Lambda readings swinging from low to high
Discussion
Having a nose round the workings of the injection system (to try and smooth out the mild low speed shunting). Checked the voltage across the lambda probe wires, seem to be getting strange reading -
engine running at warm tickover lambda shows approx 0.01v swinging right up to approx 1.10v, both sides about the same, so don't think it's the sensors themselves at fault. Already checked the AFM (1.7v) and throttle pot. Seems as though engine is over fueling (as plugs are quite sooted after only about 500 miles) giving high lambda reading then ECU is trying to (over)compensate giving very low lambda reading, also maybe the cause of the low speed shunt???
Any idea(s) of what may be the of this would be greatly appreciated.
Nick
Car is a 95 serp 500 standardno mods
engine running at warm tickover lambda shows approx 0.01v swinging right up to approx 1.10v, both sides about the same, so don't think it's the sensors themselves at fault. Already checked the AFM (1.7v) and throttle pot. Seems as though engine is over fueling (as plugs are quite sooted after only about 500 miles) giving high lambda reading then ECU is trying to (over)compensate giving very low lambda reading, also maybe the cause of the low speed shunt???
Any idea(s) of what may be the of this would be greatly appreciated.
Nick
Car is a 95 serp 500 standardno mods
Edited by selnic on Monday 3rd March 21:33
selnic said:
thanks for the replys chaps.. see what you are geting at now , having just spent the last 2 hours reading up, I've realised the bible is'nt quite as good as it seems. The info it gives is for zirconium sensors and not titania as OEM from TVR!!!
Nick
Strange as the information was consistent from several sources including Land Rover who use the same titania sensors as TVR. Nick
The probes bog standard are voltage producing type, NOT resistive. They are nothing fancy these, although some would make you believe so to keep the price high !! The normal voltage will spike up to around 1.2 volts, but this does not mean there is a fault, its a normal reading. If you want to check the ECU's ability to correct the base fueling do the following:
Get the engine / probes hot.
Reset the ECU by disconnecting the battery for 30 seconds.
Reconnect, and start the engine and watch the Lambda voltages (give it 30 seconds ignition on to heat the probes again before starting). They will start static high or low, depending on how far out the base fueling is on the fixed part of the ECU fuel map. After about 15 -20 seconds they should start to switch at about .5 second intervals as the ECU starts to alter the base fueling (This involves setting the whole fuel map richer or leaner). If you have a variable fuel pressure regulator, you can tweek the pressure, and watch the the process repeat, but its just to prove its working. This only happens at tickover. The remaining small fueling tweeks take place above tickover almost instantly as the engine mixture constantly cycles around lambda one. When the ECU goes open loop aboue about 3/4 throttle or 3000 rpm, the lambda probes should simply read .7 volts or more all the time. Its possible to overfuel the engine at this point if say the fuel pressure is to high, but you cant know how much, as it exceeds the lambda probes meaningfull range. I have run my fuel pressure up to 50 PSI on idle, and the ECU has still managed to keep it in range at this point. I hate to think what the open loop fueling would have done however. People underestimate the 14CUX, its pretty smart for its era !
Mark
Get the engine / probes hot.
Reset the ECU by disconnecting the battery for 30 seconds.
Reconnect, and start the engine and watch the Lambda voltages (give it 30 seconds ignition on to heat the probes again before starting). They will start static high or low, depending on how far out the base fueling is on the fixed part of the ECU fuel map. After about 15 -20 seconds they should start to switch at about .5 second intervals as the ECU starts to alter the base fueling (This involves setting the whole fuel map richer or leaner). If you have a variable fuel pressure regulator, you can tweek the pressure, and watch the the process repeat, but its just to prove its working. This only happens at tickover. The remaining small fueling tweeks take place above tickover almost instantly as the engine mixture constantly cycles around lambda one. When the ECU goes open loop aboue about 3/4 throttle or 3000 rpm, the lambda probes should simply read .7 volts or more all the time. Its possible to overfuel the engine at this point if say the fuel pressure is to high, but you cant know how much, as it exceeds the lambda probes meaningfull range. I have run my fuel pressure up to 50 PSI on idle, and the ECU has still managed to keep it in range at this point. I hate to think what the open loop fueling would have done however. People underestimate the 14CUX, its pretty smart for its era !
Mark
Forgot to add:
The shunting is caused by lack of airflow with the TVR setup (ie cam) through the AFM at low RPM, so the ECU under fuels the engine at around 1500 RPM. (This is above the area the ECU adjusts the base fueling). As the mixture is very lean, anything that improves the burn of what fuels there will improve things, so you will get a range of "fixes". TVR power have been working on an increased Plenum chamber that supposed to help. I would guess that it reduce the effect of the air pulses that you get back through the inlet port with a longer duration cam, that seems to upset the AFM. I tackled it by lifting the AFM output voltage a couple of hundred millivolts, by putting a low value resistor in the earth line of the AFM, that helped, by fooling the ECU to put a bit more fuel in. It skews the ECU mapping a bit however, so the resistance to be set very accurately. If you dont mind experimenting a bit, Ill send you some resistors to try. Its fustrating to me as Ive only worked on one car, and I need more testing. No one so far seems to be interested in what I have found and I appear to be fustrated by "TVR people" wanting to sell a much higher cost fix than 30p's worth of resistors!
