Etb Speedo Jumping

[Al Milton]

Hi All,

 

I've got an ETB electronic speedo which 'jumps' from above 55mph to anywhere between 90 and 130mph! The speedo is new, but has sat in the box for 2 years during the build so is out of warranty!

 

Been out with the Sat Nav, and up to 55mph it is near as damn it spot on (1mph variation), but as soon as I accelerate over 55, then it jumps up 90-130 and usually stays there until the throttle is backed off!

A couple of times it dropped down to the actual speed, (well by Sat Nav) and was again spot on for a little while before going bonkers again!

Had the car on stands today and checked the magnets and both are still there and very secure, the gap between the sensor and magnets is constant at around 1mm - backed up I guess by the accuracy up to 55mph, so all appears to be working normally.

 

What I did notice though, when turning a wheel to rotate the propshaft to check the magnets/sensor distance, that the wheel would turn without rotating the propshaft for half to three quarters of rev, then it would 'pick up' and rotate the prop. Would this have any effect to the speedo sensor and is this usual for an open diff? (Have got an LSD but as yet haven't got round to fitting it).

Going to call ETB tomorrow, but has anyone got any ideas?

 

Cheers

Al

[Guest jasonmorris]

Is the sensor bracket securely fixed, i.e. no vibration at all? I had a similar problem so moved the whole assembly to run off the near side drive shaft close to the diff. It's been very reliable since and is within 1mph of the sat nav. Other than that there has to be some sort of resistance either in the wiring or the speedo itself.

 

Jason

[sj-bradley]

That does sound like a lot of backlash in the diff. I'm not sure how much though.

 

The magnet sensors should not be affected as when you are driving forwards, this slop in the gears is taken up so there

should not be any jerky response.

 

Is the mounting bracket for the sensor REALLY secure? If it has any chance to vibrate, it will and then the gap will change from 1mm. There is no direction in which it could vibrate ?

 

I'm thinking that above those speeds, there could be a resonance frequency causing the sensor mount to move.

 

It could be that the sensor is too close. remember that the prop shaft is steel - I.e magnetic material and the sensor could be picking up small magnetisations (?) from the shaft itself - hence the high speed.

 

Try moveing the sensor away to 2mm distance instead.

 

Simon.

[David S]

What type of sensor do you have? I had this exact same issue with a Hall effect sensor as they don't really like magnets. I had to remove the magnets and reposition the sensor over the propshaft bolt heads to get the sensor to read correctly. It is still not right, but it doesn't jump any more.

[Al Milton]

Thanks for the replies everyone.

Jason/Simon, the bracket is substantial (3mm angle) and fixed to the diff via the lower mounting diff bolt. It is very secure and inflexible.

 

The sensor is the light duty magnetic sensor. Found some info on the ETB website that backs up both Simon and David in so much that the gap should be increased if the speedo 'jumps'.

 

So will adjust the gap tomorrow and give the car a blast to test, i'll post results later tomorrow.

 

Once again, thanks for the pointers chaps.

Al

[sj-bradley]

What type of sensor do you have? I had this exact same issue with a Hall effect sensor as they don't really like magnets. I had to remove the magnets and reposition the sensor over the propshaft bolt heads to get the sensor to read correctly. It is still not right, but it doesn't jump any more.

 

That's not "strictly" true as hall effect relies on magnetism to work. It's just that strong magnets can easily saturate them which is why people think you don't need magnets.

 

What these sensors actually do is generate their own magnetic field with a coil and when an iron object passes close by, this interferes with the expected field and the sensor can register the fact. It all depends on how your sensor is made up.

 

Hence my earlier post saying that rapidly moving steel parts can produce enough effect to "fire" the sensor when it is not expected to do so.

 

Simon.

[Snapperpaul]

I have a similar problem.

I'm running 4 magnets, have heard reducing to 2 and recalibrating may help

I will try a larger gap first

[Al Milton]

Rang ETB today and they confirmed the gap is too close, should be min of 2.5mm, but they suggested increasing by 1mm increments.

 

So, car on stands when I got home and loads of faffing to get the adjustment. Got it better with a final gap just over 3mm, but have no reading lower than 25mph! :sorry:

Still, have accurate readings up to 75mph and down to 25mph, guess the car will spend more time over 25 than under :spiteful:

 

Must admit to a minor cacking of myself though with the car up on stands and taking the car through the gears to register the speedo reading while doing the various tweaking, was a lot happier out on the road.

 

Once again, many thanks for the suggestions all

[fry61]

When Florin was on stands for speedo set-up she was bolted to the garage doors with a lump of wood;didn't fancy the car motoring down through the garage in 0,01sec with mrs on the go pedal looking for the stop one.

[David S]

That's not "strictly" true as hall effect relies on magnetism to work. It's just that strong magnets can easily saturate them which is why people think you don't need magnets.

 

What these sensors actually do is generate their own magnetic field with a coil and when an iron object passes close by, this interferes with the expected field and the sensor can register the fact. It all depends on how your sensor is made up.

 

Hence my earlier post saying that rapidly moving steel parts can produce enough effect to "fire" the sensor when it is not expected to do so.

 

Simon.

Simon. So if I have a Hall effect sensor, which would be the best method to get accurate readings? Is it better pointing at the diff bolts or would it be better pointing at magnets with a slightly larger gap as mentioned above? Is the Hall effect sensor or the other type better for the job? Mine was OK to about 65 with Hall/Magnets then jumped. With a change to Hall/Prop bolts it is OK to 50 then under reads over that. Just trying to find the best solution.

[knights_templar]

Hall effect is better, point at bolts, and get sensor as close as poss. (without touching) works fine. but you must have a rigid mount

 

ETB have now moved onto this type

[sj-bradley]

Hi,

 

There are two types of sensor, those that need magnets to work (passive) and those that don't (active).

 

The passive sensor could be a simple coil of wire that produces a voltage when a magnet passes by quickly. The active one needs a power supply to it as it uses a coil to produce a known magnetic field. When an iron body passes by that type of sensor, it changes the field intensity and the hall effect sensor can detect this change. No magnets are required for the active sensor, just a close fly by of iron based material.

 

Simon.