Eddy current speedometers, how does drive cable cause so many issues?

[NeoPTLD]

From my understanding, it is a torque converter that uses a magnetic field as the fluid.

The needle is basically a spring wound indicator and the further you turn, the more reaction force it makes. The higher the input shaft's rotation, the more torque it induces on the needle to read the speed.


I'm experiencing bizarre things with my speedometer.
After I put way too much grease in cable, it wouldn't read over 20 mph. The odometer was still rolling fine.

I wiped it off.. and it works, but it tends to stick around 60mph and wobbles below 40mph.

Since odometer is accumulating total mileage file, it's not a slip.

If the output and input aren't in perfect sync, the output is probably fluttering and I can understand how that would make the needle flutter, but how does it explain all the other weird behavior like delayed response?

 

[PsiStar]

Maybe the 1st thing to do is to verify that the odometer is reading accurately.

 

[Paperdoc]

Agree with PsiStar - if the odometer is doing exactly what the speedometer is doing - accumulating too FEW miles when the needle sticks - your problem is likely in the cable, not inside the speedometer. But if ONLY the speedometer is malfunctioning, then it seems somehow you got some foreign material (lube?) inside it and it is causing the needle and its drive system to malfunction.

The most common causes of speedometer failure are two associated with the cable:

1. The cable itself is broken. Even when this is true, often it appears to work, but not reliably. The broken ends of the spiral cable intermittently snag on each other so that it turns sometime, but not always. Most often when you lube the cable after disconnecting it, you only pull it part way out of the sheath and lube the upper portion only. So you might not see the break if it's farther down. Try removing the entire cable length - check your shop manual, but usually you can just pull it out, although sometimes you must first disconnect it from the transmission. If the entire cable comes out in one piece with its special fitting (often square) on the far end, you've eliminated a break as the cause. Replace the cable.

2. Less commonly, after servicing the cable, you failed to re-seat the far end of the cable in the mating fitting inside the transmission. So that special (square?) end is not seated properly, and the cable is not being reliably driven by the transmission. If you've pulled out the cable and verified it is not broken, give special attention to re-seating it as you re-install.

 

[NeoPTLD]

Odo reads fine. Checked against GPS.

 

[PsiStar]

1st para of Paperdoc's comments ... foreign material inside the speedometer. Although one would think that there would enough of a physical barrier between the end of the cable & the inner mechanics of the speedo.

Why did you feel that you needed to lube the cable? Was the speedo erratic before? What kind of vehicle is this on?

Maybe there is dirt in the other end of the cable. There are gears of some kind at each end so either end may have some crud causing the gears to slip in some way. I have little conviction in this idea tho.:\

Is the cable able to turn freely? It could handle a small amount of torque & then pop loose. This would a explanation for the odo working properly but the speedo erratic.

 

[Modelworks]

Is there a special lubricant that should be used on that cable ? Not done those cables before but I did encounter some cables in the past that were designed not to use lubricant and adding lubricant actually interfered with their functions.

 

[William Gaatjes]

I had the same problem once with my previous automobile.
When i was doing sixty kph or above , my speedometer would start to vibrate and shake and suddenly spin up to about 210 kph. Amazing, because my automobile would not go any faster then 170 kph. At the same time, a noise was present as if something got stuck and released again and slipping.

I always had this idea since i could not see anything wrong :
The eddy currents induced depend on the speed of rotation, . But if close to the end that is connected to the speedometer, the metal cable inside the flexible rotating cable assembly gets stuck inside the outer mantel, the metal cable acted as a spring. When released, the cable sweeps in rotation by creating a slow down and then a short acceleration causing the weird noise and the wrong speedo meter read out. Since the cable is similar as a bicycle brake cable, i assumed something got stuck inside it. Or one of the metal strands broke off because of wear and acted when the speed was high enough as a temporary clamp. I could not feel anything strange about it, but i always assumed that thew problem only would arise above a certain rotational speed. I bought a replacement cable and took my automobile apart. Since i was able to download the service manual in pdf from a not well secured server (was accessible by google, i did no hacking) from Hyundai automobile, it turned out to be a very easy task to replace the cable although it took me half a workday.

 

[NeoPTLD]

I'm not really asking an auto repair question or why I did certain repairs. I'm asking here rather than the Garage forum because its about eddy current motor than auto repair.

I'm not asking for auto repair tips.

The construction of speedometer coil is very different from bicycle cable. It is a coiled spring, so any change in load would cause it to spring back.

In electric motor, 5th harmonic for example would cause a rotating field that turns the motor against the rotation direction and cause a braking effect.

When speedometer cable produces a flutter (back 'n forth vibration but still moving in general forward direction. Picture you're pushing a saw in which you move 1/2", pull back 1/4" and repeat. The net effect is you're moving 1/4".

If you push 1/2", pull back 1/4" many repetitions and you move forward 10" in 2 seconds, the speed integrated over a period of 2 seconds is 5 inches per second. If you move the saw forward without any of the pull back and also move 10" in 2 seconds, its still 5 inches per second.

