Smiths digital speedo accuracy

I’ve had the Smiths digital clocks on the Commando for a while now, also fitted them to the T120. I’ve always been very happy with them.

I very carefully calculated the tyre circumference and set them up according to the (pretty terrible) instructions and left them at that.

However, recently someone posted the gearing calculator web site... I’d been thinking about my gearing as it feels under geared as I was getting 60mph at 7000rpm, so I punched all the data into the web site to consider sprocket options.

The web site said I should be doing just over 70mph at 4000rpm! I double, triple checked everything and came up with the same numbers each time.

So I went for a ride with the sat nav fitted and yup, you guessed it, the speedo was reading WAY slow. I’ve been lucky not to cop speeding tickets on this thing.

So now it seems I have to go through a load of trial and error changes to correct the speedo calibration.

Franky, I’m rather pissed off. What’s the point in expensive ‘high tech’ parts that don’t even fulfil their basic functionality?!

Maybe the rev counter is out as well? I have a hand held rev counter that pushes onto the end of the crank, I’ll give that a go soon although, obviously, that’s only a static test. The rev counter is not adjustable, so if that’s out I’ll be even more pissed off.

I suppose that on the bright side, my bike is actually quite a bit faster than I thought it was...!

Hmm, it may be that the tyre circumference is wrong - it will grow at speed, but I simply don't know how much.

Yonks ago I had a rather neat little circular sliderule which did this sort of gearing calculation. You set up the primary & secondary gearing, choose a tyre and it gave you the results. There was a note on it mentioning that the tyre circumferences given were the dynamic values, but I have no idea of how much they differed. This was as I recall, a must useful give-away thing, either from a magazine or from Dunlop, where my father worked for many years.

Perhaps you can work backwards from true speed & rpm and see how fat off your tyre dimensions are from your initial calculated values?

/Steve.

Mind you, 'published data' seems all over the place as well. In my Mk3 Workshop Manual one page shows the gearing table quoted elsewhere here, with a 'theoretical' speed figure with 20t gearbox sprocket of 97mph at 6000rpm.. (16.16 per 1000)
Following page states: Road speed at 1000rpm in gears (20t sprocket): Top: 16.92mph.

Not earth shattering but still another 'duh' moment from NT

I don’t “calculate” the circumference.

I mark the tyre where it’s touching the road, push the bike along the road for ten turns of the wheel and measure that distance.

Dividing the distance by ten is a small calculation.

I used google maps to "measure" a mile on a local road, then rode it using a GPS to check the stop point, then calibrated the speedo.
It must be perfect because i haven't checked it since so it cant be wrong.

I have never come across a worse set of instructions than the Smiths Digital Speedo and Tacho ones.

I have also never come across a ruder set of people either, they were worse than Les Emery!

When we set the instruments up on the 650SS I used an app on my mobile phone to measure out exactly one mile from a chalk line outside the house to a chalk line round the block.

It struck me that measuring over that kind of distance would provide an accurate setup.

Tested afterwards following the car (whose speedo I also verified against the GPS) and it seemed spot on at a few different speeds.

I agree on the instructions, i hadn't mentioned that it took about 14 attempts before the stars aligned.

Steve, the beauty of the gearing web site https://www.gearingcommander.com/ is that you can play with all factors, inc tyre diameter.

It seems my speedo is 10mph slow at 4000rpm.

According to gearingcommander, in order to make it read 10mph more at 4000rpm, I need to input a tyre circumference that 11.5” bigger than actual ! So, I don’t think tyre growth is the issue here.

Triton Thrasher, I have measured the circumference, and calculated it from the Avon website where they give the rolling radius. There is a small difference, but no where near the difference required to cause my error.

gtiller, ruder than Les Emery? Now we know you’re bullshitting !!


Maybe the accurately measured mile is the way to go, I’ll try that first me thinks.

I used to run my front mudguard very close to my front tyre
But I found when I was above 90mph the tyre would rub
I had a little over 1/4" clearance so the tyre diameter increased by at least 1/2"
This is a 90/90/19 road rider
I raised the mudguard a further 3/16" and the problem stopped

Fast Eddie said:

I’ve had the Smiths digital clocks on the Commando for a while now. Franky, I’m rather pissed off.

Click to expand...

I posted this on another thread - but silly me doesn't know how to deliver a link - so, hope this helps...

I have just completed the calibration process on mine and found an easier way.
I downloaded a GPS app called SpeedBox onto my phone then duct-taped it (with padding) to the fuel tank.
After each run, where the speedo was reading high, I would stop beside the road and go through the "Set PPU" process and increase the number. Started with 4705, which I got from calculations, and gradually increased to 5180, which is pretty close to bang-on for me.
Much easier and more accurate than measuring wheels or rotations per mile!
Hope this helps someone
Cheers
Rob

robs ss said:

I posted this on another thread - but silly me doesn't know how to deliver a link - so, hope this helps...

