Bigger maps to siamese sequential injection timing

This is a forum for discussing the development and testing of alpha MS2/Extra code. Documentation
(Runs on MS2 and Microsquirt)

Moderators: jsmcortina, muythaibxr

Post Reply
Vauxi
Experienced MS/Extra'er
Posts: 211
Joined: Sun Feb 19, 2017 9:06 am

Bigger maps to siamese sequential injection timing

Post by Vauxi »

Hello.
Is it possible to make like 12x12 injection timing maps instead of 8x8? Code works okay now but 8x8 is not even near enough. I don't have a clue when it comes to coding at this level.
racingmini_mtl
Super MS/Extra'er
Posts: 9128
Joined: Sun May 02, 2004 6:51 am
Location: Quebec, Canada
Contact:

Re: Bigger maps to siamese sequential injection timing

Post by racingmini_mtl »

That's not going to happen. That would take a huge amount of memory which is simply not available. Also, that's the first time I've heard such an issue. Timing should be relatively linear so it should not require that many cells. Which engine are using it for?

Jean
jbperf.com Main site . . . . . . . . . . . . . . . . . . . . . . jbperf.com Forum
Image
Vauxi
Experienced MS/Extra'er
Posts: 211
Joined: Sun Feb 19, 2017 9:06 am

Re: Bigger maps to siamese sequential injection timing

Post by Vauxi »

I kinda knew the answer unfortunately :( But had to ask. Vauxhall ohv engine. I have a bit agressive cam in it. Cam and throttlebody makes big differences within cells. It smooths out when engine "gets on" with the cam.
racingmini_mtl
Super MS/Extra'er
Posts: 9128
Joined: Sun May 02, 2004 6:51 am
Location: Quebec, Canada
Contact:

Re: Bigger maps to siamese sequential injection timing

Post by racingmini_mtl »

I didn't know that Vauxhall had a siamese port engine (but I don't know anything about Vauxhall).

If you haven't done so, you should try to use most of the table for the transition area and have as few values as possible before and after. The timing will be less critical at lower RPM since the injection window is much longer. Unless you're using the single injection pulse method (which I wouldn't recommend with a big cam).

Jean
jbperf.com Main site . . . . . . . . . . . . . . . . . . . . . . jbperf.com Forum
Image
Vauxi
Experienced MS/Extra'er
Posts: 211
Joined: Sun Feb 19, 2017 9:06 am

Re: Bigger maps to siamese sequential injection timing

Post by Vauxi »

Image

There you go. One Vauxhall siamese cylinder head fresh from the machine shop. With this setup at low rpm range its more like doesn't matter what you input than less criticall. :D It just steals the mixture to the inner cylinders no matter what you do. :lol: Single pulse is not in use.
racingmini_mtl
Super MS/Extra'er
Posts: 9128
Joined: Sun May 02, 2004 6:51 am
Location: Quebec, Canada
Contact:

Re: Bigger maps to siamese sequential injection timing

Post by racingmini_mtl »

Vauxi wrote:[With this setup at low rpm range its more like doesn't matter what you input than less criticall. :D It just steals the mixture to the inner cylinders no matter what you do.
That's strange. That would point to a setup that has a lot of wall wetting contribution. What does your intake manifold look like and what's the position of the injectors? Could you change the aiming of the injectors?

As I said, you should have a large margin for the injection timing at low RPM. But if you have a large wall wetting contribution, you will still have a large AFR difference due to charge stealing even with a perfect timing. Don't forget that you can use dual table or the trim tables to minimize the AFR difference; of course that won't work if the wall wetting contribution is too significant.

Jean
jbperf.com Main site . . . . . . . . . . . . . . . . . . . . . . jbperf.com Forum
Image
Vauxi
Experienced MS/Extra'er
Posts: 211
Joined: Sun Feb 19, 2017 9:06 am

Re: Bigger maps to siamese sequential injection timing

Post by Vauxi »

Image

Injectors are in the middle of the intake runner as intake divider is inside the head and they are pointed straight to the midlle of the channel. Injector spray type is one wide cone. I did a lot of thinking between cone and two spray type injectors and decided to go with one cone type. Wall wetting sounds good theory just because when head is cold. The stealing cylinders switch places. When clt is like less than 25'C, inner cylinders tend to go way lean in certan load points. And like from 40'C situation is back to normal outer being leaner. I grinded manifold channels as smooth as I could after welding. Surface is left rough as intake is cast aluminium. I'm using dual table and trim tables. Trim is here and there the only way to make mixtures more close each other. Another thing is to use ego as I have dual lambdas and dual inputs used in ecu. I guessed that with this setup it is not even possible to achive always same mixture all the way. Not even close. It just would be nice to have :) Atleast mixture differences calms down to the upper rpms and injection adjustments starts to make more sense.
racingmini_mtl
Super MS/Extra'er
Posts: 9128
Joined: Sun May 02, 2004 6:51 am
Location: Quebec, Canada
Contact:

