998 running a Mpi manifold and MS3X
Turbo Mini Forums

Ummm.................. I can see myself laying in the footwell adding some wires to the loom

What are your pulse widths at idle? Bit on the low side I bet.

Edited by Paul S on 26th Jun, 2014.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

This all looks very exciting. Can't wait to see more.

Edited by Minivation on 26th Jun, 2014.

Minivation
New innovations for a classic design.
www.minivation.net

On 26th Jun, 2014 Paul S said:
Ummm.................. I can see myself laying in the footwell adding some wires to the loom

What are your pulse widths at idle? Bit on the low side I bet.

Edited by dan187 on 4th Jul, 2014.

1275 N/A Sprite, 998 T2 Turbo Mayfair
1275 EFi Turbo

Looking like new



It's only took us 4 years!

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

One of my favourite specials!

If something is worth doing, it's worth doing half of.

That looks really nice. Well done.

Schrödinger's cat - so which one am I ???

afr 3, rod ,assuming that's in the connected down pipe , do you have a leak at the join ?(apology to paul for the deviation)

Medusa + injection = too much torque for the dyno ..https://youtu.be/qg5o0_tJxYM

The pulse width question was really aimed at Dan as he is running the same manifold and injectors as we are on the 998, but 4 pulses per cycle instead of our 2. If he is running 4 bar fuel pressure on the 1275 and getting 1.6mS, then we will be in a similar ball park if we swap to 4 pulses.

Rod, your AFR data is very interesting. I assuming that you are using identical sensors and controllers for all 3 readings?

If you are then I think that you are picking up a reduction in the unburnt HCs between the sample point and downstream of the turbo.

Maybe sampling at the head flange is too close to the combustion chamber

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Apologies for the hijack, the picture was really just to show that very short pulse widths (4 per cycle) are easy to handle (contrary to what quite a few people say on the MS-E forums).

Re. the AFRs, the setup is now identical controllers (14point7 SLC OEMs) on identical Bosch 4.2 LSUs.
Previously I had two TechEdges and one Innovate and the reading difference was similar (one of the reasons for changing to all 14point7 was to eliminate the possibility of different controllers, not the main reason but one reason).
The 3rd comes closer in line with the other two once the RPMs rise or any load put on the engine, but never the same, it is allways slightly higher (in steady state conditions).
I had always assumed air leak as per Robert but nothing obvious, the only join before the third LSU is sound and protected by my inclusion of a flexible link next to it (and it hasn't done enough miles for the "bellows" inside that to have cracked) so I have assumed the wastgate shaft is the source (they are really sloppy on an old T3) but unburnt HCs is something that never occured to me....

I presume you are suggesting some more "combustion" (in the chemical sense) occurs to the unburnt HCs in the hot environment of the turbo exhaust casing etc. (a bit like a catalytic convertor but without the platinum.....).

Interesting....
It would certainly explain why it idles better with inner and outer AFRs at the sample chambers of 12-12.5 (which I think is what you told me you found as well in the early days).

The design of the old cast Metro manifold does mean the sample points are very close to the head.

By using the 14point7 SLC OEMs I can discount LSU temperatures because, with the trial IOx code I'm running (part of the digital display code that Jean is working on and I'm testing http://forum.jbperf.com/viewtopic.php?f=9&t=1360 ) I can see all three run at the optimum 750C (+/- 2C in my case when +/- 25C is allowed) but what I haven't eliminated yet is pressure in the sample chambers.
I think (those of us doing this) have just chosen sample tube/chamber/pipe out sizes that seem logical but no-one has yet measured the pressure.
I have been considering doing so when I pull it out to fit the third sample chamber - if there is enough room....

Schrödinger's cat - so which one am I ???

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

thanks rod .. makes me wonder if im logging a leaner mix than reality .

paul ,do you mean your 3rd samle point after the pipes are collected , but before the turbo to avoid any additional modification of the gases by running through the turbo ?

Medusa + injection = too much torque for the dyno ..https://youtu.be/qg5o0_tJxYM

I don't use a third sensor, just the inner and outer cylinders. Previous 2 manifolds had the two sample connection at the head flanges. On the next one I'm moving them down to the collector.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

MOT later. Stressed :(

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

I hate mot's....

once its passed , supercharger instead of turbo ?

Medusa + injection = too much torque for the dyno ..https://youtu.be/qg5o0_tJxYM

:(

Didn't even get to the MOT place.

Took it around the block a few times to check that everything was OK. Front calliper seized on and got very hot.

Got it apart and got the calliper pistons moving and second attempt around the block and everything was fine, but too late for the MOT :(

Got some useful data from the ECU. Took it over 5k a couple of times :) Needs work on the mapping as expected but close enough for starters.

Try again next week.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Medusa + injection = too much torque for the dyno ..https://youtu.be/qg5o0_tJxYM

That would be why the speedo was not working



That's a new one on me. Took a while to track down the problem.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

We managed to get some fueling data on Friday. I made a couple of blanket changes to the injection timing table in areas where it was a bit out. Took it out again today to see the impact of the changes was as expected, which it was.

The datalogs were opened in MegalogViewer and viewed as scatter plots. Transient data was excluded for now.

First the raw AFR scatter plots - AFR is the inner cylinders, AFR2 the outer cylinders:


Then created a couple of calculated fields:

Av AFR = (AFR+AFR2)/2 - average of the two readings. This will tell me where I need to adjust the VE table.

Delta AFR = AFR - AFR2 - difference between the 2 readings. This tells me where I need to adjust the injection timing.

Negative Delta tells me that the outer cylinders are lean and that the injection timing needs reducing.



Obviously needs some more fuel in the 2200rpm area and some timing adjustment around 1500rpm but not bad for an hours work.

I've already made some more changes and will update once we have tested them out.

Edited by Paul S on 7th Sep, 2014.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Looking good Paul. Glad you've started using the scatter plots. They make things much easier don't they.
Is this using your inj timing figures or the ones I sent over?

1275 N/A Sprite, 998 T2 Turbo Mayfair
1275 EFi Turbo

We used to use an Excel spreadsheet with a conditionally formatted pivot table that gave the same sort of info, but the scatter plots are much easier and better graphically.

I'm still using the base injection timing table from the 998Ti. I had completely forgot that you sent over the other table. It will be very interesting to compare them once sorted.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."