need help please with 'ecumate' numbers!
Discussion
hi
So for past 10 years we have owned 500 which has had shunting issues. have tried some solutions - which to be fair havn't helped much (smooth bore inlet, TVR specialist servicing, v8 engine specialist fettle, ecu specialist and remap with new chip). Family, extension and other car issues sometimes leave the griff alone.
I am now determined to 'fix' it.
Recently fitted plenum spacer and gasket, together with returning the exhaust to standard. This has seemed to help some BUT still have shunting.
Having read various threads relating to ecu function and sensors and also mark blitzracings excellent site i was intrigued by the possibilities.
I have just bought the 'ecumate' from steve heaths site and wow, what a gizmo.!!
It has shown no fault codes retained but the live data shows an issue with lambda b (cyl 2,4,6,8). i have swapped the lambda sensors proving the issue is with that bank - not the sensor. i am not suggesting this issue has been missed by previous experts as the car has sat for long periods since (i am ashamed of some of the yearly mileages discovered come mot time!) - and i have also changed some of the mechanical constituents since ie plenum and exhaust. reason for the post is some suggestions regarding the next step?
Issue is :-
lambda a shows almost perfect numbers reducing to about 80 then switching and climbing to about 120 about every second
lambda b shows manic switching sometimes lean but mostly rich (the numbers varying widely between 0-255) with no consistency.
I would suggest this is the reason for the lumpy idle and potentially the current shunting.
Other numbers show:-
afm1.67v
stepper motor 133
water 78 c
fuel 19 c
throttle pot 6%
rpm 920
throttle .31v
tune white resistor plus cats
Any advice gratefully received
Thanks in advance
ada
So for past 10 years we have owned 500 which has had shunting issues. have tried some solutions - which to be fair havn't helped much (smooth bore inlet, TVR specialist servicing, v8 engine specialist fettle, ecu specialist and remap with new chip). Family, extension and other car issues sometimes leave the griff alone.
I am now determined to 'fix' it.
Recently fitted plenum spacer and gasket, together with returning the exhaust to standard. This has seemed to help some BUT still have shunting.
Having read various threads relating to ecu function and sensors and also mark blitzracings excellent site i was intrigued by the possibilities.
I have just bought the 'ecumate' from steve heaths site and wow, what a gizmo.!!
It has shown no fault codes retained but the live data shows an issue with lambda b (cyl 2,4,6,8). i have swapped the lambda sensors proving the issue is with that bank - not the sensor. i am not suggesting this issue has been missed by previous experts as the car has sat for long periods since (i am ashamed of some of the yearly mileages discovered come mot time!) - and i have also changed some of the mechanical constituents since ie plenum and exhaust. reason for the post is some suggestions regarding the next step?
Issue is :-
lambda a shows almost perfect numbers reducing to about 80 then switching and climbing to about 120 about every second
lambda b shows manic switching sometimes lean but mostly rich (the numbers varying widely between 0-255) with no consistency.
I would suggest this is the reason for the lumpy idle and potentially the current shunting.
Other numbers show:-
afm1.67v
stepper motor 133
water 78 c
fuel 19 c
throttle pot 6%
rpm 920
throttle .31v
tune white resistor plus cats
Any advice gratefully received
Thanks in advance
ada
Those numbers look OK to me, apart from LamB. A similar lambda problem was discussed earlier, which might give you some leads:
http://www.pistonheads.com/gassing/topic.asp?h=0&a...
Life's so much easier with a pre-cat.
http://www.pistonheads.com/gassing/topic.asp?h=0&a...
Life's so much easier with a pre-cat.
Most RV8s in TVRs with ECU 14Cux suffer to some degree some shunting on low rpm (1200+-)on a constant cruise throttle. I tolerate mine with driving technique that suits me. For town driving I use slight open throttle followed by coasting on closed, then more slight throttle followed by closed with sometimes clutch action braking when necessary. Once throttle is opened for acceleration from 1200 in 4/5th gear there is no shunting, nice smooth pick up. Cruising at 70plus no shunting. Unable to comment on ecumate numbers. Hope to get an Ecumate some day.
Edited by EGB on Sunday 15th April 18:08
Egb
I understand they all do that to a degree, but mine seems worse than it should. Coupled with the erratic lambda b I believe there is a fault, particularly as the lambda a reading alternates so methodically.
