Saturday, March 26, 2011

Wireless Cycle Computers and Interference

Today a customer came into the shop complaining that his cycle computer was giving wild readings whenever he turned his LED headlight on in flash mode. "Yep," I said, "it's RF interference. I've seen wireless computers freak out under power lines and by the anti-theft scanners at store doorways as well. The internet knows about it." The customer had tried his own search plugging in the names of his cycle computer and light together with the word "interference" and had come up empty but a quick "RF interference cycle computer" brought up pages of results showing a variety of lights and computers exhibiting the problem.

The customer was less than thrilled with my initial idea of making a little foil shield but Mike managed to make the computer behave by moving it to a location on the handlebars further from the light. Poking through a few of the pages revealed by the internet search revealed a few interesting tidbits.
  • Wired computers pretty much don't have this problem.
  • The newer, fancier cycle computers using the ANT+ digital protocol don't seem to have this problem.
  • There are still idiots on the internet and a lot of wrong info.
The final bullet point is illustrated by the page at:

The question asked is:

Is it true that wireless bike computers don't work if you have LED lights?

and the group wisdom is:

No it doesn't happen and the logic seems to be: I've never seen it, therefore it doesn't happen. My favorite is the guy who adds "Source(s): 6 years as professional bicycle mechanic."

We could apply this "logic" elsewhere. For example, I've traveled for 30 years and never been to Portugal, therefore Portugal does not exist. Or, I've ridden a thousand miles and never had a flat tire, therefore flat tires don't exist.

So, doing my bit to add a bit of signal to the noise that is our beloved internet, I'm posting these words:

Some LED lights when flashing do interfere with some wireless cycle computers. I've seen it. If it happens to you, try increasing the distance between your computer and your light or maybe work up a shield or switch to a different cycle computer.

And remember, no matter where you are or what time it is, someone is wrong on the internet.

Keep 'em rolling,

Kent "Mountain Turtle" Peterson
Issaquah WA USA


Anonymous said...

But, have you ridden a thousand miles in Portugal? I thought not.

I suspect flat tires exist, except on the Iberian peninsula. I base that conjecture on my own verifiable lack of experience. However, I will buttress my lack of factual anecdote with rather solid theory:

Generally speaking, one can presume that kilometers are easier on tread than miles (being of less length), and kilometers are units of measure in Iberia. Certainly all can agree that "flat tire" sounds more pleasing to the ear in the southern Romance tongues than in the bastardized Germanic Anglo-Saxon tongue common in Seattle, laced as it is with lingering Norman malapropisms.

We are then agreed? Then, as they say in Basque: Kumbahyah.

Such theoretical harmony would not suffer even if one approximates in vocalization the British spelling "tyre," which may or may not rhyme with "gyre," or even perhaps fancifully, but meaningfully, "wyre." If gyres can widen, and certainly tyres can, than perhaps wyres also may widen into eventual entropy.

That the computer is "less" wyred, need only concern us only to matter of degree, ipso facto. We have certainly solved the problem of the intermittent interference, at least on linguistic level. A simple thought experiment would prove that a wyre-more computer would face more interference, but while a wyre-less computer would of course show "less" interference. The burden of proof therefore shifts, does it not.

Of course, if one were to construct a logical rejoinder, one might explain the issue as "less" versus "loess," depending on soil particulate volume in the immediate atmosphere surrounding the computer device, perhaps resulting in malfunction.

But, please. Let's remain logical, shall we?

amoeba said...

LEDs are instant-on devices - unlike a filament bulb, there's no need for them to heat-up, with a LED, as soon as power is applied, light is emitted.

Lights that use filament bulbs are unsuited to being flashed, because it's so inefficient because there is a lag between applying power and the wire filament becoming hot-enough to emit light. This is prohibitive regarding battery life. Most, if not all battery lights using filament lamps would be powered direct, without any electronic circuitry. Therefore it seems highly unlikely that interference problems of this type would occur.

The interference with wireless computers caused by some LED bicycle lights is probably the way the LEDs are 'switched'. By switched, I mean that many (but not all) LED lights, even when 'on constant' are in-fact flashing - this can be seen by waving the light around in the dark. Using this simple technique will enable you will notice that many such lights, even when set on constant are in-fact flashing too fast to be noticed by the Human eye under normal conditions. Those lights that don't flash visibly when waved about in the dark would seem much less likely to cause interference. Video cameras can see the flashing and this why steady LED lights sometimes seem to 'pulse' in videos.

Flashing, whether noticeable or not, will most likely cause Radio-Frequency Interference (RFI). Whether it interferes with a wireless computer or not, will depend upon the computer's design, the light's design and the distance between the two. It may be that some lights suppress the RFI. Digital wireless computer circuitry can reject (ignore) signals that aren't correctly coded (RFI).

