Siamesed Code Trial
Turbo Mini Forums

[quote=Paul S,4th Feb, 2008My theory is that there is always some fuel left over on the walls from the squirt that goes into the outer cylinder, that starts to evaporate and goes into the inner cylinder with the second squirt. I don't think you get as much left over from the second quirt because it has sat in the port and warmed up before the valve opens.[/quote]

ah i see , i was thinking that 3 fires before 4 , so the inertia of the air fuel mix hurtles into no3 ,
then 4 opens ,
and the air changes direction and shoots down no 4 ,but the fuel with more weight ,and so on , takes a little longer to change direction ,so no 4 get a puff of leaner mix before the full fuel mix gets to it ,so it runs leaner .

thats why i felt a finer atonization of the fuel ,so making the mix more even and less heavy fuel droplets and light air molecules
would make it easier for the fuel to change direction with the air .

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

Well you may have something there, but maybe some of the first squirt sits on the back of the inner cylinder inlet valve, but then the same would happen on the outer valve, or probably it does not because it is squirted into a static air stream. ???????

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."

There is virtually no air inertia in idle. If you do a quick calculation you'll see air inertia does not really exist, only oscillations are relevant.
I'm also running 4 hole modern EV5 injectors and have to lean the inner by 20% to get the same plug color on inner and outer.
Yes Paul, wall wetting. Modern engines have 20% of fuel traveling on the walls, my experiences tend to show a much higher ratio in the 998 ports. A portion of the spray ALWAYS go to the wall because the fuel film must be restored.

Edited by alpa on 4th Feb, 2008.

std 998 A+, g295, MD266, RHF4, 109hp @0.8bar/5400rpm

any movement of any media had inertia i thought ?
for oscillations to occur doesnt a change in inertial direction effectively 'define'the oscillation.

my conception of the flow of gas in the port is a bit different ,
due to its elastisity , i see it as almost a steady stream varying in density through the flow from the filter to the valve , the events at the valve happen so fast ,and the air speed through the ports is comparitively slow ,that i dont see the flow as a in and out sort of process ,where theres a slack period like low tide , but a more dynamic situation with air in different parts of the port travelling foreward at the same time but also low velocity areas changing direction more in shear . .

the idea of the inertia effecting things would be supported by things getting worse the higher the rpm , but having said that , if the inertia builds up to a higher and higher level , would this charge the broad section of the port with more and more ram effect ,to the point where the gasses and fuel shoot down no4 with such ferocity as to overcome the increasing inertial bias ?


rambling on a bit now !

Edited by robert on 4th Feb, 2008.

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

paul , can you do the same graph 150 btdc , but with a bit more smoothing on the lines , eg .25 sec sampling etc ?

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

Just do a calculation of the air inertia given its density and speed and compare it with 30 kPa pressure. You'll see it's nothing, does not count.

std 998 A+, g295, MD266, RHF4, 109hp @0.8bar/5400rpm

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

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."

std 998 A+, g295, MD266, RHF4, 109hp @0.8bar/5400rpm

http://www.sciencedirect.com/science?_ob=A...5c0c5601eb29af6

The say: inertial at high RPM. 18000 RPM
And acoustic at low RPM.



The acoustics of racing engine intake systems

Abstract

Naturally aspirated racing engines have tuned intake systems and can now achieve volumetric efficiencies in excess of 125% and peak engine speeds in excess of 18,000 rev/min.

...

An investigation into the intake process on a single cylinder racing engine has shown that inertial ram effects make a strong contribution to the intake process at high engine speeds whereas acoustic resonance effects are more important to the rather weak wave action that occurs at low engine speeds. An acoustic model of the resonant wave action has proved useful in distinguishing between these two effects. The attributes of the acoustic model have been compared to those of more traditional time-marching gas-dynamics calculation methods.

Edited by alpa on 4th Feb, 2008.

std 998 A+, g295, MD266, RHF4, 109hp @0.8bar/5400rpm

Paul, I saw that:
http://www.canems.co.uk/siameseports.php
Their AFR seem to behave differently from yours. They are close at low loads and very far at high.
Do you know why ? If their measures are right my wall wetting hypothesis is false.

std 998 A+, g295, MD266, RHF4, 109hp @0.8bar/5400rpm

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."

Actually, just read the Canems website again and studied the log at the bottom of the page.

They are actually no nearer than me in getting this problem sorted. At full tilt the AFRs are 10 inner/13 outer. it's pretty good at idle though!

Actually, they get my goat. The cheeky bugger has applied for a patent.

Edited by Paul S on 4th Feb, 2008.

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."

Hi,
Can he apply for a patent on something that Rover already has a patent on?
Cheers,
Gavin :)

VEMs Authorised Installer / Re-seller. K head kits now available!

