EFI Coil on plugs [Archive] - DMCTalk Forum

Spittybug

12-04-2012, 10:46 AM

Still just kicking around the idea, but as most know, that ends up crystallizing into action pretty quickly if I'm not careful...

I'm seriously considering adding a toothed gear to the crank and to the cam so that I can index TDC for #1 cylinder. Using Megasquirt's oddfire capabilities, this would allow me to totally do away with the distributor and coil and replace them with coil-on-plugs. Modifications to the MS unit are fairly straightforward; adding 5 coil driver chips on a heat sink and running wires out. Not expensive either. I can go either with a gear under the timing belt cover or on a shaft into the old dizzy hole. I would prefer hidden for a cleaner look, but it may be more practical in the dizzy hole. I suppose I could get someone's old distributor, cut off the shaft and weld on the gear.....

Why? Why not? No more potential for distributor slop, rotor wear, cap replacement, static advance issue, bad VR coil, etc.. ballast resisters are already gone on mine. For someone that is looking to go full spark control right away, this would be the working prototype. Remove distributor, not simply disable the advance. Besides, I like to tinker!

My question therefore is what modern car has got COP that I should look for in the junkyards? Are COPs all created equal? Some more reliable or better than others? Alternatively I could look at a coil pack mounted somewhere central (back side of my intake perhaps) and run short plug wires. I'd have to learn more about wiring one of those up.

Productive suggestions? Thanks.

opethmike

12-04-2012, 11:37 AM

I've been toying around with the idea of going to a coil pack setup using an Electromotive ignition ECU.

vwdmc16

12-05-2012, 10:37 PM

Spittybug

12-06-2012, 04:49 AM

Worked on a Suzuki GSXR 600 today, I Think the ignition coils might be perfect for a coil on plug set up, the coil itself is quite compact and they are available cheap on Ebay Just need a bit of trimming as they are a bit too long. I even had my car at work and I could have test fitted one, too bad I just now thought of doing that.

Cool. When you say trim them, is that as simple as clipping the bottom off the boot do you think? I've never played with COPS before so I don't have a feel for it. Do they require screwing/fixing the coil portion or are they rigid enough to simply "stand" on the plugs?
Thanks.

Citizen

12-06-2012, 06:45 AM

My question therefore is what modern car has got COP that I should look for in the junkyards? Are COPs all created equal? Some more reliable or better than others? Productive suggestions? Thanks.

I think (but I'm not 100% sure), that my parent's 2006 Chevy Cobalt had them. It's a 4-banger I know, but I'm not sure of the engine displacement. Unfortunately, I can't verify anything right away for you because I don't see the car but once per week, and I don't have any real experience with COPs anyway.

Although this particular car is running just fine, you might have a better chance of finding Cobalts in junkyards than other cars.

Hope this helps.

Thomas

...

Dangermouse

12-06-2012, 08:22 AM

My old 1999 Chrysler Concorde had them too, and they are plentiful in junkyards.

dmc6960

12-06-2012, 08:54 AM

My old 1999 Chrysler Concorde had them too, and they are plentiful in junkyards.

The Concorde/Intrepid/300m coil packs have a large square coil on the top. I do not believe they would readily fit into the DeLorean's plug wells. I could always try it for fun though...

15248

DMCMW Dave

12-06-2012, 10:13 AM

Keep in mind that whatever you come up with must seal the spark plug holes at the top (like the OEM wires), unless you never plan to get the car wet. Ever.

vwdmc16

12-06-2012, 04:30 PM

Cool. When you say trim them, is that as simple as clipping the bottom off the boot do you think? I've never played with COPS before so I don't have a feel for it. Do they require screwing/fixing the coil portion or are they rigid enough to simply "stand" on the plugs?
Thanks.

Yes you should be able to cut off the bottom 2 inches. Thats just a rubber boot. The top has a nice collar that should fit near perfectly over the top of the steel sleeve of the plug well. The coil is just held in snug by that top section on the bike so I believe it should be the same in the prv. The actual coil section on top is only a bit bigger than your thumb, very compact.
I found a set on ebay, $10 buy it now plus 5 dollars shipping. Cant go wrong there. You may be able to find some still with the denso pigtail attached.

