Siamesed Code Trial - Take Two
Turbo Mini Forums

Starting a new thread as the last one got a bit political.

You may remember in February that I was making good progress but something was not quite right, so I modified the inlet manifold, got a single run out of the car then seized the engine before I could get a log.

So I've now fitted the 998 Turbo engine - less the turbo for now.

So it's basically an NA 998 with a 9:1 compression, running a 2.95 diff, 12G295 head and a secret cam.

Instrumentation as before, dual widebands etc.

Finished getting it roadworthy this morning and took it out for a quick logging session, literally just round the village. Took it up to about 40mph on the main road.

The log is attached.

I'm quite pleased with it. Between 2000-3000 rpm the AFRs are virtually identical, they spread a bit more after that.

Bit of a way to go, but good progress.

Cars looking good with the new wheels:

Attachments:

52.55kb

Edited by Paul S on 8th Jun, 2008.

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."

Paul,

That looks good! Good work!

Do you have a cam sensor or do you think it would be relatively easy to put one? I was thinking that the next thing to try would be invidual cylinder fuel trim but for that there would need to be a cam signal to identify each cylinder. There would also be a need to add 2 injector drivers if you want to have staged injection but that could be done also.

Jean

http://www.jbperf.com/

Hi,
I think it could be possible to replace the MPi cam sensor setup using the fuel pump hole.
Looking good so far Paul, give yourself a pat on the back :)
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

Jean,

Very interested in your comments - I've been following this for a while and, although I'm a long way behind, I've built my MS3 board to take a cam input (using OPTOIN) in case there was a future use for it.
My thoughts (at the time) were that, although MS only has two injector drivers, simple port injection would (kind of obviously) only require two injectors - assuming not staged - so a cam signal could allow the timing to be displaced between 1 and 2 (or 3 and 4).

I was thinking of using the dizzy drive shaft and using a very small optical trigger wheel in place of the actual dizzy with a simple slotted IR sensor. I think I can make it in not much more space than the standard blanking plug.

Rod.

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

I'm planning on sticking with the simple setup without the cam sensor for now.

The fact that the AFRs are close together between 2000-3000rpm means that I've got the right injection timing at that point. We just need to do some more work on the table.

The problem previously was that the manifold needed circumcision. This is the first log since it was modified and it is a big leap forward, results wise.

I've got one more oil leak to fix, the other tappet cover now, and we can then start the real tuning.

As off now, I'm totally knackered and am going to put my feet up and watch Hamilton.

Still on schedule for Avon or MITP.

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."

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

The best way to do this would be with a fully sequential ECU with timed injection. I think the new MS-II Sequencer will be just right.

I'm not convinced that a cam sensor and replicating the MPi setup is much of a step forward over what we have already achieved.

I'm going to explore the current Megasquirt option fully first though.

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."

Once the cam sensor is in place and the code is there to take advantage of it for fuel trim, it won't be a big step to have it fully sequential.

And I don't think you'll be able to do much more with the sequencer than what can be done with any other sequential ECU. If there is no specific siamese-port code, you won't be able to use it correctly.

But in any case, it's a good idea to try and see how good of a tune you can get with the current setup. If there's no need to go with the added complexity of a cam sensor and sequential injection then so much the better.

Jean

http://www.jbperf.com/

But surely the "sequencer" still needs a cam signal ???

I remember starting a thread on this earlier (when I was a newbie) and I caused a lot of grief.

Timed injection (between inner and outer) can only be achieved with a cam signal, can it not ???

(anyway, only 30 mins to go...)

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

Yes, it does require a cam signal.

http://www.jbperf.com/

A bit of a close up of the middle section attached.

Encouraging that the AFRs cross at low MAP and they are close at cruising speeds. I think I need a bit more injection advance at the higher speeds.

Plus the whole VE Map is on the rich side. I'll just alter the REQ_FUEL to bring the AFRs up a bit.

Attachments:

23.16kb

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."

Well, Ive now done a couple of hundred miles on the new engine with the EFi fitted.

The car runs like a dream, starts easy, idles OK and for a high geared 998 goes very well.

I spent a lot of time last weekend perfecting the fueling map and getting it running well.

