Has anyone checked speedo accuracy?

[RichDoyle]

Gen I & II are generally 5 mph off at 70 indicated....Gen III is about 2 mph off at the same indicated speed.

My Gen III is also off about 2 mph at indicated 70. I won't dispute the GPS description of RossKean above but I was always under the impression that a GPS is receiving multiple satellite signals at one time thereby giving you a very accurate reading of speed. I almost exclusively look at my GPS for speed instead of my speedo.

The more satellites the more accurate, 'tis true. But there is still some error caused by things like ionospheric conditions. And you don't always have the benefit of the same number of satellites in a proper geometry for the most precise solution. In certified aviation GPSs there is a concept known as RAIM. In these you'll get a warning when it doesn't have enough satellites in the proper geometry for the precision required for the current phase of flight (usually 4 or 5). The pure geometry aspect can be predicted before you depart. It still works but might be off by tens of meters. Not a biggee in cruise flight normally, but this condition in the newer WAAS-based receivers will cause it, in approach mode, to degrade the approach you are using to one that can be supported. You'll never see this in your run-of-the-mill automotive GPS until it can't resolve a 2D position at all, at which time it'll say something like "satellite signal lost".

And it's true it doesn't directly measure speed. It figures out how far you've travelled in some time period. But the screen update can be a larger interval.

Basically, this doesn't much matter when using GPS to assess your speedometer accuracy. You take a bunch of samples over time and come to know "my speedo reads a couple of MPH high at this speed" (or whatever) and you're done until your next tire change, maybe.

 

[Bounce]

We were discussing such things a few years back (in the midst of my 2nd Wing I think) so say mid-90s. It was common back then for manufacturers to set their speedos optimistic to (we all assumed) avoid class action lawsuits (if changing tires, etc.). Better for the rider to be going slower than they thought. Right? Honda (again, at the time) seemed to be pretty much 10% off across their product line (the faster you went the more offset there was).

We started using "Actual" vs "Indicated" when posting about speeds so it was clear what source we were using for our info.

Different manufacturers have different offsets but it continues to this day for the same reasons.

When the speedo/odo use the same source to derive their readings, modifying the speedo (upwards typically) can also modify your odo (leading to early warranty expiration). Because of that most of us just kept in mind the variance and did quick/crude math in our heads for a guesstimate.

If you want a true(er) reading (without the weaknesses described about GPS above), find a stretch of interstate, lock in at 60 mph, and time yourself along 5 mile markers. Do the math to extrapolate MPH.

T=time to complete the 5 mile test

5=constant to average the per-mile speed (miles per minute)

60=constant to get 60 minutes of the average MPM

T/5*60=Your MPH (actual)

Some times the old school ways still work. It's also a heck of a lot cheaper than buying a GPS if you don't already have one. It even works when the road isn't flat. The more closely you can lock in on 60 mph the better your results, so don't use cruise controls or throttle locks.

 

[Twigg]

That excellent post about GPS accuracy was spot on, but it is worth mentioning a couple of things.

With he exception of steep hills and very low speeds most of the inaccuracies are tiny, and even at its worst the GPS is likely to be an order of magnitude better than a speedometer.

The flatter and straighter the road, and the higher the speed, the more accurate they become, but it's tiny fractions

One thing to be aware of is the delay in reporting the current speed. It can be a couple of seconds and it is due to the fact that the measurement is historic and the software lag in the gps unit. Again, it's a minor point but worth noting.

 

[RichDoyle]

FWIW. My FJR, Prius, and Tacoma all have about the same offset from what I get from my GPS, which is 1-2 MPH high at the typical highway speeds I come to use.

The Hondas I had that I measured (ST's, NT) were all about 9% high. So 5-7 MPH high at highway speed. One calibrates and sleeps the sleep of the just.

 

[FJR Bill]

After many checks, my 09 reads 7% optimistic. Have you ever heard of a "lawyer trigger" for guns? That's used for new guns that have a heavier trigger than needed, but specified by lawyers. Most that want better performance modify the factory trigger. For motorcycles I just assume 10% optimistic. Easy to compute. If I indicate 70, I know I'm actually going about 63. If I want to be 70 actual I'll ride just shy of 80. Simple.

Bill

 

[HotRodZilla]

Short answer: No...after 12 years of production, NO ONE has EVER checked the accuracy of the FJR speedo.

 

[Fly Bye]

I just use the gps for the speedo, no muss no fuss.

No desire to warp my brain about technical equations when I can just enjoy the ride.

A +1 or 2 mph either way is not anything I am going to get all worked up about.

 

[Mr. BR]

My '05's speedo always seemed a little fast but I never formally measured it with a GPS. After I changed my final drive gearing, I added a SpeedoHealer so I would see the actual RPM drop. Going on the assumption the tach was fairly accurate, I set the SpeedoHealer to indicate 60 mph at the calculated 3100 rpms. It feels right going down the freeway and the accuracy seems better than before but I still haven't confirmed it and probably won't (I figure it can't be too far off and my 1-way odo miles to work are the same as before).

Mr. BR

 

[Rider]

I thought that the way a GPS actually calculates speed at any given moment is by measuring the doppler shift in the signals from the satellites?

