Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Doctor Gollum]

To Dave's last... That surface just looks clean, but the sensor is clearly clocked wrong. Ergo it is wrong and won't work correctly.

The magnet is, or should be, the size of the teeth on the reluctor. If it is clocked even a few degrees it will cover more than one tooth and therefore not pick up every space. Effective this fools it into see less teeth and therefore presuming the wheels are locking up. If the rear sensors are reading correctly then the ABS over reacts even more due to the fact that it thinks the front wheels are rotating at half the speed.

Undo everything you have done with actual genuine sensors and the system will work perfectly.

[bkorven]

If a part doesn't bolt straight in and/or appears different in anyway to the factory original, then it is unlikely to function correctly. Especially if it's electrical.

You missed my point re the bearings... If the reluctor teeth are different in number, size or shape to the factory originals, then they will cause an error even if the signal strength is consistent. The ECU will be looking for a specific number of teeth and if there's a mismatch between wheels or axles, then it won't function correctly. The sensors are a different issue, but both will cause faulty performance, and you've already admitted thet the sensors don't fit as you can't fit the retaining bolts. Which means that the sensors are free to move both rotationally and in/out along the casting. They're bolted for a reason.

Oh okay, I see. If the reluctor teeth would be that different from the stock ones, I'd find it startling to be honest. Am I naive to believe that parts that are sold should work? Anyways, I'm quite sure the bearing assemblies are OK. The dimensions and number of teeth was stated, and was the same for all different brands of bearing assemblies. I suppose it's harder with sensors, since they don't really state the number of windings inside them, or other such detailed information about the performance.

Although I guess it doesn't have any greater relevance to the discussion, just to clarify; the aftermarket sensors does fit perfectly. The signal is just too weak, and by rotating them it gets stronger.

To Dave's last... That surface just looks clean, but the sensor is clearly clocked wrong. Ergo it is wrong and won't work correctly.

The magnet is, or should be, the size of the teeth on the reluctor. If it is clocked even a few degrees it will cover more than one tooth and therefore not pick up every space. Effective this fools it into see less teeth and therefore presuming the wheels are locking up. If the rear sensors are reading correctly then the ABS over reacts even more due to the fact that it thinks the front wheels are rotating at half the speed.

Undo everything you have done with actual genuine sensors and the system will work perfectly.

While your reasoning with the magnet seeing more than one teeth at the same time in a rotated position sounds reasonable, the ABS-behaviour was actually improved by rotating the sensors away from the original position. I could not tell the frequency of the signal was affected, either.

[bkorven]

I've only played with Carlton/Senator ABS, not Omega B, and it might be different - but.....

There is no ground on the sensors. The two sensor wires are differential all the way to the control unit. Infact the resistance to ground of both terminals is many hundreds of K-ohms. If either terminal is somehow shorted to ground, the ABS fault light comes on.

As I say - on the Carlton/Senator.

Thank you! That is valuable information.

I have measured the resistance of the terminals to 6.8M Ohm on both sides, and I assume that reading is OK.

Yes, that sounds about right.

The sensors are called variable reluctance sensor. Their output signal amplitude is a function of both RPM and clearance. You seem to be saying that by rotating the sensor things get better/worse. Note that the clearance is important - if the sensor is too close to the reluctor ring then it'll generate too high voltage, and this can upset the control unit. Rotating the sensor will reduce the amplitude as the 'blade' of the sensor becomes mis-aligned with the teeth on the reluctor ring..

So check the clearance between the tip of the sensor and the reluctor ring on the wheel bearing. I don't know what it is supposed to be, but I'd expect something like 1mm-2mm. You say up thread that it is (almost?) touching. That isn't correct. If it is, then the sensors are wrong but you might be able to correct it by putting shims in between the sensor and the steering knuckle.

I estimate the distance to be somewhere around 0.2-0.3 mm at the moment. But the signal is too weak, and not the other way around. I don't know what the voltage should be specifically for the Omega, but reading online it seems to usually be between 0.5-1 V at 2 RPM of the wheel. I get ~50mV when the sensors are mounted in their intended position as from factory, and I can manage to get ~200mV when I rotate them.