There are standards concerning devices regarding emission of and susceptibility to RFI.

It's most likely that the RFI will fall off rapidly with the inverse of the square of the distance. So doubling the distance between light and computer will reduce the interference received by the computer to one quarter. Doubling the distance again, will reduce the interference again by one quarter and so on. (1/4 x 1/4 = 1/16).

Assuming that a headlight interferes with a wireless computer when it's two inches away. So moving it further away, by an additional foot to 14 inches, will increase the relative distance by 14/2=7. The interference received will be reduced to 1/(7 squared) or 1/49th of its original strength, that's 2% of the interference received at 2”. So there's quite a lot to be gained by repositioning.

I hope this helps.

skiffrun said...

Thanks for the humorous post.

sam said...

I had terrible interference 4-5 years ago with my wireless computer on my mountain bike when using a NiteRider HID. Switching to wired or switching to halogen lights worked fine. I've since given up on wireless computers entirely.

It's not surprising that an ANT+ computer would be less susceptible, it's the same reason you don't get static with digital TV.

rob hawks said...

To the list of sources that make wireless computers go wacky, add riding on BART through the transbay tube. While sitting on BART with my bike and travelling through the tube, the speedometer usually registers about 50+ mph even though the wheel is stationary and the magnet is no where near the pickup.

dexey said...

I really enjoyed that post :0)
Were we a little grumpy this day?

HawgFuel said...

wireless cycle computer interference when led light is on flash mode sounds like a pretty good feature to this commuter that finds those blinky white lights to be exceptionally annoying. Turn it back to the regular constant on mode and everything will be o.k.

Anonymous said...

What a coincidence! I just put a wireless computer on my bike yesterday, set off on my way to work after turning on my LED flasher and lo and behold the computer started working in fits and starts. It would update randomly and was generally frustrating to watch. As a last resort I turned off my flasher and the computer started working like it was supposed to! I tested it again several times with the same results.

Glenn said...

I'd be interested to know if Yahoo Answers has ever contained a correct answer. Not interested enough to continue with further research, mind you.

Anonymous said...

Glenn, you might try asking Yahoo Answers if Yahoo Answers has ever contained a correct answer.

If no, then yes.

Keith said...


I have an "older" Polar CS200CAD and just recently bought a new Moon XPower 300 LED front light. This light is definitely interfering with my Polar, if it is on, flashing or solid. Fact, no logic required.

I have tried moving it around, but can't seem to get it far enough away, will be trying more drastic options soon.

Doing some reading, others are experiencing similar issues with the Moon light. Obviously the light is emitting some kind of RF interference, can't afford an ANT+ unit right now, so guess I will need to find a solution with tin foil.



ge said...

I've never been a fan of wireless computers because of the extra batteries. However, I got a Cateye wireless for my Brompton because of the fold. I've had issues with my Planet Bike Blaze light even on solid, and more recently, with my new Cygolite 420 Metro, again on solid. When the light is on, the Cateye will work long enough to lull me into complacency and then display a speed of zero. I "solve" this by tossing my computer into my front bag (farther from the headlight) so at least I know my distance (and other stats).

Brian said...

Good post. In my own searches for answers, I've been surprised at how many people dismiss the issue as cheap lights or computers. I've slowly ramped up on my lumen firepower (they make more lumens now? Hmm, I could use more light...Yes, I've convinced myself I really need it even if I don't!) over the past 5 years and once I made the jump from HID to LED, I started noticing some odd behavior with my speedometer readings. I'm not suffering from cheap lights or computers.

At first I thought it was my blackberry in my front pannier (this was several years ago), which were prone to broadcasting their activities to any nearby speaker/headphone (di-da-di, di-da-di...if you've had a blackberry, you know about this). Nope, it was my beloved light. I've since gone to wired computer, and problems are solved.

Unfortunately, the wired computer market gets the runt-end of features these days. My Cateye enduro is rock solid, but I miss a backlight option (I have bar-mounted light) and I am, of course, missing out on so many of the other cool features that no wired computer gets. Wired seems to mean "entry level" these days.

It seems like it wouldn't be too hard to make a wireless computer that would accept a wired input (sell the harness separately). One can only hope that with the explosion in lower priced LED lights and increasing lumens (price per lumen has really come down in the last 5 years), that this option will be considered or other wireless protocols that aren't susceptible to interference are utilized.

Anonymous said...

I think $100 Serfas light was interferring with Cateye on dimest setting (I didn't check the other 3 levels of brightness or flash). I just moved light and computer farther apart and it seems ok now. PWM (Pulse-width modulation) or something like that I guess. Would've been interesting to see if brightest did it too but I didn't bother since I wasn't going to use that always.