Tuning MegaSquirt for 998 NA
Turbo Mini Forums

The dual table set up was the only way I could get the AFR's equal using the larger Injectors. I had to have Inj timing 1 at 12deg and inj timing 2 at around -60deg.
When I put the 33lbs/hr injectors in I just stuck with the dual tables, as it had been the best results from the larger injector tests.
That said Tick over with the 33lbs/hr is reasonable good at 0 on both tables.

I will try the single table now I am happy that I have a good fall back MSQ.
The start of pulse setting was also inherited from the larger injector tests. I just found it easier to understand the changes I was making with a start of pulse figure.
I take your point with the differences when I finally get to the larger injectors, so I’ll also concentrate in mid pulse once I’ve trialled the single table.

I took a 84 mile trip this morning up to Swindon, across to Witney and back to Hungerford with my Dad driving, so I was able to monitor the laptop, instead of doing a run then stopping and making changes. So I now have lots of logs to look at.

’77 Clubman build thread
http://www.turbominis.co.uk/forums/index.php?p=vt&tid=618189

Siamese 5 port EFI testing
http://www.turbominis.co.uk/forums/index.php?p=vt&tid=611675

The fact you did 84 miles without problems is credit to your progress.

I'll say little more at the moment as my progress is very "slow".

But I would try to stick with a single table and mid-pulse at the moment, if only to make comparisons easier.....

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

Tomorrow I will work on Single table and mid pulse.
For now here’s the best of the results from today’s last log.

Dad’s right foot is not as heavy as mine, so rev and load ranges are fairly low. But good cruise speeds...

’77 Clubman build thread
http://www.turbominis.co.uk/forums/index.php?p=vt&tid=618189

Siamese 5 port EFI testing
http://www.turbominis.co.uk/forums/index.php?p=vt&tid=611675

That's looking a lot better, but you still seem to have a noisier signal on the inner.

Also interesting how the AFRs deviate just at the end when MAP drops (assuming I'm reading all the coloured traces right) which must be foot to the floor contrary to your comment about your test driver.

But it seems to be a transient thing, similar happens mid-plot with a gear change.

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

as above its looking good and close at part throttle,

what are we seeing at the end of the log when the MAP dips? 11 and 13 afr?

Good work , I like the idea of the plot sowing diferentoial between the inner and outer AFR's. :)

’77 Clubman build thread
http://www.turbominis.co.uk/forums/index.php?p=vt&tid=618189

Siamese 5 port EFI testing
http://www.turbominis.co.uk/forums/index.php?p=vt&tid=611675

The problem with that dip at 863 secs is that it is very temperature dependant.

The only change in the load was about a 5% reduction in MAP and you see the inner cylinders suddenly get 20% more fuel. What's odd is that the outers do not lean out a corresponding amount. Hence, I conject that the fuel is coming off the port walls at the lower MAP. This would have corrected in a short time.

I would not try to tune it out yet as on a warmer day, it may not happen.

Edited by Paul S on 4th Apr, 2010.

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."

’77 Clubman build thread
http://www.turbominis.co.uk/forums/index.php?p=vt&tid=618189

Siamese 5 port EFI testing
http://www.turbominis.co.uk/forums/index.php?p=vt&tid=611675

Oh yes.

We've not made ay changes to the injection timing since last summer, but over the winter we noticed that they had moved a bit.

Last Sunday we took her out when it was around 20 Deg C and we had very close AFRs.

If we had taken her out today, 10 Deg C, it would have been different.

It's all to do with the amount of fuel sticking to the port wall and there is not much that can be done about it.

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."

The solution might to adapt the EAE or X-Tau code to take into consideration the wall contribution in the fueling for the inner cylinders.

So instead of being just for transitions, it would be for steady state operation.

I've never looked at that part of the code so I'd need to research it a bit. But that seems like what would be needed for this. And it is likely going to be needed even more if using 2 pulses instead of the single pulse since the pulse for the inner cylinders will have to be at least partly done on a closed intake valve which could increase wall wetting.

It might be a good idea to start a new thread on this if any of you think it would be something to explore and add to the code.

Jean

Edited by jbelanger on 2nd May, 2010.

http://www.jbperf.com/

OK, based on those last 2 reply's I have to ask what is probably a very basic question:

Inlet air temp sensor. - My assumption was that the MS made adjustments to fuelling based on the air intake temp. So, lower temp, more fuel - longer squirt.

I based this on the fact that when the air gets colder I see the same variations that I normally do when I increase the Fuel VE table cells. The result being that I have to retard the injection timing slightly more to equal up the AFR's


As said, it was just an assumption - so now I have to read up on EAE and X-tau code...

’77 Clubman build thread
http://www.turbominis.co.uk/forums/index.php?p=vt&tid=618189

Siamese 5 port EFI testing
http://www.turbominis.co.uk/forums/index.php?p=vt&tid=611675

You're not wrong. The MS will vary the pulse width with MAT.

However, you stated that you are seeing the AFRs spread between inner and outer cylinders with lower temperatures.

This is because a proportion of each squirt will stay on the port walls. This will evaporate and move into the cylinder. However, if the port wall is cold this will take some time. You end up with all the fuel evaporated from the port wall entering the inner cylinder if the port wall is cold.

