Lambda Sensor

[comdiver]

Just wondering if anybody has any thoughts on the best position for the sensor on the exhaust manifold. At the point where all three pipes meet or where the two outer pipes meet.

Cheers, Brian.

[FWDRacer]

You don't just want the lambda to monitor the outer two cylinders do you?

You've answered your own question, the lambda must be plumbed in after where the centre branch joins the outer two - Check out propriatory MPI/SPI LCB or 3-1 manifolds from Maniflow or Janspeed if you need further guidance on where the Lambda boss needs to go.

[comdiver]

Well yes, however, i have seen a variety of after market manifolds with the sensor in both positions hence the question.

[FWDRacer]

Mixture distribution in the siamese ports can be skewed by what type of induction system is going to be employed. An example is that Weber equipped mini's always run richer on the outer two cylinders.

If you are going to Lambda control I assume you are putting some form of bespoke injection system on the car? Is the injection system going on a 5-port head or something more exotic? If not, and you are putting an aftermarket manifold on either a standard SPI or MPI mini, I'd suggest the boss needs to be downstream of the centre branch to give optimum signal and hence mixture control back into the EMS. When you've looked at manifolds are the Lambda positions specific to SPI (very sophis Elecronic Carb) or MPI (port injection).

Mind me askin why you are welding your own boss on ?

Edited by FWDRacer on Wednesday 12th March 12:12

[comdiver]

Its a 5 port head (Swiftune 1380 SW5 on a 45 Weber at the moment). I,m going injection with Jenvey TB's and a Canems ECU, hence the need to weld a lambda boss into the manifold.

[FWDRacer]

Bloody A-series 5-port Head - makes everything compli-macated

Mixture distribution with twin Jenvey's (1 TB per pair of cylinders) Jenvey's should (in theory) be better than the single 45 DCOE - ech TB being on straight tubular manifold (?) therefore a straighter shot to the ports and no rich running outer cylinders.

Because you are in theory helping the uniformity of the exhaust charge I'd be tempted to plumb the Lambda either in or post Y-Piece. This should give the most accurate feedback loop to the EMS.

Good Luck, it sounds like a top project.

[comdiver]

Thanks. As it stands at the moment it put out 130hp and 100ft/lb's on the engine dyno, so not to worried about any further gains in that area, but i would like the control over fueling that EMS can give and oh and it will tick over quite happily at 8-900 rpm.
Lots of planning and prep ahead but at least things are moving along.

[Mr2Mike]

FWDRacer said:

Mixture distribution with twin Jenvey's (1 TB per pair of cylinders) Jenvey's should (in theory) be better than the single 45 DCOE - ech TB being on straight tubular manifold (?) therefore a straighter shot to the ports and no rich running outer cylinders.

Results by others who have gone this route suggest otherwise! The Rover MPI only works due to large injectors (far larger than engine size would normaly dictate) and some quite tricky injector timing to minimise charge robbing. Using a typical batch fired injection system is unlikely to give satisfactory results.