Jump to content
Sign in to follow this  
Execration

900Mhz Video Downlink Interfering w/ 2.4Ghz Uplink

Recommended Posts

As part of my current job, I am in charge of putting together a R/C aircraft capable of being video-piloted 4-5km (3-4 miles) via line-of-site. It will carry scientific instruments over sites that are normally hard to get to, such as the terrain surrounding the peak of a volcano.

To accomplish this goal, a 2.4 Ghz XtremeLink uplink was chosen due to its advertised 5 mile range. This is the only off-the-shelf system that could potentially supply the range I need. So far we’ve tested it to 4.5 km on the ground (after 4.5 km line-of-site was lost), beside a highway where there are lots of obstacles and potential sources of interference such as power lines etc.

The downlink chosen was a 900 Mhz 2-Watt video transmitter. The base station is using an 8 dBi patch antenna, which should be more than enough to cover a 5 km range. Here is where my problem starts.

When I went to range-check my 2.4 Ghz radio with the 900Mhz transmitter turned on, I was unable to get more than 50 feet away at full power before I lost total control. This is MAJOR interference. I was puzzled, especially considering the separation of the two frequencies. A normal range check consists of unscrewing the 2.4 Ghz antenna on the Tx, and holding a button which reduces the output power to 10%. You are then supposed to be able to get at LEAST 50 feet (150 reported by some) from the aircraft before transmission is cut off. I couldn’t even make it 50 feet with antenna on and full power, let alone 5 km! I need help in troubleshooting this interference problem.

The first thing I did was try separating all the wires and components from each other, outside the aircraft in a test area. I noticed that if any servo wires came too close to the 900 Mhz antenna, the servos would lock up to one side and wouldn’t respond until at least 5 inches of separation was given. Once separated, they worked flawlessly. After separating the 900 Mhz transmitter and its power supply from the 2.4Ghz receiver and its power supply, with no wires crossing each other, the range was significantly increased. I could now successfully perform a range check, and probably fly successfully as long as I wasn’t flying first-person. However, the range check indicated that my control crapped out sooner with the 2 Watt video Tx on that without. In other words, the 900 Mhz is still causing interference, just not as bad. If I am going to fly 4-5km however, this interference cannot be ignored, especially if a range reduction is noticed on the ground. My goal here is to have both of these components operating with each other such that no impact can be detected if the 900Mhz Tx is on or off.

I will now detail my exact setup for further clarification. For my uplink, I have the 2.4 Ghz receiver with its own dedicated battery. The elevator and rudder outputs are fed into a Co-Pilot stabilization system. This device takes readings from an attached IR sensor and adds control inputs to the rudder and elevator to keep the aircraft completely level during flight. Servo extensions are used, but they can be eliminated if need be.

The transmitter is running off the main battery pack. A DC-DC Converter is used to regulate the 14.8V 4-cell lithium batteries to 12V. Long power lines are used to reach the Tx, which is mounted just past the trailing edge of the wing. The colour KX131 camera is being supplied a regulated 5V power supply from the unused BEC of the speed controller, and thus is also running off the main batteries. A coaxial cable is taking its output and feeding it into the 2 watt Tx via an RCA cable.

I originally had a lot of wires passing each other, and I am blaming that configuration for my initial, horrible range. After the components were separated, and kept as ‘modules’ (Tx and Rx separated), the range greatly increased, however it’s still not as good as when the 2 watt 900 Mhx Tx is turned off. Is there anything that can be done to stop this interference? My boss suggests it isn’t the 2.4Ghz receiver antenna picking up the 900 Mhz signal, since it works, just only up to a certain range. He says perhaps the pre-amp or diode arrays in the receiver are being saturated by the 900 Mhz RF. Perhaps a set of in-line filters or some shielding could prevent this? Please see the attached images for further clarification of my setup. Any help is greatly appreciated, thanks.

24receiverpg0.jpg

900txvo8.jpg

Share this post


Link to post
Share on other sites

Put 2.4Ghz RX in one wing tip and the 900Mhz TX in the other.

Terry

Share this post


Link to post
Share on other sites

I've already thought of that idea, and while it does help (inverse square law), it doesn't solve the problem. Remeber, I need as much range as possible, since I require 4-5 km! Any other ideas? Could the servo extensions act as antennas and pick up the RF? If so, can anything be done to prevent this? Could the 2.4 Ghz receiver be shielded somehow without shielding the 2.4 ghz signal itself?

Share this post


Link to post
Share on other sites

why don't you put a bandpass filter on the 2.4 receive input? Or on the 900 output. Actually a low-pass on the 900 output or a high-pass on the 2.4 input.

