Why COP if stock is better???

[turboglenn]

I see so many coil on plug setups that i've really started to like the look of it and have thought about doing it myself. One thing i keep seeing is people mentioning that the stock setup is better and has more powerfull spark for higher boost apps. I don't know the truth about any of this because i haven't tested it myself, I've never had to use anything more than an MSD box on a mitsubishi to get good results.

So can some one break down the pros and cons of COP vs. Stock wastespark 2 coil setup?

 

[gsxtasy]

COP setups have the wrong dwell time and are made to operate with more time between firing. Trying to throw this on a stock igniter works fine for lower boost, but can't charge fast enough and give a good spark at higher boost levels. As you know, I have one , and my car breaks up over 25ish psi of boost. I have my plugs gap'd down to .022'' and it still wont keep up at 30+psi. Needless to say, I have an Dynatek Arc-2 CDI on the way.

With that being said, if you have a CDI box, and/or a way to control dwell time(can haltech?) you will be fine running a COP setup. If you can set proper dwell times AND have a CDI you will have one of the best ignitions available for our cars. If you have neither, you are honestly better off with stock ignition.

 

[turboglenn]

Yea, i can set the dwell times, but i normally won't run over 3.5ms, only because in the J&S knock sensor. It seems having much more dwell than that can sometimes cause there to be some electronic interference/crossover in the J&S unit causing crossfiring of the plugs after 5500-6k RPMs (his new untis don't have this issue and i could have mine modded, but i'm fine with it)

Currently, I am able to run 30psi on my stock coils and an MSD DIS-2. My plugs are gapped at .028 and i'm running 3.5ms of dwell. I haven't had any miss-fires and i've even fouled these plugs out, once with gasoline when i forgot to change teh rimer map from the E85 one and the second was when i sucked up 50 gallons of water in one of our many biblical floods we've had lately

How much dwell do they require to show gains by the way?? If it falls within what makes my J&S happy i just may give it a try for asthetic reasons when i pull all the newer parts (IC pipes..etc..) for powder coat here soon.

Gonna have to take you for another spin here soon man, i think you'll be impressed. I want to play with the tune on your car a bit next time i'm out if you're up for it. I think we could pull quite a bit more midrange out of it with a few tweaks

 

[gsxtasy]

Honestly, I don't know what proper dwell time should be.

 

[TurboTiger]

COP is best if you have a stand alone ecu capable of doing sequential ignition vs. the waste spark system our stock ecu uses. Instead of sparking twice every four cycles as a wasted spark igntion would do, you only ignite once every four cycles. This will give the coil twice as long to charge at high rpm. While it doesn't give as much spark energy as a CDI, it also doesn't suffer at igniting leaner mixtures at lower rpm due to CDI's shorter spark duration.

 

[turboglenn]

That makes sense, i just don't see why so many people do it when they don't have the means to control it properly and are losing spark energy at the same time WTF??? I mean our stock coils are good to around 450WHP reliably. Are people that caught up on the looks that they are willing to sacrifice performance just to have the look of COP ignition and the blingy little coil plate?

I have the ECU for it, but the more i've been thinking about it tonight, the more i think I'll just spend the money elsewhere until i start having spark issues. I just think it looks clean as hell though

I still need some springs and retainers next to maximize my cam's performance a little more.

 

[steve]

Are people that caught up on the looks that they are willing to sacrifice performance just to have the look of COP ignition and the blingy little coil plate?

It would seem so. I haven't looked at the specs for the coils typically used but I suspect that they are intended to be used directly with a 12v feed not in series like you are forced to when using the factory ignition. If you have a standalone and 4 drivers for the coils you could produce a better ignition than the factory setup. But the bulk of the COP setups connected to the factory PTM are actually a downgrade.

 

[GVR4592]

There's a ton of good info on the aem forum about our ignition systems limitations, and there is great information about ignition systems in general on the EFI101 forums. I spent a lot of time researching this subject and found that the best setup without spending too much money is a good cdi and stock coil in wasted spark. AEM and MSD don't make a good CDI, Autronic, and M&W make excellent CDI's. M&W also makes a nice COP setup for our cars, but it's pricey.

 

[andy4g63]

Well said. There are many brands COP on the market today and with the exception of a few, /I am not gonna mention names/ the rest is nothing but a problem. There are many examples of people making somewhere between 700-800 WHP on bone stock ignition.Personally 95% of people that have them, don't need them.
Turboglenn, you are right. You better off spending your money on something different.

