Intermodulation Distortion (IMD) in FPV - RC Groups
Thread Tools
Jun 27, 2015, 09:01 AM
Registered User
Discussion

Intermodulation Distortion (IMD) in FPV


Hi all,

With our FPV racing group, we have often noticed poor FPV performance when flying in groups. This often occurs even when we spread out channels across the 5.8 range. My father, a retired RF engineer, did some experiments and came to the conclusion that inter-modulation distortion (IMD) is at least part of the problem. As far as I am aware, this is an often ignored issue in the FPV community.

A group of pilots operating at 5.8 GHz are especially prone to IMD. The close proximity, high power transmitters, even spacing of the channel frequency, and poor receivers contribute to the problem. Per the video, there are some simple steps that can be taken to mitigate the issue. IMD is when two other transmitters combine and produce interference in your receiver at your desired FPV frequency.

I know the video is not the most exciting (the demo is at 3:45 in for those with short attention spans). I do think that if the FPV community adopted some of these strategies we could have better racing and group flying.

FPV Interference from IMD - Demo (10 min 45 sec)


Best,
Jon
Sign up now
to remove ads between posts
Jun 27, 2015, 01:16 PM
Team AlienWarpSquad
Good video covering IMD.
I have run into this issue working on Ham Radio repeater systems on multiple bands and during Club multi-station Ham radio contests.
It can be a difficult issue to work out what combination is causing the trouble.

Is the spread sheet show in the video available?

Better Receivers would go a long way to reduce IMD and also reduce adjacent channel interference.
Do any better vRX's exist?

I have been an advocate of low power vTXs. I have been running 200-250mW vTXs and use a goggle mounted 3-turn helix when I need a little more range. Tracking the copter by moving my head works very well and not that hard to learn to do.
Jun 27, 2015, 04:46 PM
Registered User
I will try and post the spreadsheet later. It is very difficult to choose frequency combinations that produce no IMD problems with one another when you are flying with 4 or more people. It is especially hard when you try to make sure the IMD products are spaced so many MHz away from any of the desired frequencies.

It was our hope that this would lead into a discussion about receiver performance. Do the more expensive ImmersionRC receivers reject IMD better? We do not know.

It is worth noting that none of the 5.8 bands available to us do anything to prevent IMD combinations. This includes immersionRC's new 'race band'
Jun 27, 2015, 04:48 PM
Registered User
Also, members of our club are working to introduce a strategy into their website pidfoo.com that will sort the racers and assign frequencies that produce no IMD products on the desired frequencies.
Last edited by Jon h2; Jun 27, 2015 at 05:03 PM.
Jun 27, 2015, 06:55 PM
Registered User
I am attaching two versions of the spreadsheet. One shows all of the two tone products and the other has those frequencies that are outside of the band removed for clarity.

The spreadsheet can be used to see the two tone third order IMD products produced by frequencies of the 32 standard FPV channels in the 5.8 GHz band. Note that any given pair of frequencies has to be fit "both ways" so to speak in the spreadsheet. That is, locate the intersection of f2 in the first column and f3 in the first row, then do the same with f3 in the first column and f2 in the first row, After a little time with these you will see how difficult it is to keep track of these frequencies! Better would be a program in which you could enter the frequency assignments that you want to try in a race and the program would calculate whether there were any IMD products on or near frequencies that had been assigned to any racer. I think one of our club members is working on something like this.

I'm looking into doing some testing of FPV receivers to evaluate their IMD performance and maybe I'll be able to make available third order intercept measurements. There are other receiver performance parameters related to interference immunity that would be helpful to know as well.
Jun 27, 2015, 07:03 PM
Registered User
To clarify, I don't at present know if any of the commercially available receivers on the FPV market offer any better IMD performance compared to the ones that I have used for this demonstration.

Waltr, you are right on target with your comments.
Jun 28, 2015, 02:27 PM
Team AlienWarpSquad
Thanks.

I remember the First HF Ham radio receiver I built. It was single conversion for 80meters and had down converters for the other ham bands. First time trying the 40 meter converted there were squeals and stuff all over when tuning. I wrote down each frequency there was a tone then learned about IMD. I did write a GW Basic program to work out all the products.
Never did get the converters working well but the 80 meter part worked great.
Jun 28, 2015, 10:03 PM
400' ..... NOT
Martin Y's Avatar
That is a great demo but can you do that same test some distance away from the rx? Dont think anybody flys that close to each others rx's. For example have all tx's about 10 feet apart and 20 feet from the rx.
IMO for racing everybody should have the same power vtx.
Latest blog entry: 1280mhz vs 1090mhz
Jun 28, 2015, 11:28 PM
Registered User
Perhaps you missed it but we did address that in the presentation. I had the same concerns, namely that in the demo, the distance between the potential interferers and the receiver was less than would be encountered in an FPV race. So I repeated the test at greater distances. I found very significant interference when the distance between the interfering transmitters and the ground station (receiver) was up to 100 ft. Its my judgement that this could be readily encountered in an FPV race. At distances over 100 ft, interference could still be seen, but was less dramatic.

Note also that in the demo we used transmitters of lower power than are often used in FPV racing. One of the interferers was 200 mw and the other was only 25 mw.

Our intention is to show that IMD can be a source of interference in FPV racing. Our experience was that despite taking adequate measures to avoid adjacent channel interference at our races, some pilots (including me) still saw significant interference in some circumstances. If otherwise unexplained interference is seen at an FPV race, IMD problems could be the cause.

Unfortunately, there are other potential deficiencies in the RF systems for FPV besides IMD that can cause problems. We would benefit from better specifications for these devices. And, of course, these devices were designed with low cost being a major consideration.
Jun 30, 2015, 02:39 PM
Registered User
Havoc256's Avatar
The problem comes in play because a lot of us use the 600mw transmitters and they stop all over the 200mw transmitters. Also when we have our race events, the pilots tend to sit right next to each other. So this does become a real life issue.
Jul 02, 2015, 09:20 AM
Registered User
Steve, please feel free to post your revision of the spreadsheet if you want to . Its an improvement and easier to use.
Sep 17, 2015, 11:22 AM
Registered User
cfischer's Avatar
So are you saying that the VRX need to be physically seperated? If so is there an ideal distance?
Nov 20, 2015, 07:35 PM
Registered User

IMD table 5.8Ghz


I have calculated all IMD ocurences in a +/- 5hz range for all 40 channels

So let's suppose I want to use frequency 5645 (Band E channel 4), I just select it on F1 filter column, then columns F2 &F3 will show the two combined channels that produce IMD, the last column will show at what frequency happens
Nov 24, 2015, 06:10 PM
Registered User

IMD on 3.4Ghz


On the same note, I also calculated the combination of channels that produce IMD on 3.4Ghz
Dec 01, 2015, 11:22 AM
Registered User
Quote:
Originally Posted by javierdiaz
I have calculated all IMD ocurences in a +/- 5hz range for all 40 channels

So let's suppose I want to use frequency 5645 (Band E channel 4), I just select it on F1 filter column, then columns F2 &F3 will show the two combined channels that produce IMD, the last column will show at what frequency happens
Good job! I like the filter.

For our racing group, we have now identified 5 frequencies that work well with one another. We use this all the time now. Assuming that pilots have 32 channel capability, they can tune to one of them:

5645, 5740, 5820, 5860, 5945


Thread Tools