## ## ##### ## ## ## ## ###
______ ( ___ \ | ( ) ) | (__/ / | __ ( | ( \ \ | )___) ) |/ \___/
_______ ( ____ \ | ( \/ | (__ | __) | ( | ) |/
## ##### ## ## ##### ## ##### ## ## #####
IMPORTANT: enter the case-INsensitive alphabetic (no numbers) code AND WRITE SOME SHORT summary of changes (below) if you are saving changes. (not required for previewing changes). Wiki-spamming is not tolerated, will be removed, so it does NOT even show up in history. Spammers go away now. Visit Preferences to set your user name Summary of change: Porsche 911 Turbo '''6 cylinder engine''', engine unknown but had the Motronic from the 3.2 engine running before vems install. The engine is similar to MembersPage/MacB ---- '''Update: the car is running on 1.1.98 fw. Good job guys!''' For the c129 signal, '''divby43, 1.1.98 (17-Jan-2012 or newer) firmware needed''' from [http://cell.dyndns.ws/f/ developer testserver] * See the http://cell.dyndns.ws/f/README.txt how to install this * (1.1.98 not released to the public yet, but actually seems better than anything before). * With the 129 tooth, we tested beyond 10000 RPM (on the bench, including ignition and injection outputs). * The change we did was basically a revert of change that was made to a certain request, but not extensively tested yet ** basically auditrigger behaviour should be now as good as back in 1.1.95 for non-3 divider also ** but with the desired changes - eg. allowing all 3 trigger signals to be connected and scoped independently with v3. ---- * '''injector outputs: only injgroup 0, 4 and 5''' were used (this might seem unconventional, it's all because the sectrig is not 720 deg but 360 deg), ** so we enabled '''2 inj outputs''' (we paired the outputs that are actually 360 deg opposite) in injgroups 0,4 and 5 (and '''left injgroup 1,2,3 empty'''). *** '''Why not use injgroups 0,1,2, and also set "Number of primary injectors" to "3" instead of "6"?''' ****'''also, should "two stroke" or "four stroke" be used in this setting?(latest config is set to two stroke)''' ---- Trigger setup: Flywheel tooth and flywheel pin triggering, flywheel is 129+1. ---- ECU Info: ECU#4215 vr/vr trigger(auditrigger non camhall) Flywheel Tooth Trigger to EC36-27 Flywheel Pin Trigger to EC18-12 ---- Here's link to 911 Workshop Manual, years 1984-1989: http://www.cannell.co.uk/911_72-89_Workshop_Manuals/84-89%20Workshop%20Manual.pdf Maybe there's some info on 911 Turbo as well.(MembersPage/PeepPaadam) ---- * '''[http://cell.dyndns.ws/f/v3_firmware_1.1.98.zip 1.1.98 (experimental)] was tested, and found good with this divby 43 config''' ** http://www.vems.hu/files/KevinBlack/porsche-1.1.98-divby43.zip with config, vemslog, triggerlog ** it seemed good on the bench beyond 10k RPM, but we're retesting it extensively, also analyzing ignition events (count and position), also sweeping sectrig and special care around 3000-5000 RPM ** Here is a [http://www.vems.hu/files/KevinBlack/Porsche_c129/kevin-500khz-rpm-up.csv.zip bitscope] file (sectrig, fixed "TDC-position" signal just for scope, 2 ign and 2 inj outputs) 8 bits packed in 1 byte, every 2 usec. (revlimit was 9900 RPM so above that inj and ign would be off). ** '''WHILE we are continuing analyzing behavior, we are more and more confident that misfire below 6500 RPM would be caused by input trigger, likely crankhome signal polarity, noise or something similar'''. Extensive onsite inspection needed, triggerlog captured at 115100 and scope will likely be needed. ---- ***'''115k Triggerlog is here, scope screen is below''' http://www.vems.hu/files/KevinBlack/3000%20OK,%204000%20Trigger%20Error%20115k%20Baud%20v3.3_u004215-2012-03-13-10.55.15.triggerlog ---- ** note: anything after "auditrigger with third trigger input software "camHALL scope" (2011-08-02) will NOT work, so '''1.1.96 and 1.1.97 will NOT work''' (at all with this trigger, see GenBoard/UnderDevelopment/FirmwareChanges ). 