 5ft !
The usual caveats on aerial positioning experiments must be emphasised.
Despite the DM Log (pointed at Emley Moor) being nearly three feet above the chimney I had still expected the signal level to increase as the aerial was moved away from the stack and this is what happened. What I didn`t really expect was for it to start dropping off again as the aerial moved more than 2ft out to the side !
Another classic example of RF being a Black Art...
At first I thought it was the presence of one the diagonal stay wires which we use to stabilise
our mast. So at slight personal risk (you have to suffer for your art....) I climbed over the ridge
and disconnected it. Result ? No difference !
Just to confirm the trend I extended the pole even further away from the stack and the signal promptly dropped even more.....
Examination of the graphic below show that as the aerial is moving out from the stack it is also getting closer to the ridge. Despite the fact that the aerial was still 4ft above the ridge, which I`d have thought was a reasonable clearance, I still think that the presence of the roof/ridge is the cause of the fall off. I say think because one cannot know anything about RF for sure.
The graphic below doesn`t show that the aerial is pointing over the ridge towards the transmitter at about 45 degrees (see picture of “Ridge Tests” ), I`m unsure how relevant this is but I thought I`d mention it anyway !
Gaining 2 to 3 dB by moving an aerial two or three foot is very significant. To put it into
perspective that`s the increase that one would get by swapping from a 46 element wideband
Tri Boom (or a Log40), to a decent grouped Yagi 18B !
Unfortunately it`s a time consuming business finding an RF “sweet spot”, and there`s no guarantee that one even exists on the installation one is undertaking. Furthermore if the antenna is pointing through trees who is to say that they`ll just grow a bit, and then you`ll have to start all over again !
How high above a roof line (or other obstruction) does an aerial have to be mounted in order for it to avoid the worst effects of it on reception of
the incoming signal ? ?
With the usual caveats about RF tests, we set up an experiment where a DM Log Periodic aerial was lowered in one foot increments from 10ft above the ridge line, down to two foot beneath it.
The results are shown below in a graph plus
(most of the readings) in tabular form, the former arguably shows the fall off in signal more clearly.
As expected the decrease in signal varies with frequency / wavelength. The largest increase in signal occurs in the first two or three feet above the roof line, then it dips slightly. This coincides fairly closely with the vicinity of the stay wire holding up our semi stayed mast, so that`s the most likely culprit, though when it comes to RF I`m not prepared to say anything for sure....
The peak reading (up to 10ft anyway) was at 7ft after which a slight downward trend was
observed. I`m unsure why this occurred.
My first guess was the effect of the aerial at the top of the mast, but I feel this is unlikely to be the culprit as even at the very top of the test mast there was still a good two foot gap, and
the top antenna is vertically polarised,
as opposed to the horizontal polarisation of the test DM Log. A more likely explanation is some sort of “RF dead spot”, but, to be frank, your guess is as good as mine, RF is a black art....
I had always advised aiming for a two or three foot clearance above an obstruction and I was pleased to see my guesstimate was vindicated, but what I didn`t expect was there would be such a huge fall off within 6” of the ridge level. Arguably if you can only get your antenna 6” above the roof line then you may be better off putting it in the loft. The fact the attenuation peaked just below the ridge proves that electromagnetic radiation does indeed travel in
straight lines (apart from when its being refracted ! ), but also that, as expected, the roof is thickest there, ridge tiles, roof trusses etc. If we could have gone lower I reckon the signal would have risen a bit more, assuming there wasn`t anything metal in the loft obviously !
Conclusion........
All these tests are investigating the effect of positioning on an aerial installation`s gain
(that is to say the amount of signal it collects) and they prove conclusively that the position of an aerial can be vital to it`s effectiveness. The tests are all outside but aerial positioning can be even more critical in a loft and inside the optimum position can even vary between aerials ! I say can be, rather than always, because the adage that RF (radio frequency)
is a black art, it`s not a science, is demonstrated time and time again in this section. Consequently none of the results published here can be regarded as definitive as every installation is different. There are so many variables e.g. the frequencies / wavelengths involved, the transmission path to the aerial, the distance to the wall / roof / trees / next door`s house etc etc. All of these factors can result in more reflection, refraction or cross polarisation of the signal. Also see Fresnel zones.
What can be established is the basic principle that an aerial will almost always operate most effectively high up and in free space, i.e far from other objects which affect RF.
All test results are averages of at least two readings.
We`re often asked how far to separate aerials mounted on the same pole. The answer very
much depends on if the antennas are of the same polarity or not and this test proves that
to be the case. Of course the potential obstruction doesn`t have to be an aerial, it could be anything metal (e.g. the aerial pole), though masonry also attenuates RF. Obviously any electrically inert object (e.g. plastic) should not have any effect on the signal.
I have been meaning to research the different losses resulting in same or opposite polarity obstructions, so whilst up at the test site (testing the new version of the XB16A) I had a bit of a mess about with a section of pole.
I taped it onto the aerial, first perpendicular,
then parallel to the dipole/directors. Thus the section of pole acted as an obstruction to the aerial of the opposite, and then the same polarity.
As expected the fall off in signal when the pole was the opposite polarity (to the aerial) was
very little. In fact it is not statistically
significant. That said, I still wouldn`t place an obstruction (e.g. an FM or DAB dipole) through my aerial if I had a choice in the matter, for aesthetic reasons apart from anything else !
On the other hand, if the polarity of the obstruction is the same as the aerial then the drop off in signal is significant, see table.
Don`t forget that 3 dB is 50% more signal, and that is more than the difference in gain between a Yagi 18A and the aforementioned XB16A.
If fitting two aerials (of the same polarity) I`d try
and separate them by two or three feet, just to be on the safe side.
The section of pole (used in this experiment) was attached to the middle of the Yagi 18A (the 8th director), but just out of interest I briefly moved it to the end (the 18th director) and
re-