Jump to content

Attention: RC-CAM.com will be closing down August 2021.

The RC-Cam.com forum was the very first online community dedicated to the advancement of wireless video cameras on radio controlled (R/C) models. This is now called "FPV" (First Person View). We are proud of the contributions that our members have made to the FPV hobby.

We've seen significant changes over the last twenty years. Initially there were a lot of eager R/C hobbyist that built their own video systems. Allowing these creative individuals to share their work was the purpose of this site. Now the FPV market is flooded with low cost systems; Sadly DiY FPV video projects are now rarely discussed.

RC-CAM.com (main site and forum) will be closing down August 2021. This was announced several months ago (March 2021) to allow our members ample time to download any information that is important to them. After the site is shutdown the information will no longer be available here.

We appreciate every member's involvement with advancing the FPV hobby. It is indeed sad to say goodbye to all our online friends. Be safe and stay healthy.

Old Man Mike

Trusted Member
  • Posts

  • Joined

  • Last visited

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

Old Man Mike's Achievements


RC-Cam'er (2/4)



  1. I'm quite suprised that you would have a test layout like that. The amount of leakage would be so bad that any test results would be suspect. While I don't doubt the possibility that the new SAW filter's 50 ohms vs 280 circuit design would provide a poor result, you really can't base it on that test. Using the TX as the signal source for evaluation of RX sensitivity would require that it be mounted in a completely shieded box with filtering on power and video lines. The first attenuator would also have to be mounted within that box. I've done quite a bit of testing to prove this as has been previously discussed in this thread: http://www.rc-cam.co...ignal-with-osd/ I'm just trying to be helpful here and hope you understand no disrespect is intended. You certainly are one of the most helpful R/C technical contributors on the net OMM
  2. I happen to capture that video entirely by chance. After determining the location of the plane, I eventually made contact with the flyer. The details are contained in a few pages of this thread starting here: http://www.rcgroups.com/forums/showthread.php?t=1147430&page=31 (this was recorded with my earlier CPOD antennas. The new stacked PinWheels measured about 2dB better than my year old stacked CPODs probably due to sunlight degrading the PCB material.) I did some specific real world tests for video signal quality improvement using low noise preamps. Here is a summary of that testing: http://www.rcgroups.com/forums/showpost.php?p=14776896&postcount=114 The bottom line is that with my current preamps (less than 1 db NF) mounted at the antenna, I'm probably about 5 dB better than the typical FPV guy. You do want to be very careful with the gain. The sweet spot for the typical FPV receivers used on 1.28 Ghz is around 6 dB. Much higher than that can lead to RX overload which will negate the improvements made by the low noise. I have about 12 dB of loss in the RX line going up the tower so I use a low noise preamp with 18 dB of gain. In addition to the low noise preamps, the triple redundancy and special processing is probably adding another few dBs. OMM
  3. Believe me, EME is much harder to get results that match the calculations. Here's video from a plane 12 Km away recorded with my omni system: I would say that is a fair amount of "moving and twisting". Except for a brief instant during the flip there is at least 6 dB of margin. The plane was running half the power I use on the quadcopter and an antenna that is 2 dB poorer. Since I don't do any flips, I think this also supports my claim that 32 Km should be possible. Of course if you do not believe math then .... I will admit that my triple diversity system is unique with the extra signal processing and low noise preamps located at the antennas so the performance is better than average. OMM
  4. I'll put the 30 dB on the RX if you wish and will post the video. How about you do the recommended test with the 50 ohm dummy on your 15 mw with 26 dB attenuator and tell us if you receive a video signal at 200 meters? I think Thomas proved the math was correct with this post of a video picture from 51 Km using 18 dB antenna & 500mw at 2.4 Ghz: http://www.rcgroups.com/forums/showpost.php?p=10526535&postcount=11 My independant 32 Km calculation seems to be remarkably close to the results scaled from his system: His 2.4 Ghz frequency compared to my 1.28 Ghz frequency is 6 dB worse His 18 dBi beam gain antenna is 12 dB better than my 6 dBi omni antenna His 500 mw TX is 3 dB worse than my 1 Watt TX Net is that his system is 3 dB better than my system so I should have 25% less range: 51 Km * 0.75 = 38 Km I had calculated 32 Km Range based on the 1 Km per 1 mw test results so it would seem quite reasonable to expect that performance. What is it about the math that you do not believe? I've used the same calculations to build systems that bounced RF signals off the moon and receive them back here on earth. That worked just as predicted and was a lot more difficult then a simple FPV system. I would certainly expect you to have a range of at least 50 meters with just a dummy load on your transmitter. That only requires about 0.01 mW of radiated power from leakage "behind" the antenna connector. Give it a try at 500 or 1000 meters. I measured my leakage at 0.1 mW of power so the only signal being received at 1 Km was from the antenna radiated 1 mW. OMM
