RidingMatthew

I just got off the phone with Garmin and they said that most people don't need a speed sensor unless you are mountain biking.. interesting.


I had been running one on my commuting bike (cyclocross) but when I switched wheels a while back it has been pause/ resume on its own even while running 10 more mile miles per hour. They suggested I put it on the front vs back and see if that works or just move it to my mountain bike because "if you are riding in a straight lines you really don't need it." Mountain biking you do due to tight turns and such.


WHat do you all think? agree disagree? thoughts.

caloso

I don't use one. You will get some wonky speed readings as you go through trees but it's not that important to me to have accurate speed second to second.

TimothyH

Curves and straight lines have nothing to do with needing a speed sensor.

Roads and MUPs have tree cover too.

I do weekly training where pacelines are part of the ride. It is important to not pull too fast or risk blowing up. Having GPS based speed tell me I'm going 13 when I'm really pulling at 22 isn't helpful.

Shimagnolo

The speed sensor makes the displayed speed much more responsive.
Without it, there is a noticeable lag when accelerating/decelerating.

Marcus_Ti

Einstein special relativity combined with spherical trigonometry calculations of radio signals....are a bastard to do fast.

Wheel sensor uses less battery power and is more responsive. GPS is good enough for a car where your speeds tend to be fairly consistent over short timespans.

Steve B.

How badly do you need to know your speed every second ?.

I use my Garmin 810 mostly on the road, with the cover and is maybe 10 seconds or so behind what the Cateye wireless is saying, but really don't need to know instantly.

I do find the Garmin to be accurate for distance to roughly 1/2 to 1 mile over 100, so good enough for my needs. My one mt. bike ride pretty much was dead on for accuracy even in heavy tree cover.

Seattle Forrest

The GPS is running anyway, it doesn't turn off if you use a wheel sensor. But, if you use a sensor, you have to power a radio receiver and maintain a connection with another sensor. How does that use less battery power?

caloso

It's a well known fact that pacelines were impossible until a reliably accurate cycling speedometers were developed.

Marcus_Ti

Because you're not having to do constant real-time differential calculus of the relativistic spherical trig to calculate speed. You already have speed. A ton less number-crunching to do.

TimothyH

I never said that pacelines were impossible before cycling speedometers. Pacelines were impossible are your words, not mine.

I said that an inaccurate device isn't useful. "Isn't useful" were my words.

The point was inaccurate devices, not pacelines.


-Tim-

prathmann

All the complex real-time calculations have to be done anyway to determine your position since the GPS records your track. Once the unit has your position at different times it's trivial to calculate speed. But, especially in areas with poor reception, the GPS speed readings will fluctuate more than those determined from the wheel sensor.

TimothyH

The bottom line is that the Garmin employee on the phone didn't know what he was talking about.

Seattle Forrest

Is this your theory, or has it been tested? If it's a test result, can you tell us how much % difference we're talking about? And which computer?

I think it has to do the constant real-time differential calculus of the relativistic spherical trig to calculate your position. Using a speed sensor won't stop it from doing that, because you're still recording a map.

JohnJ80

^this

Steve B.

What you are saying is that most people do need speed sensors ?

I'm guessing the Garmin Guy is correct in that for "most" people and except for specific uses such as pace line riding, or possibly mt. biking in certain conditions, the speed determine solely from GPS is accurate and timely enough for most uses.

I do know that when pace line riding, I'm not spending a whole lot of time looking at my computer as I'm really paying close attention to what's happening in front of me. I will glance occasionally at the 'putter, especially nearer to when I'm about to take a pull, so I know what speed to hold. I haven't yet found that the GPS is off enough for me to notice "what the hell, 12 ?". It's when I'm accerating and slowing that I'll notice the slight lag as compared to my Cateye, but when sitting in a steady group, neither computer is varying by that much.

JohnJ80

He doesn't know what he's saying and it was a silly, if not stupid, comment. The GPS speed updates much slower and is much more error prone because it depends much more on environmental factors. Why would a mountain biker have more need of precise speed and have any real difference in environmental issues than a road biker would?

My experience with Garmin support is not good. In general, I know more about their products, it seems, than their tech support (first level) does.

