MS3X MPi Trials
Turbo Mini Forums

Dan,

http://www.turbominis.co.uk/forums/index.p...id=416587&fr=25

page 2 if the link doesn't open on page 2.

No diodes, just parallel them. Just like Rover did.

So far as I can see it's the only way to run MS3 for siamese.
(but never tried it, MS2 siamese is fine for me).

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

Thanks. I'll try it as soon as I get chance. So how does ms2 Siamese differ then?

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

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

Using all four(or eight) injector outputs has some theoretical advantages, but before we get carried away, you need to consider what may happen in practise:

Firstly, if you are already firing the injector continuously throughout the outer cylinder intake stroke, then you will not get the outers any richer without increasing the injector capacity.

You then need to consider the fueling requirements for each cylinder. In theory, you may think that each cylinder needs the same amount of fuel. This is far from the case due to differences in Volumetric Efficiency (VE) and transient fuel.

The main difference in VE between the inner and outer cylinders is due to charge robbing. In your case this will be negligible with the standard MPi cam with 230 degrees inlet duration and lowish revs. As cam duration and revs are increased, then VE imbalance will also increase. However, with the right cam and inlet runner geometry the overall loss in engine VE can be eradicated: what the inner cylinders lose can be offset by a gain by the outer cylinders.

Transient fuel is the fuel that does not get into the outer cylinder at the first attempt. This happens because it sits on the back of the valve or on the port wall. It will sit in the port until the next inner inlet valve event. This means that the inner cylinder needs less fuel injected at each event and the outer cylinder needs more to replenish the transient fuel. This is a complex subject that needs more explanation but I hope you get the jist.

The advantage of using the MS3 is that you cam assign a specific O2 sensor to each cylinder and apply closed loop control to an AFR target table. However the limitation of the MS3 is that the 2 pulses relevant to each port are fixed at 180 degrees apart and work from a single VE table.

The MS3 does also allow for individual cylinder trims, so you could reduce the fueling to the inner cylinders by this method. I would start by running the inner cylinders at 90% and increase the outers to 110% to compensate.

The potential problem with the closed loop control is that the two loops could fight each other. Because the AFR of each cylinder is affected by the transient fuel, any change on fuel injected in one cylinder on a port will affect the other. You could easily have a situation where the inners will try to lean due to too much transient fuel which will make the outer loop go richer giving more transient fuel.

EDIT: Before anyone points out, I've used the term "transient fuel" incorrectly. Just need to invent a new name or does one exist already?

Edited by Paul S on 15th Jul, 2013.

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."

Why not just use wall wetting fueling or something similar? We have always talked about wall wetting, haven't we?

http://www.jbperf.com/

Thanks Rod, yeah currently running one long pulse. Presumably in terms of injector timing i'd set it for the outers and have to accept the same for the inners. In that setup would i want to run end of pulse timing (to ensure i'm done by the time the outer valve shuts)?

Paul, the 10ms pulse width limit you've spoken of, how does that come about? Is that limited by MS or the injector. (or are you estimating valve open time based on timing figures and engine speed, and maybe some overlap?).

I need to do a bit more testing to check if I am injecting for the entirety of the outer valve open period but if I am then i guess i need to change the setup. I assume I'm not as Rover were able to get it going with these injectors. I assume the power upgrades to the MPi models used the same injectors to output 70+ bhp.

Thanks for the pointers, yeah I thought i understood but... are the cylinder trims inversely proportional then? not sure i understand why inners leaning off to account for transient fuel will cause more fuel to be injected into the outers.

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

'condensed charge' Paul?

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."

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've been so busy with other things i haven't had much chance to post about the tuning.

Latest scatter plot below. Much better now. Just updated the timing based on this log so hopefully will sort out the low end mismatch. still having trouble with the top end, so either i don't have enough capacity or have completely the wrong timing. It's difficult to tune full throttle on the road by yourself. I have to do a run look at the scatter plots and then go out again (not ideal). One thing I have noticed is that to get good throttle response you need a high value of accel enrich (i guess due to the shared injection) this however, does under certain conditions cause too much fuel to be injected.
I still plan to have a go at pulse per cylinder but don't have the time at the moment and this is working well (other than top end)

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

So I haven't got round to making the manifold yet but have changed the way I inject so that it is fully sequential (with each injector doubled up on the outputs so it fires twice per engine cycle).

I just wanted to check how the 36-1-1-1-1 wheel should work and if mine is ok? I'm setting the injection timing on my MS3X to get the injectors to fire into the open valves on the outside of my siamese 5-port mini head.

Here is a trace of my composite log


Does it matter that the cam may be 180 degrees out? It's just there are different spacing between the missing teeth. Typically my injection timing seems to be working in the range -50 at idle and 70 at full throttle (N/A). However, even though the pulse width is typically only 5.4ms at 4000 rpm i can't get the outers rich enough, so just checking all avenues. The window for injection should be approximately 8ms (using same logic as... http://www.starchak.ca/efi/siamese.htm) so it seems a timing issue.

I have an adjustable fuel pressure regulator on the way so will set that a touch higher (reading some of TurboDave's old threads he suggests up to 70psi would be suitable). Also, when I've made my manifold i'll use this new regulator rather than making my fuel rail accept the stock one.

