High idle problem - I need another guinea pig
Discussion
I think I have got to the bottom of the high/slow to drop down idle problem. Using ECUmate, I have found out that the main culprit is the TVR speedo calibration/conversion unit which takes in the speedo sensor signal and converts it into a signal for the ECU. The unit is extremely prone to noise and the way it supplies a signal, the ECU thinks that the car has been doing 30+ and then done an instant stop. This confuses the ECU which wimps out and keeps the idle high and in some cases takes some time to even go into an idle mode.
I have redesigned the unit so that it has the following:
Anyone out there willing to help?
I have redesigned the unit so that it has the following:
- A 14CUX friendly speed profile so that the idle control is smoother.
- Programmable idle mode thresholds. Currently it is set to around 3 mph but this can be changed to 4 or 5 or even higher.. or lower!
- Speedo test circuit - like the intelligent speedo sensor so that you can check that the speedo is working.
- Speedo calibration - permanently adjusts the speedo reading by pressing either the + or - button. Saves having to get the speedo sent away if you change ratios, tyres or wheels etc.
- Intelligent speedo sensor input so that you can retrofit an intelligent sensor at a later date.
Anyone out there willing to help?
Edited by shpub on Saturday 7th January 10:22
As I mentioned several times in the past Steve, I find it hard to believe that nobody has attempted to install the T5 speedo drive gear so that the ECU can get it's own signal from that, like the LT77/GKN cars with separate dedicated senders for speedo & ECU. All these extra shared electronics just seem a recipe for confusion to me (not just what you're doing but the original black box/splitter from TVR) and making 2 simple circuit installations into a complex shared one?
shpub said:
I think I have got to the bottom of the high/slow to drop down idle problem. Using ECUmate, I have found out that the main culprit is the TVR speedo calibration/conversion unit which takes in the speedo sensor signal and converts it into a signal for the ECU. The unit is extremely prone to noise and the way it supplies a signal, the ECU thinks that the car has been doing 30+ and then done an instant stop. This confuses the ECU which wimps out and keeps the idle high and in some cases takes some time to even go into an idle mode.
I have redesigned the unit so that it has the following:
Anyone out there willing to help?
I live in MK - YHM.I have redesigned the unit so that it has the following:
- A 14CUX friendly speed profile so that the idle control is smoother.
- Programmable idle mode thresholds. Currently it is set to around 3 mph but this can be changed to 4 or 5 or even higher.. or lower!
- Speedo test circuit - like the intelligent speedo sensor so that you can check that the speedo is working.
- Speedo calibration - permanently adjusts the speedo reading by pressing either the + or - button. Saves having to get the speedo sent away if you change ratios, tyres or wheels etc.
- Intelligent speedo sensor input so that you can retrofit an intelligent sensor at a later date.
Anyone out there willing to help?
Edited by shpub on Saturday 7th January 10:22
Nosh
Edited by Nosh on Saturday 7th January 11:21
spend said:
As I mentioned several times in the past Steve, I find it hard to believe that nobody has attempted to install the T5 speedo drive gear so that the ECU can get it's own signal from that, like the LT77/GKN cars with separate dedicated senders for speedo & ECU. All these extra shared electronics just seem a recipe for confusion to me (not just what you're doing but the original black box/splitter from TVR) and making 2 simple circuit installations into a complex shared one?
It is not as simple as simply installing it unfortunately. The specs are not compatible and that is what causes the problem. In practice, I suspect you trade one set of problems for another as the 14CUX processes the road speed signal far more than I and other originally thought. As a result, to get the best behaviour, the speedo signal needs to be profiled quite carefully. The original TVR solution works sort off and considering when it was made, it is a cheap solution especially when compared to the mid 1990s cost of my software based solution would be.
Ribol said:
Are you describing a car problem or one that exists when using an ECUmate?
Car problem. ECUmate doesn't affect the ECU running... not unless you really want to switch the fuel pump off or move the stepper motor! For an example see here
It's where the idle does not drop correctly. Not all cars do it. Those that do have the T5 box. I also think it may have something do with the low level shunting that some see as well but that needs more testing.
ECUmate just allowed me to see exactly what was going on internally and how the ECU was really behaving. There is a big difference between the idle control smoothness on a Range Rover and on a TVR and it's that coupled with several conversations re idle problems that prompted me to look at the unit and redesign it.
shpub said:
Not all cars do it. Those that do have the T5 box. I also think it may have something do with the low level shunting that some see as well but that needs more testing.
ECUmate just allowed me to see exactly what was going on internally and how the ECU was really behaving. There is a big difference between the idle control smoothness on a Range Rover and on a TVR and it's that coupled with several conversations re idle problems that prompted me to look at the unit and redesign it.