  5. Terry, The attenuator is working perfectly. I verfied the 1mw level with my spectrum analyzer. The bigger concern is any leakage from the transmitter getting into the power and video leads which can radiate like an antenna. As mention in my post, I also measure this leakage to make sure it was well below the radiation from the 1 mw to the antenna. As for your test, you probably have a large amount of signal leakage. Leave the attenuators attached but replace the antenna with a 50 ohm dummy load. You should have no detectable signal at 200 meters. Be sure and rotate the system 360 degrees to be sure there is not leakage in some directions. You do not want to do range testing with the attenuation at the RX. The reason is that it effectively removes noise figure performance of the receiver. A good receiver is limited by the environmental noise, not it's own internal noise. When you place an attenuator at the RX input, you are no longer testing in a real environment for the receiver. It will make the performance appear better than real world. OMM
  6. Well things have improved since I originally did the 1.2 Km with 10 mw post over a year ago. The new standard is 1 Km with 1 mw Power. I have replaced the omni stacked CPOD antennas with omni stacked PinWheel antennas. There is now a Stack of two PinWheel antennas on the SW, N and SE sides of the tower. As a test, I inserted a 30 dB attenuator on the output of the 1 watt Lawmate video TX and was able to get good video at 1 Km using just 1 mw of power from the quad: Here is the video at 1 Km, first in the West direction and then in the North direction. I also rotated the quad thru 180 degrees to see if there were any deep nulls: For reference, the three signal bars correspond to the SW, N and SE antennas. Bottom line is that the 6 dBi omni system provides pretty good RX video at a range of 1 Km using only 1 mw of power. Going back to 1 watt would provide an increase in range of 32 times the 1 mw range (30 dB/ 6dB per range double = 5 doubles). So the system should be good out to 32 Km I think the 1 Km per 1 mw is a good metric for testing system performance. I measured the leakage from the TX case/wiring and found it was more than 40 dB below the output so I'm sure there is only 1 mw of power being radiated. A non-stacked single Omni antenna for reciever should get a little over 1/2 Km with 1 mw. A 12 dB beam should provide 2 Km per 1 mw which would correspond to 64 Km at 1 watt. (But I'm an Omni guy since I like to have full flying freedom without pointing and 32 Km range is good enough for my quadcopter flying). Details on the PinWheel antennas used on the Quad and stacked for the RX side is available here: http://www.rcgroups.com/forums/showthread.php?t=1147430&page=26 OMM
  7. Thanks for the detail response, especially the jello prevention approach. I would have thought just the opposite with less jello from more light rather than less light. What makes your video remarkable is that you posted nearly the entire flight. It is easy to get super smooth video when the quad is slowly ascending which is why that is the type of quad video so often posted. I've done hundreds of hours of quad FPV flying and I can tell you that you've got something to be proud of with that nice portable system. After a bit of reading I also came to the conclusion that the stock configuration is too small for me and that upscaling it would likely diminish the performance. Later this month I'll be trying larger motors and props with my normal MK system in the hopes that the extra power and lower response time will help emulate the small platform flying experience. OMM
  8. I'm even more amazed after seeing the photos. Maybe the smoothness comes from such a light weight system for HD video. And although light, the GoPro cameras are quite sensitive to vibration and will often produce a jello effect in the video. But certainly not noticable in your system. Also, those props look similar to the old Draganflyer props. Plus they look like the tips have been cut off. Did you do that or did they come that way? I like to carry extra items on the quad for my long distance flights so I wonder how well the system scales to a larger payload capacity. Guess I'll have to start reading up in some of the forums. Are there any groups you would recommend to start with? OMM
