The ABS warning light came on late last week, on the morbing the car was going in for some other work, had it diagnosed while it was in and it turns out to be a front wheel sensor. Until I get round to doing the work I thought that there would be no problem (danger) driving the car with the ABS down, how wrong I was.
Saturday morning I was pleasantly supprised to see that the warning light extinguished when the car started and I went happily on my way.
Pulling up at the first junction behind a couple of other cars, braking gently from a slow speed the ABS warning light came on. As it came on the pedal twitched and all braking ceased. There was still pressure in the pedal but there was zero braking (and I mean zero). It all happened very quickly but passengers felt the car effectively lurch forward and went into panic as there was only a few metres before impact with stationary cars. I must admit to freezing for a split second and then making a rapid grab for the handbrake. My foot was still on the brake pedal and before I got to the handbrake braking was restored and the car screeched to a halt centimetres from the cars in front.
It would appear that transition from ABS okay to ABS in fault is a dangerous time to be on the move.
Mark
Xantia ABS Fright
Moderator: RichardW
I think your problem is much simpler. - flat accumulator sphere.
What I think happened is that you applied the brakes, the abs tried to operate, and quite correctly released the pressure in the system to the locked wheels. When it does this it should be able to rely on accumulator charged with high pressure fluid to cause the brakes to pulse and retard the car. If the accumulator is flat then it can't do this properly and must wait for further output from the pump.
The pump fitted to your xantia is split with only 2 cylinders for the brakes and suspension and while this problem may be hidden on a BX where all the output should be available for the brakes the output of your pump is that much less and so the lag is noticeable.
So what is your regulator cycling speed, how old is the sphere etc.
jeremy
What I think happened is that you applied the brakes, the abs tried to operate, and quite correctly released the pressure in the system to the locked wheels. When it does this it should be able to rely on accumulator charged with high pressure fluid to cause the brakes to pulse and retard the car. If the accumulator is flat then it can't do this properly and must wait for further output from the pump.
The pump fitted to your xantia is split with only 2 cylinders for the brakes and suspension and while this problem may be hidden on a BX where all the output should be available for the brakes the output of your pump is that much less and so the lag is noticeable.
So what is your regulator cycling speed, how old is the sphere etc.
jeremy
Mark
I had very similar problems with ma brakes a while ago, the light would come on somtimes and stay on whilst I was driving, I had no problems with the brakes under these conditions, but when the light behaved correctly ( extinguised after ECU had done its checks and staying extinguished) I then somtimes had the same experience as you, the braking or lack of also makeing a sort of groaning noise, it also only happened when going fairly slowly and just before coming to a final stop. I tested the resistance of the sensors after each occurance and the resistance was always OK. I then changed the ECU, no further problems for about 6 months now.
I had very similar problems with ma brakes a while ago, the light would come on somtimes and stay on whilst I was driving, I had no problems with the brakes under these conditions, but when the light behaved correctly ( extinguised after ECU had done its checks and staying extinguished) I then somtimes had the same experience as you, the braking or lack of also makeing a sort of groaning noise, it also only happened when going fairly slowly and just before coming to a final stop. I tested the resistance of the sensors after each occurance and the resistance was always OK. I then changed the ECU, no further problems for about 6 months now.
Under NO circumstances the ABS should prevent braking [:0]
- it must ONLY release one frontwheel going slower (or blocked) than the other.
The ABS must NOT behave like this because of a dodgy wheel sensor - not even for a milli second [B)]
The ABS ECU has build-in elaborate circuits to check correct operation of wheel sensors continously - and the continuity of the ABS electrovalves.
I have my cards on acrowot - replace the ECU [B)]
If my car - i'd immediately rip the connector plug on the ABS block - ensuring the ECU does not fool up the electrovalves again.
This leaves the car with a standard brakes system.
- it must ONLY release one frontwheel going slower (or blocked) than the other.
The ABS must NOT behave like this because of a dodgy wheel sensor - not even for a milli second [B)]
The ABS ECU has build-in elaborate circuits to check correct operation of wheel sensors continously - and the continuity of the ABS electrovalves.
I have my cards on acrowot - replace the ECU [B)]
If my car - i'd immediately rip the connector plug on the ABS block - ensuring the ECU does not fool up the electrovalves again.
This leaves the car with a standard brakes system.