Always going to be tricky to work out if it is random, my car doesn't do it and I don't know anyone personally that has this issue, so thankfully not that common.ECUmate just allowed me to see exactly what was going on internally and how the ECU was really behaving. There is a big difference between the idle control smoothness on a Range Rover and on a TVR and it's that coupled with several conversations re idle problems that prompted me to look at the unit and redesign it.
What part is it you are looking to redesign then?
It's the speedo callibration/conversion box that takes the T5 cars speedo signal which is totally incompatible with the ECU input and derives a ECU compatible signal so that the ECU knows if the car is moving or not. What the ECU would like to know is the car's speed but TVR did a cheap and nasty workaround that says the car is either doing 30 or 0. This signal can be extremely erratic, affected by electrical noise and while it does work, it can and does go wrong or to be more accurate... misbehave.
I've seen on ECUmate how the 14CUX idle behaviour is very sensitive to the road speed input and that it is not a simple on/off switch which the TVR box seems to think.
The circuit is analog and its signal is dependent on analog components that degrade with time and I think this is what leads to problems. The signal gets supplied when it shouldn't, at a level that the ECU is not expecting with a profile that it doesn't want.
My replacement is purely digital so the performance does not degrade and because of this, I can add extra functionality such as the speedo calibration facility and diagnostics so it offers more than a straight replacement.
With people more than ever changing wheel and tyres, the calibration is quite a useful thing to be able to do.
There appears to be a thriving trade in these boxes from the breakers judging by the wait I had in getting hold of one so I can only assume that people are replacing them on a frequent basis. Hence the reason for doing it.
I've seen on ECUmate how the 14CUX idle behaviour is very sensitive to the road speed input and that it is not a simple on/off switch which the TVR box seems to think.
The circuit is analog and its signal is dependent on analog components that degrade with time and I think this is what leads to problems. The signal gets supplied when it shouldn't, at a level that the ECU is not expecting with a profile that it doesn't want.
My replacement is purely digital so the performance does not degrade and because of this, I can add extra functionality such as the speedo calibration facility and diagnostics so it offers more than a straight replacement.
With people more than ever changing wheel and tyres, the calibration is quite a useful thing to be able to do.
There appears to be a thriving trade in these boxes from the breakers judging by the wait I had in getting hold of one so I can only assume that people are replacing them on a frequent basis. Hence the reason for doing it.
Steve,
Do you know Richard Smith at Dulford Automotive?
My car did this with the 14cux, before and after supercharging. He had numerous maps sent to him by MA to try to get rid of it.
I replaced the 14cux with the v8d plug'n'play Megasquirt solution and it still did it - Richard finally has got rid of the problem after much tweaking.
It might be worth having a chat with him.
Do you know Richard Smith at Dulford Automotive?
My car did this with the 14cux, before and after supercharging. He had numerous maps sent to him by MA to try to get rid of it.
I replaced the 14cux with the v8d plug'n'play Megasquirt solution and it still did it - Richard finally has got rid of the problem after much tweaking.
It might be worth having a chat with him.
Yes I know Richard. I'll give him a call. Thanks.
The V8D megasquirt system uses the existing loom and sensor inputs as the 14CUX as I understand it. This means that it will also be getting the all or nothing road speed signal and ECUs really do not like this as it causes them real problems, because the input behaviour is abnormal. I am not surprised that you have seen this type of behaviour with a Megasquirt. It can be tweaked out but that is really not an option with the 14CUX which behaves itself normally, especially in Range Rovers.
I am becoming more convinced that this little box is the root of a lot of funnies/problems.
The V8D megasquirt system uses the existing loom and sensor inputs as the 14CUX as I understand it. This means that it will also be getting the all or nothing road speed signal and ECUs really do not like this as it causes them real problems, because the input behaviour is abnormal. I am not surprised that you have seen this type of behaviour with a Megasquirt. It can be tweaked out but that is really not an option with the 14CUX which behaves itself normally, especially in Range Rovers.
I am becoming more convinced that this little box is the root of a lot of funnies/problems.
shpub said:
The V8D megasquirt system uses the existing loom and sensor inputs as the 14CUX as I understand it. This means that it will also be getting the all or nothing road speed signal and ECUs really do not like this as it causes them real problems, because the input behaviour is abnormal...
I've never seen an MS with any speed signal input, it works simply off the engine speed.
spend said:
shpub said:
The V8D megasquirt system uses the existing loom and sensor inputs as the 14CUX as I understand it. This means that it will also be getting the all or nothing road speed signal and ECUs really do not like this as it causes them real problems, because the input behaviour is abnormal...
