Canems trial
Turbo Mini Forums

Well, I finally started installation of the Canems ECU that I was sent for trials. It's going on a 42 year old Cooper S engine that came out of my brothers car, which has been sitting for 17 years. I abducted the shell and it's sitting in my warehouse slowly being assembled.

Last weekend I restarted the EFI 8-port, running with a DTA ECU, ran it for about an hour so I could sort out the acceleration map, then pulled it off the dyno stand. The S engine went on for diagnostics and baseline. An hour later it was running with an HIF 6 and Lucas distributor. We then pulled the head to change the ancient valve springs and do the other engine modifications necessary for the EFI setup.

First step, of course, is the 36-1 pulley. Happiness, to me anyway, is having a full machine shop and friends who have literally tons of scrap 4130 steel. With that at my disposal, it only takes a few hours to make one of these:



The VR sensor will go on a bracket directly above the pulley so it's out of the way of the belt, but that's the job for next weekend.

The inlet manifold is already done so it won't take too long to get this thing up and running. Stay tuned...

www.starchak.ca and www.TDCperformance.ca

Marcel,

I assume you're waiting to see exactly where the sensor will be before you remove the missing tooth.

I'm curious to see how it does and what you find out. Are you going to have a couple of O2 sensors or some other means of determining how good (or bad?) the fuel distribution is between the inner and outer cylinders? It will be nice to compare your results with what Paul has seen with his Megasquirt testing in terms of fuel distribution and injection timing. Of course, your intake manifold and injector positions are different but it will be interesting to see how they compare.

Jean

http://www.jbperf.com/

Maybe a bit off topic but what is the real word advantage of timed port injection v throttle body injection?
I can see the advantage for high speed race engines with very low duty cycle injectors for correct fuel timing and port evapouration but not on a t/c siamesed head.

OK there will be some AFR variation between cyclinders but this is relatively small.
I have throttle body injected my engine with MS1 V3 and the response after some accel fuelling is really very good indeed. Absolubtly no response problems at all.
I run 0.86 at MAP>1.4bar abs which allows both pairs of cylinders to run probably within 99% of maximum torque capability, probably nearer 99.5%.

Maybe there is some HC benefit at idle but even then I am not so sure.

Any comments?

Edited by matnrach on 8th Jul, 2008.

Great to see you are back with the A series Marcel. We need all the help we can get.

It's not until you actually fit two widebands as I have done that you actually start to fully understand what is going on with the siamesed ports.

As I and Alpa have found, the inner cylinders use 15% less air and therefore want 15% less fuel. This means that you need some form of individual cylinder trimming to get the optimum results.

The advantage of port injection over throttle body is that, in theory you can address these issues. With throttle body injection you are throwing slugs of fuel into the air flow. If you are lucky, then you get good distribution of the fuel around the cylinders. However, it is unlikley at anything other than a particular load site.

This is particularly apparent at low loads, high MAP when the pulse are short, the slugs are small and have a mind of their own.

Not a problem if you do not care about fuel economy, but I do.

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,
I've no idea which is best, I went the same way as you with TBi for the easiest way in my case.
It's encouraging to hear you've had good results with your setup though :)
The way I look at it is it's can't be no worse then a carb and no one running one has any issues with fuel distribution up to over 200 bananas plus I'll get better economy on cruise
Anyway back O/T....
I'm pleased that people are pioneering with port injection and look forward to the results :)
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

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,
I've no doubt the distribution isn't fantastic, I've followed your testing, but if you've got over 200 nana's without melting something that's enough power to destroy a gearbox...
I'm going to be slightly biased in my view as I'm not going for lots of power just reasonably quick and not as noisy/thirsty as my 1380.
I'd be interested to see what Matnrach are getting MPG wise?
As I said, I chose the easy route and fair play to the pioneers. :)
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