 

[ionbeam]

Over the last weekend I had lots of time to leisurely compare the speedometer of my Gen I to the display on my Zumo 550. It turns out that at 60 my speedometer is only off by around 2 mph. It seems that there is more lag in my speedometer needle than there is in the GPS readout. Pillion discovered where the speed is displayed on the Zumo and now can read the speed over my shoulder. Now pillion knows how scared she should be

I thought that the way a GPS actually calculates speed at any given moment is by measuring the doppler shift in the signals from the satellites?

Radar guns measure speed by measuring phase angle, GPS measures position by the 'time of flight' of the signals from the GPS satellites. The communications between the satellites and the GPS is amazingly complex starting with patterns and then lots of calculating to correct for signal variations and to keep the GPS receiver in time synch with the satellites. It amazes me that it even works, let alone works so well.

 

[Fred W]

As for why the optimistic speed readings?

Consider this: Since the speed signal is derived from the output shaft's rotation speed, the speedometer's accuracy will be directly influenced by the overall wheel circumference. Not all tire manufacturers make a particular sized tire with identical outside circumference. The rim size must be 17", but the variability of the 180mm width / 55% profile can result in a pretty wide range of outside diameters A tire with a larger circumference will turn fewer times per mile and so the speedo will become less optimistic.

So, rather than risk the liability of a rider ever being shown a speed that is even 1 mph slower than their actual speed, and opening themselves up to lawsuits due to that error, they make the speedo optimistic enough that any tire of the prescribed size (or even running one of those dag nabbed car tires!) won't cause a low reading speedo.

 

[mcatrophy]

As for why the optimistic speed readings? ...

In the case of the UK (and probably Europe), the law states that the speedometer must be within a certain accuracy at certain specified speeds (used to be15% at 30 and 70 if I remember correctly, will have changed to metric now), but must never read slower than the actual speed. I would expect the USA to have similar legal requirements.
So, expect your speedometer to read fast.

 

[Twigg]

As for why the optimistic speed readings? ...

In the case of the UK (and probably Europe), the law states that the speedometer must be within a certain accuracy at certain specified speeds (used to be15% at 30 and 70 if I remember correctly, will have changed to metric now), but must never read slower than the actual speed. I would expect the USA to have similar legal requirements.
So, expect your speedometer to read fast.

It's in the "Construction and Use" regs, and I believe it is 10%. The US appears not to have similar regs or all those wide wheels would be a ticket because they are not covered by bodywork.

That is why UK cops do not ticket anyone less than 10% over. Go over that though, and you'll be done for every mile per hour you exceed the limit.

 

[mcatrophy]

...That is why UK cops do not ticket anyone less than 10% over. Go over that though, and you'll be done for every mile per hour you exceed the limit.

Doesn't follow, if you are exceeding the speed limit then your speedometer will be reading way over. Speedometer inaccuracy is no defence.

Yes, UK cops usually allow a bit over, it's unofficial, but something like 5% plus 3 mph. Cameras usually have a similar allowance. But you can be done for 1mph over. Unusual, but it can happen.

Now, in France, 1kph over and, particularly if you have a UK number plate, you will be fined, on the spot (or at the nearest ATM, they will escort you there). If you can't pay, your vehicle can be impounded.

Edited to add that my '14's speedometer is disappointingly inaccurate, around 5 or 6 mph at 70.

 

[Twigg]

The ruling is "accuracy", not optimism.

They could be 10% under-reading and comply with the regs ... Hence the allowance.

 

[HotRodZilla]

It has been a while, but if I remember, US speedos can be +5% and -0%. I may have that backwards but basically, it can be 5% optimistic but is not allowed to read under actual speed at all. So, naturally, all factory speedos are a hair optimistic.

All they have to account for is factory specs, meaning whatever was installed on it when it left the factory. If the owner chooses to change tire sizes, the change in reading is on them. Honestly, it takes some major changes for the speedo to be significantly wrong.

We see the most error on trucks that came with normal sized tires but fitted with 35s or bigger. Even then it's only a couple MPH at an indicated 70MPH. My '93 Ford truck had a process for changing the speedo ratio, dependent on tire size. It could be done 8x over the life of the truck.

Don't stress it, just understand it's there and account for it if you want. I use my GPS because it irritates me.

 

[mcatrophy]

Found the regs for the UK.

I quote:

19 Speedometers

1. The vehicle shall be fitted with a speedometer capable of indicating speed in mph at all speeds mph. up the maximum design speed of the vehicle.

2. For all true speeds up to the maximum design speed of the vehicle, the true speed shall not exceed the indicated speed.

3. For all true speeds of between 25 mph and 70 mph (or the maximum design speed if lower), the difference between the indicated speed and the true speed shall not exceed

V / 10 + 6.25mph

where

V=the true speed of the vehicle in mph.

 

[RossKean]

V/10 + 6.25

Only in a country that is not fully decided whether it is Metric or otherwise!

10% of the actual plus 10.0 km/hr

 

[mcatrophy]

...Only in a country that is not fully decided whether it is Metric or otherwise!...

Tell me about it.

I posted somewhere about the UK policy on metrication. Can't find it at the moment, I'll find the link when I'm back at home. (Touring in Scotland for a few days

)

 

[mcatrophy]

...Only in a country that is not fully decided whether it is Metric or otherwise!...

Tell me about it.
I posted somewhere about the UK policy on metrication. Can't find it at the moment, I'll find the link when I'm back at home. (Touring in Scotland for a few days

)

Ok, found the link (actually two), here and there.