Siamesed Code Trial
Turbo Mini Forums

Right,

And I would guess your theory about saturation/pooling is substantiated because the code is deliberately squirting fuel against a closed valve in a port where airflow is essentially static for about another 360 degrees, ie, only the outer cylinders get their fuel squirted into a moving air flow.

I need to think this through a bit more.

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

But it is the inners that run rich, hence, they must be getting some of the fuel intended for the outer cylinders.

I think that this is because there is time for the pooling on the inners to get closer to the valve during the static period.

I'm more inclined to think that the pooling is caused by the manifold design.

My comment above, although crude, illustrates 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."

Unless the outer cylinder's squirt is too long or too late (despite being into a moving airstream) such that some of it doesn't get sucked in and gets left to pool (over the static 360 degrees ) as well as the "dead" squirt. So the pool is bigger than it should be.
Not so much that the pool got closer to the inner valve, just that it was added to by the outer valve not using all it should have done, and the inner valve gets to use the enlarged pool but the outer can only get what's thrown directly at it after the enlarged pool has been swallowed by the inner valve ????

Just thinking aloud...

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

Can you change the duration of the squirts independently so that the second set of injectors don't fire for as long as the first set?

First set / pair feed the outer cylinders.

Second set / pair feed the inner cylinders.

Alternatively could you use two different sizes of injectors so that you have a bigger injector feeding the outer cylinders and a smaller one feeding the inner cylinders.

Although I think the later idea is a bit of a bodge.

Edited by Rob H on 11th Feb, 2008.

But,

If I've understood this right, that comes back to lack of a cam sensor.

You can't feed the inner or outer at one end of the engine without just creating a dead pool at the other end, and the inner cylinder will always get the pool first once that other end of the engine starts sucking....

I'm slowly beginning to understand this (I think).

Instead of 1,3,4,2,
Try (3,4), (2,1), (3,4), (2,1), etc
Then
(3 from pool, 4 from squirt, plus 2&1 getting a nice pool being filled), (2 from pool, 1 from squirt, plus 3&4 getting a nice pool being filled), etc...

That's the way I'm trying to visualise it at the moment.

Rod.

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."

This is why I was asking Jean if he could do somethign simple with a binary input from a hall-effect to easilly make the MS a sequential; at least what is required for the A-series...

My thinking (although I'm years behind the rest of you) is similar...
Mine is 2/3 built but I've stopped until I can fully understand how to provide a simple cam input - I was hoping to use the redundant circuitry from the choice you get between hall/optical and VR when you jumper it, but it appears there isn't that much "redundancy".
The MS-2 processor daughterboard seems to have plenty of spare capacity for I/Os but working out how to do the hardware is more of a challenge...

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

Actually, connecting a Hall sensor should be as simple as using the redundant Hall/optical circuit present on the MS board and connecting this circuit to the correct CPU pin (using one of the available JS pad). As long as you have a once per engine cycle signal coming from either the camshaft or the dizzy.

Then I need to use this signal to synchronize the injection. It shouldn't be a big deal code-wise (but not trivial) but there is one annoying side effect. There are only 2 injector channels and you will need one per intake port to be able to have different pulse widths for the inner and outer cylinders. This means the loss of the staging option.

I have a workaround for that but it will involve some small "surgery" on the MS2 daughterboard, the loss of PWM for low impedance injectors and the need to use my peak & hold injector driver board to drive low impedance injectors. But with this, the system becomes a fully sequential system with a staging capability (with some more tweaking of the code).

Jean

http://www.jbperf.com/

Jean,

We could retain the staged option if we just use one injector pulse that spans both the inner and outer valve opening. This is how I set it up initially and got reasonable results. As long as the engine started on the correct phase it stayed there and worked well, but that's not a viable long term solution.

This is also how the MPi system works, more or less.

However, I think that the code as it exists is close to achieving what we want. I just need a few more weekends of good weather and a bit of luck to confirm or otherwise.

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 Jean,

Edited by Rod S on 12th Feb, 2008.

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

I'm sure that we just need to jumper OptoOut to JS10 and connect the hall sensor to OptoIn.

This is what Jean is saying ^^^^^^.

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 thought so too at first, and the more I look at the schematic, the more I now agree.... and Optoin can be jumpered to one of the spare pins on the DB37

But Note 1 on the schematic is wrong !!! and it had me confused for a while.

I'm going to get some of those PC type jumper blocks from Maplins and put some in the Proto area to try and make all options available without having to use the soldering iron....

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

Note 1 is true in the context of the B&G code using the standard MS2 pinout. Nothing related to MS2/extra is included in the main documentation which means that any MS2/related feature and/or CPU pinout may contradict the text. The schematic clearly show the possiblity of using OptoIn and OptoOut in the manner described above.

Jean

http://www.jbperf.com/

Jean - We know that modern (high-impedance) injectors can be super-fast responding units; hence i'm not so sure losing the ability to drive low-impedance injectors is such a big deal?

A pair of MJY100460's in each inlet tract with a 4-bar fuel rail should support some big hp numbers - so if it's easier to utilise 12-16 ohm units, then this is a reasonable acceptance that anyone who wants to use MS on an a-series?

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."

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

http://www.jbperf.com/

I have one of these, Peak and hold driver PCB, going spare, if anyone is interested, you just need to put an order in at Digikey for the components. The cost to you would be what it cost me, no profits

Ok.. friendly debates

Ok.. friendly debates