Spittybug

12-15-2012, 08:25 PM

Worked on a Suzuki GSXR 600 today, I Think the ignition coils might be perfect for a coil on plug set up, the coil itself is quite compact and they are available cheap on Ebay Just need a bit of trimming as they are a bit too long. I even had my car at work and I could have test fitted one, too bad I just now thought of doing that.

http://ecx.images-amazon.com/images/I/41c19j1qygL._SL500_AA300_.jpg

You're right, these are cheap on Ebay. I would be extremely interested in seeing if they would fit. Do you have the ability to check at all? I can't think of any reason why a motorcycle COP would be any different than one for a car.

vwdmc16

12-15-2012, 11:20 PM

Spittybug

12-16-2012, 09:30 AM

Yeah I could, I have a bike at work I need to do a valve cover gasket and the coils have to come off, Just gotta be sure that I drive my D to work that day. Ill let you know when Ive done it, Ill get some pics.

That would be great, thanks. Dave S.'s comment about the top boot fitting over the hole collar is well taken too.... making something like that might be challenging, even cannibalizing a set of plug wires to use theirs.
Do you see any reason why a bike's plug coil would not be transferrable to a car? Surely they must require comparable voltage levels, reliability and longevity. Dwell time and timing are all controlled by the MS unit (or stock system on vehicle) in both cases, so the COP is serving the same function. Is the plug different? I haven't had a bike in 30 years, so I don't know.

Bitsyncmaster

12-16-2012, 02:43 PM

My guess it those coils take a lot less current than a one coil system since they would only fire 1/6 of the time and the on time "dwell" would be 6 times. You can also take a resistance reading of some different types to know what current drivers you will need.

vwdmc16

12-16-2012, 05:02 PM

Great point Dave, I think they should work, they are 12v and yeah they probably pull less current than a larger automotive one, but they still have to be strong enough to light off a high compression engine. The other point is that is a 14-16k rpm engine, those coils probably have an incredibly short saturation time to fire off that fast but these engine still idle at 1k rpm so there isnt an issue of them operating too slowly. Ill see if I can get a Ohms reading too, Ive read on line that the primary windings are 1.5ohms

Spittybug

12-18-2012, 06:03 PM

Does anyone have a good picture of one of the cam shaft timing chain gears once the cover is removed? I've never had mine off and the book pictures aren't very helpful. I need to determine if there is the means to attach a one-toothed gear and hall sensor inside the cover. If not, then I will need to get a distributor shaft (worm gear on end) and get my cam position from there. I'm sure someone has a broken distributor and would be willing to part with it. The crank wheel (35 tooth, one missing) shouldn't be too bad to mount I don't think. While the pulley is dished, I think I could use spacers and get a longer bolt. Or maybe mount it on the engine side of the pulley? Or just tack weld it on. I'll have to crawl under and take a better look.

Anyone know of an EFI'd COP'd PRV engine in a Delorean? Me neither. 8) I hate it when I get bitten by the bug.

opethmike

12-19-2012, 10:45 PM

vwdmc16

12-19-2012, 11:12 PM

I was able to get a coil off. It seems they are a bit larger than I thought but these are still must thinner than automotive COPs. With all the rubber boots off the dimensions are 5.25 x .75.

So it seems you probably cant cut them down. I think the windings go down into the base of the body not just the head. In the picture where its mounted on the sparkplug, its not fully clipped on and could go down another 1/4" but still there is no way these would clear the stock air box, and if you have a large and low mounted fuel rail that might hit too. However If you were to slide the upper rubber boot down, it would fit snug in the spark plug well, so an additional mounting brace is not needed.

http://i31.photobucket.com/albums/c396/vwdmc16/dmc/20121218_121959.jpg
http://i31.photobucket.com/albums/c396/vwdmc16/dmc/20121218_122344.jpg
http://i31.photobucket.com/albums/c396/vwdmc16/dmc/20121218_122053.jpg
http://i31.photobucket.com/albums/c396/vwdmc16/dmc/20121218_122112.jpg
http://i31.photobucket.com/albums/c396/vwdmc16/dmc/20121218_122148-1.jpg

Spittybug

12-20-2012, 04:40 PM

Thanks Mike and Wolff. Putting a timing gear under that cover looks like it would be a challenge...... looks like plan B of using the existing dizzy hole may be the best way to go.
I don't have the stock airbox or intake manifold on my car, so I have virtually unlimited clearance above my spark plugs. It is good to hear that the upper rubber boot fits nicely. These look promising!

Thanks guys!