I don't think that we have enough data yet to confirm this and we certainly need more cars running to code to verify that there is a problem.

However, the variation is only small and I not sure that it is worth code mods to sort it out.

After all, all carburetted cars must suffer the same problem.

I think that you should make sure that you have the engine fully up to temperature before logging or making ajustments. Also fit as high a rating thermostat as the cooling system will allow.

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 tend not to make changes until after I have made a run and looked at the log. And I ignore the log data until the Coolant temp is up over 80 Deg C.

I have a 84 Deg C theromstat in it. The temp normally fluctuates between 83 Deg C to 91 Deg C. I really dont want to put an 88 Deg C thermostat in until I sort out a better cowl for the Electric fan. I don't think it is as efficient as it could be. It's drawing too much air through the side,instead of through the Rad.

Graph from the latest run:



I'll find out what is causing the noise on the Lambda sensor one day...

’77 Clubman build thread
http://www.turbominis.co.uk/forums/index.php?p=vt&tid=618189

Siamese 5 port EFI testing
http://www.turbominis.co.uk/forums/index.php?p=vt&tid=611675

That looks fine to me. There is no point in trying to be a perfectionist and get them equal everywhere.

I would try and sort out the spikey inner signal in the first instance. I would also check the outer as that loks a bit too smooth. I would say try swapping them around as that is what everyone used to tell me, but I still have memories of the scrapped knuckles and aching back!

Above 4200rpm, it seems to be going rich on the inners, so you may want to try some adjustment at the higher revs.

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."

’77 Clubman build thread
http://www.turbominis.co.uk/forums/index.php?p=vt&tid=618189

Siamese 5 port EFI testing
http://www.turbominis.co.uk/forums/index.php?p=vt&tid=611675

IMHO probaly not contributing to the spikey signal, but obviously an air leak will contribute to bad AFRs.

The spikey signal is much more likely wiring/electronics, an air leak will just skew the mixture (compared to what you have set it as) and hence skew the overall readings, but not add spikes.

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

I've gone through a fair few changes now with the LC-1's trying to get rid of the spikey signal on the inner cylinder AFR trace, but with no success.

I have:
swapped the lambda sensors inner for outer
just swap controllers between sensors (recalibrated)
Re-soldered all the conectors on the LC-1 wiring looms
Installed a clean set of wiring.

I still get the spikes on the inner cylinders

However, I did notice this on a run at dinner time:



It's almost as if the spikes are caused by incorrect injection timing.
Could this be wall wetting?

The problem is that the nearer I get the AFR readings the more the inner afr trace starts to fluctuate.

’77 Clubman build thread
http://www.turbominis.co.uk/forums/index.php?p=vt&tid=618189

Siamese 5 port EFI testing
http://www.turbominis.co.uk/forums/index.php?p=vt&tid=611675

from what i can see ,the spikes are greater when on full load ,that may suggest a emf from the ht leads/plugs ,do you have good suppressed leads ? i find the spiral wound stainless ones very good .

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

Good question. I'm not sure. The leads are new genuine ford parts, but whether they are suppressed...

Would I see the effects of EMF on both AFR traces if this was the case?

Here is a chart from the same run, with a MAP of 11 kPa. The spikes are there, but perhaps not as bad.

’77 Clubman build thread
http://www.turbominis.co.uk/forums/index.php?p=vt&tid=618189

Siamese 5 port EFI testing
http://www.turbominis.co.uk/forums/index.php?p=vt&tid=611675

depends on how close the leads to the sensors are to the ht area graham .

you could try wrapping the sensor leads all the way with tin foil ,and earth the foil .

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

yes tin foil,

a quick dirty way to see if it make a difference.

I doubt that this is an interference issue.

I only ran into similar problems once the sensor cable ran with a couple of inches of the HT lead. I doubt that this is the case in Graham's car.

I suspect that the inners are misfiring due to a fueling problem. Maybe there is too much fuel going to the inners at WOT. The AFR seems to be dropping to 10:1 and then the signal goes spikey. Once of the throttle and a bit less MAP, the signal settles.

I would try to reduce the fuel going to the inners, either by improving the timing to ensure that the outers get their full share of the intended pulse or by reducing the VE of the inner map if you are using two maps.

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."

On the first plot (dinner time run) first half of plot only, the spikes, although varying in amplitude, do seem to have a very uniform timebase.

I can't make out the numbers on the scales clearly enough to work it out for myself but is the frequency of the spikes on that plot (not the amplitude, just the frequency) related to RPM - like once every revolution or once every other revolution ???

The fact you have some non-spikey plots and have done all the obvious over interence now makes me think toward fuelling or ignition issues, ie, the spikes are genuine.

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

Graham, what is the lower limit of the LC-1 output? Could it be that you have it set at 8:1 and you are straying below that?

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'll try the tin foil just to eliminate, althought the ht leads and coil pack are well away from as much wiring as I could get them.

Rod, the major grid lines for time are 2 seconds apart. There are 14 spikes per 2 seconds. I'll do the maths a bit later, else I'll be late for work.

Paul, I'm not sure of your question. How do I check this?

’77 Clubman build thread
http://www.turbominis.co.uk/forums/index.php?p=vt&tid=618189

Siamese 5 port EFI testing
http://www.turbominis.co.uk/forums/index.php?p=vt&tid=611675