Yamaha R6 trigger cam sync problem-no RPM

[Bubba Zanetti]

Hi,

Recently I finished VEMS instalation for our Formula Student 2010 project car , but I have problems with cam sync triggering.

Engine installed is 2003 Yamaha R6 engine - everything on engine is standard (Yamaha 4+1 trigger), except two things:

- we ground old cam sync tooth and machined & re-welded new tooth to 95° BTDC of #1 cyl.,
- we installed VR cam sync sensor instead std Yamaha item.

This is simplified AutoCAD  drawing of R6 std primary 4-tooth wheel and modified 1-tooth cam sync trigger:



VEMS ECU & harness was bought from VEMS webshop with this specs:

Assembled v3.6 controller:

  - 1-wire interface: no
  - Flyback: 30V flyback
  - ignition driver: 4
  - Knock and EGT: yes
  - LCD connection: yes
  - MAP connection: 250kPa (5/3mm pneumatic)
  - mounting-style: screws
  - primary_trigger: EC36/27 VR
  - PS2 connection: yes
  - secondary_trigger: EC36/13VR    

So, after I finished with installation of all sensors, I connected two  triggering sensors (both VR types) : primary VR to EC36/pin 27, and secondary VR

to EC36/pin 13, firmware used was 1.1.23 version.

After I cranked engine, there were no RPM readings - all other sensors were reading OK.

I tried to change triggering settings in Megatune but without success. Readings from both VR sensors on my handheld scope were OK - they both are

producing good signal.

As everything seems OK with wiring and triggering hardware, I guess the problem is probably in my triggering settings - here are the screenshots :





Any help will be more then welcomed 

[dnb]

I thought that the secondary trigger went to the EC18 if it was a VR sensor.  (The secondary trigger was only on the EC36 if it was a Hall type.)

I admit that I haven't had time to study the new 3.6 pinout yet, so I could be talking about old stuff... 

[rob@vems.co.uk]

Disable the secondary trigger for testing purposes.

[Bubba Zanetti]

2 dnb:

I also thought the same thing and tried EC18, pin 12 option - but without success unfortunately.

2 rob@vems.co.uk:

Today I tried this:

- disabling secondary trigger (I guess it's disabled with 0x02 in TODO! box) - no RPM,

- disconnected both primary and secondary VR sensors and connected simulated 4+1 signal to VEMS harness - simulation was done to eliminate potential sensor problems.

First I generated  *.wav file that was played in loop and amplified through small class D amp; L channel was primary trigger (4 tooth) and R channel was secondary trigger (1 tooth), p-p voltage was around 10V - again nothing, no RPM, not even some glitches, or trigger errors - nothing.

It's really interesting, I had experienced many strange things when doing ECU installations (mainly with Megasquirt), but at least I always got RPM reading on Megatune 

Hm, even if I touched ignition input on MS with +5 or +12V with resistor in series, I always got some wild flickering on tacho, it's funny that I can't even got trigger errors 

[rob@vems.co.uk]

Thats now beyond strange I've never not seen RPM with a Wav file - even with a Hall setup.  Did you use the triggergen app to make the wav?
Try the 1.1.18 firmware, that will definately work with the 4 tooth job

[Bubba Zanetti]

2 rob@vems.co.uk:

Yes, I also think that this is totally strange, I mean, I've never ever had problems with RPM detecting with other ECUs...hm, though I usually had problems with strange rpm readings, backfires, bad grounds, bad shielding, loose/broken wires, EMI noise, corrupted firmwares, wrong ignition settings, low battery voltages, etc - but at least I always got SOME kind of rpm reading

I used "Crankwheel pulser" application - very easy to use:

http://www.megamanual.com/router/crankwheelpulser.htm

As in your experience,  wav simulated signal always worked on my previous MS installations - it's clean, RPMs doesn't depend on battery voltage variations (like when cranking), and amplitude can be easily increased if needed when doing simulation on lower RPM range.

I have some questions for you:

1. Latest firmware I tried was version that was included with 1.1.23 Megatune package (VemsMT1.1.23) - is there any benefit if I try 1.1.18 firmware?
I mean, new firmware should have all options included as previous versions plus some new features & bugs sorted out, am I right, or I am missing something?

2. When I uploaded firmware, I used "Upload-firmware" application - what are other ways of uploading firmware?
I uploaded firmware succesfully at least dozen time using this simple application, but sometimes I've got errors and some strange problems.


Please, if possible, I would like some more details about both trigger settings, just to be 100% sure that problems aren't caused  by wrong trigger settings in Megatune- for example:

Primary trigger settings:

Type:  Multitooth / Coil  - I guess my application is Multitooth (4 tooth@ crank), but I've tried both options without success,

TDC after the trigger (deg):  -  if I have secondary sync signal, what to put here? I guess this setting is active only with missing tooth wheels...

Number of teeth on wheel: 4 (I guess this is OK)

Trigger tooth: 1 (probably OK)

Next trigger tooth: 2 (I guess this is also OK, as all crank teeth are sequentally fired, so the order is 1-2-3-4),

Crank min. period (usec): hm, this should be calculated for my application, if I 'm correct, please correct me if I'm wrong,

Tooth Normal relative (min): what to put here?

Tooth Normal relative (max): what to put here?

Tooth Missing relative (min): if I have only "normal" teeth, as there are no missing teeth - what to put here, zero?