Mark
The shunting is caused by lack of airflow with the TVR setup (ie cam) through the AFM at low RPM, so the ECU under fuels the engine at around 1500 RPM. (This is above the area the ECU adjusts the base fueling). As the mixture is very lean, anything that improves the burn of what fuels there will improve things, so you will get a range of "fixes". TVR power have been working on an increased Plenum chamber that supposed to help. I would guess that it reduce the effect of the air pulses that you get back through the inlet port with a longer duration cam, that seems to upset the AFM. I tackled it by lifting the AFM output voltage a couple of hundred millivolts, by putting a low value resistor in the earth line of the AFM, that helped, by fooling the ECU to put a bit more fuel in. It skews the ECU mapping a bit however, so the resistance to be set very accurately. If you dont mind experimenting a bit, Ill send you some resistors to try. Its fustrating to me as Ive only worked on one car, and I need more testing. No one so far seems to be interested in what I have found and I appear to be fustrated by "TVR people" wanting to sell a much higher cost fix than 30p's worth of resistors!
Mark
blitzracing said:
The probes bog standard are voltage producing type, NOT resistive.
I thought they were generally agreed to be titania type.http://www.lambdasensor.com/main/lrover.htm
Hi all...many thanks for the advise... update on the grif, it's now pretty much sorted, after pining out all the loom, sensors, earths etc etc have finaly found the cause of the erratic fueling, turns out the AFM was Knackered, (set the base with ign on 0.5v, start engine, check voltage - should be around 1.7/8 ish, however no increase was shown, rev engine to around 4k voltage would only increase by about 0.5v = FUBAR!!!
Fitted new AFM today an set up, car drives totaly differant, no shunting able to maintain 30mph on slight gradient with no throttle at all = Winner. Now I've got all that out of the way and am starting with something with a proper working system, might be worth trying the resistor(s) as suggested by Mark. By the way mark, what rating did you use for the resistors?
Nick
Fitted new AFM today an set up, car drives totaly differant, no shunting able to maintain 30mph on slight gradient with no throttle at all = Winner. Now I've got all that out of the way and am starting with something with a proper working system, might be worth trying the resistor(s) as suggested by Mark. By the way mark, what rating did you use for the resistors?
Nick
eliot said:
blitzracing said:
The probes bog standard are voltage producing type, NOT resistive.
I thought they were generally agreed to be titania type.http://www.lambdasensor.com/main/lrover.htm
Mark
selnic said:
Hi all...many thanks for the advise... update on the grif, it's now pretty much sorted, after pining out all the loom, sensors, earths etc etc have finaly found the cause of the erratic fueling, turns out the AFM was Knackered, (set the base with ign on 0.5v, start engine, check voltage - should be around 1.7/8 ish, however no increase was shown, rev engine to around 4k voltage would only increase by about 0.5v = FUBAR!!!
Fitted new AFM today an set up, car drives totaly differant, no shunting able to maintain 30mph on slight gradient with no throttle at all = Winner. Now I've got all that out of the way and am starting with something with a proper working system, might be worth trying the resistor(s) as suggested by Mark. By the way mark, what rating did you use for the resistors?
Nick
That must have been serious shunting, the one Ive been working on happens at 1500 rpm in the high gears and very low throttle openings. As a matter of interest a hairdryer makes a good test for the AFM, if you disconnect the pipework, but leave the connections on. With the ignition on, you can easily alter the output with a blast of air. As for the resistors there is around 100-150ma flowing down the earth wire, so you need between .8 ohms and 1.2 ohms to lift the output voltage around 150mv. The higher the resistance, the more fuel. This works out at less than .03 watts, but I doubt you would get such low wattage resistors in these very resistance low ranges. I ended up with some 5 watt ones for that reason. If you do start experimenting, you need to keep monitoring the lamda probes at tickover, to make sure they keep switching and the ECU is coping with the changed "airflow". Its quite easy to overfuel the catalysts otherwise. On the car I set up this happened at 1.5 ohms.Fitted new AFM today an set up, car drives totaly differant, no shunting able to maintain 30mph on slight gradient with no throttle at all = Winner. Now I've got all that out of the way and am starting with something with a proper working system, might be worth trying the resistor(s) as suggested by Mark. By the way mark, what rating did you use for the resistors?
Nick
If you are going to bin the AFM, Id like to take a close look at it to see the internal electronics and if they can be adjusted. I dont want to chop my good one open !
Mark
blitzracing said:
You can also measure the probe voltage outputs with them disconnected completely from the ECU as long as they are hot. (The ecu does not like it much however.)Mark
Interesting. Certainly, if they produce a voltage when heated with the ecu disconnected, i would have to agree they are not resisitive.Gassing Station | Griffith | Top of Page | What's New | My Stuff