I have just completed the calibration process on mine and found an easier way.
I downloaded a GPS app called SpeedBox onto my phone then duct-taped it (with padding) to the fuel tank.
After each run, where the speedo was reading high, I would stop beside the road and go through the "Set PPU" process and increase the number. Started with 4705, which I got from calculations, and gradually increased to 5180, which is pretty close to bang-on for me.
Much easier and more accurate than measuring wheels or rotations per mile!
Hope this helps someone
Cheers
Rob

Click to expand...

You obviously have a permanent switch wired in to facilitate roadside adjustments. I don’t !

I fitted a Smiths electronic with a GPS sender and by the instructions with each unit calibrated according to the instructions it was also well out speed wise. I then tried the ride over a measured kilometre and set the speedo accordingly. It was spot on that way. Strangely enough when I contacted the Australian agent that I had purchased from about the error, I also received a rude response. There is also a Chronometric replacement speedo they supply that is not a direct replacement for the chronometric. From memory the body is about 4mm larger than the original, so the electronic replacement will not necessarily fit original chronometric speedo holders. When I brought that to Smiths attention I also received a rude response. While I am happy with both speedos, they are overpriced and the back up is such that they will not get any more of my business.

Ando

Thanks Ando, another advocate for the measured mile / kilometre method.

When are companies gonna learn that you can’t just continue pissing off your customers and expect them to come back to spend more of their hard earned ?!

All of the R&D, set up, marketing, brand management, etc, destroyed in an instant by dick on a phone !!

baz said:

I used to run my front mudguard very close to my front tyre
But I found when I was above 90mph the tyre would rub
I had a little over 1/4" clearance so the tyre diameter increased by at least 1/2"
This is a 90/90/19 road rider
I raised the mudguard a further 3/16" and the problem stopped

Click to expand...

But not necessarily the load bearing radius.

concours said:

But not necessarily the load bearing radius.

Click to expand...

Yeah I see what you mean,

when I first heard the noise I thought it was engine/transmission related
Seemed odd it only made the noise above 90mph
Then I noticed the paint had bubbled off the mudguard in the centre!!

gtiller said:

I have never come across a worse set of instructions than the Smiths Digital Speedo and Tacho ones.

I have also never come across a ruder set of people either, they were worse than Les Emery!

When we set the instruments up on the 650SS I used an app on my mobile phone to measure out exactly one mile from a chalk line outside the house to a chalk line round the block.

It struck me that measuring over that kind of distance would provide an accurate setup.

Tested afterwards following the car (whose speedo I also verified against the GPS) and it seemed spot on at a few different speeds.

Click to expand...

I'm finding it hard to believe you found someone ruder than "the beard"

My "issue" wasn't with a 'dick on a phone' it was with their management.

When they first bought their digital tach to market, the only option for a pulse signal was to tap in to the electronic ignition.

Not an option on a magneto bike, and they weren't interested in helping.


So I bought and tested a couple of different induction sensors that could be attached to the spark plug wire to pick up the signal.

I sent them a couple of detailed mails about my tests and findings, and heard nothing back.

A few weeks later I received an email, that clearly wasn't meant for my eyes.
It was a conversation between their techies and management talking about what I was doing, how useful it would be, and not to reply or acknowledge me because they would just take a product straight to market.

Way to go boys!

[QUOTE="gtiller, post: 474206, member: 4606"
A few weeks later I received an email, that clearly wasn't meant for my eyes.
It was a conversation between their techies and management talking about what I was doing, how useful it would be, and not to reply or acknowledge me because they would just take a product straight to market.
Way to go boys![/QUOTE]

...sigh!

Looking at how the static calibration works, one wheel rotation, this seems to be the least accurate method. Effectively it is telling it a set number of pulses and assumes a set internal clock speed that would them over time, but nothing considers the variation in internal clock speed, its internal stopwatch. ie this method tells distance travelled, not that great as it does not give an accurate speed that that distance was travelled. If your DC supply to the speedo varies, the internal clock speed could vary, and thus the inaccuracy.
It seems that the 1 mile method seems to not only count the pulses but the time taken as well, giving the more accurate calibration.
I might be wrong, but it may add something.

Triton Thrasher said:

I don’t “calculate” the circumference.

I mark the tyre where it’s touching the road, push the bike along the road for ten turns of the wheel and measure that distance.

Dividing the distance by ten is a small calculation.

Click to expand...

Yes, but unless you are travelling rather fast you are still measuring the static circumference.
I wonder if there's a growth factor somewhere that can be used?