If it wasn't for the inner cylinder wideband reading rich most the time, I would have thought I'd cracked it. So I bought a new sensor just to double check the readings.

The attached log shows the sort of readings I'm getting. It's not far off at low MAP but under any sort of load, the inners go rich.

The difference is about 10-15% and I think that this is attributable to the difference in VE between the inner and outer cylinders.

I think that the Siamesed code is doing exactly what was intended. We just had assumed that the cylinders all need the same amount of fuel, but they do not IMHO.

If I was only running an NA engine, I would live with it, but I need better fueling for the turbo.

All is not lost, however, because I'm now going to fit a cam sensor. This will allow me to run the injectors on alternating ports and timed to give the correct fuel for each cylinder by injecting over the inlet valve overlap point.

If you look at my first log in the original Siamesed Code Trial post, you will see the sort of results this gave. The only problem was that it would only work every second start and would lose synch if you hit a big bump.

Attachments:

23.16kb

Edited by Paul S on 19th Jun, 2008.

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."

Very interesting.
Clear from both logs that the AFRs go well apart under load, but any ideas why the inners go rich while the outers stay a pretty constant 14ish at cruise or load?
ie, have you any thoughts why the outers don't go weak by a similar amount? When you say you think the VE is different between the inners and outers, do you mean that in the technical sense (ie, that the inners simply can't pull in as much air as the outers) or in the sense of the VE tables need to be different, inner to outer?

If it is a genuine VE difference due to the air flow regime of the siamesed port, won't that actually be much reduced under forced induction conditions?

How sooty is the exhaust? An AFR of 10-11 under load can't be doing the inner cylinder bores too much good.

Finally (sorry it's all questions...) are you running thermocouples on the exhausts yet? I seem to remember your plan was to try and correlate EGT with AFR so as to not have to use wideband sensors prior to the turbo...

Rod.

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

The outers stay around stoich as that is how the VE map has been setup. I used the Autotune feature of Megatune based on the outer AFR reading.

I think that there is a difference in the VE due to the inlet port inertia and the siamesed exhaust port helping the outers. Alpa has found similar on his setup with the inners needing 15% less fuel.

Forced induction will not change the VE. It just increases the pressure and charge density. The factors affecting VE will be the same.

No thermocouple readings on the exhaust yet. No point if the widebands show this sort of difference.

I'll just go and pull the plugs and take a photo.

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."

Isn't this interesting... I know now why you wqanted to get a new sensor to check the results - this is indeed very perplexing (but explainable I guess).

Paul - Whislt you reckon that EGT's are not required - it'd be interesting to see if they do correctly follow the O2 sensors - ie converge and then seperate at light to heavy throttle...

So is a cam phase sensor an essential then?
Is this a hall switch device or other? My weber alpha unit was basically a 2wd cosworth cam phase sensor setup. This is a nice compact sensor unit easilly fitted in a dizzy body.

Edited by TurboDave16V on 19th Jun, 2008.

Here is a pic of the cosworth setup...

http://www.bigturbo.co.uk/main5.htm

Now here is a surprise:



In order 4-3-2-1.

Hardly any difference in colouring.

But then the last mile or so was at cruise where the AFRs are closer.

The thermocouples are fitted but not connected up yet.

If I get them working then we may see some difference, but without equal AFRs we do not know how much temperature difference is due to the siamesed exhaust port.

Edited by Paul S on 19th Jun, 2008.

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."

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."

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

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."

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."

Paul, how well did the autotune work?

Did it just get you to a ball park figure or are the autotune adjustments your final settings?

How long did you keep the engine at a certain Rpm/load before moving on to another area in the table?

Sorry to go a bit O/T, just interested in knowing how accurate you found it?

https://www.facebook.com/pages/Fusion-Fabri..._homepage_panel

www.fusionfabs.co.uk



1/4mile in 13.2sec @ 111 terminal on 15psi

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've got the car running very nicely on alternating squirts, once per cycle. AFRs very close at idle and when revved.

I can't drive it because I know it will go out of synch. Then the injection point will be such that all the fuel goes in the inner cylinders.

It needs a cam sensor to keep it in synch.

The cam sensor is fitted and working, but the engine will not start with the Dual Wheel settings in Megatune.

Jean is having a look at the code to see if the Dual Wheel part is broken.

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."