Author Topic: Computer vs front light RFI problems  (Read 318 times)

Computer vs front light RFI problems
« on: January 07, 2019, 06:56:03 pm »
I have a Topeak Panoram V16 wireless "computer" on the bike that gets most use. I didn't want a wireless set-up, but couldn't find a wired device which met my requirements in mid 2016. As expected it uses (single-use :( cells/batteries) rapidly. It also suffers RFI problems, though until last week those only had a significant impact on cadence recording.

Last week I installed a shiny new B&M IQ-X front light on the fork crown, as a direct swap for the IQ Cyo (2014). When the light is switched on & I'm moving, the speed reads zero & the odometer stops recording distance. Further experimentation seems to suggest that while the supercap, that provides coulombs for the standlight, is recharging, the RFI is not as bad. That's still a very tentative conclusion.

I've read 100 Lux dynamo light. I don't think my problems should mixed up with that topic, since I'm fairly confident they're specific to the Topeak device.

Some rough geometric data may help any work-around thoughts. The distance sensor is below the IQ-X. The direct line between the sensor & the "computer" passes through the front brake caliper. The IQ-X has a shiny aluminium alloy (guess) case, which should work as a Faraday cage, apart from the hole in the rear for the on/off switch & its surrounding blue indicator. That is only about 5mm away from the (very notional) direct line from the sensor to some part of the "computer".

What suggestions does the forum have to either solve the problem or even suggest ideas that I might stretch into a solution with a bit of experimentation?

Feanor

  • It's mostly downhill from here.
Re: Computer vs front light RFI problems
« Reply #1 on: January 07, 2019, 07:01:15 pm »
I had the same issue with an old Polar wireless computer when I got a new Exposure LED light.

Basically, the wireless tech used on these things is marginal at the best of times.
As you have found, the RFI generated by the LED chopper kills it stone dead.

Only solution I can suggest is to ditch the wireless computer for either a wired unit, or a GPS-based unit.

Kim

  • 2nd in the world
Re: Computer vs front light RFI problems
« Reply #2 on: January 07, 2019, 07:09:40 pm »
You could try clamping a ferrite bead on the cables where they enter the IQ-X, on the basis that the switching noise might be propagated by the wiring.

But it's mostly because nearfield RF is a piss-poor way of connecting sensors to a computer.  Ant+ seems a bit more robust.
 Wires just work.  I think it's inevitably going to be an exercise in moving things around in the hope that you chance upon a set of positions that works.
To ride the Windcheetah, first, you must embrace the cantilever...

Re: Computer vs front light RFI problems
« Reply #3 on: January 08, 2019, 03:54:38 pm »
Thanks for your ideas. I've managed to narrow the problem down by wrapping the whole light, but not the wires, in a double layer of kithchen foil. The computer worked OK, which suggests that wiring is not the main source of interference. I presume that the light would get a bit hot, quite apart from not emitting any light. For the moment I've put the Cyo back while I think about options. A wired computer with cadence & useful statistics would be an ideal solution...

Kim

  • 2nd in the world
Re: Computer vs front light RFI problems
« Reply #4 on: January 08, 2019, 04:32:05 pm »
The general lack of higher-spec wired computers is irritating, isn't it?  I mean, I accept that sometimes wireless is reasonable or necessary[1], but the current trend of using it for everything where a wired solution would be a better option is just lazy and (where non-rechargeable batteries are involved) wasteful.

I note the new model of my preferred Speedzone Elite computer has gone wireless.  Grr.


[1] I'm using Wahoo RPM Ant+ sensors with a Bontrager Node computer on the Red Baron because the geometry is a complete pain for traditional sensors.  FWIW, that setup works fine in the presence of an IQ-X.  It's not a setup I'd recommend for start-stop riding, though, as you have to faff about waking the sensors up and re-connecting them every time you stop for n[2] minutes.
[2] For values of n that seem to be consistently about 30 seconds shorter than it takes the BHPC to start a race.
To ride the Windcheetah, first, you must embrace the cantilever...

Re: Computer vs front light RFI problems
« Reply #5 on: January 08, 2019, 04:42:50 pm »
Cateye still do a few wired computers, the Strada Cadence version is the one with the wired cadence meter. This is the same model I have been using for circa 6 years on one of my bikes. I hope it has the measurement features you want.

Re: Computer vs front light RFI problems
« Reply #6 on: January 08, 2019, 04:47:23 pm »
I have had problems with wireless computers, apparently LED lights turn on and off rapidly to regulate brightness. This creates a small magnetic field or similar which affects your computer. I had a basic battery led light which when turned on reduced the speed reading by around 3mph

Re: Computer vs front light RFI problems
« Reply #7 on: January 09, 2019, 11:51:22 am »

 Ant+ seems a bit more robust.
 

Although my ANT+ Garmin watch doesn't send my heart rate to the Etrex when the dynamo is powering my Luxos U
Duct tape is magic and should be worshipped

FifeingEejit

  • Not Small just Far Away at the back
Re: Computer vs front light RFI problems
« Reply #8 on: January 09, 2019, 01:26:11 pm »
Cateye still do a few wired computers, the Strada Cadence version is the one with the wired cadence meter. This is the same model I have been using for circa 6 years on one of my bikes. I hope it has the measurement features you want.

Strada Cadence was what I used before I went fully GPS/GLONAS/ANT+, wonder where I put them as I "never" chuck anything out.