Anders
Previously I believed ABS to be passive and for it not to be possible for this type of behaviour. I can only report what happened as I (and my passengers) experienced it. Id guess I was only braking with the front wheels at the time of the incident. I don't recall any noise when the fault light lit but the brake pedal definately did a single pulse and then all braking ceased as if I'd removed my foot from the pedal, which of course I had not. My best estimate is 1.5 - 2 seconds without brakes then braking was restored. In the mean time I had increased my foot pressure on the brake pedal so the car stopped suddenly. I'm really not sure how to progress this now. I had had the system checked the previous day and a front wheel sensor was said to be down but I've no idea of the test process.
Mark
Previously I believed ABS to be passive and for it not to be possible for this type of behaviour. I can only report what happened as I (and my passengers) experienced it. Id guess I was only braking with the front wheels at the time of the incident. I don't recall any noise when the fault light lit but the brake pedal definately did a single pulse and then all braking ceased as if I'd removed my foot from the pedal, which of course I had not. My best estimate is 1.5 - 2 seconds without brakes then braking was restored. In the mean time I had increased my foot pressure on the brake pedal so the car stopped suddenly. I'm really not sure how to progress this now. I had had the system checked the previous day and a front wheel sensor was said to be down but I've no idea of the test process.
Mark
I think Anders is defintely right - the first thing to do is disconnect the ABS completely! The possible (although unlikely) consequences are too serious to consider!
Then sort out the problem. Its unfortunate if it is the ECU - these seem to be expensive.
I've had a strange brake 'grinding' feeling in a Laguna and a Mundano*, both when there should have been no possibility of ice. Not encouraging!
* A bit unfair - the Mondeo LX TDCi 130 is really rather good [}:)]
Then sort out the problem. Its unfortunate if it is the ECU - these seem to be expensive.
I've had a strange brake 'grinding' feeling in a Laguna and a Mundano*, both when there should have been no possibility of ice. Not encouraging!
* A bit unfair - the Mondeo LX TDCi 130 is really rather good [}:)]
-
David Goddard
- Posts: 138
- Joined: 26 Sep 2001, 04:39
- Location: United Kingdom
- My Cars:
Martk, I agree with Anders, ABS systems "fail safe". Maybe there was a slug of air in the system following the replacement of the accumulator. I would lower the suspension, depressurise it with the pressure relief screw (with the engine running), tighted the screw and depressurine it two or three times. That may fix it, but to be certain I would then bleed all the brakes.
mark -
The ABS system is not exactly passive - but should be considered an add-on to the standard brakes system.
What it does - is watching each wheel's speed of the car when braking. If one of the wheels is slower (by some set value depending on vehicle speed) - this same wheel's brake force is backed off for a split second to try catch up with the other. This goes on as a constant back-off - test speed - back off - etc. - untill equilibrum has been re-gained - then the ABS does nothing again.
It's the same back-off/test sequence that makes the pedal kick back as the common ABS feeling.
Rarely in normal traffic the ABS will come in action - some sedate drivers may own a car with ABS for lifetime - with the ABS never coming to work.
The common cause to bring ABS in action is the difference in road surface friction - as you know the kerb site may be wet & dirty - while the middle of the road may be dry & clean.
If you brake firmly on such road conditions - the kerbside wheel will lock - as it easily slips over the wet dirt - hence the ABS get to work and helps keep the kerbside wheel spinning the same speed as the roadside wheel. The pedal then kicks you.
But a delay - even blocking off the pedal action - when braking - is absolutely forbidden to do by an ABS system - in any condition.
And such a condition can NEVER be caused by a dodgy wheel sensor - as any external faults detected by the ABS ECU will cause the ECU to go in no-ABS-function mode - to ENSURE it does not behave dangerous.
It can only be the ABS ECU going insanely that you have experinced.
The red line thru all this is - that the brakes behaves perfectly normal with no ABS function - since the ABS function is an ADD-ON - as a help during extreme braking conditions. Like you consider fog lamps in extreme weather conditions.
It's like fitting a spoiler to get better grip on rear wheels at extreme speeds.
But if the spoiler breaks up - it becomes a hazard - and you rip it off to drive home without that hazard - at normal speeds.
The ABS system is not exactly passive - but should be considered an add-on to the standard brakes system.
What it does - is watching each wheel's speed of the car when braking. If one of the wheels is slower (by some set value depending on vehicle speed) - this same wheel's brake force is backed off for a split second to try catch up with the other. This goes on as a constant back-off - test speed - back off - etc. - untill equilibrum has been re-gained - then the ABS does nothing again.
It's the same back-off/test sequence that makes the pedal kick back as the common ABS feeling.
Rarely in normal traffic the ABS will come in action - some sedate drivers may own a car with ABS for lifetime - with the ABS never coming to work.
