oversteer Posted January 21, 2020 Report Share Posted January 21, 2020 Have installed the larger injectors and switched to E85 ready for power tune, and are having a bit of trouble getting less fuel at idle and light cruise..... Running Siemens Deka 220lb with 550cc per cylinder on a 6cyl Subaru, switched the tune over that was running 630 Siemens Deka's with pump 98 which was tuned fairly good to run in engine, changing master fuel number from 10ms to 3ms and injector dead time numbers the car is .7 lambda at idle or cruise, give it a bit of load and rpm and its in the .9's so it drives ll right. Changing fuel map or adding -ve fuel trim does nothing to reduce inj pw any further.... Have also changed the "min effective pulsewidth" number to 0, if its set to 1.5ms the actual pulsewidth is around 3ms minimum....which is lots of corn juice out the pipes ! Is there something I'm missing here ? On the face of it I would think this may normal, big sloppy injectors on 550cc a cyl and can't get idle lean enough, however I have another subaru here, 4cyl 2.35 ~585cc a cylinder running slightly less compression(8.1 vs 8.3 for 6cyl) and a M800 that will idle in mid .9 lambda, same fuel pressure. Tune and short log attached... e85 220d s.llg Quote Link to comment Share on other sites More sharing options...
Adamw Posted January 21, 2020 Report Share Posted January 21, 2020 I suspect deadtimes as wrong. I have used similar deadtimes to you with those injectors before without drama but it was with methanol so it wouldnt have had as much influence. After a bit of a search tonight I would try deadtimes more like this: More importantly, your log shows lots of trigger errors, it would not be a good idea to give the engine a hard time like this. Can you attach a triggerscope done at idle. Quote Link to comment Share on other sites More sharing options...
oversteer Posted January 21, 2020 Author Report Share Posted January 21, 2020 The dead times I'm using were detailed by you here -> They are the same as detailed elsewhere for those injectors, yellow bullet iirc... should I go with the ones you suggest here ? Trigger errors in log i suspect is because the way it idles rich on E85, its pretty rough, last time on dyno on 98 it did throw a couple of trigger errors so levels were adjusted and it was better... I may need to move sensor air gap a bit, I did set them up at .5mm, which depends who you talk too may be a little close.... any suggestions on what gap to use on a subaru mag sensor ? Quote Link to comment Share on other sites More sharing options...
Adamw Posted January 21, 2020 Report Share Posted January 21, 2020 22 minutes ago, oversteer said: They are the same as detailed elsewhere for those injectors, yellow bullet iirc... should I go with the ones you suggest here ? Dont believe everything you read on the internet... Yeah, try the new table I suggested. 24 minutes ago, oversteer said: Trigger errors in log i suspect is because the way it idles rich on E85, its pretty rough, last time on dyno on 98 it did throw a couple of trigger errors so levels were adjusted and it was better... I may need to move sensor air gap a bit, I did set them up at .5mm, which depends who you talk too may be a little close.... any suggestions on what gap to use on a subaru mag sensor ? A lumpy idle should not cause a trigger error. Do the triggerscope first before changing anything, this should show what is wrong. Quote Link to comment Share on other sites More sharing options...
oversteer Posted January 21, 2020 Author Report Share Posted January 21, 2020 OK will do trigg scope when i can. Just looked at log, you can see the error counter only increases at idle, at any sort of load and rpm no errors get counted.... Quote Link to comment Share on other sites More sharing options...
oversteer Posted January 21, 2020 Author Report Share Posted January 21, 2020 trigger scope without any changes since last log(although i did a little touch up to ethrottle and ing to make it run better while i was driving!) After that i changed the dead time and it made world of difference....idles better than it ever as ! ... trying to upload 74K log file, says only allowed 20Kb....I assume its adding up previously uploaded files, so deleted one above, still says 20Kb ! Will convert trigger scope log file to picture ... trigger scope after changing deadtimes with a nice idle, had to add a lot more fuel back in, 4ms setting and 10+ to fuel table. Quote Link to comment Share on other sites More sharing options...
oversteer Posted January 24, 2020 Author Report Share Posted January 24, 2020 Any comments of the trigger scope ? Can better trigger levels be worked out from log ? Quote Link to comment Share on other sites More sharing options...
Adamw Posted January 24, 2020 Report Share Posted January 24, 2020 Can you share the actual trigger scope file. Use dropbox, googledrive, onedrive or something like wetransfer Quote Link to comment Share on other sites More sharing options...
oversteer Posted January 25, 2020 Author Report Share Posted January 25, 2020 https://drive.google.com/open?id=1NwGuFPcbtIPSnDFMVbRRbMY0Hgye5Q2N Should trigger scope, tune file and a log file with some data... Quote Link to comment Share on other sites More sharing options...
Ducie54 Posted January 25, 2020 Report Share Posted January 25, 2020 These are low resistor injectors do u still need a resistor box? Quote Link to comment Share on other sites More sharing options...
oversteer Posted January 25, 2020 Author Report Share Posted January 25, 2020 Not with a Thunder, it can to Hi and Low. Quote Link to comment Share on other sites More sharing options...
