VEMS Community Forum

Hi,

After many problems with  VEMS installation on our FSAE 2010 car, I finally solved almost all problems - but I still have problems with ignition outputs cut-off  >:(



This is ignition & triggering setup right now:

- modified OEM Yamaha R6 crank wheel - one tooth grinded to have 4-1 wheel (triggering cca 105° BTDC)

- OEM Yamaha R6 coils   in wasted spark mode, but connected  in series (0.8 Ohm each) : so 1st &  4th cylinder connected in series on one IGBT output, and 2nd & 3rd on other output,

- dwell around 2.0 ms




This setup worked after adjusting ignition advance with timing light, but interesting thing is happening when I try to crank engine:



- if I disconnect fuel hose, so injectors are just clicking without squirting fuel - both ignition outputs are working and I have strong spark,

- then I connect fuel hose to fuel rail, again crank engine and as engine tries to start, and immediately when it starts to cough, almost each succesfull explosion inside cylinders seems to "cut-off" spark  output!!!!!



How do I know that, you probably ask?

Because I connected 2 red LEDs in parallel with both ignition outputs , so it can be seen clearly-almost each time the engine coughs, ignition output LEDs starts to fade or sometimes even completely turn off...




OK, then I played little with fuel cranking amount, and somehow it was better, so I started the engine - and when it was running in steady state, everything was OK, no matter how long it was running...BUT if I tried blipping  full throttle - ignition immediately cut-out (LEDs aren't flashing) and engine dies.



And now  this is most interesting part - if I then try to start engine - there is no spark output at all !!!

And all other sensors and readings are still perfect,VR sensor is triggering nicely, I can see RPMs nicely -  everything is great except there is no ignition output ! ! !

This isn't firmware related, because I tried both old & new firmwares.




But another interesting thing is if I turned off ECU power, and then turned on again - voila, I then have again ignition outputs ! ! !


It seems that for some reason both ignition output is having some sort of "shut-down" enabled during full-throttle run, that can be only reset by cycling power to VEMS ECU  ???

If using MegaTune,  give us .msq and log file (.xls).
If using VemsTune, give us datalog of your event.

You ask many questions, but do not leave much information.

Which firmware do you use? Which have been tested? (old ? new ?)

What about the trigger? VR/Hall? Shielding conncted only at ECU side?
The ground wires that go between the IGBT and coils are often shielded in stock applications, with the shield is grounded at the ECU.

I would advice you to use the dual-out mode to drive the ignition coils instead, so each coil has  it's own IGBT.
The peak currents can be very tough, depending on the coil. What is the data on the primary winding ?

VemsTune will probably have a trigger log analysis function, hopefully this year.
At the moment you can see wheel errors on the 3rd page of the LCD (mlp02), if you don't have it you can access it using serial terminal (Manmlp02mll ...).

check trigger tooth width, it should be 90. And with 4-1 you should have ref table like this.
0
6
4
2
And use two more igbt for coils, serial connection is bad(more electrical noise and weak spark).

I used some Yamaha coils and had endless probelms with them blowing fuses, and were perhaps the root of all my subsequent issues. I managed to get them to work fairly well at a later date with a dwell of 0.5ms :o but it was not long before the fuse blew again. I put it down to the coils in that there were a CDI coil rather than a transistor ignition coil. I reverted to my Suzuki coils and have never had a problem. There is a topic some where with the Denso part numbers for these coils.

I was also advised for the reasons mentioned above, not to double up the coils on one output, but I thought it was more to do with the circuit being in parallel thus halving the actual power on each coil?

I have also come across a coil on a BMW 5-series (E34 with M30 engine) with a coil that required insane fuses, and didn't give enough spark even then. It was also very low resistance on the primary winding and it was replaced with a MSD 6A and a Blaster coil (= CDI) and never looked back.

The load on one IGBT should only be one coil at a time, it doesn't matter if it's a distributor system or CoP - it's about the simultaneous load. Both the initial current and the flyback energy from the primary winding that must be handled at release of the spark.