  9. Wow! Since vimeo does such a poor job on these types of videos, I downloaded the file and played it back in Quicktime. As a multiple MK quad owner, I'll say that your video so impressed me that I'm considering buying the X3D-BL. Could you post a picture of your quad and how the camera is mounted? Again, VERY impressive! OMM
  10. For the last week I've been trying to define a PinWheel antenna in 4Nec2 with variables that allow it to be optimized by the program. It was certainly one of the most difficult antennas to model but it is finally done. Here is a little video showing how the program converges toward an optimization balancing SWR, Gain (at zero elevation) and axial ratio (best multipath performance): Note that this is for a 1280 Mhz antenna and that the optimized result was: Leg Length = 67.14 mm Arc Angle = 81.8 degrees Arc Length = 121.5 mm Pitch = 47.1 degrees (this was for a 0.6 mm dia magnet wire) For those of you that know how to run 4nec2x, here is the file: http://www.af9y.com/PINWHEEL52.NEC OMM
  11. After observing some interference to GPS and FASST from my 1280 Mhz Video TX, I decided to try designing a microstrip PCB to eliminate the interference. I soon found that low pass filters were not effective since 1280 Mhz and 1575 Mhz are so close together. Instead, I ended up using notch filters. Here is the initial analysis for the filter (red is attenuation and blue is Ref Coeff which is like SWR) as designed in the Agilent Genesys program: Rusty used his CNC milling machine to fabricate the boards. We had to iterate the design a few times to compensate for the variables in the PCB material but here is how the tuned prototypes looked: I think it matches the objective design quite well. Normally it is quite difficult to get a lot of attenuation at the GPS 1575 Mhz signal and still maintain a low loss at the 1280 Mhz Video TX frequency. This tuned stub notch filter approach achieves that with only 0.4 dB loss at 1280 Mhz. In addition, the stub notch at 2440 Mhz keeps the second harmonic of 1280 Mhz from interfering with 2.4 Ghz R/C control systems like FASST. Although Spread Spectrum systems at 2.4 Ghz are usually quite resistant to interference, they can be overwhelmed by a very strong signal near the operating frequency. When you have a 1 Watt 1280 Mhz Tx next to the FASST RX on the flying platform, there will most often be some range reduction if a filter is not used on the 1280 Mhz Tx. Rusty has just announced the availablility of these filters at his website: OMM
  12. Hi Nigel, And sorry I got your 5.8 Ghz build post confused with RCC. I've corrected my post and also gave you credit in the other 3 mile omni system thread. Yes, I made it around 100 degrees but did not try too hard for an exact angle. I tried compressing and moving the elements around a bit and it did not change the minimum SWR frequency much. The frequency is pretty well controlled by the full wavelength loop. Because of that, I can see a real big advantage for 2.4 and 5.8 Ghz since it is so difficult to get things accurate at those frequencies with soldered quarter wavelength elements used in the CPOD. I like to think of it as PinWheel because they spin the way I imagine the circular polarization wave. Plus you can construct a PinWheel to rotate in reverse like you can construct this antenna to be LHCP instead of RHCP. OMM
  13. Nice work, Devonian. I'm most interested in hearing how much rejection you get between RHCP and LHCP. Given that the weather is so nice for a couple of days, I also decided to build and test a 1280 Mhz version before doing the 4nec2x modeling. I built mine in a LHCP configuration on a male SMA connector using gold plated memory wire and a small PWB used for the solder points: The resulting antenna weighs 0.2 oz (6 grams) and even with the small diameter (.025 inch) memory wire is very solid once every thing is soldered. Here it is mounted on a PWB microstrip filter which is connnected to the 1280 Mhz video TX via a right angle connector: You can also see the larger CPOD antenna in the lower right which weighs 7x more and is much more difficult to deal with on the quadcopter. In the mounted configuration shown, the SWR was 1.4:1 (Ref Coeff of 16) as shown on the spectrum analyzer: The very light weight PinWheel antenna (that's what it reminds me of) allows it to sit higher above the quad resulting in less signal obstruction. I did a flight test with the quadcopter doing a 360 degree rotation maintaining aprox 30 meters height at a distance of 1.3 Km. This was a non line of sight test with several rows of trees blocking the path. Normally this will cause a lot of multipath: As you can see, there is perfect video for the entire rotation and the signal variation on the three receiving antennas (Stacked LHCP CPODs) varied only a few dB. I only had time to do one measurement of the RHCP vs LHCP ratio. At a near zero elevation angle, there was at least 8 dB rejection of the RHCP which is as good as the CPOD. I need to build a RHCP version of the antenna to do a more accurate ratio test and I still plan to do a detail model of the PinWheel antenna to see if additional optimization can be done. Overall I am VERY impressed with the performance. OMM
  • Create New...