Motorcyle Speedometers and Magnetic Fields

I have no clear idea of what the field looks like, but I have a suggestion on an experiment to find out. Get a regular bar magnet with the poles marked. Hold one of the poles near the rim and see if the two magnets repel or attract each other. No matter what the result is, run the bar magnet around the rim and see if the behavior changes. You could also hold one end of the bar magnet near the center and see what happens.

What I suspect is the polarization runs from the center of the magnet to the rim. If you have a problem visualizing this, I'll try to work up an illustration tomorrow. By 'half round hollow' I guess you are talking about a hollow hemisphere.

Hello 3Doug:

At this point it is just an academic exercise. I did talk to one speedometer shop that seem to know a little bit about the hemispherical magnet, besides not wanting to touch it, he said it would require 4 electromagnets to work with it.

So I increases pre-load, (it was reading about 10% fast) and put it back together. So while you have a good idea , I'm past that point, I'm now simply curious as to what the field looks, or for that matter why it will require 4 electromagnets re-magnetize or de-magnetize a hemispherical magnet.

Possibly the 4 electromagnets requirement will give you an additional clue, one thing for sure I'm clueless.

I haven't started on my illustration yet, but this one from Wikipedia might give us a clue as to what we might find:




Imagine the bar magnets bent in a radius to simulate the hemispherical shape of the speedometer magnet.

Hello 3Doug:

Possibly this diagram explains why the speedometer shop said 4 electromagnets were required. The Japanese always have to figure out some weird way of doing something, US cars typically use a rotating bar magnet, that is if they have a mechanical speedometer.

I do appreciate your time in helping me conceptualize what this field looks like.

Actually, I enjoy these kinds of exercises. Creating the illustrations gives me a chance to use AutoCAD or Inventor, and to keep my skills up.

Here's the illustration of the test I proposed:



What I imagine will happen if you hold the bar magnet this way and move it around the rim, it will alternate between being repelled and attracted to the rim by the poles of the hemispherical magnetic field. I can also see the possibility of more than 4 poles, but not more than 8.

I learned something in making this - how to emboss something onto a surface in Inventor. I did have to draw the 'N' because I couldn't get it to show as a text feature.

Dear Sir.

I had been working in a magnet lab for some time & have seen magnets of a variety of shapes & poles.

This may have circumferential magnetized.

See your image with my notes.

You are well qualfied to redraw it with idea what I raised.

Regards

Hello Haajee:

Your addition to 3Doug's diagram was helpful, I think the final test as to what is actually going on with this magnet is going to be to find a similar one and physically check it, and experiment on it, in other words I something I can destroy without costing too much.

Regards & thanks for understanding my piont of view.

I can not re-draw it but 4 magnetic poles [points x by 2] will drive the speedometer pointer.

Thanks again.

Hi,

it is too late to do this now. However if you want an empirical method of finding the magnetic-field this is it.

Place the magnet under a sheet of paper, and sprinkle iron filings on top of the sheet of paper. This will give you an outline of the magnetic-field.

Hint: Use very fine filings for weak magnets.

Regards......Ralph.

"One of the methods for calibrating..." Another method (at least in theory) would be to adjust the spacing between magnet and the conductive disc that senses the rotating field. In all ignorance I would have expected the sizes of the magnet and disc to be chosen so that the magnet can be manufactured to a standard state, and the tolerance range covered using mechanical spacing adjustment (even if it doesn't need to cover a range of wheel sizes).

BTW, unless I've misunderstood you, adjusting preload tension would correct the error at one speed only - and could easily increase the error elsewhere (unless the error was in the pretensioner in the first instance).

Hello Physicist:

Unfortunately there is no adjustment available for changing the spacing between the magnet and the driven head. And unfortunately the speedometer mechanical casing is constructed of cast metal, and at $252.12, I was reluctant try and disassemble the entire mechanism.

You are also correct in the way you interpret the preload change. This entire situation ended up as an ad hoc speedometer recalibration, I intended to take it to a speedometer shop I dealt with for over 30 years, unfortunately they've gone out of business in the last six months.

