Members Page/Skassa Trigger Subaru - VEMS wiki www.vems.hu

Strobe shows different angle change than what we expect

This section seem to be VERY old comment, also without config (so it cannot be reproduced => useless, even confusing without firmware + config info). It was surely 1.0.x but apparently it might just have been measurement problem the smart strobe mislead (not config problem).

In '''recent, like 1.1.73 we tested this (using the configlet from VemsTune, with tooth_width=180 degrees) and the ignadv on scope changed exactly 20 degrees when we commanded 20 degree change. See InputTrigger/SubaruTrigger for more info on this trigger.

with (InputTrigger/ShortGapTrigger) InputTrigger/SubaruTrigger :

I downloaded the config from one subaru (where is the config ? It's unclear if that was experienced with 1.0.x or with an 1.1.x < 1.1.72 before subarutrigger support was added)
uploaded to another
ign advance is off, just like it was also off on the other Subaru
- just to clarify: the ignadv is displayed OK in megatune, (and on LCD mlp00 bottom right corner), but the strobe shows the ignadv changes half angle of what it should.
Yep, what is written in megatune sparktable, is what is showed on display,and megatune.

It is the strobe that shows half of what´s written

can it be that the advance is calculated from the cam ,and not the cas?? so it´s every time the 3tooth is passing??

I was tuning the car last night, and the advance seems to be off:

strobed at 10degrees, so this is right
if I write 40 degrees in the megatune sparktable

it changes only +15 crankdegree (to 25 BTDC), not +30 crankdegree (to 40 deg BTDC) as it should

measured on the crank, of course (not on cam)

some smart strobes are confused by the wasted spark ignition, this actually happens quite often.

Can you do measurements the 40deg and 10 deg with same strobe settings (or a dumb strobe, and see angle difference on engine ?)

40deg is 25deg measured, and 10deg is 10deg measured

Can you describe the strobe and its usage in great detail ?

type
switches
what knobs
display
...
- the timinggun is about the oldest that can be had, there is no buttons on it, i will check if i can see a name on it, but i think it´s around 20-25years old.

there is just 2wires for power and gound, and a 3rd with the inductive feeler to put on the ignitionwire.

if it lights up at the time of spark (seems likely if it has no buttons) and you read from the rotating crankshaft, than the chance of getting it measured wrong is minimal indeed.

Want to test with Sami's 1.0.77 compile that Jorgen drove for quite a while now and found good. Anyone have that firmware image ?

yes i will give it a try, anything to make it work:)
- I have now mounted the 36-1 wheel (pic at the bottum) and the car starts perfect now,

So how can it be a noise issue, nothing else changed ,but the crankwheel.

but the advance problem is still present, so i must be some thing else that have changed it, i will try to uupload one of the old files i have saved from the beginning, and see if that changes it,and upload the 1.0.77 firmware.

This way earliest is 37.5 crankdegrees BTDC (max commanded ignadv is 63.75 degrees.

This is from the helpful [VEMS FORUM]

There are trigger errors when cranking, though the wiring seems OK

and i cant seem to find any info, regarding the "advanced filtering" settings, if that maybe can help filtering that out
- GenBoard/UnderDevelopment/FirmwareChanges, (toothrel_normal and toothrel_missing ) multitooth advanced filtering so thresholds can be set for what is considered as long-gap (eg. 125-420%, in actuality 60/30 and 90/60), and what is short (eg. 16-100%, in actuality 30/90 ).
car will sometimes backfire when cranking
on the MS site, i asked a guy who run there subaru trigger,he had the same cranking problem and that was fixed with software,as he is able to see in megatune,that it was caused

by the ecu going in and out sync.

Could this maybe also be the problem here??

the "crank minimum usec" shouldn't cause that (you can set it to 370 usec, that is 20 degrees at 9000RPM, or you can even try 0). If it's too high, high-RPM triggerinput will not be believed

We yesterday tried to see if we can fix the trigger problem, when cranking, as it´s almost impossiple to start the car(it´s firing totally random,but it will suddenly , after 5min with cranking burst into life, like there is nothing wrong.

We tryed to scoop the trigger input a pin3 on the lm1815, and the output on pin12, but the scoop is not picking up much on the output, so the scop is not fast enought to pick the spikes up(4000pkt/sec),

We could then only try some differents things,

We mounted a 3.3kohm pullup on both the cas and cam, buth then the car gives no idication of being willing to try to start.

THIS IS THE MOST IMPORTEN THING TO GET TO WORK, as it´s this is not something that can be lived with

I would like suggestions to make this work fast!.

Can it run on a 4-1wheel, as i then can remove 3 teeth, or can the vems not read this as weel?, or do i have to make a 36-1 wheel?.-I'm trying to use this method on a car here nowadays.-FERO

thanks for your comment, I have been told to make a new 12volt+ wire to the vems, directly from the batteri, i will try that, but i´m positive that will make no change,

I have now had a external 12v+ directly from the batteri ,2days, and it have indeed started a lot better,so this have something to do with it,still not great.

Tonight i will get a 36-1 wheel made, as this is a costumers car, and i simply think that i cant have the car here anymore,so i will take the timing belt of again,So i can try this, and if that fail, i will then change to another box.

But i hope and belive that 36-1 will work.

Scop pics we captured the other night first pic is pin 3(yellow) and pin7 green this is from the primary trigger.

next is a closeup of the same

This does not look right as there is lots of noise on pin 7, maybe this is the problem with the cranking problem?

