Going turbocharged and injection
Turbo Mini Forums

Hi,
I have a N/A engine at the moment running megajolt and am starting to build up a turbo engine using a modified turbo manifold and single throttle body. I will inject after the throttle and upstream of the plenum volume so there should be no problems with robbing (I hope).
I would like to go mapped injection and ignition (with water injection being a possibility later when I increase boost)
Since I already use a 36-1 pulley and Edis unit/wasted spark, I thought that megasquirt would be a sensible route.
There are plenty of V2.2 units around going cheap but I wonder if this is good enough.
There is so much info around on it, the various specs/ firmwares, compatability etc and it is a little confusing (for me at least).
Has anybody had any experience of the MS V2.2, V3, MSII and the various advantages/disadvantages and functionality (boost control, idle control etc)
Any advice would be greatfully received.

Thanks.

Edited by matnrach on 1st Nov, 2007.

You will get charge robbing with the setup you describe. At low speed and light loads the short injection pulse will not get distributed to all four cylinders.

I suggest that you have a read of this if you have not already.

http://www.starchak.ca/efi/siamese.htm

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."

Maybe I described my setup incorrectly.
I will be using a setup very similar to Neilj1678.
He is using a Jenvey throttle body with twin Pico injectors and an Emerald ecu.
He seems to have no driveability issues.
Even if there is some robbing at low speed on some cylinders, as long as they do not stray outside the lean/rich limit of combustion during transients it should be OK.
From his comments it should be fine.

I think that MS has some transient fueling tables but not sure.

Edited by matnrach on 1st Nov, 2007.

No offence, but that setup is not really any better than an SU, and costs far far more, is it really worth it?

I think it should be better for a few reasons:
1. Throttle body will have less restriction than an SU
2. Fuelling can be controlled exactly. In my experience the SU is not correct across the entire speed/load range (or even at WOT v speed)
3. Much easier to adjust the fuelling than altering the needle profile
4. Integral Water injection if necessary
5. I would have thought (not sure) the the fuel preparation is better with a 5bar injector set up
6. Better cold start?

What is a better setup in terms of absolute power with a turbocharged siamesed ported engine that couldn't be achieved with a decently modified version of this setup?

Incidently I am only looking for around 130-140hp with a decent response and torque curve.

I am open to suggestions

Edited by matnrach on 1st Nov, 2007.

Tom is right. The SU is very good and unless you get the injection just right, you will not better it.

You need to control the fueling to each individual cylinder to get fuel injection to work correctly. The system you propose will not achieve that.

There is a version a the latest MS2/extra software that has been modified to work with the A series. Do a search on "Siamesed Port Code".

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."

1) Possibly, but with the SU proven to flow enough air for a 200bhp engine, is it an issue? NO.
2) The SU can be got pretty close, which is all you will achieve with a throttle body injector setup.
3) Granted, but it bloody should be with the amount you will have to spend to achieve what you suggest.
4) Proven by more than a few members of this forum to not be needed.
5) Possibly, but you then lose any advantage in terms of atomisation by sending it down an inlet manifold.
6) Not really, more consistant maybe, but you still have the cold inlet manifold walls to contend with.

130-140bhp is easily achieveable without all the stuff you propose, Alex B runs 140bhp on an engine with lower compression, a ph2 cam, a mildly ported head, an intercooler, and 11psi. Oh, and an SU too.

Please don't take this the wrong way; I am all for fuel injection, it is the modern solution, hence the DON (with the all important 4 inlet ports) runs an Emerald ECU which was chosen and installed by myself. However, the method you are suggesting really does not have an awful lot to justify the outlay required to make it work.

Edited by Tom Fenton on 1st Nov, 2007.

Having said all that, I think that it will "work" and you will find it reasonably driveable with a bit of luck.

So if that is what you want to do, go for it.

EDIT: Of course you may damage the engine by running lean on the outer cylinders.

Have a read of this:

http://www.canems.co.uk/injection.php

Edited by Paul S on 1st Nov, 2007.

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."