Time to tinker!
By the way - the information provided by the ecumate is fantastic. Certainly gives a window to what's happening with the ecu - even though my knowledge doesn't enable me to clearly see it!!
Regards Ada
I understand they all do that to a degree, but mine seems worse than it should. Coupled with the erratic lambda b I believe there is a fault, particularly as the lambda a reading alternates so methodically.
Time to tinker!
By the way - the information provided by the ecumate is fantastic. Certainly gives a window to what's happening with the ecu - even though my knowledge doesn't enable me to clearly see it!!
Regards Ada
If you are hitting 255 on the lambda its hit the limit of correction the ECU can apply. The problem is if its banging away between 0 and 255, its basically overshooting as it tries to correct, so the switching becomes slower as the ECU has to make big corrections. The problem is the Lambda probes just look at unused oxygen in the exhaust, so anything that causes a poor burn (inc misfires) will make the mixture look lean as there is excess oxygen in the exhaust. The ECU then dumps loads of extra fuel in to try and compensate, but this does not really fix it, but it may be enough to make the probe reading change from lean to rich. Then it has to reverse the process to get it lean enough again. A good map, with working sensors on a good engine will cycle quite quickly (faster as the RPM rises). Both sides should be the same.
You may find the shunting impossible to sort, as a combination of a performance cam, plenum chamber inlet and catalyst fuel map is a very bad combination for low speed behavior.
You may find the shunting impossible to sort, as a combination of a performance cam, plenum chamber inlet and catalyst fuel map is a very bad combination for low speed behavior.
Edited by blitzracing on Sunday 15th April 20:30
Thanks for the interest fellas!
I understand the shunting issues may not be resolved completely- maybe mega squirt one day BUT I seem to have a current issue with one bank of cylinders that needs resolving
Lambda a does the perfect cycle you mention - approx midpoint of 100 falling to about 80 then the ecumate shows the correction and rises to about 120 then falls to about 80 repeat. Perfect? Occurs about once per second.
Lambda b is erratic. I thought I had found something earlier having recently changed the exhaust gasket - needed re tightening. Got all excited- needn't have! No difference. Removed plugs - all consistent if a little sooty. There is clearly an issue on that bank alone. I assume the ecu uses the respective lambda inputs to provide fuel for that bank?
Did think about bypassing the loom to feed lambda signal direct to input on ecu - but the design of the ecu plug seems to prevent this.
I guess the issue could only be spark, fuel or lambda signal to ecu?
what about splitting the perfect lambda signal to both inputs?
Regards ada
I understand the shunting issues may not be resolved completely- maybe mega squirt one day BUT I seem to have a current issue with one bank of cylinders that needs resolving
Lambda a does the perfect cycle you mention - approx midpoint of 100 falling to about 80 then the ecumate shows the correction and rises to about 120 then falls to about 80 repeat. Perfect? Occurs about once per second.
Lambda b is erratic. I thought I had found something earlier having recently changed the exhaust gasket - needed re tightening. Got all excited- needn't have! No difference. Removed plugs - all consistent if a little sooty. There is clearly an issue on that bank alone. I assume the ecu uses the respective lambda inputs to provide fuel for that bank?
Did think about bypassing the loom to feed lambda signal direct to input on ecu - but the design of the ecu plug seems to prevent this.
I guess the issue could only be spark, fuel or lambda signal to ecu?
what about splitting the perfect lambda signal to both inputs?
Regards ada
I would not say it was THAT bad, its an old design of ECU with slow probes, so nothing is that precise. The easiest thing to do to see if it is lambda cycling making it shunt is to change the tune resistor to the non cat map the reset the ECU. You then need to set the mixture with the CO trim so the engine runs nicely and give it a test. Dont leave it like this however, the catalyst wont like it.
It could also be a combination of things. There could be a small misfire that is causing a bad lambda reading which the ECU then tries to adjust for but gets it very wrong. Might be worth looking under the bonnet at night in the dark to see if there is any ignition sparks jumping or tracking between leads or around the dizzy cap. Sometimes the plug lead layout can cause slight misfires.
EGB said:
I understand from another post that the Megasquirt set up gets rid of all shunting. No shunting at 20mph in 5th gear. At a price.
