VEMS Community Forum

Hi guys,

Car is a 16V 4AGE with ITB's and cams, running off of a GZE 24+1 CAS on the exhaust cam. 

I actually am starting to drive the car around and map it, but I think I may be getting hesitation up high. 

The weird thing is there is a trigger error the entire time, ever since the car first fires up...I'm actually surprised it is even running as the ignition and fuel is sequential.  I thought it would need the secondary trigger for that. 

Below is the output at idle. 

Falling
TDC after trigger - 73 degrees
24 Teeth
0 tooth is 4 teeth after TDC
Nearest tooth is 6
Angular width is 30 degrees

Secondary
Rising
Single edge while cranking
Filtering disabled

What does too many primary triggers mean and how should I try and fix a bad secondary trigger position?

First photo, I disabled one of the errors so you can see the secondary trigger line that it is on top of.
Both images were taken at the same point in time.

(http://i84.photobucket.com/albums/k23/stupidbuthey/triggererror_zps17b99ad4.png) (http://s84.photobucket.com/user/stupidbuthey/media/triggererror_zps17b99ad4.png.html)

(http://i84.photobucket.com/albums/k23/stupidbuthey/zoomedin_zps58c63666.png) (http://s84.photobucket.com/user/stupidbuthey/media/zoomedin_zps58c63666.png.html)



(http://i84.photobucket.com/albums/k23/stupidbuthey/IMG_1829_zpsd76ac37c.jpg) (http://s84.photobucket.com/user/stupidbuthey/media/IMG_1829_zpsd76ac37c.jpg.html)

(http://i84.photobucket.com/albums/k23/stupidbuthey/IMG_1821_zps5ae416a2.jpg) (http://s84.photobucket.com/user/stupidbuthey/media/IMG_1821_zps5ae416a2.jpg.html)

Thanks all!

Hello Grant,

"Too Many Ptrig" means the ecu sees more tooth per cam cycle than configured in the configuration. This can mean either noise injected into trigger signal (cross current, capacitive coupling or other) causing uneven or additional primary or secondary trigger, but since your triggerlog (visual) shows even tooth times this is less likely the problem. The most likely in your case is configuration error, could you post your vemscfg so i can review ?

Best regards,  Dave

Hi Dave,

Nice screen name.

I hate to do this to you but I can't find anywhere on this BB where I can attach a file?  How do people normally upload their configs?

Thanks,
Grant

In the mean time here is a screen capture:

(http://i84.photobucket.com/albums/k23/stupidbuthey/Screenshot_zpscbea1375.jpg) (http://s84.photobucket.com/user/stupidbuthey/media/Screenshot_zpscbea1375.jpg.html)

(http://i84.photobucket.com/albums/k23/stupidbuthey/Screenshot001_zps2d921bdc.jpg) (http://s84.photobucket.com/user/stupidbuthey/media/Screenshot001_zps2d921bdc.jpg.html)

(http://i84.photobucket.com/albums/k23/stupidbuthey/Screenshot002_zps582a1a9a.jpg) (http://s84.photobucket.com/user/stupidbuthey/media/Screenshot002_zps582a1a9a.jpg.html)

Hello Grant,

After taking another look at your configuration and picture of trigger hardware i would like to suggest the following:

1) Change primary trigger edge to rising (as suggested with vr trigger).
2) Swap wires from the secondary trigger sensor (VR+/VR-).

The secondary trigger pulse is very close to the primary (sometimes overlaps) this will lead to race condition which is seen first, hence the trigger error.
Swapping polarity on the secondary trigger should solve that. Please post another triggerlog afterwards to confirm.

Uploading of files to the forum directly is not possible, but you use the VEMS file area: http://www.vems.hu/wiki/index.php?page=FileArea and link to that.

Best regards, Dave

Swapping the polarity on the 2nd trigger make sense, as I had to swap it in the first place to get the POS AEM EMS-4 to work on a rising/falling edge signal (I forget which one it was, but AEM EMS-4 only works on either rising or falling...so I had to re-wire the CAS).  Maybe that caused the signal overlap. 

So you are saying that on a normal CAS with cam sync, it is important for the secondary to not overlap on a primary? 

I believe the car ran worse when the primary was set to rising. 

Yes, very important. Since there are no missing tooth on your trigger setup the only way to synchronize the engine is to sync to the first tooth after cam pulse, if the cam-pulse sometimes comes before and sometimes after, you can see where things will start to go sour. The ecu will however recognize this and attempt to resync (and flag a trigger error).

