I have Bell wireless computers (same model) on both of my bikes. I have a route that I ride near my office at least twice a week, and the computer on my road bike always goes crazy at this one spot after I crest a big hill, speedometer reading 40-50+ MPH and going all over the place. One time I stopped, and the computer just kept going crazy and wouldn't quit. I had to keep riding a little farther before it would correct itself. Next time I rode the road bike, same thing at the same spot, but I just kept going and it corrected itself shortly.

When I rode the same route on my other bike with the same model computer, it never blinked once. And I've had that computer longer on the same battery longer than the one on the road bike.

Any ideas? This is the spot where it happens. Nothing crazy around, just houses and trees: https://goo.gl/maps/3E0Po

It is a wireless device which has a receiver which is susceptible to RF interference. There is probably another wireless transmitting device in that area which is interfering with it. Not much you can do about it except avoid the area. Read the "FCC Warning" section of this instruction sheet, which is common to all types of unlicensed wireless devices. https://www.bellbikestuff.com/pdfs/11...lessManual.pdf

Call the guys from area 51. You've discovered a hidden source of RF radiation in the operating band of your computer. Maybe it's an alien vessel hidden in one of those innocent looking houses,

Simple wireless computers do that in the presence of EM fields from power transformers, overhead power lines, the eddy loss sensors used to trigger traffic lights, industrial equipment, some LED head lights, etc.

Upgrade to a better wireless computer which has sensors that speak the digital ANT protocol where the worst that will happen is a drop out (very rare at 2.5GHz vs the lower frequencies simple older computers used) instead of ludicrous readings that throw off your totals and averages. A few hundred dollars on a Garmin Edge 200 or 500 should do the trick nicely.

Wired computers cost less, only use a single battery, and don't get jammed by martian mind-control waves.
Ditch the silly wireless, wired wins hands down.

Just seems odd that one computer does it in that same spot and the other one doesn't when they're the same model.

I should be heading out again tomorrow and I'll see what happens there.

Wire dress has a big effect on RF pickup (I had problems with one active speaker receiving 94.9FM from a nearby 100KW radio station until I made minor rearrangements to its internal wiring) and there could be running changes in the new computer.

It is common for manufacturers to make running changes on their circuit boards so your two computers may have different "guts" even though they are the same model. Perhaps they left out a component to save a half a penny which makes one work differently, or maybe they are on different RF frequencies. As noted above a wired computer will eliminate the problem.

+1 on wired computers. Cat-eye's Mity 8 and Enduro 8 (same head, Enduro has a heavy duty wire) are wonderful wired ccomputers and now at close-out prices since they've been replaced with newer models. Buy two and never have this problem again.

Wireless devices, particularly ones that aren't properly "coded", are prone to all kinds of odd-ball interference. A friend had heart surgery several years ago and bought a cheap Nashbar heart monitor to assure he didn't exceed his doctor's advice on maximum heart rate while riding during his recovery. We rode together the first time he used it and as we rode past a power company transformer station his monitor jumped to 250 bpm and then to zero! Talk about scared! It took a moment for us to realize what happened and the reading went back to normal a few blocks later. He later bought a Polar ($$) coded monitor and that ended the interference problem.

Another +1 on wired. Additional bonus is, when you get that record cold spell of -22C in Memphis, the less batteries the better.

What does that mean? I thought traffic light sensors were all triggered by weight. Chatting with another cyclist here in San Diego, he told me over the last few years, many sensors have been tuned to detect cyclists, not just cars.

No, newer sensors detect metal passing over them, not weight.

The loops in the pavement are inductors driven by an alternating current. The resulting magnetic field causes current to flow in nearby conductive (usually metal) objects which in turn changes the loop's inductance and resonant frequency which is picked up by the traffic light controller. The circuit can be adjusted to trigger based on how big the change is, where the losses are higher when there's more conductive material nearby and a bicycle needs a much lower setting than a 4000 pound SUV.