2 mattias:


Here is the vemscfg file from VEMSTUNE:


http://www.2shared.com/file/12542256/921be924/v33_YAMAHA_R6_CDI_COP_240CC_IN.html (http://www.2shared.com/file/12542256/921be924/v33_YAMAHA_R6_CDI_COP_240CC_IN.html)


Sorry, I know that I forgot to add more details, but didn't have enough time  - so these is exact setup:

1. software: VemsTune 0.10.20 version

2. current firmware: 1.1.69

3. previous firmwares: 1.1.59, 1.1.61 (same behaviour on both)

4. VEMS triggering hardware: OEM Yamaha R6 crank sensor, triggering on 4-1 modified R6 wheel, shielded only on ECU side

5. VEMS ignition output hardware: two IGBTs, triggering OEM Yamaha CDI coils connected in series, two per one wasted spark channel.
                                               Each coil is 0.8 Ohm, so two in series is 1.6 Ohm, and in combination with 2.0ms dwell it shouldn't overload
                                               IGBTs too much...spark is not weak as I expected-compared with my old coils, it's visually the same.



I also forgot to mention one more important detail that happened one week ago - before these four CDI Yamaha R6 COPs , I used four GSX-R Denso coils

(part number: denso 129700-4800).

These GSX-R coils were inductive coils, about 1.6 Ohm each, and I connected them two in parallel, per each wasted spark channel.

Spark was good , but I had problems with finding correct cold cranking/afterstart values - so I was basically  flooding engine, wetting spark plugs,

having hard starts, etc, usual stuff and basically in that time I concentrated on fuel related problems and didn't unfortunately saw one mistake that I

made on VEMS harness when I was connecting Kliktronic gearchange system:




- for people that aren't familiar with that system: this is small electronic box that controls one linear actuator which that moves motorcycle gearbox

lever up/down with buttons on wheel, and it's connected with ignition interrupt module...so the problem was that ignition interrupt module should "cut"

+12V supply to all coils in exact moment when actuator moves lever...so I took two wires (red with black stripe) from VEMS harness for +12V supply to

coils, cut them and connected interrupt module in series BUT, I thought that I connected two +12V supplies to coils-unfortunately I took only one supply

to coil AND another wire was wire to coil No. 1 (pin 34)  >:(




(What's funny is that I already knew that original VEMS harness had two THICK red wires with black stripe for +12V coils supplies AND one slightly

thinner red wire with black stripe for Coil output, but it was late, I was tired ... etc, etc ....you know...the usual story... ::) )




And I discovered that of course when I sensed that very familiar, infamous smell - smell of toasted electronic component - in this case, burned Denso

coil  ;D

So after Denso coil melted very nicely, I then discovered that Coil 1 output was now constantly on - so IGBT on that channel probably went with coil on

journey to Ignition Heaven... :)




You can all probably now understand  my last move: it is 12 PM, I'm completely exhausted, stuck with only 3 coils & 3 IGBT outputs working, and this VEMS

installation needed to be completely finished almost one year ago ...so I took CDI coils, connected them as temporary solution and they worked...untill

the engine went first time in shutdown on full throttle...so my problems with shutdown started just in time between burning one IGBT channel and

connecting CDI coils in series...culprit is probably somewhere here.

2 MWfire:

Yes, trigger tooth width is 90°.

Here are all screenshots in one place related to ignition so you can do a quick check:

(http://img.photobucket.com/albums/v358/veliborv/YAMAHA_R6_VEMSTUNE_IGNITION.gif)


I suppose that your reference table is OK for dual output mode ...but unfortunately one  IGBT is dead, so I can't use it...hm, seriously  do you have some spares maybe ?  ;D

Here are screenshots related to cranking, afterstart and other injection settings - I'm sure that something is still wrong here, so feel free to advise:




(http://img.photobucket.com/albums/v358/veliborv/YAMAHA_R6_VEMSTUNE_FUEL.gif)

you have configured wrong for 4-1 trigger.
you need to set number if ign coil to 4.
ref tooth table:
6
4
2
0
ign out put:
ign6
ign3
ign6
ign3


and ign output table is wrong:
first 8 must be ign, second 8 to 0

also warmup is wrong

Quote from: MWfire on April 12, 2010, 03:19:50 PM
you have configured wrong for 4-1 trigger.
you need to set number if ign coil to 4.
ref tooth table:
6
4
2
0
ign out put:
ign6
ign3
ign6
ign3


and ign output table is wrong:
first 8 must be ign, second 8 to 0

Quote from: MWfire on April 12, 2010, 03:19:50 PM
you have configured wrong for 4-1 trigger.
you need to set number if ign coil to 4.
ref tooth table:
6
4
2
0
ign out put:
ign6
ign3
ign6
ign3


and ign output table is wrong:
first 8 must be ign, second 8 to 0

Quote from: MWfire on April 12, 2010, 03:19:50 PM
you have configured wrong for 4-1 trigger.
you need to set number if ign coil to 4.
ref tooth table:
6
4
2
0
ign out put:
ign6
ign3
ign6
ign3


and ign output table is wrong:
first 8 must be ign, second 8 to 0

OK, let's finally clarify 4-1 toothed wheel once for all:

This is my exact toothed wheel, with exact angles & VR sensor position:

(http://img.photobucket.com/albums/v358/veliborv/YAMAHA_R6_4-1_wheel.gif)




Is this correct reference tooth numeration according to VEMS?

