Print

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

[Kevin Wood]

A quick google revealed this: No idea how accurate :-/

12 Initiation of diagnosis

37 Fault code light low voltage

38 Oxygen sensor voltage low

98 Oxygen sensor wiring break / open circuit

17 Knock sensor 2 No change in voltage


Looks like one of the O2 sensors and one of the knock sensors are not happy. I guess this could have invoked a "safe" ignition timing hence the poor economy.

Kevin

[VXL V6]

Reckon 98 is the code to take note of, double scheck the wiring where the link was made for the LPG and at the LPG ECU end.

....Well I'm guessing!

[Jay w]

Reckon 98 is the code to take note of, double scheck the wiring where the link was made for the LPG and at the LPG ECU end.

....Well I'm guessing!

I didn't cut the wire, the shielding was bared back, the new wire soldered on and then heat wrap attached, it all looks good and solid

i reckon we have killed a Lambda, it was running for a few mins on gas, tried to accelarate and the light came on, since then she has run rough on petrol, when you put your foot down it misses, cough splutters and runs a bit naff.....

The knock sensor could have been invoked when it was running lean on one of the earlier attempts, could this be a 'stored code' or are these in order of issue with the last one being the one that is giving the biggest grief?

[Kevin Wood]

i reckon we have killed a Lambda

Make sure you've got the LPG ECU configured for narrow band (voltage) lambda inputs not wideband.

The knock sensor could have been invoked when it was running lean

It's unlikely unless you were driving it under load I'd say. The error code hints more at an "I'm getting no signal from the knock sensor" rather than "your engine is knocking" but they don't always mean what you'd expect. Can anyone expand on these codes?

Kevin

[Jay w]

i reckon we have killed a Lambda

Make sure you've got the LPG ECU configured for narrow band (voltage) lambda inputs not wideband.

The knock sensor could have been invoked when it was running lean

It's unlikely unless you were driving it under load I'd say. The error code hints more at an "I'm getting no signal from the knock sensor" rather than "your engine is knocking" but they don't always mean what you'd expect. Can anyone expand on these codes?

Kevin

yes we have it configured so as it is one wire per lambda, had i connected both to the same lambda it would have been setting it up as a wideband unit....hope that makes sense

[Marks DTM Calib]

The only sensors with screend wires I know of are the knock sensors, will double check the wire colours and pin outs tonight if I get chance.

[VXL V6]

The Lambda cables are four core unshielded on mine (four lambda's on a 3.2). Not sure what the 2.5's have.

[Marks DTM Calib]

Heres the schematic



Lambda sensors are screened and goto the following ECU pins:

1-3-5 Bank - Pin 28 - Brown Blue
2-4-6 Bank - Pin 47 - Brown Blue

I need to check some other info on the knock sensors.....I wonder if you have inadvertantly connected into a knock sensor wire.

[VXL V6]

Mark,

I presume they are only shielded within the engine bay loom then, because the lambda connection at the exhaust to the connector to the loom isn't.
Guess it'd make sense to shield it in the engine bay where it's susceptable to noise.

[Jay w]

here's how we found the wires:

located pins 28 and 47, stripped the plastic cover off the back of the ECU connector, confirmed they were brown with blue tracers.
continuity checked the pins until we found the correct wire, then did a voltage check to check for signal, on average there was one adjustment every second and the two of them were giving similar readings.

stripped the shielding, soldered the LPG wires to the lambda wires taking care not to allow the shieling to contact the wire, then heat shielded the solder and then finish with a small amouth of insulation tape as a belt and braces.

once we fired it up initially the EML light was out, it wasn't until we tried to rev the engine that it came one and stayed on.

what we have now is a misfire situation, every now and again the car will run fine and will really fly along, then it will falter, similar symptoms to dispack, it will stutter, sound like it is running on 3or 4 and then clear and shoot off again, if you try and floor it whlist it is faltering it will almost suffer all the symptoms of overfuelling, let your foot off the throttle and it will sometimes clear

[VXL V6]

Couple of thoughts.... (as I understand it so take it with a pinch of salt!)

Lambda light wouldn't come on (unless on from last time engine run) during the warm up period, I think until the CTS reports something like 62 degrees it's classed as a cold start.

Is it possible that the Dis Pack / connections have been disturbed? When my Emissions lamp is on it doesn't affect the performance or economy of the car and it certainly doesn't retard the ignition.

I think the rich / lean voltages are too fast to see on a meter or most diag equipment, only real test is with an osciloscope. What you see / read on most equipment is over a longer time base.

[Jay w]

Couple of thoughts.... (as I understand it so take it with a pinch of salt!)

Lambda light wouldn't come on (unless on from last time engine run) during the warm up period, I think until the CTS reports something like 62 degrees it's classed as a cold start.

Is it possible that the Dis Pack / connections have been disturbed? When my Emissions lamp is on it doesn't affect the performance or economy of the car and it certainly doesn't retard the ignition.

I think the rich / lean voltages are too fast to see on a meter or most diag equipment, only real test is with an osciloscope. What you see / read on most equipment is over a longer time base.

I'll take your word on the preheat on the lambdas, i have little experience of them.

dispack wasn't disturbed at all last night, we were not near the connections

part of the OBD kit with the LPG kit is an osciloscope, initially we were using a multimeter but used that as a back up to confirm.

this is going to be a series of elimination in order to get to the bottom of it

[VXL V6]

I'll take your word on the preheat on the lambdas, i have little experience of them.

this is going to be a series of elimination in order to get to the bottom of it

Actually the lambdas heat the CAT as well AFAIK, On diags the CAT heaters should go off when the sensor reads the temp as hot enough....

I think that's why the Lambdas have four wires.

You'll get it sorted soon mate, especially when you get some decent advice rather than my drivel!!!!

[Jay w]

I'll take your word on the preheat on the lambdas, i have little experience of them.

this is going to be a series of elimination in order to get to the bottom of it

Actually the lambdas heat the CAT as well AFAIK, On diags the CAT heaters should go off when the sensor reads the temp as hot enough....

I think that's why the Lambdas have four wires.

You'll get it sorted soon mate, especially when you get some decent advice rather than my drivel!!!!

not drivel mate, just assisting with the elimination process

[Jay w]

well the old girl is getting worse by the seems of it, and time is not on my side so to a small degree i am going to have to admit defeat.

disconnected the ECU tonight, reset it and then read the codes, same as before, so it would seem that we are looking at a knock sensor and possibly a lambda, in addition i need to be sure that the LPG kit is wired into the right pick ups and the nozzels need drilling out.

all of this may not seem like a lot but i only have saturday morning in which to do it, i have still to get the car back from Epsom to Plymouth on Friday evening after work, it is then going to a friends garage for the following week, i am getting a rental and he is dealing with the issues.

It's not the ideal solution but i have spent no time with the family recently, i cannot get to the bottom of this wiring and i am tired of heading down blind alleys, i just want the car done and running on gas