The substance only needs to be conductive - it needn't be magnetic too. Aluminum rims trigger the circuit.

I'm somewhat curious about how things work for people with carbon rims on carbon bikes (although the epoxy is an insulator the carbon fiber it surrounds is a conductor).

That is very interesting! So probably me on my bike would trigger it easily (if tuned for bikes not just cars), but a 300lb pedestrian with only about enough metal to trigger an airport security metal detector (belt buckle, shoelace eyelets, Vietnam shrapnel) could stand in the intersection and not trigger a green light?

Sure enough, computer went crazy at the same spot again today. It's just a residential area, with no traffic lights or any kind of overhead power lines or anything. Today it swept up past 70 MPH lol. After a few hundred feet it settled back down.

buy something less cheap?

Perhaps one of the homeowners is a ham radio operator or has some other RF source. As recommended, get a wired computer or a better quality wireless.

Crest of a hill, huh? Might be a microwave link that's designed with too little link margin. Look for two towers in opposite directions. One is probably fairly close to affect your bike computer that much.

Be glad you don't live there. Are the local kids kinda weird-looking? Six fingers, that sort of thing?

Most likely (as already stated) you are temporarily near a more powerful source of RF wireless transmissions, and in the same freq. as your bike computer. Lots of homes have wireless now for computing as well as other devices.

Why not both bikes? The wireless bike computers most likely are set to transmit and receive on specific frequencies, and the manufacture probably has a number of frequencies set in their individual devices. If all of them of that brand used only one frequency, and you rode close by the other computer, you might get the data from the wheel on the other bike. I don't know what the reception range might be, so this is speculation.

Never mind- I just looked at the satellite image. Looks like a bunch of 22000 volt power lines overhead nearby- jct of Green Tea and Palmer. That can do it. A slight bit of corona on an insulator will spew all kinds of frequencies out, including the one that the wheel sensor sends to the computer. If you hear a buzzing from the lines, that's a pretty damning indicator of corona. Also, I see a transformer on the street view nearby. A (RF) leaky transformer can put out some pretty horrendous fields as well.

Different computers will have different sensitivity even if they're on the same channel. Also the bike structure acts as an antenna or director of waves, so a different bike will induce different local fields.

Ah I see that now. It usually doesn't go crazy until several hundred yards after I pass those overhead lines, though.

Oh well, may look into different computer I dunno. I actually had planned on going with something different on this road bike, and really only got the Bell computer as more of a temporary solution so I could go ahead and start keeping track of miles straightaway after purchasing the bike. But I never had a problem with it until this came up a couple weeks back.

Do you ever feel "fried"?

I second the recommendations above for older Cateye wired. Various members of my family have a MITY 3, a Velo something, and, um, another one. One of them has only one button, that's for the youngest son. I have the MITY 3, and the wife has another one with two buttons that is not a MITY3, but it does have two buttons, and for the life of me I cannot find any difference between them. Every operation is identical, all three of them share the same mount, etc.

BTW, both of my 2-button ones (MITY 3 included) are two-bike capable, which is sweet, because I can mount harnesses on my road and mountain bikes, and just pop the computer back and forth and I have just one odometer to track everything. Just a few button presses to switch between wheel circumferences.

I have a cateye wireless on one bike and a trek wireless 8 on my other. Never had a bit of trouble unless battery needed changing. I have 15000 miles all over the place never any interference. They are dead accurate by wheel rollout method. Typically my garmin fourunner is off by less than 1/2 of 1%. I did 30 the other day on the bike garmin showed 29.96.
I will keep my wireless.

There is nothing inherently more or less accurate between a wired and wireless bike computer. They both have pulse generators and a time base clock that use a user set wheel circumference to calculate distance and speed. Typically you get one pulse per wheel revolution and if the pulse it received by the reading head, they read identically. The only problem with wireless is stray RF can interrupt the pulse count or add to it. If there is no interference or the computer is made to ignore it, it reads properly.