NOW SEARCHING FOR A DISTRIBUTOR SHAFT OR OTHERWISE UNUSED DISTRIBUTOR FOR PROTOTYPE WORK.....

opethmike

12-22-2012, 03:45 PM

Yeah, it is REALLY tight under there; I just don't see how any sensors or gears could be added. Though, I'm curious why you need a wheel on both the crank and the cam; for example, Electromotive uses a single 60 - 2 wheel on the crank.

vwdmc16

12-22-2012, 05:24 PM

Isnt it because while the crank is spining twice as much and giving twice the data, the computer doesnt know which cycle each cylinder is actually on. Is it on compression or exhaust? engines dont run when timing is one rotation off .This doesnt matter with a waste spark system that fires on both of those cycles on an opposite pair of cylinders, because one of those does need the spark and the one on compression conducts the electricity better so it will get the spark while the other cylinder doesnt. But if every cylinder has its own coil and were to fire every rotation, the chance of a missfire during an exhaust stroke is very possible and would create an explosion in the exhaust if there were a fuel charge in the cylinder that wasnt fired 360 degrees ago. That is more of an emission issue that doesnt concern us but still the computer would be firing one rotation off and wouldnt know it.

Spittybug

12-22-2012, 05:42 PM

For oddfire engine, you can't use wasted spark. Each plug is 90/150/90/150/90/150 degrees apart. You need the cam sensor to tell you if you are on compression or exhaust stroke, crank to tell you 10* increments of movement (35 teeth, one missing at TDC) to properly set the timing to #1 for TDC. I think it could be done with a single perimeter toothed gear and a 180* "groove" like a smiley face, in the face of the gear. Mount that on a distributor shaft and use two hall sensors. The groove would act a "0" for the cam and "1" when the groove is absent (identifying compression stroke) and the teeth would be for ......oh crap....each tooth would now be 20* of crank, not 10*. It would need to be a 72 tooth gear which may be too tight to put into the empty distributor shell.

Still working on the best physical setup obviously!

Bitsyncmaster

12-22-2012, 08:51 PM

There must be some encoders indicating shaft position. Probably be a simple interface and read the value even when the engine has stopped. I would think you could not use a missing tooth sensor unless your software waits to fire the engine only after a few revolutions.

Spittybug

12-23-2012, 12:34 AM

nkemp

08-08-2013, 10:36 PM

Anyone know of an EFI'd COP'd PRV engine in a Delorean?

Here is a video of one (looks more like CNP): http://www.youtube.com/watch?v=ohVm3VhkGsU Not only is it COP but it runs on propane (not sure if it is full time or a mix of propane & gas) ... but that is another story.

I contacted them about the details and here they are ... drum roll..
You have to have a cam input.

We have cars running on distributor and coil per plug.

It is configured as V6 odd fire.

More PRV V6 COP DIS ... EIEIO thoughts:

The flywheel has a sensor that reports TDC C1 and the output is located in the diagnostic plug.

Is the flywheel keyed so that when it is put back on it always has the TDC C1 located properly?
The sensor doesn't sense teeth so the resolution is REALLY low.
QUESTION: For every rotation, the sensor indicated TDC C1. Is that once for compression and once for exhaust? Even though it is an odd fire engine, the compression and exhaust are still 360 degrees apart. Correct? Something someone said made it sound as if it was not the case but it sounds way wrong.

But if every cylinder has its own coil and were to fire every rotation, the chance of a missfire during an exhaust stroke is very possible and would create an explosion in the exhaust if there were a fuel charge in the cylinder that wasnt fired 360 degrees ago. That is more of an emission issue that doesnt concern us but still the computer would be firing one rotation off and wouldnt know it.

Why can't we fire each cylinder on both the compression and exhaust (assuming that MS would do that and if so how do we config MS to do so?)?

Isn't that similar to normal even-fire wasted spark dual coil, dual opposing cylinder? In that case one cylinder is in compression and the other is in exhaust the difference being that they are different cylinders.
Set up as MSExtra or MS3, 6 ignition outputs, COP(or CNP)

If we fired on both the compression stroke and the exhaust stroke no cam position is required.

Again, can MS be set up to fire every compression and exhaust per cylinder?

At one point I thought it might make sense to config MSExtra as a V12, wasted spark, odd fire but I'm not sure that helps.

nkemp

08-09-2013, 08:58 PM

I asked DIYAutoTune about COP, DIS, Odd Fire V6 and here is the summary of their comments:
-The MS3/MS3X has a wasted spark COP option
- The wasted spark is every other stroke(compression , exhaust, compression exhaust, etc). Individual cylinder spark control.
- Use a 36-1 Crank trigger
- At this point it is unconfirmed and you may need a 1 tooth cam trigger added to the dizzy and run then as full sequential
- Anyone want to confirm this?

As such I think this sounds real promising. Coil near plug such as off the LS1 engines may be easier to install under our air filter box than actual COP. All said and done not real inexpensive but doable.

Nick

SS Spoiler

08-09-2013, 09:27 PM

I love it when the sharpies in the group push the envelope. Good thinking Nick, I think [I didn't follow half of that but it sounded neat]

Paul Cerny #2691

nkemp

08-09-2013, 09:41 PM

Good thinking Nick

Sundance Kid: "You just keep thinking, Butch. That's what you're good at."
Butch Cassidy: "I got vision, and the rest of the world wears bifocals!”

