pal Posted March 6, 2007 Posted March 6, 2007 Hi Folks, I have an issue that makes no sense to me, likely because it is a little outside my field of expertise. I got totally sick of the spotlights on the shed coming on (via the motion sensor) every time a stray cat walked down the drive, or a thunderstorm cause a power blip. I figured a better way would be to put a remote control on the lights so I could turn them on by pressing a button in the car as I approached... The "control" part of the equation is all OK - that is my field of expertise, but it is the "remote" part that is giving me grief. I bought a 4 channel UHF receiver board and matching transmitters (spare channels for later use on a door operator), which were all installed and tested OK. On initial test (for the sake of the discussion - "condition 1"), the remote system is mounted in a box on the wall inside the shed with about 100mm length of wire soldered to the board as the antenna, and the shed door is open. In condition 1, I get about 30m range on the transmitters which is approximately the same as the length of my driveway. Ideal except that the shed door is open... In Condition 2, with the shed door closed but everything else the same, the range reduces to about 10m. This isn't practical for me, since I need the light to reverse up the first 20m of the drive (not a lot of turning space at the top!) So, I bought a UHF antenna and base (about 100mm long "whip" type of antenna - identified as 477Mhz UHF antenna), mounted them on a bracket attached to the outside front of the shed up above the door, and ran a 75ohm co-axial cable from the antenna back to the control box (about 5m of cable). I cut off all but about 20mm of the original antenna wire and soldered the end of the signal wire in the co-axial to this piece of antenna wire (to save de-soldering / soldering directly onto the tightly packed receiver board). In this new "condition 3" I anticipated the range being easily 30m out to the street, and likely closer to the maximum of 300m advertised for the remote kit. However, with this antenna connected, my range has further reduced, to only a few paces from the antenna itself! What have I done wrong? I'm no radio technician, but I was reasonably sure I would have improved the range with the new antenna, not reduced it. Is 75ohm co-ax the wrong cable to have used? Any clues greatly appreciated. Best Regards, Pcal
Charlie 75 Posted March 6, 2007 Posted March 6, 2007 Hullo Pal (or is it Pcal? Not sure). Anyway, 75 ohm cable is not the correct cable but over 5m it wouldn't cause the problem you are experiencing. The correct cable is 50 ohm co-ax, W2092 is the Dick Smith Cat Number. I suspect that you may have only soldered the centre wire of the co-ax to the control box since the original antenna was just a single wire. With co-ax you need to connect the outer braid, the earth, to the appliance as well. If it was going to a typical "saddle and clamp" type TV antenna connection, like in a splitter box in a ceiling, you would fold the braid back like you were peeling a banana and put that under the saddle and then put the centre wire into the clamp, ensuring that they did not touch each other or the signal is lost. However, as I mentioned, you may have connected only the centre wire so the antenna and cable is not doing its job. May I suggest that you extend the original single wire to the outside of the shed. Also I believe that just about every movement sensor is installed at its maximum oversensitivity. If it's a clipsal type you have three adjustment screws underneath for the movement sensitivity (ie the object needs to be close to activate it), the "on" duration and the light sensitivity. Play around with all of these and it may eliminate the cat situation. cheers
pal Posted March 6, 2007 Author Posted March 6, 2007 Hullo Pal (or is it Pcal? Not sure). Sorry - pal is 1st choice, but if its already taken in a forum, pcal is 2nd choice. Just forgot which forum I was in for a moment! Anyway, 75 ohm cable is not the correct cable but over 5m it wouldn't cause the problem you are experiencing. The correct cable is 50 ohm co-ax, W2092 is the Dick Smith Cat Number. I suspect that you may have only soldered the centre wire of the co-ax to the control box since the original antenna was just a single wire. With co-ax you need to connect the outer braid, the earth, to the appliance as well. If it was going to a typical "saddle and clamp" type TV antenna connection, like in a splitter box in a ceiling, you would fold the braid back like you were peeling a banana and put that under the saddle and then put the centre wire into the clamp, ensuring that they did not touch each other or the signal is lost.However, as I mentioned, you may have connected only the centre wire so the antenna and cable is not doing its job. Hmm? At the antenna end, there was a clamp arrangement with earth connection. At that point the shield is soldered to the lug and earthed (via the bracket and shed wall). At the receiver end, there is only a single plated hole in a pc board, so that's where the signal wire goes, and the shield is floating in part because there is nothing to connect it to. Standard practice in my experience of connecting co-axial cables in other sensor systems (which I have done professionally for about 20 years) is to only earth one end of the shield on a co-axial anyway to avoid ground loops caused by differing earth potentials that can induce voltages big enough to interfere with minute signals. Again, my experience does not relate to rf systems so I'm only assuming the same principles apply. May I suggest that you extend the original single wire to the outside of the shed. It may come to that, but I bought the antenna because I didn't want a miscellaneous chunk of wire siliconed to the front wall of my shed! And now having made that investment, I'm keen to make it work rather than give up and throw it away... Also I believe that just about every movement sensor is installed at its maximum oversensitivity. If it's a clipsal type you have three adjustment screws underneath for the movement sensitivity (ie the object needs to be close to activate it), the "on" duration and the light sensitivity. Play around with all of these and it may eliminate the cat situation.cheers The sensors were adjusted many, MANY times hunting for the illusive optimum. We put up with their inconsistencies for 12 years until they finally died - at which point I decided to replace them with something better... One small error I've found that I'm not sure how significant it is though. At the suggestion of the salesman in the antenna shop, I bought the 477Mhz UHF antenna because "that's what most of the transmitters are". (I had already ruled out the 27Meg variety as I knew it wasn't one of them.) Since looking deeply into the issue after it hasn't worked as expected, I've discovered that the transmitters are actually locked at 433Mhz. Would this difference also contribute, and if so, is there an alternative antenna design I could exchange this one for that might solve the problem? I was assuming that since a piece of wire will do the job, the finer points of the antenna would not be quite so significant... Thanks for your suggestions, and may there be many more!! Best Regards, Pal
Charlie 75 Posted March 7, 2007 Posted March 7, 2007 I don't think the 40mhz difference is important in this application. The only suggestion I have left (you've covered everything else) is to find a way to earth the braid and see if that helps. good luck
pal Posted March 7, 2007 Author Posted March 7, 2007 The only suggestion I have left (you've covered everything else) is to find a way to earth the braid and see if that helps.good luck Soldered an earth wire to the shield and ran it back to the mains input terminal block for the power supply. Made no difference at all. Based on advice from a local electronics supplier that the 75 ohm co-ax was a major issue, bought some 50 ohm cable and installed that instead. Total waste of effort as the range reduced even further to only about 1m - actually had to hold the tx up in the air pointed at the antenna from right under it to get a signal through!! So I reverted to plain old wire, and soldered one end of a part coil of leftover building wire to the receiver, and with the coil laying on the floor and the door open, range shot back out to about 30m. Now here is the really freaky part! I cut off about 8m of this building wire and ran it up the wall and out one of the corrugations in the roof and ran it along the gutter line outside the shed. I lost almost all the range again!! So I pulled the cable back inside the shed and left it loosely coiled hanging from a rafter - and it works! Even with the shed door closed, I get about 25m range with the cable INSIDE the shed, while I only get 2 or 3m with the very same cable OUTSIDE the shed. It defies all logic I can apply to the situation... While I would like more than 25m range, it seems to be the best I can achieve at the moment. It would seem however, that the UHF antennas are a total and complete waste of time, money, and effort... Thanks for your help. Best Regards, Pal
Recommended Posts