Biquad on Dish Question
in [Wireless Internet]
|
From: amdx on 13 Jul 2008 14:29 I have what I call my reference antenna an MFJ-1800 spec says 15db gain (hmm ?) best signal is -72dbm to -70dbm. My test signal is in a building about 100 yards away. I built the biquad with 30cm lips my best signal strength is -74dbm . when I mount the biquad on a 18"x36" offset feed dish (looks like this) http://www.2000networks.com/accessories/dishes_lnbs/dish_directv_international.html the best signal I get is -64dbm, a 10 db gain over biquad alone. Is that the gain you would expect? I found the original feed position wasn't optimum, I had to raise the biquad about 2" and also back it away from the dish about 2" to peak the signal. What is are your thoughts about that? Mike PS. I question the the MFJs gain because the modified (110mm and lips) biquad is only 2 to 4 db less than the MFJ. If I did good, and got 8db gain from the biquad then that looks like 10 to 11db gain for the MFJ. I do understand antenna measurements can be difficult. That said, I have used the MFJ-1800 for almost a year now and think it is a good buy at $30.00 plus shipping.
From: Jeff Liebermann on 13 Jul 2008 23:06 On Sun, 13 Jul 2008 13:29:18 -0500, "amdx" <amdx(a)knology.net> wrote: >I have what I call my reference antenna an MFJ-1800 spec says 15db gain (hmm >?) >best signal is -72dbm to -70dbm. > >My test signal is in a building about 100 yards away. > >I built the biquad with 30cm lips my best signal strength is -74dbm . >when I mount the biquad on a 18"x36" offset feed dish (looks like this) >http://www.2000networks.com/accessories/dishes_lnbs/dish_directv_international.html >the best signal I get is -64dbm, a 10 db gain over biquad alone. > >Is that the gain you would expect? > >I found the original feed position wasn't optimum, I had to raise the biquad >about 2" >and also back it away from the dish about 2" to peak the signal. > >What is are your thoughts about that? Ok, let's play with the numbers. First, with a reflector type antenna, the gain of the biquad feed is not important. It's sole purpose is to "illuminate" the reflector. If the beamwidth of the feed is too wide, you lose gain in transmit. If the beamwidth of the feed is too narrow, you lose gain in both receive and xmit because you're not using the full area of the reflector. See: <http://www.w1ghz.org/antbook/contents.htm> <http://www.w1ghz.org/antbook/chap4.pdf> When dealing with elliptical reflectors, that are designed to handle multiple feeds (for multiple satellites), you don't get the benefits of the entire dish. The offset feed also means the effective cross sectional area is somewhat less than the physical area. I'm too lazy to work out the exact number, but my guess(tm) is that the DirecTV dish has an effective circular diameter of about 60 cm. In other words, the bulk of the reflector is not used. The maximum gain of a dish antenna can be calculated. Assuming you did everything perfectly, and the feed was properly matched to the dish f/D ratio, then: Maximum gain for a 60 cm meter diameter dish: gain = 9.87 * Dia^2 / wavelength^2 * (feed efficiency) gain = 9.87 * 600mm^2 / 125mm^2 * 0.4 gain = 91 dBi = 10 log(91) = 19.5dBi The 40% efficiency (that's the 0.4) is probably optimistic for your home made feed. Again, the numbers assume that everything is lossless, perfectly matched, built correctly, and properly designed. Your 10dB difference is gain is very close to what I would expect (i.e. calculate). The typical biquad has 8-11dBi gain. See: <http://802.11junk.com/jeffl/antennas/Biquad/> So if I give the biquad a 9dBi gain and the dish perhaps 18dBi of theoretical gain, then an 9dB difference is about what you would see. Close enough to your 10dB difference. Some stuff on dishes and feeds from my bookmark pile: <http://www.findmorecollectibles.com/wifi.html> <http://www.weijand.nl/wifi/> <http://www.qsl.net/ki7cx/wgfeed.htm> <http://www.engadget.com/2005/11/15/how-to-build-a-wifi-biquad-dish-antenna/> <http://www.nodomainname.co.uk/parabolic/parabolic.htm> >I question the the MFJs gain because the modified (110mm and lips) biquad >is only 2 to 4 db less than the MFJ. MFJ means "Might Fine Junk". They also no longer sell or support