Spittybug

08-09-2013, 09:51 PM

If you have both the crank and cam triggers, there has never been any debate that Megasquirt will allow you to go full sequential. I think you just confirmed a round wheel will roll..............;)

The issue has always been one of having to establish not only TDC of #1, but teaching the system if you are on compression or exhaust stroke. So yes, a form of 1 tooth cam wheel is needed. Wasted spark has never been an option since there isn't an "event" that is 180* from another.

nkemp

08-09-2013, 10:25 PM

The issue has always been one of having to establish not only TDC of #1, but teaching the system if you are on compression or exhaust stroke. So yes, a form of 1 tooth cam wheel is needed. Wasted spark has never been an option since there isn't an "event" that is 180* from another.

The "wasted spark COP" does not need a crank & a cam. Only a crank. This appears to be untested but theoretically it should work fine. The issue up until MS3 has been if the controller will fire in the right sequence (every 360 rotation of the crank per cylinder).

The idea is that MS3 has 6 ignition outputs. Each fires on each cylinders compression (duh) and also as each goes to exhaust (these are 360 apart). So every other is wasted during the exhaust. When the controller fires, it only fires one coil, the coil of the cylinder it is controlling. So our PRV fires 1,6,3,5,2,4 in each rotation. At any particular time the cylinder is either compression or exhaust

Running this way the ignition controller doesn't care where the cylinder is at other than it is at TDC + timing offset. That is why it can run crank only

Contrast that with "normal" wasted spark on an even fire engine. Each firing of the coil sends 2 sparks concurrently, one to the cylinder in compression, the other to the cylinder in exhaust. Since the cylinders are 180 degrees apart everything is in order. As you noted, our PRV odd fire does not have the opposing cylinder 180* apart and thus there are issues with "normal" wasted spark ignition on an odd fire engine.

Was I clear? Does that help clear things up?

Spittybug

08-10-2013, 12:12 AM

Gotcha. Just use the odd fire code to set the spacing once the crank sensor identifies #1. And with 6 coils, it's not like it's going to really matter that you're sparking twice as often as normal.... Do we know what the settings are in the software?

You'll need to add more coil driver chips at $25 a pop, but who cares. COP would be cool... no dizzy to worry about. No spark plug wires (at least not conventional) all over the place. CLEAN>>>>>>>>

nkemp

08-10-2013, 09:51 AM

Some coils for COP/CNP do not need drivers and can be controlled via the 5Volt logic level MS3 control thus not needing the additional drivers :-) Per the MS3 manual:

"The LS style coils are of a "coil near plug" design, they are also logic level. They are fairly easy to obtain used (in the
USA) and are available new from OEM or aftermarket. The truck coil gives the most powerful spark and is easily
identified by the heatsink.

Coil-on-plug coils
There are many different coil on plug coils used by the OEMs. Some are logic level, some require high current drivers. If
a coil has two wires it is certainly high current. If a coil has three or four wires it is almost certainly logic level."

One benefit of going logic level is that you keep all high current switching out of the MS box. So this is looking better all the time.

I'm not sure what the software settings are. I have not made the leap to purchasing yet and I'm in the "identify the solution" phase. But I think I know what the solution is:

- MS3, 6 logic level ignition outputs
- GM logic level controls (truck preferred). Still need to verify mountability in and around the std DeLorean air cleaner
-Does anyone know of a coil on plug coil that fits under the air cleaner?

Note: With this arrangement we can leave all the original ignition in place and it would be a relatively easy swap back and forth to compare performance differences. Or as a backup until we are satisfied.

Nick

DrJeff

08-10-2013, 11:10 AM

Gotcha. Just use the odd fire code to set the spacing once the crank sensor identifies #1...

Maybe I missed it, but did you figure out how to fit a crank sensor in the space available?

Spittybug

08-10-2013, 11:56 AM

Maybe I missed it, but did you figure out how to fit a crank sensor in the space available?

Haven't tried... but could be done with thin tooth gear tacked on pulley or flywheel.

nkemp

08-10-2013, 03:19 PM

Owen, When you converted to EFI, didn't you need to install a toothed wheel?

Spittybug

08-10-2013, 03:26 PM

Nope. I'm using dizzy's VR tach signal. That's all MS needs.

dmc6960

08-10-2013, 09:58 PM

Not having adventured past MS1extra and Fuel Only...