I've never seen an MS with any speed signal input, it works simply off the engine speed.
Megasquirt gets its speed signal from the CAS or the PIP signal from an EDIS module, just out of interest my speedo traducer (T5 box GKN diff) failed a couple of years back and had no effect on the idle speed or idle speed drop down when rolling up to a junction etc in any fashion with the original CUX ECU 
not sure if this helps in any way Shpub, in fact it as taken a little while to get my MS ECU to mirror how good the idle characteristics were with the CUX ECU, I cant say I am a lover of the Delco stepper design on the Hotwire RV8 as idle control anyways I much prefer the ported/rotary type idle valve's (i.e Bosch) as these are a lot less temperamental, in saying all this my idle as never caused me any concern but I know plenty of others who cannot say the same 
Nice piece of diagnostic kit for the CUX Shpub 
Edited by Simon says on Saturday 7th January 17:38
Interesting... I stand corrected about the speed sensor input - it is available but only on the later versions and as either a way of including road speed into data logs or as a traction control feature.
I don't use Road speed either but it is a nice way in conjunction with an idle control of setting a double stage idle: a higher idle to make things nice and smooth when lifting off and a low idle for when the car is stationary.
I don't use Road speed either but it is a nice way in conjunction with an idle control of setting a double stage idle: a higher idle to make things nice and smooth when lifting off and a low idle for when the car is stationary.
shpub said:
Interesting... I stand corrected about the speed sensor input - it is available but only on the later versions and as either a way of including road speed into data logs or as a traction control feature.
The transducer uses the mechanical movement of the speedo cable and "transforms" it into an electronic signal for the ECU. When it fails, the most obvious symptom is an erratic idle. The most common reason for failure is that the transducer has lost its earth. Let's assume your problem is more than that.The wisest replacement is to use a company called Speedy Cables http://www.speedycables.com/ in Wales. They are the specialist division of the company Caerbont, which makes many different instruments under many different names. If you tell Speedy Cables (try Andy Barraclough) what engine and fueling system you have, who made your speedo and the length of cable you need. He will make you one that will cover all the bases.
+8
plushuit said:
The transducer uses the mechanical movement of the speedo cable and "transforms" it into an electronic signal for the ECU. When it fails, the most obvious symptom is an erratic idle. The most common reason for failure is that the transducer has lost its earth. Let's assume your problem is more than that.
The wisest replacement is to use a company called Speedy Cables http://www.speedycables.com/ in Wales. They are the specialist division of the company Caerbont, which makes many different instruments under many different names. If you tell Speedy Cables (try Andy Barraclough) what engine and fueling system you have, who made your speedo and the length of cable you need. He will make you one that will cover all the bases.
+8
I think you are getting Morgans/Range Rovers confused with TVRs... The wisest replacement is to use a company called Speedy Cables http://www.speedycables.com/ in Wales. They are the specialist division of the company Caerbont, which makes many different instruments under many different names. If you tell Speedy Cables (try Andy Barraclough) what engine and fueling system you have, who made your speedo and the length of cable you need. He will make you one that will cover all the bases.
+8
There is no mechanical speedo cable on the Chimaeras or Griffs or V8S.
The signal is generated on the LT77 box by an internal speedo transducer which generates the correct signal for the 14CUX ECU. On the T5 equipped cars, an external sensor is used to generate the signal for the speedo but is incompatible with the 14CUX ECU hence the TVR conversion box. It is this box that is the bodge and causes a lot of problems.
Edited by shpub on Wednesday 11th January 09:09
shpub said:
piemanpaul said:
Steve,
I have an early Chim 430 with the Rover SD1 ‘box, has similar high idle behavior at times. Is cause and fix the same or different for the SD1 ‘box?
Paul
Probably related but as your car doesn't have the TVR box that I am redesigning, it is different. I have an early Chim 430 with the Rover SD1 ‘box, has similar high idle behavior at times. Is cause and fix the same or different for the SD1 ‘box?
Paul
Even for the TVR specific minds, it is quite a minefield understanding which particular permutation of bits to drive speedo & ECU that folks have in their cars, and an area where you see many posts talking with their wires crossed
shpub said:
piemanpaul said:
Steve,
I have an early Chim 430 with the Rover SD1 ‘box, has similar high idle behavior at times. Is cause and fix the same or different for the SD1 ‘box?
Paul
Probably related but as your car doesn't have the TVR box that I am redesigning, it is different. I have an early Chim 430 with the Rover SD1 ‘box, has similar high idle behavior at times. Is cause and fix the same or different for the SD1 ‘box?
Paul
Gassing Station | Chimaera | Top of Page | What's New | My Stuff