Odd-fire coil control

[AdmiralSenn]

Hi guys. I'm not yet an MS-Extra user - heck, I'm still working on getting my standard fare MSII V3.0 installed and working - but I may be upgrading to extra in the future, and I like to plan way ahead so I know what I'm doing in advance. My ultimate goal is to remove the ignition computer from the car and run as many functions as possible through the MegaSquirt box. Ignition is the last one I need to figure out how to eliminate.

I'm still a newbie, so I apologize for any stupidity in this post.

I guess I'm a bit behind the times, as I just found out recently that MS2-Extra supports odd-fire direct coil control. This is good news for me since I already have a recently purchased, quality coil and I'd rather not have to run EDIS and buy and fabricate everything to make it work - not that I couldn't do it, I'd just like to keep what I have if possible.

Unfortunately, the documentation that I can find doesn't really cover the odd-fire setup too well, or indeed at all. I'm just looking to verify that if I switch to extra, it will work with my distributor/coil setup. My concern is that I don't know how the system will know when it's seeing cylinder 1. I've had the distributor apart, but not recently, and I just can't remember anything in the mechanism that would differentiate cylinder 1 vs the others, either by a stronger or weaker or missing pulse, or any other method. (I'll feel really stupid if the system doesn't need to know TDC #1, but if that's the case I'll need it explained since I can't figure out how that would be possible - somehow it has to know when to fire the small crank angle vs the large one, right?)

I believe the distributor does use a VR sensor, which tells the stock ignition computer how to calculate dwell (all advance is mechanical or vacuum). The car does not have a camshaft or crank sensor.

Here I'll show my advanced noobishness: what do I need to know/learn in order to figure out whether my distributor will work as-is?

Thanks!

[jsmcortina]

EDIS wouldn't work as it is even-fire only.

What engine is this? I know that the oddfire code has been used on a V6 Chevy that has a 150/90 deg firing angle. The code looks for the different sized gaps between the input pulses and figures out which cylinder it is on. So if the reluctor in your dizzy has uneven lobes on it, that is probably the answer.

As you are hoping to move to MS2/Extra in the future I would suggest you start with it now. There may be slight differences in tune between the firmwares due to the calculations and if you start with MS2/Extra now (get the 2.1.0 release) then you will have a fuel tune that is already done.

hope that helps

James

[AdmiralSenn]

Ah, I didn't know EDIS was even-fire only.

This is for the PRV V6, better known as the Volvo B28F, that lovely paragon of bizarre engineering. Small angle is 90, I can't remember offhand the other one and I don't know how to calculate it (doesn't seem to be in my manuals either) but 150 does sound familiar. Seems like the code just needs the small angle anyway, though, right?

I just did some reading on odd-fire engine theory (http://home.off-road.com/~merls_garage/oddfire.html for anyone who might need it) and I feel kind of silly now. Don't ask me why, but for some reason I forgot that the sequence goes large angle - small angle - large angle etc, so it doesn't matter which cylinder is firing, just that the system can tell the large crank angle from the small. I had somehow gotten it in my head that the engine fired 90-90-90-150-150-150, don't ask me how, and never mind that it is probably impossible to get an engine to run in that configuration, even if someone was deranged enough to build one!

This is why I don't work on the car when I'm tired... or after I get home from work.

So that solves that question!

I guess I'll download the extra code and start tuning with that once I finish my harness and sensor setup.

From the documentation it looks like if I'm not running ignition yet then I don't have any hardware mods to make to convert over, right? I'll have to open up the box and make modifications when I start fiddling with the ignition anyway since coil negative input uses the wrong input circuit (hall/opto vs vr), so I can make any changes for ignition control at that point (jumpers, and I don't think I installed the BIP373 either). Once I get the fuel working I'm looking forward to eliminating a few more old components from the car and streamlining it further.

Thanks!

[jsmcortina]

Sounds like you have it covered. For oddfire fuel only, you can set the spark mode to "Fuel only" and ensure you enabled Oddfire on the engine constants.

James

[ScoobyCoder]

James,

Following on from this, I have been given the task of installing a new ECU on a Metro 6R4 I'm going to keep using MS2 extra as I prefer it over the Ms3 stuff... anyway.. It currently runs a dizzy and is apparently a 90/150 firing order (the dizzy has odd spacings)

So two questions

Can I run this with the MS2 Extra code, retaining the dizzy and use a 36-1 trigger disc (would I need a CAM sensor or would this be suffienet)? and if so, any idea on base ignition settings.

AND

Could I go over to wasted spark, does the oddifre code support this,. would I need anything more than the 36-1?

Any help, as always very much appreciated

Mark

[jsmcortina]

Wasted spark won't work on an oddfire V6 as the cylinders are too far apart. (Too odd.)

Yes I think you will need a cam sensor to pick the correct phase.

James

[ScoobyCoder]

James,

OK. Just didnt want to start cutting the loom up without being sure. There is a cam sensor on the engine, so with phasing detected, odd fire (of this type) is supported in the MS2 EXtra 3.1.0 firmware?

Thanks,
Mark

[jsmcortina]

Use the current firmware, but yes it should be supported.

James

[ScoobyCoder]

Ok James thank you once more

We will be fitting a 36-1 to the bottom pulley.

We will also fit a VR sensor to the cam, and have a single cam trigger tooth.