 

[1gDSM4g63]

Needless to say, I have an Dynatek Arc-2 CDI on the way.

I've heard good things about Dynatek. Update my on the Dynatek and how it work out with the your COP.

 

[polarmoment]

part of the issue is the stock power transistor unit. i don't have a DSM anymore, but from the calculations i've done from info over on stealth316 for the DSM coils, the power transistors have something on the order of almost a full ohm of on-resistance, and drop as much as 2V out of the coil circuit. that's just fine if you're charging the stock coils, since they have moderately high coil resistance, but very low inductance (3.5mH, compared to others like GM's ~5mH, 0.35ohm). because of that limitation, you'll only ever really see about 6A through the circuit. this is compounded when you run the COP coils in parallel and essentially introduce about .4ohm more resistance into the already limited circuit. the ignition power circuit just doesn't have low enough resistance to make the coil's current rise time fast enough to charge it to a decent current level.

12V/2.2ohm= 5.45A charge current. that means the energy stored in the primary side of the COP coils (assuming an inductance equal to stock- we know they're lower because they're smaller coils with smaller cores) is eqqual to 1/2L*I^2, or (1/2*3.5)*5.45^2, or about 53mJ. that's not much spark energy. hell, even the stock coils only put out about 67mJ, assuming the stealth316 articles are correct and the coils charge to 6A.

honestly, i think the answer is to use a seperate (and more efficient) power transistor unit for each coil, with each cylinder pair's igniter being fired by the stock transistor unit.

i'm going to be using DSM coils on my MR2 in wasted spark mode with a stock honda P72 ECU and a splitter circuit driving two honda igniters, and the calculations with the honda igniters show the coils, in the same ~4mS the DSM ECU uses for dwell, charging the coils to about 8.5A, allowing for about 126mJ of energy stored in the primary- that's almost DOUBLE the energy for only 2.5A more charging current.

i think that the only way to run the COP setup inductively is to use a seperate igniter (honda igniters would work very well, and already have a tach output on each one- you'd only need to connect them together through diodes) for each coil. the trigger side of the honda igniters needs to be pulled low to charge the coil and high to fire it (the trigger pin on the igniter has an internal pullup resistor to 12V), which is what the stock power transistor already does. wiring would be very simple and the honda igniters have extremely low on-ressitance, as evidenced by the ability to charge the high inductance, medium resistance (5-6mH, 0.6ohm) honda coils to over 8A in 4mS.

running the COP setup this way would likely more than double its spark output.


the formulas i used for my calculations are:

energy stored in coil primary= 1/2L*I^2

dwell time required to charge an inductor to specific current= (-L/R)*ln(1-((R*I/E)

where

L= inductance in henries
time is in seconds
R= resistance in ohms
I= current in amps
E= voltage in volts.
ln is natural logarithm function found on any scientific calculator.

 

[turboglenn]

polarmoment...since you seem to know quite a bit about teh technical aspect of the ignition system I would like to run some things by you(i know how to hook it up to a stand alone and that's about my knowledge of it's detail )

I am currently running an MSD DIS-2 and instead of getting rid of the transistor and using the MSD as teh ignitor/transistor, i wired teh MSD between the stock transistor and the stock coils. What is your opinion on this setup?

I had it hooked up without the stock transistor, (since i have an ECU i can configure the trigger and firing edges on) but i noticed it didn't like to start as well that way and sometimes i would loose tach signal for a moment or two. After putting the stock transistor back in it's started and ran perfect since with no tach issues what so ever..

 

[polarmoment]

it was probably due to the fact that the MSD unit uses a fairly low resistance pullup resistor to 12V on the trigger line (on the order of 10-20ohms). in order to pull it low, then high to trigger the unit, whatever is triggering it needs to be able to easily sink a little over an amp (think 2x safety factor and power dissipation). i doubt the trigger line from the stock ECU is capable of this (there's probably a resistor on the output that limits sink/source current to that line to prevent damage to the ECU), but the power transistor can accomplish this quite easily.

with the ECU triggering it, the signal would never get pulled fully to ground due to the input impedance of the trigger line. think of it in terms of a voltage splitter- take 2 1kohm resistors, connect them in series and connect either end to 12v and ground. if you tap between the two resistors with your DVM, you'll see ~6v, since each resistor drops pretty much half the circuit voltage. with the ECU triggering it and sinking that signal through the input resistor, instead of pulling the signal to ground, it pulls it, say, halfway down- to 6v. if that 6v is right near the threshold voltage for the MSD's internal trigger circuitry, you'll occasionally get ignition events that don't fire the unit, and get some that do. the power transistor still has low enough resistance between it and ground that it makes the MSD's internal pullup drop nearly ALL the voltage, resulting in a very low, close to ground voltage on the trigger pin inside the MSD, triggering it cleanly.

this would be especially prevalant at low battery voltages (ie. cranking) since the voltage split would be even lower. if you wanted to ditch the power transistor unit, you could accomplish the same trigger signal with any cheap power transistor from radioshack. a TIP120 would work.