But to be frank it is imperative that cam pulse never overlaps the primary trigger pulse. I suspect the primary rising caused a bit more overlap (without changing secondary) but vr= rising is preferred and recommended.

Regards, Dave

Thanks a lot Dave for explaining this.  This will certainly help me in the future (I do about 3 installs with various systems a year). 

I ill be sure to update after I get this sorted out.

Thanks,
Grant

I got home after work and proceeded to swap the secondary trigger wires.  I also set primary and secondary to FALLING edge.

Now the Secondary is no longer overlapping a primary trigger event, but the same errors were still being reported - Too many primary triggers and secondary trigger errors. 

The other clue was that there were only errors showing as soon as the engine fired...while cranking there were no report of errors.  I was beginning to think it was a noise issue.  I added a variable pot accross 2nd VR+ and 2nd VR- and adjusted between 0-500ohms with no improvement. 

I was pretty stumped. 

Then I started messing around and found that I had "filter" enabled on VR 2 settings, show below:

(http://i84.photobucket.com/albums/k23/stupidbuthey/filtervsjustshow_zpsb700f652.jpg) (http://s84.photobucket.com/user/stupidbuthey/media/filtervsjustshow_zpsb700f652.jpg.html)

As soon as I set it to "Just show error - Default", the trigger error messages became erratic, and not on the whole time. Also, I would only occasionally see an error on the trigger log!  It seems like a bug to me, when filtering is enabled, the car drove pretty well but it reported an error every secondary trigger event.  Now, I only see an error when quickly revving the engine right off idle, the rest of the time it seems fine.

Below is where I am now: Trigger errors when "blipping" the throttle suddenly off idle. Also low speed driveability problems

This image shows primary trigger dropping out when quickly revving off idle.  After 2500 RPM, trigger is rock solid it seems.

(http://i84.photobucket.com/albums/k23/stupidbuthey/revvingaftersettofallingedge_zps57e9f900.png) (http://s84.photobucket.com/user/stupidbuthey/media/revvingaftersettofallingedge_zps57e9f900.png.html)

(http://i84.photobucket.com/albums/k23/stupidbuthey/blippingthethrottleaftersecondsettofallingnosecondfilter_zpse266033f.png) (http://s84.photobucket.com/user/stupidbuthey/media/blippingthethrottleaftersecondsettofallingnosecondfilter_zpse266033f.png.html)

Note:  there is still a 500ohm resiter across the VR2.  Tomorrow after work I will remove it and see if the signal is OK when revving up.

Thanks!

Please use rising for both primary and secondary trigger, this should not change timing but might be causing your trigger problems. The way the input circuit is designed on the VEMS (when using VR) the rising edge is synced to zero crossing of VR signal, while the falling is variable time after. Variable time is dependent on VR signal Vpp and Slope result -> phase shift with rpm (variable).

In review i missed the Filtering enabled, good catch. It should be disabled in your application (just show). Depending on when your camsync occurs this might show some false trigger errors BAD SECTRIG POS, these can be ignored for the time being, will be fixed in up coming firmware release.

Best regards, Dave

Darn, I forgot you told me to set to rising-rising, not falling-falling.  The thing is, I know the CAS works as falling edge in the OEM format so I confused myself!

Another interesting note, my guide through this whole install has been my friend Keith, aka Sly.  My understanding is that he had a hand in designing the V3 hardware.  He mentioned that he didn't think changing from falling or rising would make much of a difference as the LM chip conditions the signal anyways and only gives a logical output anyways.

I will switch from falling to rising tonight and will report back.  Thanks!

Hi Grant,

The LM chip conditions the signal you are correct, but if you take a look at the LM1815 datasheet page 8, figure 18 you'll get an idea what i am referring too. When using both VR sensors it should be less of a problem but still, we recommend rising when using VR. Please pass Keith my greetings,

Regards, Dave

Hi Dave,

I spent a couple of hours tonight in the 42F weather trying to push forward with the project. 

I did indeed switch the triggering for both primary and secondary to RISING, but didn't see much of an effect.  However, since you mentioned there may be some problems with leaving the triggering on falling for VR, I continued my testing in this state. 

Since that didn't help much, I started messing with the number of teeth between the 0 tooth and trigger tooth setting.  It seemed to help if I leave it at 4 or 5 (with corresponding timing change to suit), and I did have to remove the dizzy several times to turn the gear to allow for the correct timing range (Distance between first tooth and TDC). 