I think your boss is correct about the input stage being saturated. One other thing you can try is to cut your output power to 1W (which should still work).

Daniel

Share this post


Link to post
Share on other sites

There are a lot of ways to attack this problem but separation is the most effective and should be the first way to tackle it.

Keep wires short, use separate batteries, filters, chokes, capacitors and lots more.

As Daniel says a bandpass filter may help. This could be both 2.4Ghz on the RX and 900Mhz on the TX, do you know how clean the RF is from your TX ?

You will need to do a fair bit of testing to sort this out I suspect.

Terry

Share this post


Link to post
Share on other sites

I guess it will be very hard indeed.

I'd try sipmplifying at the most first. Eliminate co-pilot etc, i.e leave just the RX with 1 servo connected (without extensions) and remove all the rest of the gear that is on the control side. Make a comparison now. If you still get range reduction (I guess you will) that means you won't be able to solve the thing only by caring with wiring.

I wouldn't share any battery (control, camera, TX all separated, maybe even use a RX battery instead of the BEC if it helps).

Actually a low-pass on the 900 output or a high-pass on the 2.4 input.
Or even both.

I was puzzled, especially considering the separation of the two frequencies.

That separation is actually small. Even my high-end 35MHz RX that gives me more than 2800m reliable range will go down to ~600m with a 200mW 2.4Ghz video TX at 5cm distance. That's nearly 2 decades separation, while 900 and 2400 are only a fraction of one, hard to get rid of even with a good filter...

Share this post


Link to post
Share on other sites

I'd try:

- put the entire R/C installation (Rx & flight pack) in the tail, 2 mini servos right next to the elevator & rudder, autopilot on the top of the rudder, Rx aerial pointing directly away from the video transmitter aerial (i.e. in the dead zone), shorten all leads to minimum, apart from

- a long lead to your ESC at the front (+ flight batteries + motor), running it through a ferrite ring or two.

- put the video transmitter & battery underneath the motor in the nose.

- perhaps some good old foil shielding between the transmitter + battery + aerial and the motor & its battery.

(This might be a challenge to balance!)

And (as Kilrah says) 0.9 - 2.4 GHz still isn't much separation...

Share this post


Link to post
Share on other sites

Execration,

I've been following this thread since a few days.

I'm also into long range RPV and still looking for a satisfying solution for both long range video and rc control.

Meanwhile I try to get the info some FPV-ers are sharing with us.

Although he's not using 2.4Ghz for controlling his plane,this guy might have an answer to your questions:

http://www.rcgroups.com/forums/showthread....t=511550&page=6

Have a look at post #79 and keep us up-to-date.

Share this post


Link to post
Share on other sites

Thanks for the response guys. I haven't updated this thread for a while because I've been so busy. I followed Doofer's advice and put the receiver and servos in the tail, with CoPilot back there as well and the IR sensor on the tip of the tail. I have two wires running from the tail section to the midsection; one is to the receiver battery pack, and the other is to the ESC at the very nose. I made these twisted pair to help with any noise.

My 2 Watt 900Mhz Tx is located in the nose, antenna pointed down. It is tapping into the main battery packs that the ESC uses to supply my motor and the BEC for my mini camera. The ONLY link between these two systems is the wire from my receiver to the ESC, and since the ESC is supplying the camera, which is connected to the 900Mhz Tx... it could potenetially be a problem but not likely.

I shielded my 2.4 Ghz receiver with copper tape, and I soldered all the joints together. I then grounded this shield to the battery pack. The antenna, of course, is not shielded, so there is an area surrounding its base where some noise could leak in I suppose. I opened the receiver module itself and there is indeed a shielded box inside. Other circuits are still exposed, however.

I tested this modified setup today and found that I had much better range than before. The test was quick, in a noisy environment with lots of cars and overhead wires, so it probably isn't entirely accurate. I still notice a slight range reduction, but I'm not sure how relevent it is. For example, if I can get 150 feet from the aircraft (with my transmitter antenna removed and power attenuated at 10%) with the 900 Mhz off, I can get 140 - 145 feet with the 900 Mhz on. There is definitely a reduction in signal quality though, and I suppose if this was a full range test (coming soon) I might lose a range of almost a kilometer, but we'll see. I am going to add some more shielding and move things around for further tests, and if any of you have anymore suggestions I will be glad to try them. I might be putting a low-pass filter on my 900Mhz, I'll look around for one. Is there an easy way to attenuate the power to 1 Watt instead of 2? Even if just for testing, since I can always order another 1 Watt Tx instead of 2 Watt if need be.