The stronger I manage to get the signal (by rotating), the better the behaviour of the ABS.

The reason ABS is activated is likely due to the signal being too weak to be registrated by the ECU, so that it is interpreted as the wheels locking up.

Have you ground or filed the surface that the sensor sits on ,to make the sensor CLOSER to the reluctor ring 
the surface looks way too flat and clean compared to the rest of the casting in your image


have you tried shimming the sensor out ,to see how that effects the signal 
though i suspect the after market sensors may be the issue .

Yes, I have filed the surface that the sensor sits on. It was an attempt to make the signal stronger... and a successful one at that 

In my desperate troubleshooting I saw a possible cause of the problem that corrosion had increased the clearance.

I have tested the signal with a larger distance from the reluctor ring, yes. The signal gets weaker, as expected.

[Doctor Gollum]

Which just goes to show that the sensors are wrong.

You might try fitting them in the correct position, and see if the driveability improves.

The sensors might work electrically but the information they are sending to the ABS ECU is clearly not right.

[bkorven]

Thanks for all the input, from all of you! I really appreciate it.

It seems highly likely that the sensors are bad, after all. I actually did not know the sensors I had ordered were not genuine OEM. The brand of the ones I ordered was pretty well hidden under an "extra info"-tab that needed to be opened to show. It would have been better if I had provided you with information about the manufacturer early on, ofcourse.

I'll order new ones, and will report back with the result.

The problem now is that the only sensors that seem reasonable, that I seem to be able to get online, are of brand "Vemo", model "V40-72-0472". No genuine sensors anywhere to be found. Just alot of other shady "beer"-named ones and similar.

I think I'll just do it the old way, and give my local Opel-dealer a call on Monday.

[bkorven]

Which just goes to show that the sensors are wrong.

You might try fitting them in the correct position, and see if the driveability improves.

The sensors might work electrically but the information they are sending to the ABS ECU is clearly not right.

Might indicate that they are wrong. Not 100% certain of that.

I have tried fitting them in the correct position. It gets worse, and the ABS-light comes on again.

I agree with you about that, the information they are sending is definitely not right.

[LC0112G]

Both the frequency AND amplitude of the sensor output will increase with RPM. You say you are measuring at 2 RPM. Assuming you're using 235R45x17 tyres, the rolling circumference is 79.57 inches (2021mm). 2 RPM is 120 revs per hour, which is therefore 9548 inches per hour, which is only 0.15 miles per hour.

I don't think the ABS is supposed to work at all below 5mph. Therefore the voltages you are seeing from the sensors are not representative of what you should see if the car is travelling at > 5MPH (8 Km/h).

Your 50mV at 2RPM might be correct - Try it again at a more realistic 200RPM (15 MPH), you should get a much larger signal - probably 20-50 times larger. Remember the ABS needs to work all the way up to above top speed (say 2000RPM/150MPH) so the signal can be massive at that speed. If you adjusts the sensors to give large signals at very low speeds, then at normal/high speed you can overload the ABS inputs.

I've got some old Carlton hubs in the lockup, still with the ABS sensors fitted. I'll dig them out tomorrow and measure them up for clearance. 0.2-0.3mm sounds way too close to me.

[bkorven]

Both the frequency AND amplitude of the sensor output will increase with RPM. You say you are measuring at 2 RPM. Assuming you're using 235R45x17 tyres, the rolling circumference is 79.57 inches (2021mm). 2 RPM is 120 revs per hour, which is therefore 9548 inches per hour, which is only 0.15 miles per hour.

I don't think the ABS is supposed to work at all below 5mph. Therefore the voltages you are seeing from the sensors are not representative of what you should see if the car is travelling at > 5MPH (8 Km/h).

Your 50mV at 2RPM might be correct - Try it again at a more realistic 200RPM (15 MPH), you should get a much larger signal - probably 20-50 times larger. Remember the ABS needs to work all the way up to above top speed (say 2000RPM/150MPH) so the signal can be massive at that speed. If you adjusts the sensors to give large signals at very low speeds, then at normal/high speed you can overload the ABS inputs.