Correct mean velocity is 5 inches per second, however the question is would an analog eddy current meter still register 5 inches per second or does certain stroke combination or frequency cause it to over-register or under-register?

My guess is that the if the forward lash occurs at higher velocity than back lash, it would cause the meter to over-read and vice versa.

After getting a new speedometer cable, there's no needle flutter, or temperamental inaccuracy. With the old cable, it would sometimes over-report and sometimes under-report (in reference to GPS)

An ideal speedometer produces no flutter whatsoever and behaves like a solid shaft with infinite flexibility. In real world, I believe the effect is affected by the friction bending causes on the spring surface and perhaps spring constant is affected when you bend the cable.

 

[William Gaatjes]

Acceleration indeed. That is also what i was thinking, these short burst of acceleration when the rotating cable gets stuck near where it is connected to the speedometer assembly, the cable gets wound up and then released. You would get bursts in rotation speed.
Although your analog of a saw makes sense, it does not make sense for inducing currents.
The faster the outer magnetic field is changing, the more current is induced in the case of an electromagnetic coil, and the more eddy currents are induced to oppose the external magnetic field.
You can see the speedcup as a single turn electromagnetic coil for simplicity.
When you move a magnet inside the coil slowly, your making slow changes in the "neutral" magnetic field the coils is in, you induce a little current. The rate of chance is low.
When you move a magnet inside the coil fast, your making fast changes in the "neutral" magnetic field the coil is in, you induce a lot of current. The rate of chance is high.
And the rate of chance is what it is all about when it comes to inducing high currents or strong eddy currents. The faster the disturbance, the more dramatic the effect is.


I forgot to mention that the strength of the external magnetic field also matters. But within the speedometer it is a constant of course.
The faster the change in the magnetic field or the stronger the magnetic field, the stronger the induced eddy currents.
Read about the Lenz' law.
From how stuff works :

http://auto.howstuffworks.com/car-driving-safety/safety-regulatory-devices/speedometer2.htm

The drive cable consists of a number of superimposed, tightly wound, helical coil springs wrapped around a center wire, or mandrel. Because of its construction, the drive cable is very flexible and can be bent, without fracture, to a very small radius. This is handy because the cable must snake its way from the transmission to the instrument cluster, which houses the speedometer. It is connected to a set of gears in the transmission, so that when the vehicle moves, the gears turn the mandrel inside the flexible shaft. The mandrel then communicates the rotational speed of the transmission down the length of the cable to the "business end" of the speedometer -- where the speed measurement actually takes place.

The speedometer has other important parts, as well. The drive cable attaches, via a spiral gear, to a permanent magnet. The magnet sits inside a cup-shaped metal piece known as the speedcup. The speedcup is attached to a needle, which is held in place by a hairspring. The needle is visible in the cockpit of the car, as is the speedometer face, which displays a range of numbers from zero to an upper limit that can vary by make and model.

Now let's look at how this relatively simple device actually measures vehicle speed.

As the magnet spins, it sets up a rotating magnetic field, creating forces that act on the speedcup. These forces cause electrical current to flow in the cup in small rotating eddies, known as eddy currents. In some applications, eddy currents represent lost power and are therefore undesirable. But in the case of a speedometer, the eddy currents create a drag torque that does work on the speedcup. The cup and its attached needle turn in the same direction that the magnetic field is turning -- but only as far as the hairspring will allow it. The needle on the speedcup comes to a rest where the opposing force of the hairspring balances the force created by the revolving magnet.

What if the car increases or decreases its speed? If the car travels faster, the permanent magnet inside the speedcup will rotate faster, which creates a stronger magnetic field, larger eddy currents and a greater deflection of the speedometer needle. If the car slows down, the magnet inside the cup rotates more slowly, which reduces the strength of the magnetic field, resulting in smaller eddy currents and less deflection of the needle. When a car is stopped, the hairspring holds the needle at zero.

Since i am a bit of a geek who gets of on this, here is the full explanation and another example :
http://en.wikipedia.org/wiki/Lenz's_law

An induced electromotive force (emf) always gives rise to a current whose magnetic field opposes the original change in magnetic flux.

If the magnetic field of current induces another electric current, , the direction of is opposite that of . If these currents are in two circular conductors and respectively, then the currents and must counter-rotate. The opposing currents will repel each other as a result.

Currents bound inside the atoms of strong magnets can create counter-rotating currents in a copper or aluminum pipe. This is done by dropping the magnet through the pipe. When done, the descent of the magnet is observably slower than when dropped outside the pipe.

When an emf is generated by a change in magnetic flux according to Faraday's Law, the polarity of the induced emf is such that it produces a current whose magnetic field opposes the change which produces it. The induced magnetic field inside any loop of wire always acts to keep the magnetic flux in the loop constant. In the examples below, if the B field is increasing, the induced field acts in opposition to it. If it is decreasing, the induced field acts in the direction of the applied field to try to keep it constant.