The common cause to bring ABS in action is the difference in road surface friction - as you know the kerb site may be wet & dirty - while the middle of the road may be dry & clean.
If you brake firmly on such road conditions - the kerbside wheel will lock - as it easily slips over the wet dirt - hence the ABS get to work and helps keep the kerbside wheel spinning the same speed as the roadside wheel. The pedal then kicks you.
But a delay - even blocking off the pedal action - when braking - is absolutely forbidden to do by an ABS system - in any condition.
And such a condition can NEVER be caused by a dodgy wheel sensor - as any external faults detected by the ABS ECU will cause the ECU to go in no-ABS-function mode - to ENSURE it does not behave dangerous.
It can only be the ABS ECU going insanely that you have experinced.
The red line thru all this is - that the brakes behaves perfectly normal with no ABS function - since the ABS function is an ADD-ON - as a help during extreme braking conditions. Like you consider fog lamps in extreme weather conditions.
It's like fitting a spoiler to get better grip on rear wheels at extreme speeds.
But if the spoiler breaks up - it becomes a hazard - and you rip it off to drive home without that hazard - at normal speeds.
My money is firmly on the ECU, this fault is too similar to mine for it not to be. I tried the citerobics etc. to no avail, it is fairly common for these ECU's to go faulty, especially the ones fitted to the earlier Xantia's pre 1995 I think,( these have a white faceplate on the ECU,the later ECU,s have a black faceplate) but having said that my ECU was the later type, it is a very worrying fault.
I'll go against the flow on this one.
I had a similar experience in my BX16V & it ain'y funny I can tell you.
I had a dodgy wheel sensor, in fact it was the wire to the sensor that had an internal break. I managed to get it to hold in position one day when I went to test it and as a result the light stayed off.
Approaching a "T" junction I put my foot on the brake pedal and....nothing[:0] I put it on a bit harder, then a lot harder aand was down to second gear and reaching for the handbrake when it all started working & the car stopped. As it did, the light appeared on the dash.
I disconnected the offending sensor & subsequently I replaced the faulty sensor whereupon I took the car for a test (filled with apprehension,) but it worked as it should & has continued to work correctly ever since.
I'd suggest firstly getting the crook sensor replaced & take it for a drive before diving in with new ECUs as I tend to agree with the theory that these ABS ECUs do seem to have a period be it ever so short where their little brain seems to get a bit confused when sensors go wrong come right & then go wrong again.
Whether it's a defect in their make up or design or just some strange inexplicable glitch I don't know, but my car has like this one only done it the once & has travelled several thousand Klms since.
Alan S [:D]
I had a similar experience in my BX16V & it ain'y funny I can tell you.
I had a dodgy wheel sensor, in fact it was the wire to the sensor that had an internal break. I managed to get it to hold in position one day when I went to test it and as a result the light stayed off.
Approaching a "T" junction I put my foot on the brake pedal and....nothing[:0] I put it on a bit harder, then a lot harder aand was down to second gear and reaching for the handbrake when it all started working & the car stopped. As it did, the light appeared on the dash.
I disconnected the offending sensor & subsequently I replaced the faulty sensor whereupon I took the car for a test (filled with apprehension,) but it worked as it should & has continued to work correctly ever since.
I'd suggest firstly getting the crook sensor replaced & take it for a drive before diving in with new ECUs as I tend to agree with the theory that these ABS ECUs do seem to have a period be it ever so short where their little brain seems to get a bit confused when sensors go wrong come right & then go wrong again.
Whether it's a defect in their make up or design or just some strange inexplicable glitch I don't know, but my car has like this one only done it the once & has travelled several thousand Klms since.
Alan S [:D]
-
philhoward
If the sensor sticks in an "on" or "off" state, then the ECU will (correctly) assume that the particular wheel is locked and stop the braking until it sees a state transition on that particular sensor (wheel starts moving again). It is possible for a sensor to retain a particular state and appear to any diagnostics as OK. The only stumbling block is that the ECU should have detected a sticky sensor when the brake are not being applied. Perhaps the ECU only detects for either a pure short circuit or a pure open circuit, not just a slightly higher/lower resistance whch could result in either an "on" or "off" state being registered.
If owners (like our very experienced member AlanS [:)]) keep submitting these frightening tales of dangerous ABS behaviour ...
- then my first action after getting a "new" car with ABS would be to rip the ABS circuit [:(!]
I will NOT for one milli second accept a dodgy ABS preventing the brakes functioning [:0]
There are NO excuses for the ABS computer to behave like this [:(!]