 

[jshuman]

Thanks for sharing that information polarmoment. I think that is the most valuable cop info that has been posted in these forums.



I thought I would throw in my 2 cents on why I run a COP in waste spark knowing that its inferior to the stock ignition. The number one reason that I made a cop setup was to solve the issue of not having a place to mount my stock coils after I installed my jmfab smim.

 

[Brianawd]

I have been running my COP for 3 years now with no problems. I have never once had a miss. Even when I was running 32psi on my FP3052 with e85 and 1600cc injectors. Spark plug gap is at 28.

 

[turboglenn]

it was probably due to the fact that the MSD unit uses a fairly low resistance pullup resistor to 12V on the trigger line (on the order of 10-20ohms). in order to pull it low, then high to trigger the unit, whatever is triggering it needs to be able to easily sink a little over an amp (think 2x safety factor and power dissipation). i doubt the trigger line from the stock ECU is capable of this (there's probably a resistor on the output that limits sink/source current to that line to prevent damage to the ECU), but the power transistor can accomplish this quite easily.

with the ECU triggering it, the signal would never get pulled fully to ground due to the input impedance of the trigger line. think of it in terms of a voltage splitter- take 2 1kohm resistors, connect them in series and connect either end to 12v and ground. if you tap between the two resistors with your DVM, you'll see ~6v, since each resistor drops pretty much half the circuit voltage. with the ECU triggering it and sinking that signal through the input resistor, instead of pulling the signal to ground, it pulls it, say, halfway down- to 6v. if that 6v is right near the threshold voltage for the MSD's internal trigger circuitry, you'll occasionally get ignition events that don't fire the unit, and get some that do. the power transistor still has low enough resistance between it and ground that it makes the MSD's internal pullup drop nearly ALL the voltage, resulting in a very low, close to ground voltage on the trigger pin inside the MSD, triggering it cleanly.

this would be especially prevalant at low battery voltages (ie. cranking) since the voltage split would be even lower. if you wanted to ditch the power transistor unit, you could accomplish the same trigger signal with any cheap power transistor from radioshack. a TIP120 would work.

I'm running a Haltech E6K ECU, it can fire pull up or pull down ignition modules and run in either constant duty or constant charge and has dwell control. It runs through a J&S Safeguard knock sensor and timing retarding ECU ( the module also is able to be setup for any type of ignition) It just seems i have better luck using the falling edge trigger for some reason *shrugs* Of course it doesn't bother me to run the stock ignitor, but being that tuning and setting up cars and ECU's is my 2nd living I feel a need to know as much as I can about every angle of the job. Even after 12 years of doing it, I am constantly learning everything i can. Even if i know how to setup a system one way, if there's a second way to do it, i want to know that way too


I guess what i'm saying is, i'm not having issues, but trying to understand exactly what was going on and how to repair or avoid similar situations. To which your knowledge has been rather helpfull, thank you!

 

[polarmoment]

you bet. the more you know, the easier it is to improve things. i get sucked into this stuff really and end up absorbing as much as i can. part of my issue is that i'm not allowed to spend much $$ on the cars, so i have to build a lot of my own stuff- case in point: instead of spending the $1500-2k for a hydra EMS, i modified a honda distributor to bolt into the 3SGTE so i could use CROME/eCtune and the honda ECU as my EMS. i've mostly looked into this stuff as a method of going distributorless WHILE making it a worthwhile ignition upgrade for my car, but knowing that the stock DSM ignition was pretty good on its own, i was mostly looking for ways to improve the coil control side of it. i won't know for sure how successful it is until it's on the car, but thus far my track record for designing things and having them work exactly as i planned is damn near 100%, so i'm optimistic. i figure if i spend enough effort on the 'R' side of the R&D, the 'D' side won't cost as much in terms of time or money.