One of the main reasons is you set the target AF ratio to what you want, not what the anti polution lobby want and the poor old 14CUX tries to achieve. You can get them to shunt less in some cases by richening the mixture slightly, but its near impossible if you still run the catalyst fuel map, the ECU just keeps correcting any changes you make to sensor inputs or fuel pressure. The non cat fuel map is much better, but it could damage the catalysts as the AF ratios used are over a much wider range, both richer and leaner. The nice thing about the MS is it uses lambda feedback all the time to set the target AF ratios, whilst the 14CUX has to make do with its fixed map settings and CO trim, and hope its near correct. This can be made to work prefectly well, as long as the base map is good, and engine / sensors in good nick. Just a thought about the manic switching- I think what you are seeing is possibly a slow lambda response (old probe). I think the ECU mate is simply displaying the voltages, but as they are not switching rapidly, I think you are seeing many more voltage steps between the cycles than you would with a new probe.
This may not help as others on here are far more familiar with the Rover ECU but as only one bank is affected it surely has to be something common to that bank. From memory the 2 injector banks are fed seperately from two separate pins on the ECU plug/socket perhaps one connection is not 100%. Maybe worth a check.
Bank 1-3-5-7 Pin 13
Bank 2-4-6-8 Pin 11
Bank 1-3-5-7 Pin 13
Bank 2-4-6-8 Pin 11
Edited by Loubaruch on Monday 16th April 15:44
Problem you have is the probe reads the sum exhaust gas from 4 cylinders, and just averages it out, so it one is wrong, then it skews the results for the other 3, and appears to be a bank fault. I dont think these probes are fast enough to show bad combustion from just one cylinder at a time. (You would need a 'scope to see it on the wave form). I think I tried pulling one plug lead off when I did scope the signals, and it still just averaged out. Modern probes are fast enough to pick up one cylinder at a time.
Hi
Done some more tinkering!
Swapped lambdas again - just to be sure. Issue stayed with rhs bank ie 2,4,6,8. No difference.
Convinced myself it might be wiring as I couldn't test consistent voltage on the lambda +v input. Took power from the good side and fed lambda signal direct to pin 24 ecu. no difference.
Inspected all the female pins on the major ecu connection. Found pin 24 had bent female connection. Winner??! Pin 24 is the offending lambda input. Straightened out connection making good again- No difference.
All lights off - checked for sparking- none seen.
was going to swap dizzy cap and rotor for my new 'Lucas' kit. Found to be half the weight of my current Lucas items, had no embossed manufacturer name and the centre graphite pin had massive lateral play. Cleaned contacts on my current kit with a Drexel brass brush. No difference ( will return the 'Lucas' kit as paid top dollar.
I have heard about rough soldered ecu's. And wonder if I can swap pins for both lambda and injectors. Effectively moving bank a pins to bank b and vice versa. If issue remains I guess I have narrowed it to mechanical issues on the offending bank.
Will this work or am I asking for trouble?
Not sure if it was apparent on the video but the ecumate occasionally flashes ecu error before it returns to ok. Is this something to investigate further steve?
Regards
Ada
Done some more tinkering!
Swapped lambdas again - just to be sure. Issue stayed with rhs bank ie 2,4,6,8. No difference.
Convinced myself it might be wiring as I couldn't test consistent voltage on the lambda +v input. Took power from the good side and fed lambda signal direct to pin 24 ecu. no difference.
Inspected all the female pins on the major ecu connection. Found pin 24 had bent female connection. Winner??! Pin 24 is the offending lambda input. Straightened out connection making good again- No difference.
All lights off - checked for sparking- none seen.
was going to swap dizzy cap and rotor for my new 'Lucas' kit. Found to be half the weight of my current Lucas items, had no embossed manufacturer name and the centre graphite pin had massive lateral play. Cleaned contacts on my current kit with a Drexel brass brush. No difference ( will return the 'Lucas' kit as paid top dollar.
I have heard about rough soldered ecu's. And wonder if I can swap pins for both lambda and injectors. Effectively moving bank a pins to bank b and vice versa. If issue remains I guess I have narrowed it to mechanical issues on the offending bank.
Will this work or am I asking for trouble?
Not sure if it was apparent on the video but the ecumate occasionally flashes ecu error before it returns to ok. Is this something to investigate further steve?
Regards
Ada
Gassing Station | Griffith | Top of Page | What's New | My Stuff