If that's correct numeration then we have situation like this:



We have 4cyl engine, with 1-2-4-3 firing order, wasted spark ignition enabled for four IGBT outputs....so we
need four ignition events during complete 4 cycles (720 crankshaft degrees):


1. FIRST IGNITION EVENT:

H (2)=0 (1st & 4th cyl firing)          Ref. tooth = 0




2. SECOND IGNITION EVENT:

After 180° we have next triggering event:

H (2)=1 (2nd & 3rd cyl firing)          Ref. tooth = 2



3. THIRD IGNITION EVENT:

After 180° more we made full circle, so trigger wheel is back to start position:

H (2)=2 (1st & 4th cyl firing)          Ref. tooth = 0



4. FOURTH IGNITION EVENT:

And after last 180°, we have this situation:

H (2)=3 (2nd & 3rd cyl firing)          Ref. tooth = 2





So, according to above we should have this table:

Reference tooth according to ignition outputs:

H(2) = 0         Ref. tooth=0

H(2) = 1         Ref. tooth=2

H(2) = 2         Ref. tooth=0

H(2) = 3         Ref. tooth=2


And according to your ref. tooth table my numeration is totally wrong-this is your recommended table:


H(2) = 0         Ref. tooth=6

H(2) = 1         Ref. tooth=4

H(2) = 2         Ref. tooth=2

H(2) = 3         Ref. tooth=0

So, if I understand correctly VEMS is reading like this:

ref. tooth 6 is in fact "real", physical Tooth No.#2, and

ref. tooth 4 is in fact "real" , physical Tooth No.#0   ?

Quote from: Bubba Zanetti on April 12, 2010, 05:50:45 PM

So, if I understand correctly VEMS is reading like this:

ref. tooth 6 is in fact "real", physical Tooth No.#2, and

ref. tooth 4 is in fact "real" , physical Tooth No.#0   ?

Yes 6th tooth is physical Tooth No.#2.
btw help:
"low toothcount wheels like 4-1 (m031) and c004, c008 supported, without camsync too For 4-1, use h[1]=00 06 04 02 00 00 00 00 wether with or without camsync ("camsyncpretender" inside)."

OK, so that means that ONLY in second event, physical tooth 0 becomes Ref. Tooth No. 4, and
physical tooth 2 becomes Ref. Tooth No. 6 ?

This ref. table (at least to me) doesn't have any sense, beacuse with 4-1 toothed wheel, ECU knows exact position of missing tooth and can calculate easily 1st cylinder TDC position, so software can use only two opposite tooth with SAME numbers (0 and 2 for example), but as it's triggering in WASTED spark mode it should of course DOUBLE the events  - and too clarify this even better, I made four screenshots in AutoCAD for four different triggering events - it can't be easier to visualise triggering events then with this:

1st trigger event:

(http://img.photobucket.com/albums/v358/veliborv/VEMS_WASTED_SPARK_1.gif)



2nd trigger event:

(http://img.photobucket.com/albums/v358/veliborv/VEMS_WASTED_SPARK_2.gif)



3rd trigger event:

(http://img.photobucket.com/albums/v358/veliborv/VEMS_WASTED_SPARK_3.gif)




4th ignition event:


(http://img.photobucket.com/albums/v358/veliborv/VEMS_WASTED_SPARK_4.gif)



This seems logical triggering event sequence, correct me if I'm wrong...


Also, please, can you explain what's wrong with warmup enrichment - some setting aren't very clear, so I guess they are way off,?

Call me and i'll explain all.

OK,

Everything is solved  - problem was caused by new Denso spark plugs.  :o

It seems that problems started at time when we flooded and fouled our original, old NGK-R spark plugs and then we exchanged them
with new, fresh Denso set...

And when we started using Denso plugs, engine was immposible to start, run very badly, sometimes on all cylinders, sometimes on few cylinders, backfiring,
reseting ECU -all in all in last six months, we had very nasty & totally unpredictable engine behaviour that we unfortunately never related to new spark plugs...

That was realized after we connected our complete harness & VEMS ECU to another Yamaha R6 engine and we still had same problems, and then we somehow decided
to try with old, NGK-R plugs and suddenly engine started & worked flawlessly each time...