Share this post


Link to post
Share on other sites
The ONLY link between these two systems is the wire from my receiver to the ESC, and since the ESC is supplying the camera, which is connected to the 900Mhz Tx... it could potenetially be a problem but not likely.

A shared battery often causes problems. So, you might want to start off with isolated supplies (R/C not shared with video), then transition into a shared supply after any other bugs are worked out.

Share this post


Link to post
Share on other sites

You have been busy! Glad to hear the (initial) test results are so encouraging. As Kilrah has said, you just aren't going to get 100% RC range with something transmitting from the plane, 80%+ would be really impressive.

If things are worse in practice, I'd 100% endorse Mr RC-CAM's suggestion about power supplies, though. You've just 2 servos, right? A modest (and LIGHT - the plane has to balance! e.g. 2 cell LiPo @ 450MAhr) Rx battery could be popped back in the tail area, and should work for ages. Thus you could have 900MHz Tx + camera + ESC off the big pack, with a single cable (2 core, signal & ground) going forward.

(I guess the completely obsessional approach would be to use opto-couplers to send the ESC signal forward along an optical fibre)

Share this post


Link to post
Share on other sites

OK, another update. After fooling around with some wiring, I was able to eliminate my problem. My switch harness for the radio was too close to the main battery and camera wires, and in that jumbled mess something was causing interference. After moving it away, the problem was relieved. I went to a large open field with no obstacles around and after doing a range test, I noticed no difference in my radio's performance whether the 900Mhz was on or not! I walked about 147 steps away from the aircraft, to the point where I couldn't see the tail deflection anymore, then marked the spot. I then walked to the same spot with the 2 Watt 900Mhz Tx on and it worked perfectly.

HOWEVER, this test was done without the speed controller connected, since I was alone and didn't have the aircraft tied down, and if interference were to occur I didn't want to risk my aircraft flying away on me. Today I plugged the speed controller in, and revved the engine. Everything seemed good. Even the camera's picture was near perfect, seemingly unaffected by all the noise present. A range check, though, showed that some major interference was occuring. I narrowed it down to the speed controller. After looking more closely, I noticed that the only wire going to my receiver from the very nose of the plane is from the speed controller. On top of that, the power lines to my motor are travelling right around (touching) my 2 Watt Tx and coming within centimetres of the antenna.

After seeing the importance of proper placement and shielding, and realizing RF isn't like Bigfoot after all (I assumed everyone was afraid of it - but it was rarely seen hehe), I am shielding every wire in the aircraft to ground, and installing an RF choke on the lines entering the receiver. Copper braid will be used for shielding. I am confident this will solve the problem. My ultimate range, however, will only be tested when I can get an area of 4-5 km line-of-site. I will update when I do this.

Share this post


Link to post
Share on other sites

Is your 900mhz tx by any chance a ebay purchased mobicom tx? I am currently chasing interferance from one using a std pcm radio. I have moved the tx and it's battery back to the tail, thinking it would be cleaner there, the camera is in the nose, with a video only feed traveling through the center of the plane. I am still getting jitter, and only about 60% of normal rc range.

Share this post


Link to post
Share on other sites

Execration,

I read through this tread with some interest.. In particular the TX reduction when using a 2.4gig TX and 900MHz Video TX.

If in fact you find there is no solution to you current range problem there may be another solution and that is using a UAV system depending on your cash :)

Have a look at this Aussie site: Slivertone or go straight to an example UAV . This is a very interesting site and is very helpful.

Just thought I would drop this in the thread as another option :)

Share this post


Link to post
Share on other sites

Lupy - My video Tx is indeed a Mobicomm from eBay. I've spoken to the man behind the transmitter, and he is a very knowledgeable and helpful guy. Unfortunately, it seems that these transmitters are pretty noisy if you are indeed having the same problems. I know it is interfering with the XtremeLink 2.4Ghz module, but I haven't yet tried my expensive JR 72Mhz PCM receiver, which I will get around to eventually. I suppose the same thing will happen. Once I shield everything in the aircraft (doing that as we speak) and getting rid of the interference caused by the speed controller being so close to the video transmitter, I'll test the system and see my actual range reduction. I'll post some pics as well for everyone to see. I am hoping I will be successful.

Bear - The purpose of this aircraft is to be cheap, since conventional UAVs run in the 10s of thousands of dollar range. We are also developing a GPS system to guide the aircraft, but this is more of a return to home feature if the uplink or downlink fails. I'm not sure how far this will go, but thanks for the links.

