Injection and turbo
Turbo Mini Forums

I know DTA dont work on the 5-port, but is there another enginemanagmant systen who work with injection and turbo??

www.shag.no

emerald?

None (apart from the weber alpha based - ie user unfrinedly) unit that Keith Calver keeps plugging.

At least as far as i'm aware anyway. Give it a little time and Megasquirt will be however.

Megasquirt would be fantastic I think :)

www.shag.no

Motec M4 (or current eqivalent) with advanced tuning enabled will work on a 5 port turbocharged A-series.

In fact other EFI computers should do it so long as it has:-
a) sequential injection (you need this for b) to be effective)
b) individual cylinder trim
c) end/start of injection (firing point) adjustment

Cheers

Daniel

Edited by DOZ on 1st Sep, 2006.

First 5-port EFI turbo in the world (built 1997, running in 1998)

that holley projection looks remarkably similar to the spi ,....

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

It all depends where you are going to mount your injectors. Forget mounting them on the inlet tracts, you'll need to sequentially fire the injectors and the firing window will get very small when you get anywhere a decent power output. I've got a 'wet' manifold system and an Emerald M3DK ECU on my mini and it works fine. I removed the carb and fitted a Jenvey SF45 throttle boby with 2 injector ports (one sits above and one below the TB). A pair of 420cc pico injectors were used to keep everything compact and these run at about 80% duty cycle at max power.
138 bhp @ 6100 rpm and 144 lbft @ 4000 rpm, confirmed on Dave Walkers rolling road.

It all depends where you are going to mount your injectors. Forget mounting them on the inlet tracts, you'll need to sequentially fire the injectors and the firing window will get very small when you get anywhere a decent power output. I've got a 'wet' manifold system and an Emerald M3DK ECU on my mini and it works fine. I removed the carb and fitted a Jenvey SF45 throttle boby with 2 injector ports (one sits above and one below the TB). A pair of 420cc pico injectors were used to keep everything compact and these run at about 80% duty cycle at max power.
138 bhp @ 6100 rpm and 144 lbft @ 4000 rpm, confirmed on Dave Walkers rolling road.

Neil, do you have a turbo or is that normally asperated

Team www.sheepspeed.com Racing

Do you inject before, or after the TB ??

9.85 @ 145mph
202mph standing mile
speed didn't kill me, but taxation probably will

It's a T3 turbo at the moment although I might go smaller to try and get rid of some lag. The injectors are mounted on the TB, (check out the SF45/0/2 on Jenvey's website: www.jenvey.co.uk), so effectively I'm injecting after the butterfly but just before the inlet manifold. 138bhp is certainly not the biggest output in the world but it is quiet, smooth and has a decent tickover. It's currently running 10 psi boost with a small intercooler but when I get round to it I'm going to get an electronic boost modulation valve and set up the ecu to give me more boost at higher revs. If anyone would like any pics let me know and I'll post some.

Hi Neil,

Can you please send me some pictures of your setup, and also some details of who made your pipework to connect the SF throttle body to your intercooler. I have almost the same setup running NA at the moment and am looking at options for turnocharging it. I have a lead on a NEW metro turbo engine that was bought by a racer and never used

Also if you dont mind I would like to take a look at your ,fig file from the ECU. I can swap mine but not sure if it is of any use as its NA

Cheers

Attachments:

274.52kb

Check out the (slow) progress at.......
http://www.turbominis.co.uk/dn89mini

Hi,

I'll dig out some pics tomorrow and get them posted. The cooler pipework is made up of short lengths of aluminium 2" tubing TIG welded together, I used to have a contact for this sort of thing but he's not in business anymore.
You're welcome to a copy of the .fig map, I'll pm it to you tomorrow with a bit of an explaination on the MSPB value I used.

Cheers

Neil

A few pics as promised. You can't see it from the pics but there is ducting below the intercooler to try and direct cool air from the behind the grille up through the core and then out of the bonnet via some Escort RS Turbo bonnet vents. It seems to work quite well although probably not as good as a front mounted intercooler. I've got a front mounted Metro rad so I don't have any room left there for the intercooler.
You can probably just make out a Rover K series IACV on the offside of the modified turbo plenum which gives me a decent tickover from cold.