Tom,
I sort of agree with all you said. Not sure about the atomisation down the inlet manifold. Wall wetting with a carb must be awful.
Nevertheless it is a nice project to do and not too expensive if you have the parts already which was my original point about using a 'cheap' MS V2.2 ecu.

On the water injection front, how rigorously has this been investigated. Are you sure that the detonation limit does not change when injecting the correct mass ratio of water/fuel.
In WRC it is absolubtly vital although the bore of the A series is much less.

Glad to see you've taken the comments in the spirit in which they were intended, although the ECU may appear "cheap" having recently been through the process of fitting injection to a car that did not have it before, it is almost certainly everything else that soon starts to add up, suitable fuel pump (even more so if you are planning on 5 bar fuel pressure, why so high?), fuel filter, regulator, injectors, throttle pot, throttle body, sensors for the ECU, intercooler pipework etc etc etc.
Carl was more than suprised at the amount of money needed to turn an engine fully built and ready to go in the car into something that would run, the Emerald ECU may be £600, but you will spend nigh on £1000 in order to buy all the other gubbins required just to go with the ECU side of it, then probably the thick end of £200 to do a decent job of wiring it all.
At the end of the day its your project so do as you feel, but I think there is a significant amount of cost involved that perhaps you have not thought about.

Tom,
I sort of agree with all you said. Not sure about the atomisation down the inlet manifold. Wall wetting with a carb must be awful.
Nevertheless it is a nice project to do and not too expensive if you have the parts already which was my original point about using a 'cheap' MS V2.2 ecu.

On the water injection front, how rigorously has this been investigated. Are you sure that the detonation limit does not change when injecting the correct mass ratio of water/fuel.
In WRC it is absolubtly vital although the bore of the A series is much less.

im seriously thinking of this approach mat , i have a 1310cc running 150 bhp at around 9 psi ,and am happy with the fuel curve of the carb on full power throughout the range ,but to get it to run the air fuel rations i want at cruise and transient ,is proving dificult , iam pushing the lean run cruise envelope though ...
using a cheap squirt ,and running 4 inj into a body ,seems to me to give some huge advantages ,mapping with no leaning over the engine and hoicking out the needle ,all parts available from the scrappy for naff all ,easy to make a su body into a throttle body ,fuel pump off any fi 200 bhp car ..
with ref to water inj , i think its great , i built an experimental 9.25 to 1 cr v8 in 87 to use water specifically ,and it would run 20 psi boost no probs on normal unleaded ,v simple water on off and flow adjustment only .im very suprised it isnt used more on high hp turbo a series .
go for it !!

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

Robert,
Very interesting to hear your comments although I was thinking if there has been any real dyno work done on looking at spark limits with boost and water injection quantity.

As you say the parts are not that expensive if you already have them or you go to a scrappy
At the moment I need a fuel swirl pot and a throttle body. I have no pressurised SU carb so taking this into account it shouldn't be too bad for the functionality.
Also the possibility of launch,traction control etc,etc etc and endless hours of dicking opportunity

Edited by matnrach on 1st Nov, 2007.

Roberts sentiments echoed here as regards fuelling at cruise...

I gave up trying to optimise my SU a few years ago - it just won't give me what i want (AFR wise) on cruise. Sure - the bananas are plentiful with the right foot welded to the floor, but at all other times they seem lousy. I'm also looking forwards to losing the choke - I can easilly overfuel mine on a cold start.

More power to you Matnrach. If spi does what you want - then go for it.

SPi rules

Now we all seem to be agreed that single throttle bodied injection is at least a sensible way to go, has anybody got experience of the MS V2.2 or MSII
Is there any real benefit of the later (more expensive) unit?

Edited by matnrach on 1st Nov, 2007.

Hi,
+1 for TBi
I'll say VEMS .. £330 fully built ECU, uses Rover connectors, nicely water proofed :)
Ask Sprocket about his
Wideband built in, will map the fuelling itself
3 Bar is all you'll need.
Good luck with it :)
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

Thanks for info
I already have a LC-1 wideband. Thinking about it I will put a sensor on one outer and the inner branch and a look for myself how bad the robbing is over the entire speed/load range (and under transients)
Before I go digging, does MS have closed loop lambda, and does it work OK?