Need a little help to set this car up.
Car Porsche 944 2.5 N/A 1984, 4 cyl
Firing order = 1342
using only 1 coil + distributer
16/11-2010
Config and logs
16/11-2010
We are sweeping the secondary trigger position
- if changes multiple tooth we can see a trigger error and then goes everything good
- if the changing is little no problem detected
15/11-2010
Testing of today
Loaded your tested config and nothing changed still can´t fire group 1 and 2
if I set it up to 130 teeth and no divider results is the same still can´t fire group 1 and 2
I wonder haw [h2] can be A7 A7 A7 27 87 ff D6 ff
I mounted leds on all outputs for better overview and when I use group 0 and 4 all light´s up as expected
maybe it could be smarter to hack 1.1.80/82 to avoid all the new settings/functions.
Hope Hq will find the error when bench testing today and I will give et another try.
Thank you
Peter
Corrected config added: http://www.vems.hu/files/MembersPage/PeterJensen/porsche_944/v3.3_n001212-A-2010.11.16-13.20.20.vemscfg
The next VemsTune will check these problem.
We tested 1.1.85 with auditrigger divby5 and c130. works like a charm.
- 1.1.85 (not yet released), November 15
- re-download if you have older version (old one might have incompatible real-time comm)
- all 4 injectors and 1 ignition (as configured) works
- we used [this config]
- note that we disabled config-switching. Always disable, and only enable when engine runs properly.
- we used Tools/Play Trigger, and we manually changed signal pattern to c130 before pressing play.
- note that trigger play shows twice RPM compare to ECU, because it thinks the 130 tooth is 1 engine camshaft rev, while in this case it's 1 crankshaft revolution. The ECU shows correct RPM.
Trigger
Assuming 130 crank tooth and one crank ref tooth (Auditrigger) see trigger log
No camhall masking is used. Since there is no cam-HALL inverter in the box, EC36/13 camHALL input can be left unconnected (>3V).
12/11-2010
Result of today's testing.
Loaded firmw 1.1.85
changed toothrel_missing=04 manually in the config file (did not find tag in VT)
- perfect !
div by 2
number of teeth on wheel 26
first trigger tooth 1
next trigger tooth 13
width of tooth 13.75
- The car did not fire and somehow I lost real-time data in Vt (Still can changes parameters).
- unfortunately - there was a misunderstanding, sorry - the new real-time data format (that can be interleaved with triggerlog). It was not yet fully implemented in the unreleased 1.1.85 on Friday
- Now we added command 126 to send real-time data in triggerlog.framing format. Changed back so 'D' sends in old format (so megatune compatibility is still possible). Should be fixed in the (still unreleased) 1.1.85 now - we're also testing it. I don't know if it cooperates with Friday nightly VT, but we upload new VT nightly soon anyway
I due not hope it´s the “I button immobilizer” that's cursing these problems. => no, just don't enable I-button immobilizer. Use the default mode: "prohibit according to firmware version" (this SafetyMode was very important to prevent more flooding when not following the procedure).
BAD config: http://www.vems.hu/files/MembersPage/PeterJensen/porsche_944/1185/v3.3_n003352-2010.11.12-16.48.05.vemscfg
(eg, configswitch should never be enabled during setup).
Triggerlog: http://www.vems.hu/files/MembersPage/PeterJensen/porsche_944/1185/v3.3_n003352-2010-11-12-16.14.46.zip
Tank you
Peter
- If so, than each trigger period is one crankrev, 130 primtrig pulse.
- c130 trigger pattern ("c130" can be entered in play trigger dialog, right ? Enter before start "play")
- tooth_wheel=41 ? 130 divby2 is 65 might be good after all
- but hey... 65 tooth is NOT divisible by 2.
- so another_trigger_tooth=21 (33, vs 65/2) is rounded up, so needs some trick
- Emil had a brilliant idea: another_trigger_tooth=21 (decimal 33) - rounding up seems better
- definitely needs benchtest (or strobing at the very minimum), but igntrim could easily adjust cyl2,3 that should be timed by the tooth neglected because of divby2 ( 2.75 deg igntrim is perfect).
- But even than, with this setup RPM reading will fluctuate +-1.5% (not a big problem, but would hide individual power and even some more important information)
- you can only delay with igntrim (not advance). But you can delay the OTHER cyls, eg. cyl1,4 if you like
- alternative2: Maybe we have to try with normal coiltype trigger ? (not auditrigger-divby2). This might limit max RPM.
divby5 allowed in 1.1.85: divby2 setting now means divbyANY mode
See 1.1.85 in GenBoard/UnderDevelopment/FirmwareChanges (a testing 1.1.85 into cell.dyndns.ws/f/ )
- toothrel_missing=04
- together with divby2 (will be renamed to divbyANY) this means divby5 (reducing 130 to 26 tooth)
- another_trigger_tooth=0d (13 decimal)
- reftooth should be 0,13,0,13 withOUT camHALL
Benchtest is needed. The 135 tooth worked up to 11k in earlier tests with same trick (a higher divider).
Symptoms
The car starts and run but it will not fire injector group 1 and 2 (group 0 and 4 is ok) so we used 0 and 4(see config)
You mean 0 and 3 injector groups are fired ??
Trigger problems first show up as trigger errors, than ignition missing, than injection missing. Ignition OK but injection missing is weird, usually colliding outputs or similar (use Validate !).
All outputs tested ok and injectors also firing when ignition go high (primepulse).
Possibly just a error is settings but I can´t simulate this type of trigger so need help.
Config: http://www.vems.hu/files/MembersPage/PeterJensen/porsche_944/porsche.vemscfg
Trigger log: http://www.vems.hu/files/MembersPage/PeterJensen/porsche_944/v3.3_n003352-2010-11-10-14.52.53_trigger_log.zip
We start reviewing now, and we can give it a full benchtest on Tuesday. We could have done this week had this page existed when we were asking at the time of processing the order. (we spent a lot of time with this at the time of the order, mostly wasted because this page didn't exist than. Jorgen gave a lot of help, at least we could find out what HW is suitable: ECU was equipped with auditrigger but withOUT cam-HALL inverter).
- I am sorry to anyone that have wasted time, this is not my intention. I did not have the car at the moment I ordered and no other info ether. Stage3.dk called if I could help them make the car this week because they are busy installing Vems to other project cars. I did say yes because we due not like to louse the order.