I've got some old Carlton hubs in the lockup, still with the ABS sensors fitted. I'll dig them out tomorrow and measure them up for clearance. 0.2-0.3mm sounds way too close to me.

I'm comparing my voltage at 2 RPM to what others have measured at that same speed, so what voltage the sensors generate at higher speeds is not really of relevance for that matter. I'm expecting 0.5V-1V at 2 RPM, but I'm measuring much lower than that at 2 RPM.

Overload the ABS-inputs, as in causing damage, or as in disrupt the function? I assume there is binary switching inside the ECU, for a threshold value of the signal. In that case, the signal just needs to be strong enough, and all that matters is frequency/consistency of the signal. I could be (and probably am) wrong.

I am convinced that 0.2-0.3mm is a shorter distance than what it was originally, but I'm not convinced the distance can be too short. I also still believe my signal is too weak.

Either way, I'm grateful for you for taking the time to check it out.

[Doctor Gollum]

Perhaps you are over thinkng the whole thing.

Given that the system doesn't work correctly with those sensors and that the sensors don't fit correctly, the simple fact is that you need the correct sensors. And possibly bearing assemblies. (Omega A and Omega B hubs may seem to be similar, but that doesn't guarantee that they are the same and it would be all too easy for the parts suppliers to make the same assumption).

Anything else is a waste of time and effort unless you're retired and extremely bored.

Trying to position the wrong sensors in the correct position is akin to trying to make a rear door fit the front door aperture... You might make it latch, but it will never function correctly.

[Doctor Gollum]

Both the frequency AND amplitude of the sensor output will increase with RPM. You say you are measuring at 2 RPM. Assuming you're using 235R45x17 tyres, the rolling circumference is 79.57 inches (2021mm). 2 RPM is 120 revs per hour, which is therefore 9548 inches per hour, which is only 0.15 miles per hour.

I don't think the ABS is supposed to work at all below 5mph. Therefore the voltages you are seeing from the sensors are not representative of what you should see if the car is travelling at > 5MPH (8 Km/h).

Your 50mV at 2RPM might be correct - Try it again at a more realistic 200RPM (15 MPH), you should get a much larger signal - probably 20-50 times larger. Remember the ABS needs to work all the way up to above top speed (say 2000RPM/150MPH) so the signal can be massive at that speed. If you adjusts the sensors to give large signals at very low speeds, then at normal/high speed you can overload the ABS inputs.

I've got some old Carlton hubs in the lockup, still with the ABS sensors fitted. I'll dig them out tomorrow and measure them up for clearance. 0.2-0.3mm sounds way too close to me.

I'm comparing my voltage at 2 RPM to what others have measured at that same speed, so what voltage the sensors generate at higher speeds is not really of relevance for that matter. I'm expecting 0.5V-1V at 2 RPM, but I'm measuring much lower than that at 2 RPM.

Overload the ABS-inputs, as in causing damage, or as in disrupt the function? I assume there is binary switching inside the ECU, for a threshold value of the signal. In that case, the signal just needs to be strong enough, and all that matters is frequency/consistency of the signal. I could be (and probably am) wrong.

I am convinced that 0.2-0.3mm is a shorter distance than what it was originally, but I'm not convinced the distance can be too short. I also still believe my signal is too weak.

Either way, I'm grateful for you for taking the time to check it out.

The hall sensor on a square cut reluctor wheel produces a binary signal, it's either on or off. The frequency relative to the other sensors is what makes the whole thing work. It might be intelligent enough to allow for steering input ie slower wheel speeds on one side of the car vs the other.

If the ECU receives anything other the the expected on/off then it presumes there is an issue and behaves accordingly.

[Doctor Gollum]

Incidentally, what hubs did you fit? Whilst they are both have 48 teeth, they look very different in your photo. The new reluctor face looks to be off centre.

[bkorven]

Incidentally, what hubs did you fit? Whilst they are both have 48 teeth, they look very different in your photo. The new reluctor face looks to be off centre.