1.1.95 should work (we didn't tested it) * Trigger Edge: Rising * Trigger Type: Coil Type * Special Trigger Type: Auditrigger DivbyN * Audi trigger '''divider: 43''' * Number Of Teeth On Wheel: 3 (NOT 6, because only 3 per sectrig. 3*120=360 as sectrig pulse comes every 360 deg on this engine) * First Trigger Tooth: 0 * Next Trigger Tooth: 1 * Angular Width Of Tooth: 120 * Filtering and below: Disabled * reftooth: 0 2 1 0 2 1 x x (because teeth 3,4,5 does NOT exist) Secondary Trigger: * Mode: Cam Sync * Edge: Rising * Edge While Cranking: Single * Use: Cam Sync '''are we certain that sectrig VR is not accidentally inverted ?''' * scopeshot suggests sectrig is OK * primtrig pulses == 129 for sure ? '''Picture of Trigger on scope:''' http://www.vems.hu/files/KevinBlack/Flywheel%20and%20Dowel%20Pin.JPG The primtrig has a lower-frequency component (which can cause runout at higher RPM and could explain the symptoms) of unknown source. * was the scope powered from 60 Hz or battery ? ** '''Scope was powered by wall charger, battery is not working.''' * any idea what could be the source ? * '''does the pattern repeat with multiple engine rotations?''' ** if yes, it might be mechanical abnormality of the steel wheel. Is the sensor mounted axially or radially ? *** '''Yes, the up-down-up primary pattern repeats.''' * it seems that this can/should be cured in HW. ** eg. more regular wheel ** or tuning the '''VR signal conditioning (higher R181 ? Maybe 270k ?)''' '''I was able to change the R181 to 235k, with no real improvement(still trigger errors beginning at ~3,500 rpm). I also applied a 1k to ground on the sec trigger (per email) with no noticeable change in results. The sec trig scope trace shows a slightly better dip below 0v now.''' '''new scope trace:''' http://www.vems.hu/files/KevinBlack/photo4.JPG ---- We also tested with no divider (tooth width=2.75 deg): it worked, but peak-RPM would likely be limited. '''This special case of auditrigger worked on bench''', including the per-cyl delay, but '''only after tricking''' "divider-off, than "divider-on" (so we're working on initialization change, needed so it syncronizes without this trick, currently stuck in tooth_cnt=FF and wouldn't spark without this hack). Per-cyl ignition delay is needed (because 43/3 != 14). Note: RPM reading might fluctuate, so divby43 is better choice * divby3 * toooth width=8.5 * Number Of Teeth On Wheel: 43 * First Trigger Tooth: 0 * Next Trigger Tooth: 14 * Angular Width Of Tooth: 8.5 * Filtering and below: Disabled * reftooth: 0 28 14 0 28 14 x x (because teeth 3,4,5 does NOT exist) * '''per-cyl ign delay: sg. like 0 5.5 2.75 0 5.5 2.75 0 0''' degrees ** needs verification with strobelight + adjustment if necessary * '''injector outputs: only injgroup 0, 4 and 5''' were used (this might seem unconventional, it's all because the sectrig is not 720 deg but 360 deg), ** so we enabled '''2 inj outputs''' (we paired the outputs that are actually 360 deg opposite) in injgroups 0,4 and 5 (and '''left injgroup 1,2,3 empty'''). ***'''see question at the top of page''' Optional: Add document to category: Wiki formatting: * is Bullet list ** Bullet list subentry ... '''Bold''', ---- is horizontal ruler, <code> preformatted text... </code> See wiki editing HELP for tables and other formatting tips and tricks.