Well after last years attempts and running R1 Throttle Bodies on a pinto using the TPS and getting dire economy regardless of how much welly you were giving it we've decided to return to trying to use the MAP sencor again.

We currently have the MAP pipes going through Y connectors and finally joining together like so:

1+2 3+4
Y Y
Y

We then had a 0.7mm restrictor in the last connection, this gave a very bouncy MAP signal of approx between 80KPA and 95KPA at idle (approx 800rpm). Just as a test we when put in a fuel filter as a resevoir (it was just at hand) and this then steadied out at about 82kpa.

No amount of altering fuel or timing seemed to make the KPA go any lower and all Idle air bypasses we knowof are closed, the only other thing we've done over the standard R1 setup is prop the CV slides open and fitted wedges to the throttle butterflies to try and get better part throttle cruise.

Anybody have any suggestions or information as to what we should try or where we've gone wrong?

My zetec is running the same sort of set up and I get a 45kpa idle when warmed up.There isnt a lot of resolution between closed and open throttle but I have had good results so far and get between 20 and 30mpg depending on driving.
I think MAP has got to be better if you can get a reasonable vaccum signal as it represents the engine load directly and is easier to visualise when tuning.
I think there wassome talk on the megasquirtforums of software smoothing of the map signal but it may have involved extra map sensors and taking an average from them.
I would have thought that the CV slides would be more lokely to help with the vaccum signal rather than hinder it, what do you mean by wedges on the butterflys?
Paul.

We've tried to replicate the Vauxhall XE main throttle butterfly idea. They've a small part trottle seperate butterfly about the size of a 5p and then when it starts opening the main big throttle body there's a wedge on it like a section of chocolate orange, presumably to smothe the transition between the two.
And it'd hopefully allow us the part throttle cruise we'd like.

Its shown half way down this article:
VX throttle Body

We propped open the CV slides to try and take yet another variable out of the equation.. we probably thought at the time it was them causing the fluctuating MAP reading.

This may be of some use. I did a bit of an investigation when I was using a MJLJ ignition (still use the same setup with MS)

http://www.autosportlabs.org/viewtopic.php?t=1171

Datalog the MAP values then set the parameters to capture the busy areas.

My fireblade engine looks like an on/off switch if I count 10 20 30 40 50 - 100

So I joggled the RPM values with the KPA so I could get something that worked effectively.

Megasquirt is clever in that it joins the dots, and carries values forward so although the engine will rev to 13,000 rpm I stopped at 10,000 so saved a bit, the zero's down the bottom mean that on overun foot off throttle it uses no fuel.

[Edited on 14/4/09 by mark chandler]

It surprises me that TBs seem to give much worse fuel consumption than carbs. I'd have thought it would be the other way round.



Stu

Laguna

In the pic above there are 2 traces. The top one is from a 2.0 Pinto with longer than standard duration cams and the bottom one is an internally standard 2.0 Duratec.

Both engines ran on the same GSXR throttle bodies

The MAP is the Cyan trace. Yellow is RPM

You can see that the pinto has a much higher idle MAP an goes close to 100 kPa much quicker than the Duratec.

The Duratec has a much smoother and lower idle and is slightly more progressive

Since the throttle bodies are the same on both engines I could only conclude that the type of engine and valve overlap has a much greater effect on the MAP than damping it or altering the butterflies.

I experimented with pnumatic and electrical damping on the MAP signal but all that ever did was delay the load signal getting to the ECU which caused accel issues.

In the end I just had to put up with a less than ideal MAP signal on the Pinto and did like Mark says, respaced the load sites so more bins were between 80-100 kPa

Alpha-N would have been the next thing to try with the Pinto.

Paul.

[Edited on 15/4/09 by Doofus]

[Edited on 15/4/09 by Doofus]

Hmm some very interesting findings, it hasn't got too crazy of a cam in it,
only a Kent FR32 and i've mapped it sucessfully using MAP on a Megajolt with the original carb on the same engine.

Narrowing up the load sites doesn't give me much scope when its such a high idle value there's little or no chance of it telling the difference between cruiseing and full throttle.

Will check tonight how exactly the lines from each plenum run and join, Paulf looks to have some pretty good MAP values to tune with so we can surely improve on ours.

Average idle MAP value will depend on a number of things. Mainly it is ignition timing, then inlet valve closing.
If you run retarded then you need more airflow to produce the same indicated power from the cylinder.
Also if the inlet valve closes late it will back flow more charge out of the inlet (hence you need more airflow to trap the correct quantity of air at idle).

Well we've had some sucess atlast.. mainly re-evaluating everything and realised that in the process or re-spacing them we had the Vacuume pipes connected to the ports that must measure the amount idle air bypass, thus half of it being open to the almosphere and giving a crap MAP value.

Were seeing much better MAP range now.. and in the few goes i've had it only fires 6" flames out of the exhaust when changing gear rather than foot long flames at random