Thread: 2.4ghz standard range testing

Originally Posted by scallybert

I assume you're referring to DSM2, rather than DSMX ?

Should see some improvement on DSMX - its taken them 5 years to catch Futaba but Spektrum say they are now the leaders in spread spektrum DSMX: Spektrum - The Leader in Spread Spectrum Technology

They'll be telling us DX8 is brilliant next


Stuart/Adieuk - worth watching the Spectrum video in the link above - there's some interesting information about how many TX's can be used together.

Originally Posted by Mslater

Should see some improvement on DSMX - its taken them 5 years to catch Futaba but Spektrum say they are now the leaders in spread spektrum DSMX: Spektrum - The Leader in Spread Spectrum Technology

Would Futaba have released FASST if DSM hadn't forced them ?

They seem to have been leap-frogging each other on this. As a modulation scheme, DSMX is probably better than FASST. There seem to have been some implemementation issues, however...

[FWIW, some older guys down the flying field recently were saying Futaba have had their share of similar problems in their time.]

They are going to need to go some to beat Futaba but I really hope they do, it would be the end of TX problems as we know it.

Futaba had a problem with the Zero code on the 6ex, 7c and one other that I can't remember, and made a hash of putting it right in the states but that was down to an ID fault not a problem with FASST.

I'm not sure about newer pilots to the game but when testing equipment for myself I expect it outperform the recommended distances by a rersonable margin. Looking at my 27AM..27FM & 35mhz gear (supposedly 30 paces) all goes 125m+ with one section of aerial. My Dx7 (a very early one) easily 75m and at least a couple hundred meteres at full power at ground level. My Futaba 12Fg is similiar and in excess of 200 in full power.

My DX8 low power range check is a lot less about 30-40m. This is the first radio I've not actually walked out its full range potential relying on the telemetry to show any short comings. I've had two instances of un-explained crashes possibly caused by signal loss using a Parkflyer Rx, approx 35m out at 1m or less above ground. To date the AR8000 combo has been good showing only some frame fades at approx 80m out at 2m or less altitude although it dropped frames a couple frames. I do fly in a quiet enviroment though.

I like the DX8 but would not trust it to fly any long range flights, (12FG was fine) even a plank that covers a bit of ground would scare me on the DX8. A F3A glider 1/2 mile out no chance, but I wouldnt choose any EU 2.4ghz for that.. 35mhz yes

On 27mhz I've flown gliders to extreme range sat in a chair cuddle box style using binnoculers (spelling?) untill I could no longer tell which they were flying spending many anxious minutes trying to get it home. 2.4 doesnt have a look in at EU power levels

is there need for such a test?

To get anything meaningful out of it, you'd have to test each receiver antenna one-by-one. Otherwise, you won't notice if up to two out of three Rxes are broken (with main Rx and two satellites), making the test result practically meaningless for the receiver.

In my opinion, GHz range tests don't make sense. They will not spot the majority of "real" radio failures, like flying too close to a cellular basestation.
Sure, there is the occasional loose antenna wire in the Tx. But how often does this lead to crashes, compared to for example bad Rx batterys, including those that aren't tied down properly?

Instead of range checks, how about making it a rule that pilots need to cross-check their shoelaces before flying?

The problem I feel is that we rely on what the manual says if you look at either the Futaba or Spektrum they say roughly the same yet we appear to have a variety of different results from various tests carried out.

So to me it would seem to make sense that say we set a test based on the following.

1. Place Heli on table or held at aprox 1m
2. With heli rx turned on and low power mode set on the tx walk with tx between pilot and model counting paces until connection is broken = how many paces


Report back with findings - but alse we could collect info such as local weather info / was the grass wet etc etc.

Whilst its not an exact science but if atleast over 100+ reported back we can start to get a picture if we collate all the information and all carry out the same test - then from this we pick decide on a safety margin.

At the end of the day there is a flip side to this in my view, that should someone's heli fall out the sky and cause harm, because we failed to try and come up with a half decent workable test it could get messy with insurance people. You could say it is like doing a mini MOT on your TX, you do it once or twice a year for peace of mind.

But atleast if the stuff ever hit the fan we can say to our insurers we took what we saw as a potential issue and came up with a method that gave us a degree of comfort. Its never going to be perfect, like a cars MOT it can pass one day and fail the following week it only takes a light bulb to blow!! - but I would rather do something that nothing.

We can make it an online report so that ALL can see how the data as its building and maybe it might give us a few surprises.

I'll have a go down the street tomorrow, ran out of field on Sunday and had to run because I only get 88 seconds of low power on the f8, Failing that I'll post you one of my RX's and we can test 'over the phone' lol:

I think you need to turn the heli and see if there are any weak spots where the rx or rxs are being shielded by carbon or metal or getting reflections causing destructive interference. As far as checking out the air range based on a low power test, the principles are the same as 35Mhz. 30 yards low power ground range is out of sight full power air range. Perhaps the fpv flyers could comment on this as some of them fly at some distance. I'm a bit concerned that Spektrum say that 20 or 30 fades in a flight is normal. Is this with only one rx? If it is then one rx and one satellite should give no fades. 20 or 30 fades to me is 20 or 30 times that both channels failed. Or again is it a fail on a single channel?