Does the software in the MS2/3 support dynamic odd-firing entries? Not all odd-fire are 90-150. I'd imagine the firing interval would need to be manually set. Without that info I don't see how a COP wasted spark odd-fire is possible. The software I have for MS1 does not have such information.

nkemp

08-10-2013, 10:49 PM

This only works on MS3/MS3extra (does not work on MS1 or MS2 or MS2Extra). As I understand it you input the smaller of the two angles. Beyond that someone else needs to jump in.

Spittybug

08-10-2013, 11:06 PM

That's the way MS2 works... we just tell it 90* oddfire and it calculates the rest. Remember, it isn't having to figure out WHERE to send the spark, only when. He waits for a couple of VR pulses, therefore knows if it's on the short or long angle and then starts counting. The dizzy does all of the directing of that spark.

With sequential however, MS will need to know not only WHEN but WHERE to send the spark. The concept of sparking on both the compression and exhaust strokes is fine, but I'm still not familiar just how (software wise) one will teach MS the firing order. I guess it would be as simple as inputting it and once it detects TDC#1 from the crank gear ring's missing tooth, it counts 90* on the crank tooth gear and fires off the next spark, then 150* and so on.... so if that is true, no cam sensor required, that spark just gets wasted on an exhaust stroke.

I am not familiar with MS3 code however.

nkemp

08-10-2013, 11:27 PM

Is there anyone with a MS3 and a Simulator? Could we test this before installing? For that matter, is there anyone with a MS3?

nkemp

08-11-2013, 08:20 PM

One more thing to consider ...
HEI ignitions can draw significant current at start. Some reports are 25 amps or more. Considering the current carrying of our OEM harnesses it often leaves something to be desired. Having six smaller coils means that each will draw less than one large coil but the total could be as much or more.

Josh

06-26-2014, 01:32 AM

Just reviving this thread. I would definately like to do a coil on plug setup evenually. I am currently running edis wasted spark with a common coil pack, but eventually it would be neat to transition to sequential spark and fueling. Really interested to fid a coil on plug setup. I know the car you posted Nick had it, but i would like something with no sparkplug wires whatsoever. Just the power and ground to the coils.

I designed a crank trigger setup, its for a 3.0 but would work well for a 2.8, the only thing that may change is the pulley inner face depth. Uses a common ford VR sensor.

http://i84.photobucket.com/albums/k38/jschatte/Facebook/Project%2082%20Delorean/10308175_833408566672658_4604407445116557359_n.jpg (http://s84.photobucket.com/user/jschatte/media/Facebook/Project%2082%20Delorean/10308175_833408566672658_4604407445116557359_n.jpg .html)

Spittybug

06-26-2014, 11:06 PM

And how would you do the cam trigger? A disc in the distributor? A toothed gear under the cover? This is the sticky one!

Josh

06-27-2014, 02:00 AM

I was thinking about it and it could be done. I have eliminated the distributor and removed the housing. It runs off the passenger side cam (1-2-3). However I could re-mount the housing and instead of installing the rotor, install a toothed gear of some sort. Or an indexed wheel. Another gentleman actually did this with his 3.0 swap instead of using a wheel on the crankshaft. He did not go the edis route either, he basically did this to get an rpm signal from the distributor which in its stock form has no such thing its simply a cap and rotor. So this would require a 3.0 timing cover and distributor housing, and some of the associated nuts and bolts. You would have to convert over to a serpintine belt too then.... hmm. Maybe a hole could be punched out of the 2.8 timing cover.

Hopefully you can use your imagination and piece these pictures together...

admirall2

06-27-2014, 11:20 PM

Someone on another board I visit has used megasquirt on a Pontiac OHC-6 and a 455 Buick. I believe he has used the distributor as a cam position sensor to allow for this signal only.

Josh

09-08-2014, 09:18 PM

Well I found a suitable coil for the PRV. Installed them this afternoon and they work perfectly fine. Still running wasted spark for now.

Coil:
Motorcraft DG500 / DG513
Application: Ford 3.0L DOHC Duratec V6 (2000-2005 Taurus)

Here is a picture of everything disassembled. Remove the boot off the coil and throw it away. Cut the internal spring to the length shown in the picture (about 1/3 of the original length).
Remove the boot from a set of delorean spark plug wires. I reemed out the "bore" of the boot a little, just enough so the spring could slide through nicely. Cut off the top portion of the boot.

30645

Press the boot on the coil and install just as you would a spark plug wire. I will be making a retaining bracket but for not they are extremely sturdy, I highly doubt they would come loose or pop out.

30644
30646

EDIT: I guess the eagle premier and delorean spark plug wire boots are different. Both should work though. The extra girth the premier boots really does help keep the coils securely in place.