| Page: |
| Home > A-Series EFI / Injection > Tuning MegaSquirt for 998 NA | |||||||
 (2)[/url] by [url=https://www.flickr.com/photos/150672766@N03/]Rod Sugden[/url], on Fli) 5988 Posts Member #: 2024 Formally Retired Rural Suffolk |
12th Aug, 2010 at 09:07:44pm
That looks really good Graham.
Schrödinger's cat - so which one am I ??? |
||||||
 608 Posts Member #: 1106 Post Whore Hungerford, Berks |
13th Aug, 2010 at 07:36:14am
On 12th Aug, 2010 sturgeo said:
Great Stuff! 
AFR's keeping nice and close as the revs increase. Are you doing all this on your own? Yes just me, at this end... I’m using the 42Lbs/hr injectors, which appear big enough up to 5000rpm. Not risking more than that on the old girl. I’m actually thinking about trying the 33lbs injectors again. It may help at tick over. I’ve not got better than around 12.5 – 13.5 AFR at tick over. Any leaner and the AFR’s get erratic. I’m using single fuel VE table and Single timing on mid pulse. Looking at the last plot, I’m going to try fixed timing. Injection timing 1 is set to 0 (there is a +5 deg cell at tick over, but that’s not required). I am using Injection timing 2 to make adjustments, but apart from the blip at 2000rpm, 100Kpa and the work I have done at 25KPa to eliminate the off throttle lean outs, the table is actually fairly flat: 
I think, if I get all the table to around 13.1 – 13.5, the timing throughout would be the same. Perhaps… What I have found is that increasing Fuel VE table values means I have to bring injection timing 2 further into the injection window, so bring it further into the –ve values. To get the AFR down to 13.1ish at full throttle, I had to increase fuel by 8% and -5 on injection timing 2. For the time being though, I want to get AE sorted because I’m get lean out at pull away and gear change. Rod, I’m using V0.999.7, which seems ok. I’ve got the data rate screwed down at 5 frames per second at the moment. Anything above that and its erratic my AFR plots were. Edited by Graham T on 13th Aug, 2010. ’77 Clubman build thread
|
||||||
|
5988 Posts Member #: 2024 Formally Retired Rural Suffolk |
13th Aug, 2010 at 08:43:46am
On 13th Aug, 2010 Graham T said:
Rod, I’m using V0.999.7, which seems ok. I’ve got the data rate screwed down at 5 frames per second at the moment. Anything above that and its erratic my AFR plots were. V0.999.7 is where my problem started - they don't seem to force upgrades as mandatory anymore but I was fed up with the annoying message so I let it upgrade on my desktop first. No problem, worked fine so I foolishly said yes on my laptop too... It either freezes while loading or if it does load, no comms. If it freezes while loading and I submit the error report to Microsoft their automated message says it's a problem with Java. Re. injector sizes, what pulse widths are you getting at different RPMs - if you look at pulse width as a percentage of an engine cycle (ie, convert RPM to milliseconds) it gives a good idea. Schrödinger's cat - so which one am I ??? |
||||||
 8604 Posts Member #: 573 Formerly Axel Podland |
13th Aug, 2010 at 11:32:40am
On 13th Aug, 2010 Graham T said:
I’m using single fuel VE table and Single timing on mid pulse. Looking at the last plot, I’m going to try fixed timing. Injection timing 1 is set to 0 (there is a +5 deg cell at tick over, but that’s not required). I am using Injection timing 2 to make adjustments, but apart from the blip at 2000rpm, 100Kpa and the work I have done at 25KPa to eliminate the off throttle lean outs, the table is actually fairly flat: So that's using the dual timing tables, not single. I've found that my single table for the hybrid mode is fairly flat, but because I'm using a high duty cycle, I don't think that I could get away with fixed timing. Any use of the term "I" also means "We". Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
|
||||||
|
608 Posts Member #: 1106 Post Whore Hungerford, Berks |
13th Aug, 2010 at 12:43:13pm
On 13th Aug, 2010 Paul S said:
So that's using the dual timing tables, not single. Yes, sorry, I Meant single VE fuel table, dual Injection timing tables ’77 Clubman build thread
|
||||||
|
608 Posts Member #: 1106 Post Whore Hungerford, Berks |
13th Aug, 2010 at 12:50:15pm
On 13th Aug, 2010 Rod S said:
It either freezes while loading or if it does load, no comms. If it freezes while loading and I submit the error report to Microsoft their automated message says it's a problem with Java. I sometimes still have connectivity issues. Load tunerStudio and the ms does not connect. I go into communications, settings and test port. Some times it takes a couple of tests before it connects. Are you running the latest java? ’77 Clubman build thread
|
||||||
|
608 Posts Member #: 1106 Post Whore Hungerford, Berks |
13th Aug, 2010 at 01:04:28pm
On 13th Aug, 2010 Rod S said:
Re. injector sizes, what pulse widths are you getting at different RPMs - if you look at pulse width as a percentage of an engine cycle (ie, convert RPM to milliseconds) it gives a good idea. I'm not sure that these calculations are correct, but they seem to be. 
