Siamese-port MS2/extra mod

[ptegler]

During a general smoke break b.s. session with a fellow engineer at work(who spends all day calculating condensation drop out due to pressure differentials, oscillations and compression reverberations in gaseous distribution systems) he convinced me building my twin bodied TBI setup was not off base and should work just fine.

So I went ahead and built up this version using two TBs (one injector per body), on a stock 1.75" twin HS6 manifold for the B engine. It has the cross over tube just like stock. http://www.teglerizer.com/fi/MGB_twins/index.html

Now reading this thread... I'm confused again. The irregular air intake pulse deviations that people are talking about (due to the non-linear rep rate of the intake pulses) would create an unequal intake draw from the carb as well. So why would this not effect the consistant fuel rate intake from the SU (or ZS) carb as well? Yes...the air intake velocity is pullng in air and fuel through the carbs...supposedly continuously, but if the pulsations in the intake are that big of an issue you'd have puddling at an unpredictable time related to intake valve openings and the 'non-linear' pulsations in the intake tract pulling fuel through the carbs.

So should we be looking at it as with the twin carbs... there is a slight time period where the intake column of air is not pulling as much fuel and other times it is. So in that vain.... what it you simply ran four pulses simutianeously. this would fill the whole column of air regardless.

Now the next question.... How does the MG Mestro and or Moss Motors setups do it? Why wouldn't two injectors right at the port do the same thing as one ijector poert port (cyl) as long as you run the max number of pulses possible per engine cycle?



ptegler

ptegler[/url]

[jsmcortina]

Didn't the Maestro use the O series engine?

James

[ptegler]

duh... that was suppose to read the later (but original) Mini The late models did have FI on a bunch of them.

I'm just wondering how hard would it be to have the fuel injector pulse train...trigger off the crank sensor the same as the ignition. Then a simple table like the wasted spark setup could be set for all four intake positions/squirts for the siamesed ports.

ptegler

[RickRST]

The later a series injected Mini's used a single throttlebody system. There is a patent knocking about somewhere.

The way i understand it, is that the TBI system is just like the carb. Fuel is mixed with the air further up the intake. At the port, there is a continouus amount of fuel/air mix which is sucked through whichever inlet valve is open.

With port injection, the injection events aren't normally timed for a specific crank position. Therefore the injection event would tend to favour being drawn into one of the cylinders rather than being evenly distributed.

What this code does is time the injection events so that the injector fires when inlet valve of cylinder 1 is open, then fires again when inlet of cylinder 2 is open, so each gets the same amount of fuel.

Jean - the injector staging idea is intersting. This would definetly help to keep the injector size down. I think this will become important for high power Mini boys, as you only have 2 injectors. With the second set of injectors, it is better to have the outer set doing the majority of fuelling in high load areas - this is where the extra power really comes into play.

I'm trying to thing whether 2 different sized injectors on the same channel would work? I think it would, but at the crossover point you would have to reduce the pulse width by a percentage equal to the difference in size between the small and large injector?

Rick.

[borris]

I've been a long long time reader. Here's my first post!

I've got a 1275 engine that I have been trying to turbo and FI for years now. I have some small CNC equiptment that I have produced all the part for it. I've designed and built two systems.

1. TBI unit, single injector with carb cap for pressurization

2. Port injection unit with 2 injectors on a standard plenum.

Question:

I'm at a turning point. I need to go one way or the other. Is there a solution for the port injection? Can I sample my distributor pulse for the trigger? Sorry if this is redundant, but it isn't really clear to me where this is at.

Thank you all for your hard work.

[racingmini_mtl]

I have a version of the siamese-port code for port injection which is based on an early version of MS2/extra which works on the bench but has yet to be tested on an engine. I'm still working on a version of the code which will be more general and will have sequential/semi-sequential code for other engines as well as siamese-port engines (A-series Mini engine). This will be based on the latest MS2/extra code.

With either code you will not be able to use your distributor for trigger because you will need to have a timing reference for synchronizing the MS with the engine position. This can be any missing tooth wheel or some type of dual wheel setup. With the distributor, MS cannot know which cylinder the trigger is coming from.

If you want to use the TBI, you can use any code and the distributor trigger because injection timing and engine position synchronization are not needed.

Jean

[borris]

Couldn't a person use a distributor body, lock the advance, and install a optical trigger inside it to represent the timing? Granted it would be a 4:1 Ratio, but you could put 4 flags on the distributor making it the same pulse as a crank trigger.

I'm very interested in your development. Of course I'm just learning VB, so I wouldn't be much help..... or I would be wanting to help out more.