Cant see the pics?

Check out the (slow) progress at.......
http://www.turbominis.co.uk/dn89mini

You're right neither can I! I think they may be too large to attach so you'll find them here:

http://www.turbominis.co.uk/index.php?p=20&id=3279

Attached is my fig file from my Emerald ecu. Normally they are supplied set up so that 255 in your injection map equals 100% injector duration. As the ecu is a TPS vs speed input with MAP correction and I wanted to run 15 psi max boost, that meant a max injector duration of 50% in the map (i.e. in NA mode) to allow the 100% (approx) correction on full boost. This made my map numbers around 110 max which is losing resolution so I resized the MSPB number (to 31 I think) so 255 in my map is about 50% duration and I still get fine control.
I hope that makes sense!

Regards

Neil

Attachments:

1.25kb

Edited by neilj1678 on 17th Feb, 2007.

A nice install.

Do you have any way of getting an AFR reading from the centre and an outside branch of the exhaust? I;d say not from the pics but then again I am sure I'm going mad so you've probably got it covered.

I am interested in how even the mixture distribution is.

Every day is a school day ...........

How fast and how expensive ...... the same question...

Looks great mine is only a test setup at the moment but it looks soo poor compared to some of those posted on here Thank you for taking the time to post this info. I will review the Fig file and hopefully all will become clear. Thanks again

Check out the (slow) progress at.......
http://www.turbominis.co.uk/dn89mini

Neilj,

Is that the standard MPi Air Control Valve?

Did you convert it from Unipolar? Or is it standard?

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."

Unfortunately there's not a lot of space to get a couple of lambda sensors on the manifold, it's a stock turbo manifold so everything is pretty cramped. Also I'm led to believe that the backpressure from the turbo would give you false AFR readings anyway, plus they would probably get fried! It would be nice to know though wouldn't it?
There's an A series project on Emerald's website in a Midget, DW tried two lambda sensors on that with an LCB and didn't see any mixture bias so I've just followed the same design as that but with the turbo.
The Idle valve cost me 8 quid new off eBay, it was sold as a K series type but I believe it's the same as the MPi apart from the lower housing. It interfaced directly with the ecu without any conversion; 4 wires between the valve and the ecu and the 5th was a 12v ign controlled feed to the valve.

for a very rough idea of whats going on you may want to try an EGT probe in one outer and the centre runner. 2 k-type thermocouples and one decent thermometer should be overly expensive and should hint at the fuel distribution. obviously 2 widebands would be best ,but certainly not cheap, and like you say pressure skews the readings. Have you done a hard run, shut off and pulle dthe plugs for comparison?

turbo 16v k-series 11.9@118.9 :)

Denis O'Brien.

It's true that pressure skews an O2 sensor's output (not just wideband) but what you mostly want is relative data. If both the side and centre branches give the same results then you're ok. You can then get the correct AFR with a sensor after the turbo.

As you say, the temperature is an issue but if you do some quick tests you might be ok. And the sensors themselves are not that expensive. It's the controller that gets you.

Jean

http://www.jbperf.com/

You could remotely mount the lambda sensors on a small diameter pipe.

You could either restrict this, then dump to atmos, or back into the exhaust after the turbine.

YES, it will make the turbo system less efficient, but for the purposes of testing....it may be worth it ??

9.85 @ 145mph
202mph standing mile
speed didn't kill me, but taxation probably will

Excellent ideas. I've just bought a LC-1 wideband controller at 150 quid inc the sensor, at that price I may just buy another and fit them both just to see what's really going on.
Thinking about it, doesn't the SPi set up have the lambda sensor on the manifold, near the downpipe mounting face? Maybe they won't get too hot in that case?
My ecu has a datalogging function and it has a couple of 0-5v auxilliary inputs that I could use to log the two AFR's so I could set it all up, start the datalogger and go for a run down the road.
I'm just putting the finishing touches to an inspection pit in my garage (got fed up of crawling around on the floor!), as soon as that's done and I can get the car back in I'll have a serious look at this.

Ref the pulling the plugs question, I had it on Dave Walkers rolling road last year and after it got a good spanking on there I pulled the plugs and they all looked even.

Edited by neilj1678 on 19th Feb, 2007.