These are the hubs I ordered: https://www.autodoc.co.uk/ridex/8054400
And these are the hubs I received, which is not what I ordered: https://www.buycarparts.co.uk/esen-skv/18164200

I suspect that one manufacturer produces parts for several brands, and the only difference is the logo on the box.

[bkorven]

Perhaps you are over thinkng the whole thing.

Given that the system doesn't work correctly with those sensors and that the sensors don't fit correctly, the simple fact is that you need the correct sensors. And possibly bearing assemblies. (Omega A and Omega B hubs may seem to be similar, but that doesn't guarantee that they are the same and it would be all too easy for the parts suppliers to make the same assumption).

Anything else is a waste of time and effort unless you're retired and extremely bored.

Trying to position the wrong sensors in the correct position is akin to trying to make a rear door fit the front door aperture... You might make it latch, but it will never function correctly.

I definitely agree. I'll order new sensors on Monday.

P.S.: And I hope I won't need to order new bearing assemblies as well.

Both the frequency AND amplitude of the sensor output will increase with RPM. You say you are measuring at 2 RPM. Assuming you're using 235R45x17 tyres, the rolling circumference is 79.57 inches (2021mm). 2 RPM is 120 revs per hour, which is therefore 9548 inches per hour, which is only 0.15 miles per hour.

I don't think the ABS is supposed to work at all below 5mph. Therefore the voltages you are seeing from the sensors are not representative of what you should see if the car is travelling at > 5MPH (8 Km/h).

Your 50mV at 2RPM might be correct - Try it again at a more realistic 200RPM (15 MPH), you should get a much larger signal - probably 20-50 times larger. Remember the ABS needs to work all the way up to above top speed (say 2000RPM/150MPH) so the signal can be massive at that speed. If you adjusts the sensors to give large signals at very low speeds, then at normal/high speed you can overload the ABS inputs.

I've got some old Carlton hubs in the lockup, still with the ABS sensors fitted. I'll dig them out tomorrow and measure them up for clearance. 0.2-0.3mm sounds way too close to me.

I'm comparing my voltage at 2 RPM to what others have measured at that same speed, so what voltage the sensors generate at higher speeds is not really of relevance for that matter. I'm expecting 0.5V-1V at 2 RPM, but I'm measuring much lower than that at 2 RPM.

Overload the ABS-inputs, as in causing damage, or as in disrupt the function? I assume there is binary switching inside the ECU, for a threshold value of the signal. In that case, the signal just needs to be strong enough, and all that matters is frequency/consistency of the signal. I could be (and probably am) wrong.

I am convinced that 0.2-0.3mm is a shorter distance than what it was originally, but I'm not convinced the distance can be too short. I also still believe my signal is too weak.

Either way, I'm grateful for you for taking the time to check it out.

The hall sensor on a square cut reluctor wheel produces a binary signal, it's either on or off. The frequency relative to the other sensors is what makes the whole thing work. It might be intelligent enough to allow for steering input ie slower wheel speeds on one side of the car vs the other.

If the ECU receives anything other the the expected on/off then it presumes there is an issue and behaves accordingly.

My Omega does not have hall sensors, they are variable reluctance sensors. They are passive, and the signal they generate from the reluctor ring is a sinus wave (see my attached images in my earlier post). The ECU probably just monitors the passing of a threshold value for the voltage.

[Doctor Gollum]

That's a mistake on my part but the principle stands.

[Enceladus]

EPC says that Omega A, Senator B & Omega B/B2 all use the same front hub assembly, GM# 90486467 / Opel KAT# 16-03-194. Not carried forward onto newer GM models.

And you're right, it's very likely that there is only one source supplying the hub assemblies and they are rebranded with whatever the buyer requests. It's also very likely that actual production ceased years ago and that anything available of whatever brand is actually old stock, dumped on the market to the highest bidder or bought up by old stock specialists. There are simply not enough Omegas or Cateras left to make it worthwhile for anyone to tool-up and manufacture any new parts.

The Ridex parts you ordered and the ESEN-SKV parts you received should be equivalent to each other and the factory fit and should work.

How did you clean out the bores for the sensors?