Re: Bigger maps to siamese sequential injection timing

Post by racingmini_mtl »

I hadn't realised in the first picture that these were the intake ports. You seem to have a good setup and you seem to have a good handle on what is needed to try to make it work.

That doesn't look like the easiest physical configuration to work with and it might be possible to have the injectors closer to the valve and with a slightly different angle. But you already gave a lot of thought to come up with a good solution and I don't know if what I mention would make a significant improvement.

While the increase in the timing table size could make it easier to tune, I think you would also need other changes to the code that would take into consideration the wall wetting factor. And unfortunately, there is no space for the table and even for much more code.

Having said that, I'm glad to see a different setup being used and I hope you find a way to make it work a bit better even with the current code constraints.

Jean
jbperf.com Main site . . . . . . . . . . . . . . . . . . . . . . jbperf.com Forum
Image
Vauxi
Experienced MS/Extra'er
Posts: 211
Joined: Sun Feb 19, 2017 9:06 am

Re: Bigger maps to siamese sequential injection timing

Post by Vauxi »

Yeah.
There are not many us who squirt this Vauxhall engine at this level. I know few others who had tried basics and leaved them there.. I did consider to install injectors few centimeters closer to valves but I was afraid that fuel spray amount then isn't going to be even for both inlet valves. As you likely know injector spray, cone or not doesn't get wide straght away from the tip. Moving injectors few cm more away from valves does increase wall wetting a lot. What is significant factor too is camshaft. As it is big cam with 103 lobe angle. It does pump air backwards at low rpm what makes tuning notably more difficult. =)

Yes. I'll try tune it to perfect as it reasonably makes sense. :lol:
MS2tester
Master MS/Extra'er
Posts: 606
Joined: Sat Apr 11, 2009 12:54 pm
Location: Aalsmeer the netherlands
Contact:

Re: Bigger maps to siamese sequential injection timing

Post by MS2tester »

4 pencil stream injectors angled towards the valve by 4 channels?
Vauxi
Experienced MS/Extra'er
Posts: 211
Joined: Sun Feb 19, 2017 9:06 am

Re: Bigger maps to siamese sequential injection timing

Post by Vauxi »

I think that setup won't work at all. Code doesn't support it and there is no room for 4 injectors. Intake pipe is only 40mm ID.
stevemgbgt
Helpful MS/Extra'er
Posts: 112
Joined: Tue Sep 02, 2008 12:36 pm
Location: north lincs UK

Re: Bigger maps to siamese sequential injection timing

Post by stevemgbgt »

007.JPG
I have siamese ports but as I run MS1 I cannot use the siamese code.I have thought of using MS3 and going sequential with the 4 injectors I have but I am unable to grasp fully the effect on the pulsewidth required.Not much help to you but I admire your work.Steve.
Vauxi
Experienced MS/Extra'er
Posts: 211
Joined: Sun Feb 19, 2017 9:06 am

Re: Bigger maps to siamese sequential injection timing

Post by Vauxi »

Sweet dual injector setup. 8) I think I have it now tuned as close as it can be. AE is still impossible to tune to work all the way. There is way to make tiny pedal move to squirt a lot when opening throttle from closed position and much much less when throttle is already opened. It is what you get with throttle bodies. I have it now configured to work ok at open positions.
It would be awesome to have seamese code in ms ultimate with all possible siamese code features what we are missing now.
Vauxi
Experienced MS/Extra'er
Posts: 211
Joined: Sun Feb 19, 2017 9:06 am

Re: Bigger maps to siamese sequential injection timing

Post by Vauxi »

I changed from Alpha-N to ITB. What a difference :!: It drives soo much better at primary drive area of the map. What is mostly welcome is that nasty fuel stealing behavior seems to be cured to way more easier to tune. Had to re tune all but ignition and afr maps. Now this is good place to start fine tune all over again... :D

BTW I might found a bug in TS. Accel Enrichment green dot when using 100 MAP doesn't work as it should. Looks like it's following mapdot value. AE works okay but the dot is completely useless.
Post Reply