Thanks,

Richard

[racingmini_mtl]

If you put a missing tooth wheel in the distributor then you provide enough information to cpu for it to time the injection (and ignition). If you just have 4 triggers in the distributor that correspond to the 4 cylinders then there is no way for the cpu to distinguish between each trigger so it can't time the injection. When a pulse comes in the cpu it wouldn't be able to say if it's for cylinder 1, 2, 3, or 4.

Also, if you go to the trouble of fitting a missing tooth wheel to a distributor you might as well fit one to the crank because you'll have better accuracy due to not having the slack of the timing chain and any distributor slop.

Jean

[Mini Sprocket]

So I went ahead and built up this version using two TBs (one injector per body), on a stock 1.75" twin HS6 manifold for the B engine. It has the cross over tube just like stock. http://www.teglerizer.com/fi/MGB_twins/index.html

DUDE!!

I am drooling over your injection manifold set up, looks well smart!!

This is the kind of thing im aiming for in the next few months, as cheapo cheapo as possible (I have other plans afoot for something special) while Siamese code is progressed

Out of interest, what TB's are those, they look the ticket

Im seriously liking your work

[6040solder]

Jean,

Whats the current status of the 4 cyl sequential code, and would it be possible to use that code with a modified ms2 card with the nand gate removed or disabled such that the four pwm outputs could be used for a channel each and an external p&h board or module used IF low imp injectors are required?

I'd be very interested in that if so :-)

Have you ported your Siamese code over to the latest ms2e base yet? Which ms2e revision was it based on?

Is it possible to get a copy of it to have a read through?

Sorry for all the questions, I've been doing a lot of 9s12 reading yesterday and today :-)

[racingmini_mtl]

Jean,

Whats the current status of the 4 cyl sequential code, and would it be possible to use that code with a modified ms2 card with the nand gate removed or disabled such that the four pwm outputs could be used for a channel each and an external p&h board or module used IF low imp injectors are required?

I'd be very interested in that if so

There hasn't been much progress in quite a while because I've been keeping myself busy with a bunch of other projects. And that was pretty much the plan (and still is).

And it's actually not PWM outputs but rather output compare and the removal of the gate is not needed but rather just a couple of jumpers from the gate's input to 2 unused pins on the MS2 card (the crystal pins are nice for that). That's for the 2 PWM pins (which are now used in OC mode) and the 2 flyback pins can be reused as is.

Have you ported your Siamese code over to the latest ms2e base yet? Which ms2e revision was it based on?

Is it possible to get a copy of it to have a read through?

Sorry for all the questions, I've been doing a lot of 9s12 reading yesterday and today

I haven't yet finished porting the code to the latest ms2e code because it's going to be integrated with the sequential code as just an option.

The "working" siamese code is based on an early alpha version of the ms2e code. Since it hasn't been tested on an engine yet (but it's coming soon), I'm not completely sure it's functional yet even though it works on the bench. That's also why I'm a bit hesitant to make the code public because there has been interest in the code from the "wrong" crowd.

I don't know if you've heard of MTech but they seem to have been involved in "acquiring" code and design left and right (in addition to having shoddy workmanship in their assembled boards, see Phil's post). They seem to have "acquired" the design for the mini-MS, the MS-AVR (VEMS), and I suspect their WBO2 controller, and are still selling MS board even though they can't get it from B&G. They were "working" on port injecting the Mini siamese-port engine so I assume they're looking to "acquire" some code.

To make a long story short, until I have some results from an actual engine run and the code is integrated with the MS2/Extra, I will keep the code availability very limited. I'm not saying that you would distribute it or have anything to do with Mtech but I'd rather keep it private for now. Of course, once the code is public, people will do as they please with it even with the copyright notice and all that but that's life.

Jean

[6040solder]

No, thats totally understandable. I've had some communications with him (initiated by him) and I certainly wouldn't be handing out code to him to say the least. I only wanted to fiddle for now anyway, so its no big deal.

Thanks for the information :-)

Do you know which section Phil's post is in? or have a link? I'd like to read it.

[racingmini_mtl]

You can have a look at this: http://www.msextra.com/viewtopic.php?t=27074 and this http://www.msextra.com/viewtopic.php?t=23526.

I have also received an email saying they are no longer working on the A series project and that they do not infringe on any IP (intellectual property) under the English law. Whatever...

Cheers,
Jean

[jsmcortina]

You can have a look at this: http://www.msextra.com/viewtopic.php?t=27074 and this http://www.msextra.com/viewtopic.php?t=23526.

I have also received an email saying they are no longer working on the A series project

i.e. you haven't published any code for them to copy.

and that they do not infringe on any IP (intellectual property) under the English law.

Their interpretation is that any copyright or patents outside of the UK are of no interest to them and therefore they will copy anything they can.

James