Very good, this measurement might get us closer. Can you try with WBO2 disconnected ?.

We did last night some measurements again, and there is no change to unplugging the wideband, or disconneting the power to it, or removing the iac plug,You do realise that it´s pin7 and not pin12, that this measument is from?

we put a kondensator on the output,so we could measure the pin12 output, and it looks fine,and it does not look like there is noise on it,

So i do suspect it to be someting we the camsync.

Tonight i will on the Impreza mount a 36-1 wheel, then we see what happens.

The best way to start first, is actually with no WBO2 (use a failed old sensor to plug the exhaust). Dumping fuel during initial startup attempts can decrease sensor lifetime by dozens or even hundreds hours.

The car have allready been mapped, and run fine, it just takes 2-5min to start it!
- also make sure that HV spark wires are away from trigger cables.
They are weel away from the wires on the Impreza, and the Legacy runs cops.

next is secondary trigger pin3 yellow and pin 7green

This looks more right

today i mounted a potientiometer on both cas and cam ,to ground, to make pulldown,

But it seems random how long it takes to start the car

Can any of you that know comment on if this triggerwheel looks like it can be read be a VEMS ecu, as i dont have more time to mock around, with a car that cant start probbely, becourse of the vems not being able to read the oe trigger wheels like i hoped it could,but cant.

I have just had this made, if all other fails, 36-1 subaru triggerwheel, can now be made really easy.

Link to some info to how ms extra does subaru, can any of you see what the vems is missing to filter the noise, if this ms extra setup works, interesting that they trigger on raising egde

http://www.msextra.com/ms2extra/MS2-Extra_Subaru.htm

Another link with a sulution to ms setup with the exact same symtems as i got when cranking.

http://www.msextra.com/viewtopic.php?f=91&t=28092

We scoped the oe dwell time, and that is 3ms.

Now cranking 200 RPM (OK),

car runs fine, when it is started, only cranking is out of sync

but sometimes backfires during cranking?

this is worse when the weather is colder, and better when warmer, like now.

This is a cranking scope shot

Crank and camsensor signal scopeshot

CH1 is crank signal

CH 2 is cam signal

- looks OK. VR (LM1815 input) triggering on falling edge of the EC36/27 and EC18/12

Also measure DC voltage of both triggers (engine off, but ignition on).

EC36/27 DC measurement: ...0.50volt (primtrig sensor: 1.9k Ohm)
EC18/12 DC measurement: ...0.01volt (sectrig sensor: 1.9k Ohm)

This is measured on a Legacy turbo 92´ and a Impreza turbo 99´

This subaru legacy is running the 6/7 trigger setup

as it seems no one have had this running with a VEMS yet, i hope to have some help setting this up,

I will, then document it,so it will be pnp for other users.

D.10/7-2008 :This is the current setting.

Firing order 1-3-2-4

primary_trigger=29

secondary_trigger=59

tooth_wheel=02

trigger_tooth=00

another_trigger_tooth=03

crank_minper=BC

BAD: tooth_wheel_twidth1=00

BAD: tooth_wheel_twidth2=00

cam_sync_r_edge_phase=00

cam_sync_f_edge_phase=00

reset_engphase_after=04

Date:6marts 2010

Subaru trigger firmware 6/7 related stuff, link to another ecu manufactor, the G10 Impreza is the classic one, with the 6/7 trigger setup.

Please read before making firmware,as this can come in handy

http://www.syvecs.com/forum/viewtopic.php?f=6&t=110

http://www.syvecs.com/forum/viewtopic.php?f=2&t=118

date:16april 2010

screendump from Vemstune 1.1.72 primary and secondary trigger,

is it right that secondary trigger is off ?
- no. As you write above, secondary_trigger=59 (enabled). Try the following:
- choose the "subaru" configlet (the others will NOT work for a subaru). Use newer vemstune if needed. Do not forget to send and burn values.
- check that "next trigger tooth" == 3
- check that tooth width is 180 degrees

Also the ign outputs cant be stepper as this is used for steppermotor on my99/00 Impreza(that also runs wastedspark)
- true. Anyone will need to adjust his ignoutputs anyway. 2,3,2,3 might be a reasonable default ? (assuming wasted spark transformer on ign output channels 2 and 3)

date.24april 2010

Is the ref tooth table in the default subaru trigger right?

yes ! Regardless of firing order. Tweak the ign outputs, and not the ref tooth table !

Firing order is 1324,and NOT 1342 as normal!!

so pair the outputs as 1-2 and 3-4 (not 1-4 and 2-3 as usual)

I use only 2ign outputs, ch0 and ch1 as it runs wastedspark.

Output 2-3 is recommended if you drive the coils with IGBT (not logiclevel coils), because that is available even on a vems with only 2 IGBTs. But it should work anyway. Set the ignition outputs for the outputs you use. Wasted spark is no problem at all. Any trigger with secondary trigger is suitable with fully sequential COPs, wasted spark or distributer. And VEMS firmware supports them all.

Map sensor - is there a page for other sensors, or should we make one ?

From irc (peter_j):

" map scaling bigger value. We can´t scale the oem map enough" ** use kpa_conf=42
- if subaru MAP sensor has higher than 1024 kPa / 5V slope (!!?!), than give use more details
  - the oe subaru mapsensor is 280kpa max, the problem is that -the offset cant be low enought, the offset only goes to -32, and it needs to be lower to get the reading in line with reality(skassa)