MS3X Starting Problems - MPi Trigger Wheel Sync (Now Working)
Turbo Mini Forums

yeah wired to A & C. i'm just looking into crank sensor angle

1275 N/A Sprite, 998 T2 Turbo Mayfair
1275 EFi Turbo

I found this:

http://www.msextra.com/forums/viewtopic.ph...t=rover#p334413

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

thanks paul. it seems my sensor is about 30 degrees before the first missing tooth but i think that is how the pattern works. so should be correct. (i need to find out who made the pattern)

so still not starting. to me it seems like it isn't sparking during cranking. I am running wasted spark using an extra bip on the main board. The LEDs only flash occasionally when cranking. If i remove the plugs and place them on the block and then crank they spark perfectly and the LEDs flash as i'd expect.

Why would my crank signal go to pot when under compression? too slow cranking speed? its about 240 rpm with plugs in and 360 rpm without plugs.

this is driving me crazy

1275 N/A Sprite, 998 T2 Turbo Mayfair
1275 EFi Turbo

Schrödinger's cat - so which one am I ???

thanks rod have been playing.

looks like the second gap each 180 degrees is the problem it doesn't pick it up as well as the first

Edited by dan187 on 22nd Jun, 2013.

1275 N/A Sprite, 998 T2 Turbo Mayfair
1275 EFi Turbo

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

hmmm i did wonder but have been going off how they wired the coil in this
I'll change that then.

Edited by dan187 on 22nd Jun, 2013.

1275 N/A Sprite, 998 T2 Turbo Mayfair
1275 EFi Turbo

Hmmm....

That's certainly not how my MS2 install is wired. As Paul says, my fuel pump relay only powers the fuel pump.

In practice it won't make much difference as the "fuel pump" relay won't energise (apart from the initial 2sec priming pulse) until sync is achieved.

Lack of sync is the root cause of your problem (if you've still got the indications on the red line of the composite log).

Have you got a photo of the back of the flywheel, there is something else I don't understand on your new tooth log...

EDIT - typo plus added a line.

Edited by Rod S on 22nd Jun, 2013.

Schrödinger's cat - so which one am I ???

no photo of my exact one but it is spaced as per this lotus one...

here is a tooth plot with no plugs in. you can see the 4 teeth clearly. the time between normal teeth is approx 5ms and missing tooth double this (as expected).

the trouble with the plot with plugs in (above) is that the first missing tooth lasts for 80ms (the second is 12ms and each normal tooth 6ms). so it is that 1st tooth of the pairs.

1275 N/A Sprite, 998 T2 Turbo Mayfair
1275 EFi Turbo

OK, the log immediately above and the photo make sense (the photo explained why the gaps ar not symetric) BUT, the log further up (plugs in) isn't showing all the "teeth".

Try it again on a longer timebase.

Don't worry about the height of the lines at the moment, I just want to see the correct number of the cycle repeating after 32 times (36 - 4) if it is possible to see the correct number under compression.

I don't really like the composite logger (I prefer a scope) but what would really help would be a datalog, both for the error codes and simple things like battery voltage.

I'll PM you my email address so if you want to send it over I'll have a look.

Schrödinger's cat - so which one am I ???

Thanks. I'll sort a log and send it. I've logged before and was getting error 22. Rover #2 missing tooth error.

1275 N/A Sprite, 998 T2 Turbo Mayfair
1275 EFi Turbo

OK, apart from the error code 22, I can't see anything else wrong (but I'm no expert...)

Hopefully Jean will see this and explain what "22" means on the MPI flywheel.

But the fact you are missing some of the normal teeth on compression does suggest it's an external (like wiring) issue rather than the MPI wheel code.

Battery voltage logs more than enough for a start so your loss of signal on cranking really does seem external (noise, grounding etc.)

Schrödinger's cat - so which one am I ???

22 is missing tooth error,

http://msextra.com/doc/ms3/syncloss.html

sugests the pots still need some work?

Yes Joe, just wondering if it has any specific meaning with the MPI wheel entered.

The blue logs seem to show the teeth after the long gaps are missing, but why ???