I'll be practicing my aluminium welding skills at the weekend in prep of making the manifold.

FYI: Also posted this same info on MSExtra forums
http://www.msextra.com/forums/viewtopic.ph...=393628#p393628

Regards
Dan

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

If the cam sensor is 180 degrees out then it will shift everything. I'm assuming that you are talking cam degrees, not crank degrees.

If you have not swapped over the injector wires, then they are not firing in the correct place.

EDIT: Having said that, your timing values seem to have accounted for that.

I still think that you need larger injectors.

Have you tried using the individual cylinder trims to increase the fueling to the outers?

EDIT2: Just read it again.

Your timing range is huge. I only see a variation of around 30-40 degrees from idle to full throttle.

I now have the MS3X and MPi manifold on a 998 running. Hopefully get it MOTed this summer. See if the values are comparable.

Edited by Paul S on 6th Mar, 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."

Just checked the tune for the 998 running the MPi manifold and injectors.

To give a good fuel balance at idle, we are running 470 degrees injection timing. That's with the cam sensor set as the MS instructions.

This is the same as I ran it on the 998Ti.

If you are getting good fuel balance at idle at -50 degrees then something is wrong.

EDIT: I'm using mid-pulse timing and aiming for balance across both cylinders, so you should still be in the region of +/- 90 degrees from my setting. ( unless you are using start or end pulse timing)

Edited by Paul S on 6th Mar, 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."

Hi Paul, thanks for the reply.

Yes cam degrees, although i'm fairly confident I installed it in the engine with the timing dots aligned and then tweaked my vernier timing gear to give 106degrees, so it should be as rover intended (which may not be what MS wants). My cam lobe seems to be located at BDC (not 120 BTDC as suggested here: http://msextra.com/doc/ms3/trigger-wheel.html)

Should I just swap the injector plugs over or does it not really matter if the software has taken account with the lower numbers?

Yeah the plan is to double up using a pair of standard injectors per port when i make the manifold.

Yeah the fuel trim is maxed at 25.4% adjustment positive on the outers and 25.4% negative on the inners for full throttle past 3000 rpm

I'm using mid-pulse fully sequential. Are you using the long pulse spread either side of valve overlap?


At idle PW is around 1.2ms. This is around 15 crank degrees. With a window of about 180 degrees to inject, the timing can vary from by up to 165 degrees and still get inside the window. Travel time at idle should be around 7.5 ms from injector to valve, corresponding to 112.5 crank degrees. Which suggests I am getting fuel in the cylinder from 155 to 170 degrees ATDC.
Is that correct thinking?
I'm using the chart here as reference: http://www.starchak.ca/efi/images/figure2.jpg

So if i'm right 470 degrees injection timing should correspond to 110 degrees BTDC? So if you're doing 1 long pulse that corresponds to fuel at the valve around the overlap point (112ms travel time). So you're injecting into the opening valve whereas my fuel is going into the closing valve?

I'll post up some data logs of today when I get home to show AFRs PWs etc.

Edited by dan187 on 6th Mar, 2014.

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

Actually that assumes 100% VE, which is more likely to be 20% at idle. So my logic is probably wrong.

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

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

So I took some logs driving to and from work today. I think it is simply a case of not enough fuel and matters being complicated by unknown scavenging amounts. It doesn't seem to be a timing issue. Because although i'm injecting 6.6ms which is on the limit of the time window, I've only got 4.4ms going into the inners. Yet the AFRs are 14.6 and 10.3 respectively. Yet it seems the injection timing I have at 4000 rpm, 69, gives the lowest outer AFR. So some of the outer fuel is spilling into the inners.

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

Did you have the injectors cleaned? It could just be an issue with flow.

I agree with your analysis of the injection timing. As you are injecting into each cylinder then your timing will b different from mine and at idle it is not so critical as there is plenty of time.

If you consider that you would expect the mid pulse of your injection event needs to enter the cylinder 90 degrees after TDC then the maths stack up similar to mine.

Allowing an extra 360 degrees for the incorrectly phased cam sensor:

70 + 90 + 360 = 520

I was running around 520 degrees at full chat under boost.

EDIT: Thought that was wrong, changed it, but now believe it is correct.

Edited by Paul S on 6th Mar, 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."

Edited by Rod S on 6th Mar, 2014.

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

Is your FPR working correctly? Have you checked the fuel pressure?

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."

Edited by Rod S on 6th Mar, 2014.

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

i think the cam position is working ok then based off Paul's calcs. Clearly either rover or myself put the cam in 360 (crank) degrees out.

I fitted the after market fuel pressure reg this evening and been out for some runs. I didn't measure fuel pressure before but now it is about 3.3 bar (seems that is max that the MPi pump can provide).
Now, for the same rpm (4150) and PW i get an AFR of 13 rather than 14.3 so a little extra fuel. I guess this is pretty much all i'll get until i put in the second injector. I'll have a little play with the timing tomorrow to see if i can get a tad better. Feels so much better to drive.

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

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

Glad to see you're making progress.

10% extra fuel is all that was missing. The standard FPR must have been running at around 2.5 Barg.

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."