Tooth Missing relative (max): same as above


Secondary trigger settings:

Active: enabled (obviuously) 

Edge: Falling / Rising (I tried both, without success)

Filtering: Enabled / Disabled (same as above)

Type: Multitooth/Coil (same as above)

Use: cam sync (that seemed like reasonable choice, as I have cam sync sensor )

Rising edge: hm, I didn't found more details about this setting, so I put 0, as I found that value  on some cam sync configurations-more details, please?

Falling edge: same as above, more details?

Angular width of tooth: hm, I put 15° - I guess it should be OK, as I have one tooth, that has roughly 15° width

Angular width of missing tooth: If I have only one tooth (there are no missing teeth) what to put here?

Engine phase, when to reset: I guess this is angle when sensor is aligned with sync tooth - in my case 95° BTDC of 1st cylinder

Is there some other settings in Megatune, that can influence on RPM readings ?

[rob@vems.co.uk]

We have a tool in the tools directory that will generate a wav, unlike anything MS based, it is all command-line based and a pain in the arse to use.

When the VEMS detects any signal on the primary VR it will switch on the fuel pump and wideband heater.

There is a trigger debug function - again its not very usable but it's written in a rambling form here:
http://www.vems.hu/wiki/index.php?page=InputTrigger/TriggerLog

Don't assume that a certain firmware will give you the triggering option that  you want, 1.1.18 was released specifically for N+1 triggers.

You can use VEMSTune to upload firmware and do the other bits:
http://www.vems.hu/wiki/index.php?page=VemsTune

[Bubba Zanetti]

OK, I still can't get RPM reading, and this really isn't funny anymore

Things that I tried today:

1. first I connected 2 new wires for primary trigger input and 2 wires for secondary DIRECTLY from ECU connector (to eliminate harness completely!) and then connected that wires to simulated wav output - nothing.

2. then I played with both primary/secondary settings, turned off cam sync - nothing,

3. I installed VEMSTune, uploaded 1.1.18 firmware, uploaded config - still nothing,

4. uploaded 1.1.62 firmware again just in case, played with trigger settings - nothing,

5. OK, as this was last thing that I can test outside ECU, it was time for dismantling.

So I completely dismantled VEMS box, carefully pulled out mainboard and placed it on the desk-first thing that I checked was if all jumpers and connections for both primary and secondary triggers are soldered in place as according to V3 Manual - everything seemed OK, except jumper that should be placed under both LM1815 chips (marked in yellow rectangle on this pic from manual):



Hm, on my board that jumper wasn't installed, but when I looked carefuly once again, there was no need for jumper as this two points were already connected via pcb track - so I guess this is just one of the changes in V3.6:



Next thing was checking if signal is actualy coming on both chips - with wires from sound card connected directly to ECU connector, my scope showed that simulated wav signal is indeed coming to pin 3 on both LM1815 VR conditioners.

Also it revealed that output of the chips (pin 12) also has some signal coming out...but RPMs are still missing 


Hm, is it possible that some other jumpers/connectors are missing, from output of LM1815 to some other pin, that aren't mentioned in V3 PDF Manual, or anywhere else ?

What about pin 5 (adaptive mode) ?

I'm too tired & really out of ideas right now...

The trigger saga continues...

[MWfire]

That Formula is from FSB?
I can explan and solder vems to read that trigger. I'm from zagreb.

[MWfire]

btw set trigger toth to 0 and you will have rpms

[MWfire]

Try if p259 outputs works?
Did you connect analog ground to power ground?

[Bubba Zanetti]

Finally solved 

MWfire gave me great hint about possible problems with P259 chip & RPM reading - if the P259 chip died somehow, it can pull 5V supply for both LM1815 VR conditioners to ground, causing lack of supply for both VR input chips and obviously lack of RPM reading...and I already check every connection EXCEPT +5V supply to both LM1815 chips...

And when I measured them yesterday, both supplies were around 1.7-1.8 V. 

So, I tried turning on & off P259 outputs, and they didn't worked - then I unsoldered supply for P259, and made jumper from +5V supply directly to LM1815 pins-voila, I've got good, clean RPM reading from wav file 

Tnx MWfire for help, one more time!

P.S.
One thing that worries me is that I still don't know real cause of P259 "death" - I can't see possible problems with ground connection, as I bough harness from VEMS web shop directly (only one ground wire is coming from the loom) - right now, I connected VEMS ground directly to battery negative - shortest distance possible, with 10mm2 cable, crimped and soldered.

Second thing is that now, with disconnected supply, P259 turns on/off output relays very fast, in harmony with RPM - very strange - I will have to disconnect all output pins.

[rob@vems.co.uk]

The P259 dies when the sensor ground is not tied in with the power grounds.

[MWfire]

I always make connectio between analog and power ground in vems.
yelow line

[rob@vems.co.uk]

I always make connectio between analog and power ground in vems.
yelow line

If it were that simple they'd have done that in the design.
The reason for the ground wire rules is because of the amount of current that the grounds have to deal with.  The resistance of the wires is tiny, but when the coils, injectors and solenoids are all firing away the grounds can be handling many 10s of amps.  Say for example the resistance of the grounds is 0.05Ohm and you've got a load of 10amps being grounded, then you can expect a voltage rise of V = 10 * 0.05 = 0.5v  Not much... but unless the sensor ground is not biased by the voltage increase it will be subjects to false readings - temperatures and pressure values are very discrete and having them change as your revs rise would not be a good thing.