Does the relative position of the cam trigger matter? Looking through oddfire settings, we can specify the small angle of 90, and I'm guessing its toothed wheel (Dual) with missing tooth as overal settings. Is the cam sensor only used to determine which cycle of the 36-1 we are on, using the 36-1 as the angle finding, so in effect position of the cam sensor does not matter?

If thats not clear let me know and I will reword, but I think you probably get what I mean

Cheers,
Mark

[ScoobyCoder]

James,

After reading the dual wheel missing tooth guide, it looks as though indeed, it acts as I thought, with the lobe on the cam indicating to the MS which phase we are in, with no timing effect. (The missing tooth determines this, as per normal.)

My next question...

This engine as you know is a '90/150' firing type. How does the ECU know which comes first after the missing tooth/offset calculation. Is it 'the norm' for it to be the small angle first, so it goes MISSING TOOTH, CALC ADV. AGAINST TDC using the 90*, then next time 150*... or am I just over complicating? Do all these odd fire engines have the small angle first (between our missing tooth and next event, or is it the big angle first? Does it matter, as the dizzy is directing the spark?... arghh confused, though this is the last area of confusion as today I have made up the cam bracket and sensor, and the trigger wheel will be here tomorrow for me to fit, following the guide to get the rough angles correct....

Thanks in advance

[ScoobyCoder]

in addition to the above...

I gues if we find that the gaps are, and Im talking firing of plugs here not actual firing order No1....90*....No2.....150*....No3......90*.....No4.....150*....No5 etc then yes, base don our missing tooth offset, and the ECU firing small angle first then that will fine, if it is the opposit... (No1....150*.....No2....90*....No3....150* etc) then I guess we can just change the crank offset, or time to what is in effect No2, such that the small angle occurs first... am I right in thinking that?

Cheers, Mark

[karlo]

I have run the Maserati C114 odd fire V6 successfully with both fuel and ignition. This engine is very similar to the PRV (both are 90/150), so you shouldn't have any particular issues. I have run them with a single trigger wheel running at cam speed, and crank speed trigger wheel + cam trigger (just a crank trigger wheel won't work, as you need to know the phase if you want to run ignition). Both single distributor and dual distributor work (although single dizzy has a dwell issue, that I have reported). I haven't tested COP, but I expect that should work fine. Wasted spark will not work on these engines.

[ScoobyCoder]

Karlo,

Thanks for the positive feedback

Could you attached your MSQ for a single Dizzy, 36-1 on crank with cam single lobe?

In fact I am guessing its ust a case of small angle 90, odd fire, 6 cylinders, 36-1 with tooth no1 as per usual etc?

regards,
Mark

[jsmcortina]

"Oddfire small angle" is actually angle after cyl#1. In your case I think that's still 90 deg.

James

[ScoobyCoder]

I have run the Maserati C114 odd fire V6 successfully with both fuel and ignition. This engine is very similar to the PRV (both are 90/150), so you shouldn't have any particular issues. I have run them with a single trigger wheel running at cam speed, and crank speed trigger wheel + cam trigger (just a crank trigger wheel won't work, as you need to know the phase if you want to run ignition). Both single distributor and dual distributor work (although single dizzy has a dwell issue, that I have reported). I haven't tested COP, but I expect that should work fine. Wasted spark will not work on these engines.

Karlo,

We are still having issues apparently.

We ran the engine with a 36 tooth bottom trigger (no missing tooth) and single top cam trigger. The engine would rev lovely up to around 4000rpm then power would just drop off to almost nothing for the rest of the rev range, and this was on the dyno at castle combe, not a mapping issue, as the fuelling and ignition remained fairly steady. No loss of sync or anything obvious. Is this the dwell issue perhaps?

There was actually an obvious falter at around 4k...

Next we tried removing every other tooth from the bottom trigger (crank), left cam as single trigger, to make an 18 tooth wheel, now the engine struggled up to around 4k (almost no power) but then went like a rocket, making nearly 400bhp on the dyno. So now the top end is OK, the bottom end is not right.

We have now gone to a 36-1 on the crank and single cam trigger, and the engine runs, but we appear to not be able to time one side of the engine up. Once side is dead on the commanded advance up to 7k, the other side is 60* out.

Also, the plugs on the right back are one colour., the ones on the left are a diff colour, indicating that one side is firing 'on time' the other at a different, again pointing me at the oddfire stuff not working correctly.

We are using 3.1.0 release.

Could you send over your maps you used so we can at least go from a goo starting point, and what version of code were you running?

Regards,
Mark

[ScoobyCoder]

I have just looked at your bug you found karlo,

viewtopic.php?f=91&t=43924&p=311001#p311001

This would appear to be EXACTLY what was happening on the rolling road, power completely dying at higher RPMs (ie dropping a bank). This may well be the exact problem we are seeing!

If you could please post your compiled S12 file, and basemap, I will give it the acid test on this car, unless James this has been fixed in the new revision?

Mark

[racingmini_mtl]

Use v3.2.3: the fix for the problem in your link has been implemented.

Jean

[ScoobyCoder]

jean,

Thanks, I will give it a go tomorrow and report my findings. The dyno is booked for the end of the week.

Mark

[karlo]

3.2.3 may well fix your issue. The bug only affects single coil setups and will drop a bank entirely once dwell has to be shortened due to high rpms. (Be sure that you have a hot enough coil to light the fire at higher rpms).