The ABS system is fitted to cars for one single purpose :
<font color="blue"><i>to aid the driver on extreme road conditions during braking</i></font id="blue">
<font color="red"><i><b>under no circumstances the ABS must ever prevent or delay braking</b></i></font id="red">
- would you accept any mechanical device that have a mind of it's own ?
What about a speed pilot that thinks red lights are the time for speeding up the car ?
- then my first action after getting a "new" car with ABS would be to rip the ABS circuit [:(!]
I will NOT for one milli second accept a dodgy ABS preventing the brakes functioning [:0]
There are NO excuses for the ABS computer to behave like this [:(!]
The ABS system is fitted to cars for one single purpose :
<font color="blue"><i>to aid the driver on extreme road conditions during braking</i></font id="blue">
<font color="red"><i><b>under no circumstances the ABS must ever prevent or delay braking</b></i></font id="red">
- would you accept any mechanical device that have a mind of it's own ?
What about a speed pilot that thinks red lights are the time for speeding up the car ?
-
philhoward
Easy Anders!!
I can see your point though...i'll disconnect the ABS on any car of mine which starts to play up! I've lived for years without it, so can do again. It's not a deciding factor on any car i've bought; if it's there, it's there. I think there's only been once when i wish i did have ABS, but it was my own fault for approching a bend too quick (in a Renault Laguna Estate fully loaded at the back...)
Back to the original problem; sounds like an ABS ECU playing daft sods, rather than a sensor. The only other avenue could be a sticking valve block...is the car an early (non-anti-sink) one? If so, could it have had an exploding hydro pump at some point and left a little bit in the valve block? If not, perhaps a hydraflush might help?
I can see your point though...i'll disconnect the ABS on any car of mine which starts to play up! I've lived for years without it, so can do again. It's not a deciding factor on any car i've bought; if it's there, it's there. I think there's only been once when i wish i did have ABS, but it was my own fault for approching a bend too quick (in a Renault Laguna Estate fully loaded at the back...)
Back to the original problem; sounds like an ABS ECU playing daft sods, rather than a sensor. The only other avenue could be a sticking valve block...is the car an early (non-anti-sink) one? If so, could it have had an exploding hydro pump at some point and left a little bit in the valve block? If not, perhaps a hydraflush might help?
-
Dave Burns
- Posts: 1915
- Joined: 14 May 2001, 05:30
- Location: United Kingdom
- My Cars:
- x 2
Sticky sensor, no not at all possible, they don't work like that.
They are Hall Effect Devices, when a ferrous object passes close by, in this case the reluctor ring machined onto the drive shaft or hub, they generate an electrical pulse as each segment of the reluctor passes by, the faster the car goes the greater the frequency of pulses.
The ecu monitors each wheel sensors pulses and from them it can work out the speed of each wheel and compare it with the other wheels for a known normal state.
Any signal from a sensor that suddenly drops in frequency compared to the rest of the wheel sensors, and also compared to the last known speed of that particular wheel, will cause the ecu to assume that lock up is about to occur and relese the brake on that wheel, when the signal is re-established from that wheel as it begins to rotate the ecu will re-apply the brake to that wheel at the force dictated by the drivers foot on the brake pedal, this situation will continue rapidly until the car stops or the brakes are released or reduced by the driver.
If a sensor has or gone open curcuit then a fault will be registered, similarly if the integrity of the wiring or its connectors is compromised, then any signal generated by a good sensor will not get to the ecu and a fault state will again be registered.
Dave
They are Hall Effect Devices, when a ferrous object passes close by, in this case the reluctor ring machined onto the drive shaft or hub, they generate an electrical pulse as each segment of the reluctor passes by, the faster the car goes the greater the frequency of pulses.
The ecu monitors each wheel sensors pulses and from them it can work out the speed of each wheel and compare it with the other wheels for a known normal state.
Any signal from a sensor that suddenly drops in frequency compared to the rest of the wheel sensors, and also compared to the last known speed of that particular wheel, will cause the ecu to assume that lock up is about to occur and relese the brake on that wheel, when the signal is re-established from that wheel as it begins to rotate the ecu will re-apply the brake to that wheel at the force dictated by the drivers foot on the brake pedal, this situation will continue rapidly until the car stops or the brakes are released or reduced by the driver.
If a sensor has or gone open curcuit then a fault will be registered, similarly if the integrity of the wiring or its connectors is compromised, then any signal generated by a good sensor will not get to the ecu and a fault state will again be registered.
Dave