The problem still remains that the car starts and revs up fine it seems, but on quick acceleration blips the trigger drops out, causing a hesitation and a report of a trigger error.  My main goal is to solve this issue so I can start mapping on a dyno and return the car to my customer.

Does it matter if a spark event happens near a secondary trigger event? I noticed in some of the logs the two were awfully close, which was another reason I was messing around with the 0 tooth and trigger tooth settings.

Keith's on a road trip right now but he suggested I pull a complete data log with me revving to try to get the symptoms to show them selves and ask here.

I also have some resistors I can try to pull the whole VR signal down in case noise is an issue.

Any input would be greatly appreciated =)

Thanks,
Grant

Could you post a logfile (vemslog) of the engine running with trigger filtering disabled (show only) at idle and triggerlog files (the files iself not screenshot, one during cranking without ign fuse (no start) and one during idle) for review ?

This will help me to narrow it down,

Best regards, Dave

Hi Dave,

Here is the current config:

http://www.vems.hu/files/GrantHendricks/risingrisingcoldstart%20tuning.vemscfg

Here is a trigger log file:

http://www.vems.hu/files/GrantHendricks/v3.3_u008216-2013-12-02-22.38.58.triggerlog

I'm sorry, I missed the part where you asked me to get you a log file while cranking with w/o ignition. 

I just have a log of the car running, and trigger errors at 5K and above.  I'm glad I looked at the file, I didn't realize I was getting errors above 5!

http://www.vems.hu/files/GrantHendricks/v3.3_u008216-2013.12.04-23.17.58.vemslog

I'll get you a the requested log tomorrow night.  Sorry =(

Thanks,
Grant

Hello Grant,

Thanks for uploading the files, no further log taking required. I have reviewed your current set of files.

The trigger error BAD_STRIG_POS can safely be ignored for simple trigger plus cam sync. It sometimes gives false reading based on position of initial sync and does not cause resync but inform only.

I do not see any errors in running that might cause missfire (from trigger standpoint). It is more likely the engine is misfiring because its just running very rich. Please start by tuning the low rpm/low load section before attempting any higher rpms and work your way up gradually. I suspect this will solve any misfire issues you might have.

Best regards, Dave

As shown by my post about "injector voltage compensation", actually I do pay attention to the PW time  ;)

Grant, I noticed that in in your log 'Injector RampUp" is 0.5ms, battery compensation @14V is +0.1 ms (??) for a PW total time of 2ms à 2500rpm. Hence the speed density rule enters for only 63% of the total time.
This seems very strange to me. But I may be wrong. Indeed this makes the low load part of the VE map quite useless and in any case too rich.

( btw Does someone knows any technical reference for this "traditional" strategy.
Where injector closing time or "injector Rampdown" taking into account. What are the physical differences for an injector self between closing and opening?)

Hi Dave,

There is NO LAMBDA SENSOR hooked up yet =) 

I still have an external AEM Wideband that I am referencing. 

As far as the injector opening times, I know nothing about that.  How does it only effect the ve table 63%? If you have any suggested base line value to use I am all ears.  The injector config was mostly just what was in the default config file. 

The injectors are 250cc injectors I believe (even though I have it set as 195cc) and are high Z injectors.

Dave, is there any information I can try and provide?  The mixture is in the 11.9-14.0 AFR range for 80% of the map, and not where I'm having trouble with the engine cutting out.

Thanks!

Hello Grant,

Thats handy info, it was assumed the lambda reading from your log file was the actual measured value. I'll do a more extensive review when i have a bit more time.

Regards, Dave

Hello Grant

"As far as the injector opening times, I know nothing about that.  How does it only effect the ve table 63%? If you have any suggested base line value to use I am all ears.  The injector config was mostly just what was in the default config file.  "

I read those values that form  the VT view "Calc Model", where battery compentation is added after the speed density value to give the totla PW time. As you can guess from my post "injector voltage compensation", I do know much how to handle this correction

I will still pull a log file of cranking with coil packs hooked up, and with coil packs turned off.  I can also do the same with injectors hooked up and turned off as well.

Thanks!
Grant

PS, coils are grounded to the head individually and power wire originates form the firewall.  The injector power goes all the way back to the ECU area and is run along side the VR wires.  VR wires are shielded, shield is hooked up to ground on one end. 

Hi Dave,

The I've mapped the car some more, it seems to drive pretty well.  There doesn't seem to be as much power after 7000rpm for a 280 duration camshaft as I would expect, but that could be from other factors other than trigger. 

