Odd-fire trigger preferences

[MRRPM]

I have a 80 degree, odd-fire V2 engine, 4stroke, it has a 36-2 tooth wheel on the crankshaft (VR), and a one tooth cam sensor (HALL). I had it running on the stock triggers for years, ran great, but I had to change trigger wheels, for reasons not relating to the ecu. My question is, what would be the best trigger setup to fire this thing with? What would be everyone's dream setup, if they were gonna run this? I can make whatever trigger wheels I need, no problem. Should I go right back to stock triggers? I know this ecu can't do wasted in this setup at all, so we need the cam trigger.

Right now, I have a 36 tooth wheel, none missing, and one tooth cam sensor, can get rpm signal, but timing is anywhere, when spun over with starter, consistent as long as cranking, but stop cranking, and spark is anywhere. I was gonna do a trigger like this, but it doesn't seem to work.

[Simon]

The ultimate would be.

A cam sensor hall effect that is looking at a very wide tooth a 180deg tooth is ideal.

On the Crank then have 12 teeth with one removed (11 remaining teeth)

The missing tooth needs to be clear of either edge of the cam signal.

[MRRPM]

Ok, thanks Simon.

If I understand you correctly, I would mount a cam wheel that is basically a half moon shape, but the ends of this half moon shape would or could be anywhere, but not occurring when the crank sensor is reading the missing teeth gap.

In other words, the cam sensor would need to be halfway in the moon shape, or 180° opposite that,, while missing teeth are going under the crank sensor.

I will try this trigger setup sometime along, this is a new engine, having other issues at the moment, gotta pull the engine out now & fix a few things, then back in, with the new trigger setup.

[Simon]

Yep spot on.

[Gsab]

If I may piggy back here guys.

I had a 24 tooth wheel made for the crank and a single reset hall trigger on my RB25 exh cam.

I did the 24 tooth as it was recommended elsewhere on this forum.

24 teeth are equidistant around the wheel.

It never was totally successful and I then went back to the std Nissan cas.

My crank wheel does NOT have a missing tooth. Could this be why the setup was not quite that successful ?

The symptoms were bad starting, poor cyl firing and misfiring at anything over 3 to 4 k revs.

On the std cas it is perfect.

[Simon]

Hi Gsab

On a 24 with 1 sync the most likely issue would be if the sync edge is occurring close to a crank trigger.

You want to make sure the sync signal is halfway between the teeth on the crank.

Shouldn't need a missing one.

[Gsab]

Thank you Sir. Will look at that.

[MRRPM]

Ok, I have my engine back together now. I just slapped the 36-2 crank setup on, and the stock cam trigger, just to get it running. I get a lot of trigger errors, and lotsa popping & banging, and, of course, the errors coincide with the popping & banging. All the errors occur at around 5-6000rpms. What I don't understand, is why it runs good, until rpms come up? Why would errors occur at speed? What exactly is a "trigger error"? What is happening that the ecu does not like? I have read & learned a lot about trigger setups over the years, still don't quite understand how they work, & what is an error.

Tomorrow, I am gonna put a trigger setup on as Simon suggests, and I hope it runs with no errors. If it does, what have I learned? I would like to try & learn why the ecu likes some triggers better than others, & what are the do's & don'ts of trigger setups. I have read everywhere, and have yet to see anyone elaborate on triggers very much, like what works & what doesn't.

Maybe the factory triggers work good because their signals on the factory ecu are handled in a different way? Maybe the Vi-Pec doesn't read the triggers the same way? Maybe there's a better setup on the Vi-Pec that I'm missing? I know the latest thing on factory ecu's is using crank acceleration, & deceleration to calculate where to put the spark. The factory setups are good enuf that they need no cam sensor, on some models, and are non wasted spark. I don't think very many aftermarket ecu's can do this, yet, if any can. I think there is a readout on the Vi-Pec that shows crank acceleration? Or am I misreading something? Is this a coming feature?

[Dave Kriedeman]

Hi MRRPM,

What type of sensor are you using on the crank mag reluctor or hall effect.

If you post your PCL file and explain exactly your crank tooth position relative to TDC and the crank sensor and the same for your sync or cam sensor relative to TDC.

We should be able to work this out.

Also do you own an oscilloscope.

As for trigger tooth counts and errors etc.

A basic introduction for you, the crank tooth count must be able to be devided into 360 degrees evenly, no decimal points.

If you are using odd ball tooth counts with missing teeth or , strange sync tooth counts etc, then firmware must be written if possible for that particular tooth decoding to work correctly.

For example your new chopper wheel has 36-2 teeth, 360 / 36= 10

but 36- 2 = 34 : 360 /34 = 10.588

So how can this work, the software algorithyms have been written to except this particular tooth count and recognise the missing teeth as tooth counts.

This is why we set it up as missing tooth.

We have to tell the ECU the number of teeth that SHOULD be there (36) and then we tell the ECU how many teeth are missing (2).

Your sync sensor being hall effect, may have to trigger on the rising edge or the falling edge or on some setup the sync must trigger on the rising and falling edges.

Having a good understanding of different trigger patterns and setups can be very complex

I have been doing this stuff for 20 years now and every now and again I get caught out myself.

However, I must say that setting up triggers custom wise ViPEC has been the easiest ECU to create a trigger pattern to suit custom wise physically and then make it work via the software, being so configurable.

Regards

Dave.

[MRRPM]

Hi MRRPM,

What type of sensor are you using on the crank mag reluctor or hall effect.

If you post your PCL file and explain exactly your crank tooth position relative to TDC and the crank sensor and the same for your sync or cam sensor relative to TDC.