J.

cydewaze

Funnily, I use one on my road bike (it does cadence too) but not on my mountain bike.

Seattle Forrest

Yep. Having/not having a speed sensor (above and beyond GPS) makes no difference in my cycling life.

njkayaker

???

The GPS is already determining position. It's easy and fast enough (especially, at short scales) to determine speed from that using simple geometry.

woodway

Between tree cover and twisty trails, my GPS distance record is pretty consistently 10-20% low when mountain biking (I still have an old hardwired cateye computer on my mountain bike for a side-by-side comparison). GPS for road riding seems pretty accurate but sometimes under heavy tree cover the speed reading will go wonky.

John_V 

Reading some of these responses, it appears as if there is some confusion here as to how a GPS device works. In 2007, I was the primary software developer for an engineering firm that was awarded a FEMA project that required accurate GPS positioning and tracking so I'm quite familiar as to how they work. I'll try and explain this as simple as I can and not get too technical.

GPS receivers scan for satellite signals and lock onto the strongest one. The signals are nothing but radio waves and like all radio waves can be blocked by any kind of interference. For this reason, GPS devices usually look for at least 4 satellites to get data from. The radio signals are converted to binary data by the receiver. The binary data is again converted into a byte array of multiple elements. Each element contains specific data that is grouped by type. For example, one array element may contain time data and another lat and long data and so on. The data in each array element is length specific so, as an example, the first 4 bytes of the time element may contain the date and the 5th byte will start the time of day and so on. The firmware in the GPS device will have to parse the byte array into each element than pull whatever data it requires from a specific element from the exact starting and ending position for it to do whatever calculations it needs to do in order to display the data on the screen. Cycling software on a phone has a few more conversions and steps to do to get the same results; so processor speed makes a difference on lag time. Using GPS, speed is determined by distance, calculated from a position point and time data, taken from two of the array elements, between each read of a satellite signal. All of this is going on at the same time the firmware/software is using the lat and long to pinpoint your position on the map and anything else that is required. If the satellite signal is weak or lost, the receiver will have to scan for another signal that is stronger than the previous one so during that time period, nothing is being parsed and your display screen may be showing the same data until the receiver locks on to another signal. This is one hell of a lot more processing and calculating than calculating wheel rotation and circumference. When using speed sensors with a GPS device, almost all devices will use the speed sensor over the GPS so that it doesn't spend as much time parsing and processing speed and distance. If the speed sensor loses connection or fails, the device will automatically switch to GPS until a signal from the speed sensor is found. This is true of cycling apps as well.

Steve B.

Thanks for the info., John but a question.

Does anybody know for sure how a Garmin 500-520, 800 - 1000 series device, that is providing on-screen mapping (thus needs GPS) as well as speed, actually works ?.

If it's providing an on-screen map, it's going to be doing all the GPS data collection to generate position and thus mapping, as well as and as a side effect provide speed.

I would suspect that the software will use the speed sensor data as the "better" data, in place of the speed data coming from the real time calculations of GPS "every one second" breadcrumb style tracking. But I also suspect that the GPS speed tracking is going to occur even with the presence of a speed sensor, if only in the event the speed sensor sends erroneous data or craps entirely.

Thus I suspect there's no saving of battery life by using a speed sensor as I suspect the unit is GPS data processing for speed at the same time, just for giggles.

JohnJ80

I don't think I'd agree with your assumption. If the GPS isn't having to provide instantaneous speed readings, it is going to be possible to relax the reporting interval/duty cycle and use the speed sensor data in between the relaxed GPS fixes. That should pay huge dividends in battery savings. Generally, in portable devices like this, it's important to take advantage of all the tricks one can because there is never enough battery and it always must be conserved.

J.

Steve B.

One problem with your premise about saving processing power, is if the device has a map, that would make it necessary to have location as up to date as possible and shown on the map. Are they then going to throttle the speed section of the processing ?. Doubt it especially as it's possible to view speed and map on the same screen. Somewhat academic questions as we really don't know how Garmin or the others do this.

njkayaker

I suspect it deals with GPS data at the same rate with or without the wheel rotation sensor.

I doubt there would be "huge dividends" in battery savings either.