Does the code count these normal teeth on an MPI wheel ???

Schrödinger's cat - so which one am I ???

Thanks gents. I'll check through all the wiring to make sure it's clean. I tried swapping the polarity of the sensor but it looked the same. I may have to borrow a scope from work on Monday.

I've left this issue for the moment and welded in the bung for my second lambda sensor (was going to do it in the week once I had simply taken the car for mot today). All in place, I'm just wondering which sensor gets wired as the main reading? Inners or outers. I'm using Paul's base map so hopefully he can shed some light.

1275 N/A Sprite, 998 T2 Turbo Mayfair
1275 EFi Turbo

it looks a bit like its counting the gaps, not the teeth. if you take the spikes as long gap

yeah that chart is counting time between teeth.
These charts show teeth (with time as the x-scale)
As a test i've spun the engine with 2 spark plugs in 1,2 then 2 plugs in 3,4 then all 4 plugs in.
The errors occur 30deg BTDC of a relevant compression stroke (see the gaps move with the plugs)

Edited by dan187 on 22nd Jun, 2013.

1275 N/A Sprite, 998 T2 Turbo Mayfair
1275 EFi Turbo

It does count the teeth but it also checks for the gaps and if these are not at the expected tooth number, the error 22 is flagged.

This certainly looks like there is a need to adjust the pots on the crank signal but it would be very useful to actually see the signal on a scope because there might be more to it. One of the problems is the VR circuit: it is very adjustable so can be use on many circuit but due to the adjustability it can also be difficult to set because there are so many wrong settings for a specific setup.

And the wiring as shown is definitely the recommended setup. You should have your coil(s) and injectors powered by the fuel pump relay (plus any additional relays if needed). This way you always have the code in control of these even at startup: they are powered once the code has been initialised and the fuel pump has been activated. There are many examples of people powering coils independently from the main relay who were seeing sparks and a resulting ignition of the remaining fuel when turning on the power.

Jean

http://www.jbperf.com/

Are you actually letting enough air through to spin the engine with the plugs in?

What have you done about the IACV? If it is not working then you will need to use the throttle.

Have you also correctly set up the temperature sensors and TPS? If not then that will affect fueling at startup.

I always put the primary O2 sensor on the outers.

Saul Bellow - "A great deal of intelligence can be invested in ignorance when the need for illusion is deep."
Stephen Hawking - "The greatest enemy of knowledge is not ignorance, it is the illusion of knowledge."

Yeah throttle open (IAC no plugged in at the moment anyway)

It's not a matter of no fuel it appears not to sync.

I've borrowed a scope and attached the trace of input and conditioned signal.
There is 1.1V at the point where the signal gets lost in the conditioning circuit.

I suspect a bad component or solder joint. However, how best can i fault find this?

1275 N/A Sprite, 998 T2 Turbo Mayfair
1275 EFi Turbo

Matches the TS logs, try and shortern the timebase to see if the missing tooths are correct.

The drop in input voltage, and increase in duration on the compression stroke is tiny so I can't see how the output is being lost (yet).

Schrödinger's cat - so which one am I ???

That is really strange: your conditioned signal seems to be following the peaks instead of the zero crossings. That should not be. How are your pots set? Are they turned completely counterclockwise? Are they installed the correct way in? That would result in needing to have the pots turned completely clockwise.

Edited by jbelanger on 24th Jun, 2013.

http://www.jbperf.com/

In addition to Jean's thoughts, try this whilst you have the scope borrowed
http://www.diyautotune.com/tech_articles/v...rt_vr_index.htm

Schrödinger's cat - so which one am I ???

i'll investigate your last few points bear with me.

edit: wrong image

Edited by dan187 on 24th Jun, 2013.

1275 N/A Sprite, 998 T2 Turbo Mayfair
1275 EFi Turbo

pots fully clockwise gives no sinal on VRout.

what i've noticed from the close up plot is the conditioned signal falls when the input is at it's peak, so it then follows that the lengthy high on the conditioned signal hasn't fallen because the next peak on the input is too low

1275 N/A Sprite, 998 T2 Turbo Mayfair
1275 EFi Turbo