the MFJ-1800 antenna. The link: <http://www.mfjenterprises.com/products.php?prodid=MFJ-1800 > is dead. My guess(tm) is about 13dBi gain at best. >If I did good, and got 8db gain from >the biquad then that looks like 10 to 11db gain for the MFJ. I do understand >antenna measurements can be difficult. Ummmm..... yeah, that's an understatement. Actually, it's not difficult to make the measurement. The problem is finding a place to do the testing that isn't full of reflections and interference. I've found the deep dark forest to be fairly good as the trees absorb all the reflections. The only problem is that I have to elevate the test antennas well above the roadway to prevent ground bounces. >That said, I have used the MFJ-1800 >for almost a year now and think it is >a good buy at $30.00 plus shipping. I have a bad attitude about yagis. They have their place but are a PITA to build, maintain, install, and design. I much prefer a panel antenna which is much simpler, easier to mount, waterproof, wider bandwidth, and only slightly more dollars. How about 13dBi gain for $36? <http://www.fab-corp.com/product.php?productid=893&cat=0&page=1> -- Jeff Liebermann jeffl(a)cruzio.com 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558
From: amdx on 14 Jul 2008 11:41 Thanks for all that info Jeff, I'll take some time to digest it, especially chapter 4. Also thanks for for the link to the panel antenna. I think I'll purchase that just to compare to the yagi. Oh, I find the yagi link does work. http://www.mfjenterprises.com/Product.php?productid=MFJ-1800 Two more questions. Regarding you formula; > Maximum gain for a 60 cm meter diameter dish: > gain = 9.87 * Dia^2 / wavelength^2 * (feed efficiency) > gain = 9.87 * 600mm^2 / 125mm^2 * 0.4 > gain = 91 > dBi = 10 log(91) = 19.5dBi Where does the 9.87 come from? At my boat location, I find I need to have a signal strength of -76dbm to log on. If any weaker it won't work. At home I logged on with -90dbm with good speed. One big difference between locations is this, on the boat I have 10 to 14 signals that I can see, at home it's one signal. Why do I need the stronger signal on the boat. (interference?) Thanks, Mike
From: Jeff Liebermann on 14 Jul 2008 11:59 On Mon, 14 Jul 2008 10:41:07 -0500, "amdx" <amdx(a)knology.net> wrote: >Oh, I find the yagi link does work. >http://www.mfjenterprises.com/Product.php?productid=MFJ-1800 So much for my bookmarks. Note that the MFJ-1800 is NOT listed on the antennas page: <http://www.mfjenterprises.com/Catergories.php?sec=14> > Two more questions. Groan... > Regarding you formula; >> Maximum gain for a 60 cm meter diameter dish: >> gain = 9.87 * Dia^2 / wavelength^2 * (feed efficiency) >> gain = 9.87 * 600mm^2 / 125mm^2 * 0.4 >> gain = 91 >> dBi = 10 log(91) = 19.5dBi > > Where does the 9.87 come from? See formula at: <http://en.wikipedia.org/wiki/Parabolic_antenna> It's Pi squared. >At my boat location, I find I need to have a signal strength of -76dbm >to log on. If any weaker it won't work. > At home I logged on with -90dbm with good speed. >One big difference between locations is this, on the boat I have 10 to 14 >signals that I can see, at home it's one signal. Welcome to the joy of interference. You might be able to improve things at your end using a highly directional antenna. However, the interference will still be present at the access point end of the link, resulting in no real improvement. Also, high gain antennas have narrow beamwidths, which causes problems as the vessel rolls in the waves. Also, see: <http://wireless.navas.us/wiki/Wi-Fi_on_a_Boat> <http://groups.google.com/group/alt.internet.wireless/browse_thread/thread/debebce6300cd6d2/> > Why do I need the stronger signal on the boat. (interference?) Probably. The big question is interference at which end of the link? If it's at your end, there's plenty you can do with antennas. If it's at the access point end, there's not much you can do. -- Jeff Liebermann jeffl(a)cruzio.com 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558
From: amdx on 14 Jul 2008 14:09
Thanks Jeff, Sorry, I missed the Pi^2 on that page, I looked several places and couldn't find it. I'll study and try to leave you alone,,, for awhile! Thanks again, Mike |