Below is a data log of me sitting and "blipping" the throttle right off idle.  This is where the engine stumbles and loses trigger badly:

http://www.vems.hu/files/GrantHendricks/v3.3_u008216-2013.12.05-21.23.51.vemslog

Probably not that useful, but here's a log of cranking without firing up....trigger errors while cranking:
http://www.vems.hu/files/GrantHendricks/v3.3_u008216-2013.12.05-21.21.23.vemslog

I  believe this is a trigger log file of me just revving the engine up and down:

http://vems.hu/files/GrantHendricks/v3.3_u008216-2013-12-05-18.35.27.triggerlog

This is cranking with coils connected
http://vems.hu/files/GrantHendricks/v3.3_u008216-2013-12-05-21.17.29.triggerlog

This is cranking with coils disconnected
http://vems.hu/files/GrantHendricks/v3.3_u008216-2013-12-05-21.19.48.triggerlog

Please let me know if there's anything else!

Thank you,
Grant

Change to 1.2.15 or more recent firmware. Simple trigger had issues when used with cam sync. I noticed back in August on my Volvo engine with 4+1 trigger and it was fixed. Issue was very slow to start (or at all) and took many seconds cranking before I saw rpm even though I did see "cranking on" flag turn on from first signal.

You can check the "pump on after powerup", increase that to many more seconds and if you crank the engine immediately it will probably have an easier time starting. Or better yet, upgrade firmware. It might not solve everything ..

You have very short ignition dwell and no scaling with battery voltage. What coils are used?

Just a one cent remark (I am not by far an expert).
I would put the dwell alright, the acc. enrichment = 0 and try first to idle good at 900 rpm. I noticed that the IACint could be for a while at about -49, actual rpm about 760 with IACTarget 900.
(the IACInt value may be allowed to keep increasing or decreasing over the increase and decrease limits, that is behind the scene )

Hi Guys,

Thanks for looking over the files I posted.

I didn't realize I wasn't using the latest firmware!  I will try looking for the version you posted and see how that goes.  The car seems to drive well other than the stumbling at quick throttle openings.

Coil dwell is low as I just wanted to get the car running safely.  I don't know what ideal coil dwell is yet as I haven't researched.  Coils are Toyota 1NZ coil packs (smart coils) from a Toyota Yaris/Vitz.  We are driving them directly with logic level outputs off the VEMS board, no FETs are installed on the ignition out puts!

Accel enrichment seems fine, I have the fueling pretty good right now. 

IAC is not used...I just have the ITB's cracked open at idle, and idle speed is controlled by ignition advance/retard.

Thanks!

Just to get a little more personal, here is the car and the engine bay:

(http://i1113.photobucket.com/albums/k520/ethergore/aa5c1b69.jpg)

(http://i1113.photobucket.com/albums/k520/ethergore/0362a4c6.jpg)

(http://i1113.photobucket.com/albums/k520/ethergore/B7D01741-4E38-462D-B047-EED334B655C7-1338-000000AD838EB7C2.jpg)

280 duration cams, $1600 USD header (work of art), and a few other upgrades.  Engine has a full ITG filter set up now. 

This is a customer car, not mine.  I wish I had a car this nice. 
Thanks for the help so far guys.

Just an update: 

Upgraded to latest firmware 2.17 I think it was, and that did not solve the problem.   :-\

It is interesting to note that when bench testing this distributor, there were no trigger issues.  I actually took the distributor out a couple of days ago, spun it with a drill (injectors unplugged) and got trigger errors right in the 1-2K rpm range where I have been having the most trouble.  It does this regardless of the direction I spin the drill. 

I'm wondering if I'm getting too much noise from the coil packs?  Power for them is provided from the firewall, and the CAS wiring is properly shielded and grounded on one end.

Firmware upgrade tool is very slick, this is a far cry from my1.1.44alpha release and megatune days when I started using VEMS in 2009!

Quote from: Grant on December 13, 2013, 07:37:19 PM
Just an update: 

Upgraded to latest firmware 2.17 I think it was, and that did not solve the problem.   :-\

It is interesting to note that when bench testing this distributor, there were no trigger issues.  I actually took the distributor out a couple of days ago, spun it with a drill (injectors unplugged) and got trigger errors right in the 1-2K rpm range where I have been having the most trouble.  It does this regardless of the direction I spin the drill. 

I'm wondering if I'm getting too much noise from the coil packs?  Power for them is provided from the firewall, and the CAS wiring is properly shielded and grounded on one end.

Try repeating your test with the coils unplugged.
That would indicate a relationship between the coils and  your missfires.

Good idea, I'll give that a go.  ;D