Megasquirt Injection question
Turbo Mini Forums

I'm about to build my wiring loom and install my megasquirt, and have been familiarising myself with all the documentation on setup etc... Most of it I understand just fine, I'm just trying to decide how to run the injectors.

Basically, I will be using 2 injectors after the throttle body, but in a single point wet manifold arrangement, with both injectors firing simultaneously for simplicity. What I'm trying to work out is how often to fire them - obviously I have the choice of firing them once, twice or 4 times for 720 degrees of crank rotation on the megasquirt - but what would you guys suggest?

The engine is 1293cc, 8.2:1 compression, will be running upto 20 PSi boost, running 2 x 650cc injectors.

Any idea's would be greatly appreciated.

All I can say is.......

I don't know what to say - have you not understood the problem with injecting the A series?

Don't waste engines when the code for port injection is available.

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."

Whilst I agree with Paul in principle, ie, port injection is the way forward.... if you are trying to replicate wet manifold, I would fire them as often as the basic code allows.

That, I would think, would give the best possible mixing of fuel/air before it hits the siamese ports and starts to split.

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

Although I understand the principals of a wet manifold it is only now that I am thinking about how it actually works.

Injectors controll fuel flow via pulse widths, but cars do it by air flow. So to make a wet manifold set up do you simply pulse the injectors at a set rate and then increase the on to off time as the load increases? If this is the case then I can see real problems at low loads with port robbing.

http://www.twin-turbo.co.uk
http://www.hillclimbandsprint.co.uk/default.asp

A man without a project is like a like a woman without a shopping list.

I use 4 injections per cycle on a wet manifold with 2 EV6 injectors (can't remember the flow rate but think its around 450cc)
Have a look in the cars section for pics of my setup.
I consider that it works very well with little ot no transient problems.

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

Paul, I understand what you are saying, but I don't have the time nor inclination to do what you're doing.

I simply wanted to replace the carb with something more reliable and adjustable - to my understanding, running injection in a wet manifold arrangement is basically like having an electronic carb. I can't see the port robbing being any different to how it is running with the carb.

I just wondered whether it was beneficial to pulse the injectors once for every spark or if I could get away with less.

The best fuel distribution in a wet manifold setup will be with 4 injection per 720 degrees but you'll still have uneven fuel distribution.

However, since you have 2 injectors, you would be better off firing them in alternate mode. This way you have fewer pulses per second for each injector and at higher load, the pulses will be able to overlap if needed to response to the fuel needs.

But to really benefit from fuel injection, you should consider going with a port injection setup similar to Paul's or Rod's. This way you'll be able to tune out any fueling discrepancy between the cylinders.

Jean

http://www.jbperf.com/

I've tried alternate and it is worse. Just use 4 injections simultaneous per cycle with MS1 V3 and its perfect.

Job done

I really don’t see a problem with running Wet manifold, Paul what you are doing is fantastic there is no denying that I am very excited by your results so far and the more you do on it the better it is getting. But I also see what Coupe is saying, I have been running SU's and Webers for years with very good results where there is no kind of phasing.

In the world of Power tuning the best power is not found at stoichio it is found around 12.5:1 to 13.5:1 engine critical, so for all out power in my mind the level of accuracy that can be achieved by the New mega Squirt Code will only come to real benefit on road cars.

Just to back up what I am saying in relation to stoichio, I was involved in some Bench marking of the Irish Punto championship they use the Tepo engine with a single over head cam 8 port head.
these engines are lightly ported and run a compression in the Reagan of 10 to 1 we found that at 12:1afr we where getting 96 Flywheel HP and at 14.5:1 we where getting 98 FWHP. we ran them like this for a season until one day during a dyno session we found we had 101 FWHP we could not understand this but when we checked the fueling we where running 13.3:1 at full chat. It turned out that when we set fuel pressure original the feed fuel line to the pump had a restriction in it, (a cable tie pulled to tight) when the restriction was removed the fuel pressure rose some 10 psi, seen as we where not running closed Lupe on the fuel map there was no comp made for this fuel pressure causing an increase in AFR.

this is not the only time I have seen this happen eather

sorry for the orr topic

The issue is not in running either at 13.3 (or whatever is best) or 14.5. The issue is that is if you want to have this 13.3 with a wet manifold setup, you won't be able to have it on all cylinders. You will either have it on the inners or the outers or you will have it on neither even though your O2 sensor tells you that you have the correct value.