As I previously mentioned a speedometer was 10% fast and pretty much linear. My test apparatus was very simple, I used a battery drill motor which would drive the speedometer to about 28 miles an hour, after attempting to manually increase tension on the spring which is extremely delicate I ended up actually reducing the tension, I repositioned the indicator needle, which rests against the stop at the zero mark to increase the preload.

When all was said and done the speedometer was indicating 25 miles an hour when driven with the drill motor.

I initially tested the accuracy of the speedometer on a radar speed sign, the readings agreed with the sign in the 35 to 40 mile an hour range.

Having not been prepared for this I had no written down figures that correlated speed versus RPM in any given gear, so I had to go from memory.

On a desert trip a couple of years ago I had the motorcycle indicating 160 mph, at red line in fourth gear (or pretty close I wasn't concentrating on the tac reading).

Currently in fourth gear at 80 miles an hour if I were to double the RPM indicated on the tach I would be well over redline.

So these are the current results as to my recalibration. I too share your concern that the speedometer may not be linear, however thus far it seems to be. It seems to be accurate enough to keep me from a getting tickets around town, and not to have to do math in my head to figure out how fast I'm going at normal cruising speeds.

Intend to goad to a friend who has a GPS that attaches to his motorcycle into meeting me at a more suitable location to test the speedometer at higher speeds, of course his motorcycle's top speed will be the limiting factor.

PS. I intend to send him a link to this post, to encourage him to get the motorcycle out of the garage, this should really get his goat. And possibly even get him to take the dog for a walk.

Being pedantic - I think the speedo should still be linear - just have the wrong scaling factor, and so read too slow at low speeds and too fast at high. As regards not attracting speeding tickets, that's probably the way you want it??

It's a shame you hadn't asked this question a while back. I had an old speedo from an 83 yamaha. It had the same kind of magnet. I trashed the speedo because the speedo was trashed. The lens was broken from hail and the needle was broke off from sulight degredation. A toothpick served as the needle for a while.

I'll rummage through my junk and see if I kept the magnet or any speedo parts. I could then do the iron filings trick....

Hello DAG:

It's actually a shame that I didn't make this post before I put everything back together, as I explained in my reply to physicist this became an ad hoc recalibration.

I am now more curious than ever, however I do not intend to disassemble the motorcycle or destroy its speedometer, but there is a motorcycle wrecking yard around here, and from what I understand this hemispherical magnet is typicaly use in Japanese motorcycle speedometers. So hopefully I can find a damaged one for a few bucks to satisfy my curiosity and put this knowledge to use.

Thinking about what physicists had written, obviously I couldn't change the spacing between the magnets and a round driven element, but it might have been possible to retract the magnet out of the driven element, which probably would've been a more proper way to calibrate the speedometer.

The problem with having a speedometer that is inaccurate is that you don't pay any attention to it.

I generally cruise with a faster flow of traffic on the freeways, and completely disregard the speedometer, this has left me shuffling to do some fast math when I've been by myself and noticed that ominous CHP vehicle.

I sent the friend that I referred to a link to my post and have goaded him into being used as a speedometer calibration device (being a motorcycle guy you might be interested in the picture he sent back in regards to taking the dog for a walk, an angry pit bull ) so hopefully my ad hoc recalibration is close enough, I'll be satisfied if it's relatively accurate in the 40 to 80 mile an hour range.

Cool!

I must have chuncked the magnet I had...I went on a "junk reduction spree" not too long ago.

I would be interested to see what he sent back...

There is a film you can buy,,,can't think of the name,6 inch wide

showing magnetic field quite conssistantly,or should I say impressive?

Here is a reply I made to another thread about a year ago. You may benefit from it. These were magnetically driven speedometers. Around 1970 their was an article in an automotive magazine that stated due to the design then in use a speedometer was accurate at only one speed so it was common of manufacturers to calibrate to 35 mph. It was known that below this speed it would indicate slow and above this speed it indicate fast. A 1969 Corvette with 160 mph speedometer was a good example at an indicated speed of 160 mph the actual speed was closer to 135 mph. The calibration being performed at 35 was thought to be a happy medium that would cause the fewest number of violations due to the average speeds driven by most drivers. I hope maybe this helps clear up the logic applied to indicating a vehicles speed. Jerrell