I haven't done any mpg calcs yet. I depends very much on the duty cycle of the engine obviously.
My old N/A induction system with around 100-110 bhp was giving on average 40mpg with a reasonably well setup twin 1.5 SU and MJ ignition.
I still do not believe that the AFR differences will loose much at max load, maybe 1%, but I could be wrong.
I have mapped the engine at around 0.9 low load to 0.84-0.86 at full load to compensate for possible large lean transient AFR excursions so I'm not really expecting it to be great, especially given the extra power it now has.
However for me, an increse in fuel consumption in line with the power increase is totally acceptable.
N.B At the moment I have the N/A cam fitted which has too much exhaust overlap and the inlet valve closes too late so that will also affect consumption. This will be rectified shortly.

Too lean or too rich is all the same: the power is lost. The overall efficiency as well. My engine is clearly unhappy under 11:1 AFR. And to stay under 1% CO in idle (880 RPM) it needs 14:1 or more.

std 998 A+, g295, MD266, RHF4, 109hp @0.8bar/5400rpm

Yes you loose power either side of optimum lambda but how steep is the loss as you move away from it?
I chose 0.86ish average so that the loss wasn't too bad on the lean or rich cylinders.
If all were the same I agree it would give more but I doubt it would be that much. This generalised graph shows it could be a bit more than I first thought but not much.

Attachments:

5.26kb

Edited by matnrach on 10th Jul, 2008.

When going under 11 I feel the engine really sluggish.
Of course all that will depend on the entire setup. I used modern EV6 fine spay injectors and it's an MPI, so combustion is certainly better than with old-school injectors. Also the head is a 998 std, the chamber if very compact, also better for combustion efficiency especially at low RPM. With a poorer combustion the impact should be lower because HC exhaust level would be higher.

I've already presented my theory about siams. I think with a TBI you are creating a kind of wall wetting reservoir, this may cure the problem at high RPM. Under some RPM level all the accumulated fuel would be consumed by the inner as evaporation rate is rather constant.

I don't think all that really matters on small 4 cyl engines used on light cars like Mini. I'd like to try my system on an old US engine, some of them are also siams. This should impact MPG much more. I'm moving to the US by the end of July :) .

std 998 A+, g295, MD266, RHF4, 109hp @0.8bar/5400rpm

All I can say is that at lambda 0.86 (12.6:1) at high load and reasonable speed there is no problem as I see. I doubt the rich cylinders will see 11:1 (lambda 0.75) or lean ones 14.5 (lambda 0.99) so all in all the loss will be quite small compared to a perfectly set up (unlikely anyway) port injected engine.

At low load all that matters at least to me is response and I cannot fault my setup as I doubt you could make it any better especially if you look at the average tailpipe lambda on transients.
I too use the latest 4 hole EV6 injectors which must make a difference.

I agree there is copious wall wetting at low speeds which will increase consumption but how much in reality, especially if you use high load most of the time, is questionable.

All I am saying is that in my opinion the gain from port injecting over TBI on a turbocharged A series head is very small.

Edited by matnrach on 11th Jul, 2008.

In the beginning my setup was not phased, so all cyls were getting the same amount of fuel. It worked fine, but plugs were of clearly different colors.
I think in an FI application the most important is to avoid ping. Richness is only one aspect. All these odd things like pockets and sharp edges are also sources of ping. In my case since I've been using phased MPI I've been unable to make the engine ping at stoichio, while it was easy before.

std 998 A+, g295, MD266, RHF4, 109hp @0.8bar/5400rpm

Unless the duty cylce is really short, phasing should not make much difference to knock for a given mechanical specification.
Sure if the AFR is wildy different between cylinders there will be a reduction in knock limit. I cannot see this is true with a well mapped TBI.

My plugs with TBI are pretty similar in colour although this is only subjective as I have not measured the lambda in each of the ports.

I wonder if Marcel ever ran the Canems?

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."

Yeah... Besides what they have on their site, I haven't seen anything else about someone running it. It would be good to have some independent results.

Jean

http://www.jbperf.com/