You will always have a mismatch between the inners and outers AFR. And it can be a full 2 points spread (or more). So if you want an AFR of 13, you'll have 12 on the inners and 14 on the outer or 13 inners and 15 outers or 11 inners and 13 outers depending on where you make your measurement.

The only way to even come close will be by using port injection with independent fueling for inners and outers as does the latest Megasquirt code.

Having said that, if you're happy to have a sub-optimal setup that won't be worse than a carb then a single point injection setup will be easier to set and will work. But you can also easily end up with something worse than a carb due to the way fuel is introduced in the air stream compared to a carb. And the dual throat shower-type injector is probably one of the worse setup for that in my mind.

Jean

http://www.jbperf.com/

http://www.jbperf.com/

Sorry I gave you duff info.
Just looked at my MSQ file and it indeed says 4 injections per cycle and the injector staging is alternating.
Its been a while since I went through all the options!
Port injection is the best option no doubt but I wanted to get mine running quickly without going through too much development so I opted for the safe wet manifold option.
As mentioned before, after a bit of mapping mainly with transient fuelling, it works really well.

http://www.turbominis.co.uk/car/99/

what trottle body is that you are using Mart from your pic it only looks like you have two injectors have you changed that if so any pic,
regards,
Paul.

I have put a couple more pics in my album showing the throttle body.
It is a Rover K series 48mm throttle body which has been modified to accept two injectors side by side.
I did it this way for simplicity so I needed only a single fuel rail.

Nice setup. To me that's a very good way to do injection with a wet manifold.

Jean

Edited by jbelanger on 26th May, 2009.

http://www.jbperf.com/

Awesome Matnrach, I will be copying you for my 998 Injection system realy a very nice simple way of geting the package in there.

Edited by PaulH on 26th May, 2009.

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 totally agree with Paul and have also never seen anyone back up their claims with hard data.

And if you're going to the trouble of putting 2 injector pockets, you might as well put them one per port close to the head and run the "right" code. Even if you use a single SU intake manifold and an SU as the throttle body, you'll be able to tune the AFR distribution to what your setup needs, provided you also get the sensors needed to monitor it.

Jean

http://www.jbperf.com/

I have to say that’s a good point Jean but looking at what Paul and rod has had to do I, Fear I just don’t have the time to get all this phasing together. My understanding from what I have read of Pauls posts is it no where near as simple as sticking two injectors near the port loading some code to mega squirt and setting up your fueling according to two AFR gauges.
where as wet manifold although it is not simple it is alot quicker to get a good result quickly. correct me if I’m wrong.

I would love to run the 5 port code my self but cant see my self having the time to get it all working.

It will be simpler and you MAY get something that feels like a good result but you won't know unless you measure it.

And even if you do measure it, all you'll be able to do is live with it and set it to the best compromise that won't hurt your engine.

Jean

http://www.jbperf.com/

Very true Jean, Mor thinking need I think,
Paul.

As I have said previously, i did this because I did not want to go down a development route that would take time and money and end up with something that was, at best marginally better for my particular application.
I accept it is theoretically better to have all cylinders running with equal AFR but to what extent this is superior to a wet manifold with two cylinders running at say lambda 0.92 and two others at 0.8 is unkown.
Maybe you might be able to get 1degree more spark on average but this would take individual cylinder knock sensing to fully evaluate.
I accept my opinons are unsubstanciated and go on feel but unless you back to back the two solutions on a dyno, both of which would be fully optimised, you can't say to what extent port injection would be of any measurable benefit.
Also my solution has been up and running for nearly a year now that only required base and transient fuelling to be sorted.
Maybe if I wanted to a development project I would have gone down the port injection route.
Just my opinion, sorry.

Matnrach,

As one of the "few" going down the port injection route, I hope I didn't come across, on this thread, or any other thread, as anti-wet manifold.

I think my first comment in this case was inject as often as possible to achieve best mixing before it hits the ports - ie, do the best replication of the carb.

Your TB solution looks ideal, and a damned site easier to tune than filing needles on an SU.....

I personally wanted a "development project" and that shows in that yours is up and running and mine only "nearly" is......

So, back to Coupe's question, stick with it, pick the injection timing/pattern that best replicates the carb and at least know you can adjust it all without a file, well not the kind that removes metal, just the kind you find on a PC !!!!

Rod.

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

Yes, the wet manifold is the easiest solution. But using the ECU also allows full 3D ignition control, closed loop average lambda (so at least it maintains its cylinder error rather than exaggerating it), control of charge cooler pump/fan, rev limits, closed loop boost control etc etc
So its not just a replacement for a carb.
But you knew that anyway!