I was little sceptical about fact that this all problems were caused by spark plugs, so I checked part numbers and compared specs, but both NGK & Denso spark plugs have same specs, and they both are designed for this application...I suppose that with original Yamaha CDI ignition module it doesn't matter which plugs are inside, but as we are using inductive ignition fired directly from VEMS, maybe some spark plugs are little bit sensitive about this...

Btw, MW fire kindly corrected few things in our config and gave us few good advices concernig triggering settings- tnx!

How did you come to a value on the coil charge time?  You might need to increase the coil charge a little or even a lot.
Also running on fouled plugs can cause all manner of problems, we were doing a track car and couldn't get the CO right at all, put in new plugs, got the tuning spot-on...
After a quick grit blast the old plugs went back in and ran faultlessly again.

Quote from: [email protected] on June 09, 2010, 07:29:31 AM
How did you come to a value on the coil charge time?  You might need to increase the coil charge a little or even a lot.
Also running on fouled plugs can cause all manner of problems, we were doing a track car and couldn't get the CO right at all, put in new plugs, got the tuning spot-on...
After a quick grit blast the old plugs went back in and ran faultlessly again.

For now, configuration is not completely done - it is still very "raw", but at least engine starts right now each time, every time without problems - and that was impossible with Denso plugs...

We are unfortunately many months behind schedule, and right now our car needs lot of testing in extremely short time - next  FS 2010 event is starting in July 15th @ Silverstone, and that's only one month away  :-\.


Our complete ignition system is right now really, how should I say, exotic, but at least it works flawlessly to 16 krpm  ;)

- 24-1 toothed wheel (laser cut from 3mmm steel plate, tooth shape & depth exactly as std 4-tooth R6 wheel, missing tooth part was made slightly shallower and with smooth edges , hopefuly to lessen effect of large missing tooth amplitude with VR conditioning circuit ),
- Yamaha R6 VR sensor,
- Ford Zetec ignition coil (this is strange choice, but as I didn't had suitable inductive COP's, this was only wasted spark coil that I had in the moment).

As ignition dwell - it's not adjusted yet - it's around 3ms, but I think we could get with 2.5ms, we should see how it will perform on testing track next week.

Before, we also thought that problems could be also caused by too short dwell - and with Denso plugs we tried  up to the 4.5ms dwell but with no effect.





(http://img.photobucket.com/albums/v358/veliborv/FS_RITEH_2010.jpg)

^^ That looks like it could be real fun :)

Quote from: Sprocket on June 11, 2010, 07:46:48 PM
^^ That looks like it could be real fun :)

Yes, you could say that 200kg car + R6 engine +  RWD with Torsen diff  sounds like fun  ;D



http://smg.photobucket.com/albums/v358/veliborv/?action=view&current=FS_RITEH_RRCX1.mp4

http://smg.photobucket.com/albums/v358/veliborv/?action=view&current=FS_RITEH_RRCX1.mp4

I WANT ONE ;D

Whats the score with spectators at Silverstone for the Formula Student events?  I might see if I can sneak an afternoon off work and get over to see it.

Quote from: [email protected] on June 16, 2010, 06:58:06 AM
Whats the score with spectators at Silverstone for the Formula Student events?  I might see if I can sneak an afternoon off work and get over to see it.

Sorry for late reply: yes, the spectators can come and take a look, free of charge, this is C/P from Formula Student rules:

Spectators:

The event will be open to the public, free of charge, on Saturday 17 - Sunday 18 July. Entrance to the event will be signposted from the main entrance and there will be a designated car park (free) for spectators. As well as watching the cars undergo a series of dynamic tests, including Acceleration, Sprint, Skid-pan and the Endurance event, you will be able to look around the paddock and see the teams in the garages. There is no need to register in advance. Teams are welcome to bring guests on Thursday 15 and Friday 16 July but they will not be issued with a badge and do not need to register or pay.


If you've never been on Formula Student event, I encourage you to come and take a look -  you could be really,really  amazed about many hi-tech details, great amount of creativity (and sometimes great amount of money as well) that is included in making these little racing cars...all judges are from F1 world (or some racing related companies)...hey, some judges are even coming to Silverstone driving its own personal custom kit cars  :)

Feel free to visit us at paddock, and to take a look at our car (or the pieces that will left from next  Friday skidpad circuit testing :-) we'll be glad to offer you some beer  ;D

So here are the details:

- team number is 75,
- name of university: "University of Rijeka",
- and our car number is 055.

http://www.formulastudent.com/events/FS2010/teams.aspx