Independent tests prove lack of frequency hopping with XPS

[Kiwi]

Guys,

I'm using two different systems to test these radios and Rx's with.

Your going to see a whole lot more on the two companies who supplied us with the test software and hardware. These two companies have really bent over backwards to help us ( FG ) bring as clean an answer as possible to you all regarding the 2.4 Ghz RC systems available.

Todays testing has been a fruitless event in so much as no matter how hard I try there is no way I can get XPS to hop.

Here's the basis of my tests so far.

I am using a device called AIRHORN supplied by www.nutsaboutnets.com

This device is a selective transmitting device that can:
Transmit on one selected channel ( 1 thru 11 )
It can traverse all the channels sending a fast or slow transition between channels (Traverse mode)
It can pulse on a selected channel (1 thru 11) with a selected duration between 5 and 60 seconds and a selectable wait time between pulses 5 thru 60 seconds.

To see what is happening the guys at Meta Geek supplied us with a Wi Spy unit and a spektrum analyzer. This Chanalyzer as they call it is a full on data recorder and you can leave this thing recording in real time all day if the need arises. That way if you had an event at your field you could easily go back and see what was happening in the area when it happened. Its likely something not many people need to do or want to do but its there if needed. At big events I honestly think it would be good idea while all the 2.4 issues are high on the discussion list.

However the Wi Spy is absolutely a must have tool for those who want to know what the environment is (2.4 Ghz) around them or at the field. It comes in two version. One with a detachable antenna and the other with a built in antenna. The top of the line unit costs little more than a top digital servo so go figure if its worth the effort.

You can find out more about them at www.metageek.net

OK so what has happened so far.

Lets just stick with XPS for a start so we dont things all mixed up.

XPS does look for clear air on start up. So far I have been able to have it start on 6 of the 11 WiFi channels available. Zigbee channels appear to have 16 on the data logger but I'm not at all knowledgeable enough to go into what the differences are.

OK with the XPS on channel 1 I can bring the noise level up to within 20 DBm of the XPS unit and it will not hop. Pulsing or just a straight hit, it will not move no matter what I throw at it. I dont think I have many more options to make it hop as I have maxed out the power of my AIRHORN. The channels either side of 1 were clean and had limited or no traffic on them at all yet the Tx and Rx would not move.

Given the circumstances in the other tests XPS did not hop when swamped, and it does not appear to hop when the noise floor is ramped up either. I think by pulsing an interfering signal all around the working channel would induce a hop if anything could. Well so far nothing.

I will post a video this evening of the setup and the way I attacked the XPS so you guys can be the judges.

I need a beer, its hot down here guys.

[klhoard]

Kiwi,

Any chance you can run some tests with the 2.4Ghz cell phone jammers to see what their effects are on these systems?

[Kiwi]

Funny you ask. I have one on the way but I will save that for last as I dont want to muddy the water too much before I have a chance to do all my tests in an operational type environment.

[klhoard]

I've read a few blog and forum postings from guys that are **really** proud of their cell phone jammers and love bragging about how they enjoy going into crowded settings and shutting down folks cell phones. I can only imagine what these jerks will do if they find out they can go to a fly-in and make model airplanes crash.

[Foamaholic]

Is the Maxstream module supposed to hop, or is it an XPS modification that is supposed to make it hop? If the Maxstream hops, then it looks like XPS did something to it to make it not hop.

[XJet]

	Quote: Originally Posted by Foamaholic
	Is the Maxstream module supposed to hop, or is it an XPS modification that is supposed to make it hop? If the Maxstream hops, then it looks like XPS did something to it to make it not hop.

No, the XbeePro modules are not designed to frequency-hop, nor are they particularly frequency agile when driven by user software that attempts to make them hop.

That is why any significant change to the XPS frequency agility will require a return-to-base upgrade, probably involving reflashing the XPS and MaxStream firmware (assuming MaxStream is willing to write the new code), or perhaps even a total RF module replacement. This will *not* be a low-cost operation for XPS so I suggest that it will never happen.

If JD has shipped the "tens of thousands" of units he claims (which I strongly doubt) then it could be an even bigger nightmare than the Futaba ZGUID problem -- given the relative resources of each company).

I repeat my belief that JD will end up trying to resolve this situation by saying "XPS 2.4GHz has only limited hopping abilities" and those who want true frequency agility should upgrade to our 900MHz gear which is a FHSS system"

This will leave the 2.4GHz users with systems that either won't hop or will only hop when Mars is in the seventh house with Jupiter ascending and a 5knot wind blowing exactly from the Southeast (ie: not often if at all).

It's pretty obvious that JD has always known he'd have hopping issues with the XBeePro modules, hence the *very* careful wording of his frequency hopping claims "only hops when potential interference is encountered". If you examine the very precise semantics of that claim, the system may well be working as advertised (who can really define "potential interference"?) -- however, when it really matters (ie: when *actual* interference is encountered) XPS appears unable to change frequencies... but he never claimed it would hop in response to *actual* interference so how can you complain?

Even the ideas that JD has mooted so far for "improving" the XPS frequency hopping are marginal in their practicality and if he can pull off a practical implementation that actually works effectively in the field then I'll take my hat off to him.

[Foamaholic]

Thanks. That's what I wondered about. Lends credence to the tests that show it doesn't hop, and I don't think JD should have claimed that it did, unless he has done something to make it hop, which it doesn't look like he did.

[sweetpea]

great testing Kiwi. It will be interesting to see how the other systems respond to the exact same test setup.

[1bwana1]

I don't expect the other systems to Hop in response to interference.

By design the Spektrum/JR picks and sits on two frequencies. It may be interesting to design a test to see how intelligently it picks these frequencies.

The Futaba FAAST system will hop continuously regardless of how busy an individual frequency is. It depends on being on a given frequency for so short a time that the interference doesn't matter.

The only system that depends on hopping to cope with a changing environment is XPS.

[sweetpea]

I don't expect that either.

I just want to see how each system reacts with the same type of interference.

I want to see Spektrum work on one freq
I want to see FASST hop through the interference

And have both with no ill effects.

[Eflyer_75214]

Hey Kiwi, You buy very expensive servos eh. Per the web site, the Wi Spy 2.4 is $399 and the v1 is $199. Course, when comparing that to a large size model, it is not an expensive item I would guess.

For us guys that go with smaller models, would be a good club purchase.

Really appreciate the time you are putting in to testing the radios. Will you be able to obtain one of the new Airtronics RDS8000 units to add to the test group at some point?

Thanks again,

Ron

[XJet]

On the subject of 2.4GHz spectrum analyzers (SA), I'm just waiting for some parts to arrive before I complete my next version of a two-part SA (and the test-rig sig-gen of course) which can be flown aloft in a model (where it logs the data to memory) or used in hand-held form at ground-level.

My first one used a plain 128x64 LCD display to graph the data in realtime or playback the logged data from the airborne unit but was somewhat limited in spectral resolution. I'm just waiting on the 2.4GHz chipset and an OLED display that would allow a color display to be created.

The new 2.4GHz chipset I'm using also allows me to scan the entire band in 1MHz increments which will carry an overlay that indicates the JR/Spektrum, XPS and FASST channel assignments.

I think these would be a useful pre-flight or airborne black-box device for those who are planning on committing to 2.4GHz.

Once the bits arrive I'll post a video showing the next prototype in action.

I don't plan to manufacture these (where would I find the time?) but if there's a demand I'll probably knock out a few (pretty much at cost) for those who really want them. Price-wise I'd expect them to be less than the WiSpy 2.4GHz USB device and have the benefits of being fully self-contained (no PC necessary), as well as providing in-air data-logging to memory.