like i said though, even with the haltech it's designed to trigger ignition modules, which typically have a higher ohm pullup resistor. the honda module's internal pullup resistor is 1k ohm, so the ECU only has to sink about 12mA to trigger it. most aftermarket ECUs are similar in their driver abilities, with a few actually being able to directly fire the coils. most manufacturers stay away from directly firing the coils via the ECU because of the EMI from the inductive kick. it's easier for them to move the power handling side of things to an external module, away from all the crank and cam position sensors. the only mfr that i can think of that directly fires the coils using the ECU is chrysler.

so i think what i posted earlier still applies. i'd bet that if you used a P-channel MOSFET as your external power transistor, the rising edge trigger would work the same as how you're using the falling edge with the stock N-channel transistor. it's just that the EMS can't sink enough current itself to accurately drive the MSD box.

i actually have a PDF schematic of the MSD 6A somewhere on my other computer. i'll have to dig it up and see exactly what the value of the trigger pullup is.

 

[bastarddsm]

Would you be willing to sned me that pdf? i'm kind of information hungry myself.

 

[turboglenn]

Yea, i'd love to see this PDF if you're willing to email it to me at [email protected] or [email protected]

Thanks for all the info as well, that's a little over my head though honestly, i can solder things up from schematics, but i have to use "cheat sheet" to know what everything is I love what you are talking abou using the chrome/honda ECU for your other car, that's almost like teh guys using old GM ECU's and reprogramming them to run as stand alones on other cars, but too much "3exu76 address, value = bl;a bla bla" And i can't fully understand it, but i've enver taken the time to really delve into learning it either

 

[TurboTiger]

If you're willing to go through the trouble of hooking up a honda ecu, you might as well go to a megasquirt 2 setup. You can do sequential ignition on a 4 cylinder with the ms2extra.

 

[polarmoment]

the honda EMS is cheaper, more refined, has better control of the engine and already has things like cranking enrichment, warmup, etc programmed into it, plus the hardware has millions of dollars of R&D for things like thermal protection, vibration/shock etc. it's hard to find a more reliable piece of electronics than the denshigiken units. the car runs better on the honda ECU after only a few hours of tuning than it ever ran on the stock ECU. i also consider having individual cylinder trims/sequential injection a necessity on a turbo engine equipped with a centerfeed intake manifold, since the center two cylinders tend to run leaner. wiring the honda ECU was also MUCH less involved than you'd think, and far far less complicated than a megasquirt.

switching it to sequential ignition is a matter of a simple circuit with about 5 components. i could easily do full COP with it by adding one chip, but wasted spark is less complex and cheaper. consider too, that the old buick 900HP turbo indycar engines ran stock GM wasted spark DIS coils.

the MSD schematic is available at: www.dainst.com/info/circuits/msd6a_02.pdf which is the site i originally found it at. the unit in the schematic is well out of patent, so it's pretty much fair game, but i don't think the design has changed much.

looks like the pullup resistor is a 40ohm, 5watt. that means that the ECU trigger needs to be able to pull 300mA, or 25x the current it would need to sink to trigger an ignition module.

 

[polarmoment]

i also wanted to add this link as well.

www..apexology.com/files/schematics/siliconchipCDI.pdf

this was an article in silicon chip magazine (australia) from i think 1997. basically it's an electronics magazine whose authors happen to design a lot of cool car gadgets, and they designed a multi-spark CDI box and described the design in detail. it gives you a good idea of how a multispark CDI box works, and there's a schematic for the circuit and a bill of materials if you want to try building one yourself. obviously, with a wasted spark ignition setup you'd have to modify the circuit to have multiple inputs and an extra capacitor, but it's more an excercise in how they work.

the MSD box schematic, as i said, is probably still pretty close to what comes in a 6A ignition. they're still completely analog (as opposed to others like crane, mallory, accel that are digital and run by a microprocessor). most CDIs nowadays though don't use the SCR (silicon controlled rectifier- kind of like a diode with an avalanche switch) to dump the capacitor through the coil, they just use a high voltage rated IGBT (insulated gate bipolar transistor). a single capacitor could be dumped through multiple coils in the waste spark method by simply adding another IGBT and trigger circuit. i'm kind of surprised that no one has made a cheap multichannel CDI box out of the MSD units yet.

i'm kind of a nerd when it comes to the electronics side of things. i'm not an EE, but i learn more everyday. it's kind of amazing how simple a lot of the very expensive electronics we use really are.