## ##### ## ## ##### ## ##### ## ## #####
## ## ##### ## ## ## ## ## ## #####
## ## #### ## #### ## ##
# # # # # # # # ##
## ### ## ## ## ## ## ## ####
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: '''Nissan CAS trigger (360 pulses)''' '''Old solution was a 24+1 slot trigger wheel''' as described here: VehicleFitment/Nissan ... plenty of cars are running on these trigger disks. Still a good option, especially above 1.6 million pulses per minute (above 9000 RPM). ---- '''Leaving the 360 impulse disk''' - implementation works well * See "Nissan special settings" in [http://vems.hu/vt/help/v3/v3_secondary_trigger_settings.html VemsTune Help] on how to measure the length of the longest primary trigger window (if you don't know, or want to verify) We'll adjust the TDC-delay (or is it close to 60 crankdegrees which makes most sense ?) in the userfriendly "1-click" triggersetup configuration (bottom of primary trigger dialog) * after someone writes exact pattern info '''here''' (in place of the "..." below, '''relative to TDC''') - from a reliable source or measurement. Fill in this form (and correct if necessary): signal is normally 0V (according to patent FIG.3 at least), but 5V in the small slots: * ... - ... deg BTDC cyl1 (high for 24 deg) * ... - ... deg BTDC cyl3 (high for 16 deg) * ... - ... deg BTDC cyl4 (high for 8 deg) * ... - ... deg BTDC cyl2 (high for 16 deg ???) Is this correct or 32 deg instead (=16 pulses) ** the picoscopeshot says windows are 16,12,8,4 smallpulse (=32,24,16,8 crankdeg) with the rising edge (start of the window) is evenly placed every 180 deg. TDC info unknown. In any case, the order seems decreasing window (HIGH) length. There is also the 6-cyl wheel: * ... - ... deg BTDC cyl1 (high for .. deg) * ... - ... deg BTDC cyl5 (high for .. deg) * ... - ... deg BTDC cyl3 (high for .. deg) * ... - ... deg BTDC cyl6 (high for .. deg) * ... - ... deg BTDC cyl2 (high for .. deg) * ... - ... deg BTDC cyl4 (high for .. deg) Best I can do right now (Gunni) Download this http://downloads.picotech.com/winxp/PicoScope6Automotive_r6_5_78.exe Then open this file http://www.megaupload.com/?d=SJVJBNIK It´s a SR20DET 1997 engine captured trigger input Should be easily viewable in the Picoscope program. The low (4/6) pulsecount connected to primary trigger, the 360 tooth divided by 4 (or 8?) connected to sectrig is being implemented. The exact positions "relative to TDC" (!) would still be needed (firmware+configlet will be released when that is written above). ---- '''Geek notes''' A hardware divby4 divider can help. We have divby4 / 8 implemented with HCF4024 circuit (under PCB manufacturing, originally for high toothcount ABS wheels). Since this circuit has (active-high) reset input, this can be useful to make a precise 180, 90 or 45 pulsetrain from the 360 tooth. * '''Please help to find this out so our stimulated pattern is exact, and we can recommend good ignition trimming values if necessary !''' - Gunni : Please find the patent information for the nissan trigger here (http://www.freepatentsonline.com/RE34257.pdf) It explains the syncing methods used by nissan, hope it helps. ** because the sync pulse is mostly 0 (and high for only a short time), the active-high reset input of the HCF4024 seems suitable ** is the sync pulse falling edge not racing with (the rising or falling edge of) one of the 360 pulses ? (I assume not, because that would be very stupid. This would make a counter-type implementation impossible, but a helper "assistant CPU" could still help) ** configuring sectrig input=both edges and '''firmware modification is necessary''' with the 4-window sync pulsetrain ** by blocking 3 of the 4 windows, (and using a HW divider) firmware mod would not be needed (good for experiments, but this should not be needed finally) * TODO: actually try the resettable divby4 counter (on bench, and on engine) ** with divby4 (360 => 90 pulses).firmware could (should) be configured to divby3 or even divby15 (via divby-ANY setting) ** divby8 (360 => 45 pulses) for 4 cyl engines requires a small trick of 4 crankdegrees ignition trimming (because delaying by 23 teeth is not exactly the same as the required 22.5 teeth) ** according to MembersPage/JannoKusman/NissanTrigger both '''sectrig edges must be sensed because of the 4 slot syncpulse''' Besides any necessary HW, suitable config will be provided (configlet to be able to upload to suitable firmware with 1 click). ---- Alternative solution would be (without HW divider) a '''software divide by the main processor'''. The processor is definitely capable (up to at least 17000 RPM). But unfortunately we have some tasks in interrupt that would need to be rewritten. (not the trigger input capture interrupt, but other interrupts that could interfere). From auditrigger bench experiments that worked at 11000 RPM (pulse=2.66 crankdeg), we know that 8300 RPM (pulse=2 crankdeg) is possible without major rewrite. Going above the "all pulse always sensed" threshold could delay ignition by 1 tooth (2 crankdeg) or more (not advance that could blow engine). We speculated that without HW divider going above 9300 RPM '''could be too much work that would delay other developments unacceptably'''. So we will only implement fw mod that works well with HW divider. * we think >9300 RPM is minimum reasonable goal to say that some trigger is supported (even though much lower than the 11k RPM of 135 tooth auditrigger with sufficient divbyANY config or 25k RPM of lower toothcount wheels) ---- '''See also''' * guys interested in this ** MembersPage/FredrikJagenheim ** MembersPage/JannoKusman ** MembersPage/FredrikBruland ** MembersPage/MattiasSandgren/MattisNissan 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.