MegalogViewer reports the duty cycles as half of my calculated duty cycle. But I have found that it is dependent on how you set up squirts per engine cycle and injector staging in the engine constants window ’77 Clubman build thread
|
||||||
|
5988 Posts Member #: 2024 Formally Retired Rural Suffolk |
13th Aug, 2010 at 01:41:22pm
Looking aqt your table above, I agree with MegaLog..... ie, your duty cycle figures should be half what you are calculating.
Schrödinger's cat - so which one am I ??? |
||||||
|
608 Posts Member #: 1106 Post Whore Hungerford, Berks |
13th Aug, 2010 at 08:50:39pm
I'm now happy...
’77 Clubman build thread
|
||||||
|
Forum Mod  1927 Posts Member #: 1761 Stalker Bristol |
13th Aug, 2010 at 09:10:14pm
That's just showing off! Good work Graham, hope to see you on Sunday if you are coming to MITP. |
||||||
|
5988 Posts Member #: 2024 Formally Retired Rural Suffolk |
13th Aug, 2010 at 09:26:48pm
That is absolutely brilliant - and I thought I was finally getting somewhere this evening on just getting AFRs right at tickover on my new setup.....
Schrödinger's cat - so which one am I ??? |
||||||
|
1267 Posts Member #: 831 Post Whore Montreal, Canada |
13th Aug, 2010 at 10:04:14pm
That looks very good.
|
||||||
|
608 Posts Member #: 1106 Post Whore Hungerford, Berks |
14th Aug, 2010 at 08:06:02am
Rod, did you need to change the injection timing much from the setup you had with the previous injectors? ’77 Clubman build thread
|
||||||
|
608 Posts Member #: 1106 Post Whore Hungerford, Berks |
14th Aug, 2010 at 08:44:31am
On 13th Aug, 2010 Carl S said:
That's just showing off! Good work Graham, hope to see you on Sunday if you are coming to MITP. No, I’m not going to make MITP. Now I have the injection fairly well sorted, it’s time to get back to my roadster project. I’m half way through making the fiberglass rear section and it’s been on hold for a while. So before I get stuck into the turbo engine, which will eventually go in the roadster, I need to get a bit closer to having the shell to put it in. 
’77 Clubman build thread
|
||||||
 16540 Posts Member #: 4241 King Gaycharger, butt plug dealer, Sheldon Cooper and a BAC but generally a niceish fella if you dont mind a northerner Rotherham, South Yorkshire |
14th Aug, 2010 at 08:48:03am
So... exactly how many calendars with semi naked/naked girls on them do you have in your garage? On 11th Feb, 2015 robert said:
i tried putting soap on it , and heating it to brown , then slathered my new lube on it 
|
||||||
|
608 Posts Member #: 1106 Post Whore Hungerford, Berks |
14th Aug, 2010 at 09:03:56am
1 Semi naked and 3 fully naked. There's another one just out of shot to the right. ’77 Clubman build thread
|
||||||
|
5988 Posts Member #: 2024 Formally Retired Rural Suffolk |
14th Aug, 2010 at 10:41:44am
On 14th Aug, 2010 Graham T said:
Rod, did you need to change the injection timing much from the setup you had with the previous injectors? Yes, still working on it.... It looks like quite a significant change too but I "think" it's down to the injector characteristics (ie, spray pattern, exit velocity, openning time, etc) being different rather than them being a different size. Schrödinger's cat - so which one am I ??? |
||||||
|
8604 Posts Member #: 573 Formerly Axel Podland |
14th Aug, 2010 at 10:59:03am
What is interesting here is the fact that you have effectively moved the pulses closer together. The inner and outer cylinders work 180 degrees apart, but you have timed the injector pulses arong 140 degrees apart.
Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
|
||||||
|
5988 Posts Member #: 2024 Formally Retired Rural Suffolk |
14th Aug, 2010 at 12:12:30pm
On 14th Aug, 2010 Paul S said:
If you can set the AFR balance by adjusting the timing of Table 2, then you must be getting some of the inner cylinder pulse going over into the outer cylinder. That is my belief too, it's what I did (for all the wrong reasons and before I properly understood it) and had the same effect last year when I got good AFRs for the first time. It's also what you said to me at the time :) But I've been thinking about it recently and if you look at "charge robbing" the opposite way around to normal, ie, consider the outers run weak because they draw in more air (as opposed to drawing in less fuel) then the outers need a larger slug of fuel injected and the only way I can see of doing that with individual pulses it to time the inner pulse slightly late so some carries over to the outer, or get the timings spot on and do the tuning on VE tables instead. Schrödinger's cat - so which one am I ??? |
||||||
|
608 Posts Member #: 1106 Post Whore Hungerford, Berks |
3rd Sep, 2010 at 05:02:58am
On 13th Aug, 2010 jbelanger said:
That looks very good. I'm with Rod on wanting to see that on a 180+ HP turbo and it would be nice to see that on a dyno shoot out. Now what is needed is to get the same results at a few partial load sites to see how much fuel economy can be maximized. Jean 
There is still a bit to do at partial load, but for the most part it runs fairly well. I won't be doing much more testing with this engine because it's now seen better days. The underside is covered in oil and to top it all I pulled one of the Manifold studs clean out of the head on Saturday when I was changing the steering rack (removing the exhaust). So, now I have to change the head and keep it going long enough for my new engine, gt1752 and mirage manifold to arrive.
In the meantime I’m thinking about how to install a second set of injectors on the MPI Manifold. One thought I had was to make some bosses and install them on the outside of the standard injector bosses, and use some BMW motorbike injector holders for clamping and fuel delivery. The question is, if there is enough room to do this, how much effect would the angle of the injector have on setup. The injectors would be pointing slightly inward, towards the inner cylinder. These injectors would be the primary’s, so the smaller injectors for tick over and the larger injectors would be in the standard MPI Injector boss. That’s assuming staged injection is the way to go… ’77 Clubman build thread
|
||||||
|
5988 Posts Member #: 2024 Formally Retired Rural Suffolk |
3rd Sep, 2010 at 07:44:42am
On 3rd Sep, 2010 Graham T said:
The question is, if there is enough room to do this, how much effect would the angle of the injector have on setup. The injectors would be pointing slightly inward, towards the inner cylinder. These injectors would be the primary’s, so the smaller injectors for tick over and the larger injectors would be in the standard MPI Injector boss. That’s assuming staged injection is the way to go… I think the answer is that we don't know yet. I put the same question some while ago when I did a trial manifold with two per runner, one pointing inwards, one outwards. Until/unless someone tries it, the magnitude of the effect (if any) won't be known. You've obviously seen Paul's solution, put them all parallel, but that requires pencil thin injectors and custom made manifold/fuel rail parts. My thoughts were that, if there is an effect, it could be compensated for by altering the timings and/or VE tables but that would limit you to switching the staging on RPM alone which I don't think is best with the multiple options available. Schrödinger's cat - so which one am I ??? |
||||||
| Home > A-Series EFI / Injection > Tuning MegaSquirt for 998 NA | |||||||
|
|||||||
| Page: |