[VHFcontesting] Band Factors in proposed UHF contest rules

[James Duffey]

Just a reminder, although there is time left to make comments on the proposed UHF rules if you haven’t made comments you should do so. Now is probably better than later. It is all well and good to make comments about the rules here on the reflector, but to have a real impact they should be sent to the committee at < vhf-input at arrl.org > by June 15. If you haven’t made your comments yet, here is my rumination on the proposed band factors. 

There have been some questions here about whether or not the band factors suggested for the new contest are reasonable. I have had the same doubts myself. At first glance they seem high. So, I set about to see if there is a quantitative way to determine what the best band factors are. The answer is “sort of”. 

It is straight forward to calculate the path loss on a troposcatter path, and combined with an assumed station capability, one can calculate the distances at which the SNR is equivalent for the same power and same physically sized antenna (increasing capture area with frequency) and use that to normalize distances on the varying bands to determine at least part of the band factor. There can be variation in troposcatter path loss calculations, so don’t take these as gospel. I used the SM6FHZ excel spreadsheet to calculate the expected troposcatter distances over a smooth earth. The distances calculated do show a reasonable and mostly correct dependence with frequency though. Here are normalized factors to adjust for the varying distance capability on the bands up through 10GHz:

Band (MHz)  Normalized distance (dn) for 3dB SNR in 500Hz bandwidth
222		1
432		1.1
902		1.4
1296		1.9
2304		2.4
3456		2.5
5760		2.3
10368		3.1
 
Three dB SNR in a 500 Hz bandwidth is a copiable CW signal to most experienced weak signal ops.

The way to interpret these factors is that, with similar equipment, one can work 3.1 times as far on 222MHz than on 10GHz, so the band factor, based solely on distance, should be the reciprocal of that. As the contest is distance scored, these should be a significant component of any band factor. 

 
I stopped at 10GHz as the water absorption grows above that and I don’t have a good way to calculate that now. I feel pretty good about recommending band factors based at least partially on these numbers.


One of the goals of the committee is to increase activity. Thus, I think it makes sense to calculate at least a part of the band factor by the level of activity on the various bands. This data is easy to obtain from the contest results on the ARRL web site. I just divided the number of total entries, which was 175 for the 2015 UHF contest, by the number of entries per band to get a factor by which to multiply the distance to encourage activity in inverse proportion to current activity on the given band:


Band (MHz)  Normalized participation (pn)
222		 1.3
432		 1
902		 2.1
1296		 1.6
2304		 4.7
3456		 6.3
5760		10.9
10368		 6.3 


No big surprises here, except that 5760 has the lowest participation and hence the largest factor, which may dismay some who would like a smoothly increasing factor with frequency. I thought that the 222MHz factor would be higher, but it turns out that there were 138 222MHz entries out of the out of 175 total. Interestingly enough, not all of the entries included 432MHz as there were only 172, out of 175 total, 432MHz entries. 


A third thing to consider is that it is harder to get on the higher bands than it is the lower bands and that the band factor should include something to compensate for this difficulty. It is hard to quantify difficulty, so I took the capitalist way out and determined the cost of similarly capable stations on each band, including equal capture area antennas, and normalized them to 432 MHz. The station had 60W out to a 17ft Yagi or equivalent, and was background noise limited. ON 5GHz and 10GHz, I assumed a 60cm (24 inch) dish. 


Here are those factors:


Band (MHz)  Normalized cost to 432MHz (cn)
222		1
432		1
902		1.2
1296		1.1
2304		2
3456		2
5760		2.1
10368		6.1


So, you can see that there is a jump at 2304MHz, and one at 10GHz. The jump at 10GHz is due to the large cost of a 60W amplifier from Kuhne; $4500. Accepting lower power would reduce the price and put the difficulty in line with the other microwave bands, but would affect the distance factor. You pays your money and you takes your choices. 


Now, how best to combine these factors to compute an overall band factor? Damned if I know. 


Well, I do have some ideas. One usually does not want to normalize normalization factors, that is, the various factors should be added in some fashion rather than multiplied. Depending on one’s preferences, not all of the factors are of equal importance. How best to weigh the factors? The distance factor should be used as is and given highest priority. How best to weight the other two? The participation and cost factors are somewhat dependent on each other as the activity is driven somewhat by the cost of a system. 


My first Band factor generated was simply adding the distance and participation factors with equal weighting, ignoring the cost contribution as it is partially included in the participation factor. That is shown in the second column. The third column is a band factor calculated by using the distance normalization weight of 1, participation weight of 1, and the cost weighting of 0.5. The proposed Committee band factors are listed in the last column. 


Band (MHz)  	Band Factor 1		Band Factor 2		KK6MC Band Factor  Committee Band Factor

		(dn+pn)			(dn+pn+1/2*cn)		MCFactor
222		 2.3			 2.8			 3			 1
432		 2.1			 2.6			 2			 1
902		 3.5			 4.1			 5			 4
1296		 3.5			 4.1			 4			 4
2304		 7.1			 8.1			10			20
3456		 8.8			 9.8			10			20
5760		13.2			14.7			15			20
10368		 9.4			12.5			15			20


The KK6MC Band Factor column? That is a Band Factor I generated after doing these calculations, combined with what I feel the Band Factor should be, influenced a bit by these numbers, all rounded to nice integer numbers that look good in the rules. They are intuitive, part objective, part subjective, and neatly rounded off to integers. Here is what I was thinking when I generated these numbers: I think activity should be incentivized more on 222MHz and 902 than either my computed numbers or the committee’s band factors indicate, as 222MHz is underused but has great capability, and 902 is underused but can be a more of a struggle than 1296. I also think the transition from the lower bands to the microwave bands is too abrupt in the committee’s factors. 


So, to answer the question raised in the first paragraph, the committee’s numbers are fairly reasonable, but I think they have too much spread in them, probably are a bit high for the low microwave bands, and have breaks in the wrong place. I suspect that the committee can make a reasonable case for them. They probably went through an exercise similar to what I did, only with different emphasis, which results in different numbers. But they aren’t way out of line. 

These assume that the dominant propagation mode used by competitors will be troposcatter; line-of-sight and EME will distort the Band Factors. 

If anyone wants more details of what I did, I can provide that, but the details will probably give the impression that the calculations have more validity they really do. - Duffey KK6MC
--
KK6MC
James Duffey
Cedar Crest NM