Edited by Execration

Share this post


Link to post
Share on other sites

all transmitters generate harmonics,

900 x 2 = 1800 bad for GPS

900 x 3 = 2700 bad for 2.4 RX systems,

why: 2.7 is inside the RX filter open frequency

and will be attenuated and mixed,

a video feed is WIDE 20-30Mhz easy, so blocking and mix results will occour.

Add really cool low-pass filter to the TX,

and seperate the two systems as much as possible,

antenna locations and their phasing is a simple way to get another 26db.

I saw you have solved it with cable harness coubling,

you see at this super hi frequency a good coubling will happen even if cables are just a bit close to each other, TX modules also send out bad stuff on their powersupply and video signal line.

BTW: that plane you use ?? link ??

I see the fuselage can be taken apart, neat for payload and hardware install,

you must tell us more.

I liked the idea about installing nasty stuff in each wing tip,

but the harness and all the extra plugs is not too nice.

Edited by ThomasScherrer

Share this post


Link to post
Share on other sites

I wasn't expecting so much noise from this tx. I may just bail and step over to a 500mw 900mhz tx, or look into the future hobbies 700mw one. Using a std shielded video and power line dons't help much, as the ground seems to broadcast noise too. I tried various caps from ground to power, and small filter caps across the servos, without much imporvement. I have had some sucess moving the tx to the rear and running a three conductor (video, power, ground) with a shield not attached to anything. I wrapped the camera in foil, but did not connect the foil to ground or the shield. I don't like this location much though, as the heat sink, tx and battery make a pretty big package with lots of drag.

I am running a pan tilt head with 180 deg servos, which seem really sensitive to whatever the 900mhz tx is putting out.

I am using a pcm futaba. still getting about 60-70% range.

Share this post


Link to post
Share on other sites

Details about the aircraft I am using can be found here:

http://www.resonance.on.ca/Mike%20UAV%20Bu...ter%20Build.htm

This wireless video issue is becoming a real pain; I am thinking of moving to a different video Tx if a low-pass filter is impractible or it doesn't work at all. I am still looking into it. I did some extensive shielding yesterday and am ready for some more testing. Hopefully I'll see some results today.

Share this post


Link to post
Share on other sites

The results of the shielding aren't very good. I got everything out to test and my range check couldn't even pass 10 feet! Alright, so I knew something was obviously wrong, and re-arranged some wires to get it roughly back to 'normal'. my range was still reduced when the speed controller was on though. I eventually narrowed teh problem down to the wire that goes from the speed controller (front, next to 900Mhz Tx) to the receiver in the tail. The really strange part about all this is my range is fine when the speed controller is OFF, but that wire is still connected. Once the speed controller is switched on, I get reduced range. This is why I assumed that wire was not the cause of my problem. A colleague and I were playing around with the wire, and noticed that there was a sweet spot, that when moved into, resulted in no noticeable range reduction. Moving the wire as little as an inch was the difference between full range and half range. I still haven't figured this out completely, it's actually driving me insane, but I have a few things to try:

1. Complete shielding of this wire from ESC to receiver, using aluminized duct tape to enclose it along the outside of the fuselage, isolated from everything possible. It will be grounded to an appropriate spot (whatever works) and will have a network of ferrite beads, aluminum firewall, and capacitors to try and eliminate any pesky RF.

2. Use an opto-isolator with a run of fibre-optic cable to make the connection, since there is only the signal wire (throttle signal) and a ground. Only one fibre will need to be run.

3. Low-pass filter on the 900Mhz Tx to try to elminiate any emitted unwanted RF.

So far I have no evidence of overtones or the 900Mhz signal getting into the receiver via the antenna; most problems seem to be noise entering the receiver by means of the servo leads. Until I do a full range test (when I get rid of the noisy servo leads) I won't know if the 900Mhz is really swamping my 2.4Ghz receiver. Naturally, I can't perform this full range test until my small-scale attenuated range-check passes. I'm getting closer, but not quite there yet.

Share this post


Link to post
Share on other sites
So far I have no evidence of overtones or the 900Mhz signal getting into the receiver via the antenna; most problems seem to be noise entering the receiver by means of the servo leads.

Don't write it off completely, at least not yet. The RF can indeed be conducted through the servo leads into the Rx. But, these very same leads can also become unwanted radiators of the EMI/RFI noise. Which can then be coupled into the nearby antenna input. So the entry points can be complex, which may require more than one type of solution.

... realizing RF isn't like Bigfoot after all (I assumed everyone was afraid of it - but it was rarely seen hehe)

The fear often comes after several ugly battles with the invisible RFI beast. :)

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

×