_  __
 | |/ /
 | ' / 
 | . \ 
 |_|\_\
       

 _   _ 
| | | |
| | | |
| | | |
\ \_/ /
 \___/ 
       
       

  ____  
 / ___| 
 \___ \ 
  ___) |
 |____/ 
        

         
         
 # ###   
 ##   #  
 #       
 #       
 #       
         

         
  ####   
 ##  ##  
 ######  
 ##      
 ##      
  ####   
         

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:
This Mitsubishi has a HALL trigger on the camsync. unfortunately * resolution is trash (due to cambelt sloppyness) * period is same as if it was on crank (less information than normal cam signal: not enough for fully sequential, but enough for wasted spark) The trigger is Coil type (one pulse per event ), but alternating positive and negative to distinguish 1-4 and 2-3 cylinders (are you sure?). Somewhere we apparently have a '''a series capacitor''' (input highpass filter !!! ) that causes the very characteristic sharktooth signal. * maybe in the notebook: I recorded the trigger signal with a notebook. Almost sure that this is the case since the signal measured (with DVM) DC=2.5V amplitude: ~2.5V which suggests a normal 0/5V signal. Once measured with engine off it was 5V. http://www.bormodell.hu/IgnitionTrigger.png The signal is currently processed with LM1815 and series condenser so it is theoretically universal considering trigger type (hall and VR). In 50% of cases, LCD displays right value, otherwise rough half of proper RPM. if we reconstruct the original signal, we can see that * pulse for cyl1,4 has a longer width * pulse for cyl2,3 has a shorter width (or the other way around) Note that * falling edges are equidistant ** good for triggering ** store timestamp in ptrig_capturedtime_prev * rising edges are not equidistant (only useful for checking cyl1,4 vs cyl2,3 and updating igncount). pulsewidth=time since ptrig_capturedtime_prev (last falling edge) ** if this pulsewidth is (significantly, or with any amount) longer/shorter than previous pulsewidth than igncount=ignchmax Find the correct cyl: * camsync * using bothedge ( LM1815 not suitable) and consider signal length: firmware modification ** modify primary_trigger_init() so IC1 initialized (similar to IC3 in secondary_trigger_init() ) that both edge triggers. Maybe setting just one edge and changing to falling/rising in interrupt is necessary ** detect edge in PRIM..TRIGGER interrupt ** Any ideas, how to condition the signal? * remove the 10k pullup on trig1 input inside v3.x (if not already removed) ---- '''Setup is not VR anymore, but the symptoms are the same''' I converted the TRIG1 input to hall setup and removed 10k pullup as well, but I don't have stable RPM signal yet as I described above. On the other hand I found an other trigger. ---- '''An other trigger found''' I've found an other trigger probably from crank-shaft because 3 wires are coming from the bottom of engine (I will check it soon). Wires: 0V, 12V, signal (0V or 5V). I don't understand why 12V and why not 5V is on feed. Signal levels are almost the same than in the previous case. DC=2-2.5V, AC=2-2.5V, same frequency (measured by DVM). Since I use this trigger I have stable RPM signal (coil type setup in config). It is good for setting up injectors, but not enough for ignition. Observation: AC and DC voltages are measurable between the trigger described here and above. * Does this facilitate to get the trigger needed for wasted spark? * Is it worth to connect both triggers to primary and secondary inputs, and try to determine which cylinder to ignite with alien which is already implemented in the firmware? ---- Check bottom of DocsPage to upload files to wiki.
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.