Right I`m going to fit the lambda probe tomorrow, need to drill the manifold and weld a boss in. I have 2 choices really, the first is over 1 pipe coming down from the cylinder head, loads of room but it will only `sniff` 1 cylinder (does this matter apart from if I have a dodgey injector or plug / missfire?) The second is to drill after the joint between the manifold and centre section under the car, space may be limited due to the tunnel and the probe will be quite a way back from the fresh exhaust gasses, will this be a problem? Plus if I balkls it up here I don`t have to buy another manifold What do you recon`s best?
What type of sensor is it? If it's a 1-wire sensor then it needs to be close to the head to keep it hot, otherwise it'll read lean all the time and die. If it's a heated sensor then you can move it further down the exhaust. Obviously the further away it is the longer it will take to respond to changes in fuelling, but then if it's only a narrowband it's not going to need to be that speedy, it's only working at low engine speeds. If it helps, here's a pic of my G60 downpipe with the stock lambda location. Bear in mind V8 cars happily run a lambda probe an the 'H' pipe, miles away from the heads.
It`s a 4 wire heated jobby from a corraddo, will try it at the joint under the car, seems the best place for it although I notice the manifold in your pic means it`s only reading from 2 cylinders?
Depends what ECU you're using and the transport delay in the software. Also make sure it's at at least a 10degree angle so water doesn't collect in it.
Using the genuine bosch digi1 ecu with a bbm chip, looks like the normal g60 has it up near the cylinder head then? Will it make a difference sniffing 1 rather than 4 cylinders? may be a problem angleing it under the car due to the tunnel, the pobe will stick out 2 inces or more from the centre pipe. Don`t mind fitting it at the top if it`ll work, just don`t want to drill the manifold and then find out it doesn`t.. Cheers for the help.
You should try and match the distance from the exhaust port to the probe of the original fitment. There's nothing wrong with putting it in number 1 cylinder apart from the points you've made about having problems with No1, this also works the other way round, if you've got a problem with 2-4 the ECU wont try and solve it.
I might risk it then and go with one of the runners, seen some cast kr manifolds with an o2 hole in them (maybe for setting up mixtures?). think about a foot down from the port should be about right, saves chopping the wiring as well having it at the top.
Had a look today, decided it`ll just fit before the centre silencer, need to take the middle section off to do it properly and it`s started raining [:^(] Also I need to ground the silencer somehow, maybe weld a nut onto it (sensor grounds though it`s body).
About a foot back from the steering rack? Only ran it for 2 mins to get it out of the garage and onto the ramp and it was quite hot underneath! See your point though. Can`t do the top unless it`s close to the ex ports, unless I take the manifold off which is a nightmare without car lift! Suppose if it came to that I`ll be heat wrapping the f**ker at the same time!
Keith I think that may be too far back for the exhaust temp [:^(] What downpipe/manifold/system are you using? From martyn's pic you really need to be looking somewhere close to the G60 placement. Even if it means the probe only sniffing one cylinder better that than a bogus figure from all four.
Yeh think your right, it`s a full magnex system, 4 branches merge under the car, need to drop the subframe to remove it but I may as well do it right first time round. Will have a look to see which runners got the best room for the sensor, looks like it needs to be 1.5 ft from the port judging by the g60 pic. Think taking the head off will make it easier to get out, ideal time for stacked g60 headgaskets to lower the compression, may as well do it all in 1 go! Any preference on the cylinder? Does it use 1 as a reference point? don`t suppose it matters really as they should all be producing the same gasses.
I would go for cylinder 2 or 3. They are more likely to overheat before 1 or 4 so that may be safer in the event of a problem.
Thinking about you could always run 2 probes and then put a resistor in to bring the voltage down so it averages between the two, haven't put that much thought into that though
Doesn't work quite like that. The narrowband sensors swing very quickly from nearly 0V when lean to 0.8-0.9V when rich. What the ECU is looking for is the swing between the two - i.e. crossing points past 0.45V. At lambda = 1 the sensor output should be constantly fluctuating. If you just connect the two sensors together they could be fluctuating out of sync because of the different event timings of each cylinder giving you all sorts of interference patterns, very few of which the ECU will accept. Some ECU's will have two separate O2 channels for separate banks on V-engines etc, but obviously not digi1. Edited by: martyn_16v
Was thinking of welding another boss on to another runner anyway in case I need a wideband input for mapping at any point in the future, the voltage diff between rich and lean is 0.8v or less so not a lot to play with. I`ll try 1 to start with, always an option to add another later.
I did say I hadn't given it much thought Having thought about it for 5 minutes though I thought the ECU just took a sample voltage at different intervals and compared this to the ECU's map rather than comparing rate of change of voltage?
Cheers for the help lads, looks like I`ll be doing something I vowed never to do after fitting the exhaust, taking it off . At least I can wrap it and it gives me an excuse to take the head off and sort the compression ratio out while I`m there!
I`ve bought my 25 meters of heat wrap now the manifolds comming off! I`m going to stick it where the original one goes, maybe different layouts have different software for working the lambda out? More room up top as well, don`t want to break the wires off the probe.