Adamw Posted January 25, 2020 Report Share Posted January 25, 2020 Ok, the problem with your trigger is shown below. Notice the cam tooth crosses zero volts at exactly the same time as a crank tooth crosses zero (where my yellow cursor is). This is very bad, it means as the cam tooth bounces either side of that crank edge due to cam belt slap etc. your ignition timing will suddenly shift by 30 degrees. So you could potentially have 30 deg more advance at random times which probably wont end well. What you have to do is move either the crank wheel or cam tooth so that the cam edge crosses zero roughly halfway between two crank edges. Thanks to my photoshop mastery, pic below is roughly how you want it to look; notice the cam falling edge zero crossing (purple dotted line) is roughly centered between two crank falling edge zero crossings (yellow dotted lines). And the official help file explanation which shows the aim is to be anywhere within that 50% zone between teeth: Quote Link to comment Share on other sites More sharing options...
oversteer Posted January 25, 2020 Author Report Share Posted January 25, 2020 That's not easy to do to this motor, only way i can think of right now is to hope there is enough adjustment on cam wheel degree slots and maybe move the cam belt over a tooth......any way to tell how many degree's i need to move cam sync ? Pretty much impossible to degree cams with engine in car.... its drop engine trans and cross member to get it out, hmmm. further.. This engine actually has 2 tone rings on crank pulley, one is 12 tooth and the other is 1 2 3, an called "sync" in FSM, does the cam sync have to come from the cam wheel ie half crank speed ? any way to use a single tooth on crank 2nd tone ring to sync ? Would have to be a better signal with not cam/belt issues.. Quote Link to comment Share on other sites More sharing options...
Adamw Posted January 25, 2020 Report Share Posted January 25, 2020 The teeth on the crank are 30deg apart so to center the cam tooth between them you would need to move the cam tooth 15 crank degrees or 7 cam degrees. Sync needs to be on the cam for sequential. You will notice the cam tooth is roughly centered in the FSM pic so something has changed on yours, if it has adjustable cam pulleys that will be plenty to fix it, it doesnt matter which way you move it (you will need to check trigger offset afterwards as ign timing may change depending on which way you go). Quote Link to comment Share on other sites More sharing options...
oversteer Posted January 25, 2020 Author Report Share Posted January 25, 2020 OK thanks, it would be the cam grinder that didn't base time the cams very well.... seen it before with his cams, always trouble to set-up ! So 7 degree's on cam will get me dead in the middle of zone, so maybe if i modify the cam sensor mount I can move it 2-3 degree's and it should be enough !? Looking at trigger scope the best direction to go to get away from crank trigger is to fire the cam sync later ? So roll it around clockwise from current position around 4mm from a quick scribble on a piece of paper, something that can be done in car ! Quote Link to comment Share on other sites More sharing options...
oversteer Posted January 26, 2020 Author Report Share Posted January 26, 2020 OK cut and re-welded cam sensor bracket, moved sensor ~5.5mm anti clockwise, result looks pretty good on scope... good enough Adam ? Moved scope ; Original Scope ; Trigger scope log in google drive link folder above. Quote Link to comment Share on other sites More sharing options...
Adamw Posted January 26, 2020 Report Share Posted January 26, 2020 yep looks good to me. Check your trigger offset is still ok and you will be winning. Quote Link to comment Share on other sites More sharing options...
oversteer Posted January 26, 2020 Author Report Share Posted January 26, 2020 Plug lead break-out thing to get timing light to work with COP is at dyno room, so I can't check that right now.... unless you know of some trick to get timing light to pick up on cop with out putting lead between coil and plug !? Car seems to run the same after moving cam sensor.... the link does use the crank 12 teeth for actual timing events does it not ? So it must still be using the same ref otherwise it would be 30 degree's out ? Quote Link to comment Share on other sites More sharing options...
Adamw Posted January 26, 2020 Report Share Posted January 26, 2020 33 minutes ago, oversteer said: unless you know of some trick to get timing light to pick up on cop with out putting lead between coil and plug !? They will often work if you just hold the pickup on top of the coil. 34 minutes ago, oversteer said: the link does use the crank 12 teeth for actual timing events does it not ? So it must still be using the same ref otherwise it would be 30 degree's out ? Yes, the crank teeth are used for timing, but the ecu still needs the cam tooth to know which tooth on the crank to start counting the offset from. So by shifting the cam tooth you could potentially be 30 deg out. I would expect it would barely run if it was 30deg out so likely it is fine but still needs to be checked. Quote Link to comment Share on other sites More sharing options...
Ducie54 Posted January 27, 2020 Report Share Posted January 27, 2020 These are the dead time figures i found for 43.5 Psi. Any reason why you have your base fuel pressure at 50 PSI? 8v---- 1.52 10v--- 1ms 12v----0.71ms 14v----0.55ms 16v---0.43ms Quote Link to comment Share on other sites More sharing options...
oversteer Posted January 27, 2020 Author Report Share Posted January 27, 2020 Ok they are quiet different than what Adam listed above, I take it that 8v should be 1.52ms ? Where did you find those figures ? Always run higher fp with large injectors it can help produce a better atomisation at small openings and also for higher capacity, calcs say I need maybe 2600cc to for hp goal ! Quote Link to comment Share on other sites More sharing options...
Ducie54 Posted January 27, 2020 Report Share Posted January 27, 2020 Yep 1.52 sorry. Those figures dont take into account your wiring or LINKS ECU injector setup. https://www.golebysparts.com.au/products/siemens-deka-2400cc-injectors Link has a good injector test function. So worst case you can work it out that way. Quote Link to comment Share on other sites More sharing options...
oversteer Posted January 27, 2020 Author Report Share Posted January 27, 2020 Sooo many different figures for those injectors !...wildly different figures, at this point I'm inclined to believe Five-O motorsport snippet info which is .640 @ 14v... Will have to look into this test function of which you speak. Be real nice if link would get one and characterise it with G4+ so it can be settled once and for all, Link tech support pllease !??? Quote Link to comment Share on other sites More sharing options...
Ducie54 Posted January 27, 2020 Report Share Posted January 27, 2020 If u send me one I can do it on my test rig. Quote Link to comment Share on other sites More sharing options...
oversteer Posted January 27, 2020 Author Report Share Posted January 27, 2020 Where are you located ? Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.