WB/EGT gauges. Click here for customers write-up

Visit www.doyouneedabrain.co.uk

My Mini build diary

Edited by alpa on 4th Feb, 2008.

std 998 A+, g295, MD266, RHF4, 109hp @0.8bar/5400rpm

He can apply for a a patent if he wishes - but in this world of megasquirt, it won't make a jot of difference...

This is great stuff, but i'm a little lost in it all.
Can we back up and see exactly which code we're using as i know Jean was looking at several.
Is this the one that (in theory) triggers something like 90 degrees ATDC and the pulse duration expands symetrically about this point, or the one that triggers the injectors at (theoretically) 40 deggress ATDC and lengthens the pulse forwards in the 'traditional' stylee?

Last thing i want to do is offend, but are we sure the code is doing what it should be?

Damn, I wish i had my MS built and time to assist :(

I don't know the rules for getting a patent in the UK nor have I looked at what the claim in the patent is but I don't see how they could patent what Marcel has had on his site for years or what I have had in my code and discussed here in details for a long time.

They may get a patent but defending it would be a futile exercise.

Jean

http://www.jbperf.com/

Dave,

You can set the injection timing to Start, Mid or End Pulse. You can enter a fixed value or use a RPM/MAP map.

The above tests are all at fixed timing.

All my references to angles are "Mid Pulse" as I found that this gave the best starting point. Once we know a bit more about what is needed then we can try start or end pulse timing.

I checked the actual timing on Saturday with a strobe and at 0 dgrees, the missing tooth was visually at TDC. The injection pulse at idle is about 1.2mS, which represents approx. 7 degrees. So without knowing exactly what point that the strobe is picking up, I am happy that it is working correctly. I also set it at +30 and -30 and the missing tooth strobed correctly.

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 can control anything in my code (not MS), but I only have one AFR sensor. If someone could tell me how to know when I'm REALLY stoichio (14.7:1) I would try it for different conditions.
I'm able to run on 2 cylinders, however the O2 measure becomes wrong.

I saw siamese ports can be found on some inline-6 US engines. Does someone know an example of a running MPi ?

Edited by alpa on 4th Feb, 2008.

std 998 A+, g295, MD266, RHF4, 109hp @0.8bar/5400rpm

I've now read through this four times, and each time follow it more (LOL)!

Just thinking out loud, what is the status of a cam phase sensor on the MS?

If nothing is in place currently can we use a simple hall switch (a nice simple binary output for you non techies) on a modified dizzy to feed a trigger to the code as to which phase we are 'in'?

I'm wondering if this is easilly implemented (from Jean's behalf of using a 1/0 trigger to make things happen in the code)- that it'll make things easier in the long term?


Here is a scribble of what i mean:
http://www.davecoxon.co.uk/cam_phase.jpg

I've used these sensors at work when instrumenting vehicles before. They are automotive grade, and aside from a 'pull-up' resitor, are about as easy as you can get for a robust Hall-effect.

http://www.davecoxon.co.uk/1gt101dc.pdf

Of course, this might all be easy inside of the current MS developemnts?

I made my phase sensor with an ignition distributor. Left one cam of four. Works fine. If it's worn the ECU detects that and goes to unphased mode.

std 998 A+, g295, MD266, RHF4, 109hp @0.8bar/5400rpm

Did two runs this afternoon. One at 160 degrees and the other at 180 degrees.

Getting worse now, so seem to have passed the optimum for WOT.

I'm a bit disillusioned, this was supposed to be getting better.

So its back to 110-150 to see if I can get the inner AFRs above 10:1 and get something decent at idle.

I've staged out two of the injectors to see if that helps.

Had to stop due to the oil slick.

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."

Mine open 10 degrees after TDC up to 4000 RPM. Seems to work well, especially if inner are leaned by 15%. I'm still searching for the best way to compute the balance coeff.

std 998 A+, g295, MD266, RHF4, 109hp @0.8bar/5400rpm

After feeling a bit despondent yesterday, today has given some better results.

I staged out two of the injectors so that they would not operate within the useable rev range.

I also re-flashed the firmware to the LM-1 reading the AFRs on the inner cylinders and re-calibrated the sensor. Not sure that it has had any impact on the results though.

We then did a series of logs at fixed mid-pulse timing from 50 to 110 degrees.

From these results, I generated a table.

That attached datalog shows that we are getting much closer. Still some work to do, but heading in the right direction.

I think that the two injectors per port dumps far two much fuel into the port at the same time and creates pooling, particularly at the cool temperatures at present.

It may have something to do with my manifold design and the narrow hole through which the spray has to pass. So I'm going to get the manifold off and open that area up.

http://www.jaservices.co.uk/photos/Dual%20...20Table%201.pdf

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 would have thought that two 460cc injectors was more than enough to provide fueling for well over 100bhp, I think you are heading in the right direction taking two of the four 460cc injectors you have, out of the equasion