I think it would be useful to log if the rx is in a conductive frame or not and also the size of the heli, as the distance from the aerial to any conductive surface is critical on 2.4Gz. Also the orientation of the tx aerial.

Chris

Okay the more info the better - I'll get to work on a db input form..

Originally Posted by scallybert

Would Futaba have released FASST if DSM hadn't forced them ?

They seem to have been leap-frogging each other on this. As a modulation scheme, DSMX is probably better than FASST. There seem to have been some implemementation issues, however...

[FWIW, some older guys down the flying field recently were saying Futaba have had their share of similar problems in their time.]

Futaba was already producing frequency hopping 2.4ghz systems for industrial use long before 2.4ghz for R/C cars/planes/heli's was even considered, plus they use their own chips with their own software, Spektrum, like almost all other manufacturers, uses off-the-shelf chipsets and just tweaks the software. After a couple of issues with FASST, Futaba 2.4ghz systems have been rock solid and I have every confidence in mine. The quote from Spektrum that they are 'leaders' in the technology is marketing hype, you have to ask why Japan didn't allow DSM2 to be used so forced Spektrum to come up with DSM-J which looks very similar to DSM-X! I've heard quotes that if Spektrum were just coming to the market now, DSM2 wouldn't be allowed - might just be forum gossip but I'd guess there's a grain of truth in it.

As said we need in my view to build a true and reliable picture of actual results, take the branding away from all of this and get to the nuts of the issue. What is safe and what isn't collect as much data as possible based on a set test with a few variables so that we can built up a true picture. If all manuals state that 30 paces is the aprox distance for testing then that tells me that both/all systems are working off a similar power level for testing but what we can do is build up a picture based on facts and get real results that matter not what's printed on paper.

I'll work on a db input form based on the test and we can add in some extra variables such as heli orientation etc.

I think you are probably right about DSM2 not being allowed, but that does imply that someone is in control. It might be true in Japan but there seems no control over the sets sold in the uk. You might have a good working tx, but if someone else has a badly designed one or one that is faulty that detects your channels and instead of avoiding them, transmits on them, then you are dead. Anyone on 2.4Ghz should switch on their tx in such a way that it can see all the other txs. That means as high as possible and not on the ground. In my view this should be BMFA advice and all clubs should follow it. Txs shouldn't be left switched on when on the ground. If other people are flying and you want to switch on your tx, you should give a warning and check to see if you cause a problem. I know it's all automatic and bullet proof, but if a simple check can stop an accident then it must be worthwhile.

I'd also encourage every club to monitor the general 2.4G environment as analyzers are now cheap. All new txs should also be checked to see if they are doing what they should and that they behave cooperatively.

I know that most people just want to fly and automatic systems make that easy. In my experience things don't always work as well as they should, especially when built to a price with little regard to regulation.

Chris

Hi

Well, having had, possibly the most expensive DX8 issues thus far, and having a degree in electronics / maths and being an electronic designer for most of my adult life, and having designed avionic systems, missile guidance systems etc etc I feel that I may be able to offer a reasonably educated opinion, hopefully for the benefit of all.

I am not an RF engineer, that is an area of electronics that is extremely specialised and only people that have studied RF for decades and practised the art for even longer can make a true RF engineer, for the simple reason that is is as much about randomness as it is about science. Understanding what is going on in an RF link is an incredibly complex thing. One of the worlds most respected RF engineers (from AT&T) wrote a book, the introduction to the book quotes "When the moon is full, wear a garlic necklace and commence RF work another day".

In short, it is simply not possible to create a simple test that confirms if a spread spectrum system is working as it should. Tests can be created to test if the system is faulty in certain areas, but not if it is all working.

Very little about spread spectrum system problems are about the RF, it is mostly about the software - and very complex it is too!

I wish I could point this discussion to an definitive solution, but it simply is not possible.
Buy and use your equipment based on your gut feel and the experience of others, this is the most scientific approach that you can have.


BTW, Spread spectrum broadcasts a data packet that is prefixed with an ID and suffixed with a CRC checksum over multiple channels pseudo simultaneously, the chances of getting signals from another set are nil, and the chances of all channels being occupied are very, very low to the extreme, if they are, then no link will be possible.


I hope this helps.... I for one would oppose any structured field testing method because people will end up relying on this as a full test, which it cannot be.

Gareth

Sorry dont agree to ignore the issue and just say its impossible to measure I think is not a justified reason not to try and find a solution. We are not privy to the methods or no of tests done my manufacturers. Weve always assumed that it was reliable yet findings across wide range of members show different.