We should be able to work this out.

Also do you own an oscilloscope.

As for trigger tooth counts and errors etc.

A basic introduction for you, the crank tooth count must be able to be devided into 360 degrees evenly, no decimal points.

If you are using odd ball tooth counts with missing teeth or , strange sync tooth counts etc, then firmware must be written if possible for that particular tooth decoding to work correctly.

For example your new chopper wheel has 36-2 teeth, 360 / 36= 10

but 36- 2 = 34 : 360 /34 = 10.588

So how can this work, the software algorithyms have been written to except this particular tooth count and recognise the missing teeth as tooth counts.

This is why we set it up as missing tooth.

We have to tell the ECU the number of teeth that SHOULD be there (36) and then we tell the ECU how many teeth are missing (2).

Your sync sensor being hall effect, may have to trigger on the rising edge or the falling edge or on some setup the sync must trigger on the rising and falling edges.

Having a good understanding of different trigger patterns and setups can be very complex

I have been doing this stuff for 20 years now and every now and again I get caught out myself.

However, I must say that setting up triggers custom wise ViPEC has been the easiest ECU to create a trigger pattern to suit custom wise physically and then make it work via the software, being so configurable.

Regards

Dave.

I have a VR crank sensor, hall on cam.

As far as crank trigger wheel, I can make whatever I need, as I have a couple Mitsubishi lasers here in house to cut whatever I need. This is why I am asking what is the perfect setup in this thread.

I do have a scope, pc based, works pretty good.

I am still not quite understanding how the cam sensor works to sync up. Does the ecu keep up with crank position via crank trigger teeth, then know if it's on compression by the trigger being tripped, on the cam? If so, why does Simon want the cam trigger wheel to be half way around? This would likely work good, but would be a real pain to get balanced, as the cam turns 5000rpms, and a piece of steel half way around would throw it out pretty good.

[MRRPM]

Here is my PCL file.

I have a 12 tooth wheel, 1 missing, 1 tooth on the cam, now, as suggested by Simon. It is not a full half moon, like he said, but it's about 1/2 inch wide, and I am getting trigger signals, both crank & cam, according to the ecu. My spark is erratic, and not in the same place when winding it over with the starter, no plugs in. What am I missing?

1288 turbo.pcl

[Simon]

The benefit of the 180 sync is that we are able to sync the engine with in one turn of the crank.

We look to see if the signal is High or Low on the cam as the gap on the crank goes past. We then know if #1 cylinder is on an exhaust or compression stroke.

Where as on a narrow tooth we have to look for one edge and depending where the engine stopped it could take up to almost full 2 turns of the crank before we get a sync position.

Looking at your PCL file the arming voltage on trigger one is really low and will make it susceptible to electrical interference. You ideally want to measure the signal with a Scope and set the arming voltages to 50-70% of the peak voltage at each RPM.

Are you seeing a stable RPM reading with no trigger errors. Is the falling edge signal of the cam clear of the tooth after the gap on the crank?

[MRRPM]

UPDATE:

I was having erratic spark, reversed VR sensor polarity, got rock solid timing, with no trigger errors. I adjusted the ignition offset, fired right up.

Makes sense about the half moon cam wheel. It does, at times, take a rev or two to fire up, but this is all race, & that doesn't matter too much. I do get trigger errors as it's syncing up, if it winds over much. If it fires right up, I don't get errors. After it's running, there are no errors, as of now. I will be dynoing it tomorrow, see what it does.

The 12-1T crank wheel seems to deliver as solid timing as my 36-2T did. I would have never thought that. Will update with any problems that come up, during dyno session.

[Dave Kriedeman]

Hi,

12-1 works well.

I ran 12 - 1 on a V6 2 stroke with sequential fuel and sequential ignition via 2 x M & W CDI units.

Like I say the capabilities of these ECU's is outstanding.

Great to hear all is panning out for you, good luck.

Regards

Dave Kriedeman.

[MRRPM]

Hate to resurrect an old thread, but this is tied in with this setup. 

I have this thing back on dyno, once again, odd fire V2, turbo, burning methanol, runs great until around 6000rpms, then lotsa popping, sounds like a rev limiter, put timing light on it, #2 cyl timing seems correct, where #1 starts to advance up to 45˚, where I have it locked at 30˚, in ecu. No trigger errors, when running, which is normal. It will make errors cranking, if it kicks back, but I've always considered that normal.

  What could cause ecu (or something) to advance timing? I could see sporadic timing, but not solid advancing. I'm running 2 triggered coils, one for each cyl, could a coil be bad?

Edited October 31, 2015 by MRRPM

[Simon]

Does it move slowly up or does it do it in a jump?

And is it a fault that has developed or one that might have always been there?

[MRRPM]

Moves slowly, it has a blown head gasket now, I can't test anything right now, for a fix. 

Not sure if it has always been there, but it has ran good, with this new trigger setup. I don't run this thing much, it's a dyno queen, I do mostly stock ecu stuff, on customers machines, I use this to try things.

It's weird that the #1 cyl could be off, and #2 be spot on, seems if #1 is off, #2 would also be off. I guess timing is not calculated like I think it is.

Maybe timing is actually retarding? Going all the way around to overlap? Maybe advancing until it flips to overlap? It's been a year since I've ran this, I'm prolly using outdated firmware too.

[Simon]

It would be good to get the latest firmware in there just in case. But I don't know of any previously found odd fire bugs.

The main issue around odd fire originally was that we couldn't do it in a wasted spark format.