AEM High Output Inductive Coils

[Moonlight GSX]

It would seem that no one here has ever used the high output inductive coils made by AEM. Is there a reason why? The only problem I could see from these would be trying to get the original signal from ECU to all 4 coils as they are individually connected meaning you'd need 4. But with that being said, these are slightly cheaper than running a COP system and also seem to be able to be a direct replacement of our coils without the need for a standalone ignition system like MSD or Dynateck.

With that being said, would it be worth giving it a try and seeing how well it actually works? I'm honestly all too curious as to how something like this would work. The only reason I'm looking into it is I'll be needing upgrade my ignition system soon to match the demand of the fuel and turbo system I plan to soon run (GT4088R and a much upgraded fuel system) at or above 650whp, shooting for 700whp. But yeah let me know what you all think, sounds legit but I just want to see what your thoughts are.

The "smart" coil is listed here.

"Dumb" coils are listed here.

 

[Moonlight GSX]

Now I'm kinda confused, does anyone know how the two coils within the coil pack are separated during the dissipation of energy? The new coils are separated into individual coils allowing each plug to have it's own "coil" but they, the dumb coils, require a +12V constant, trigger and ground. I know that our stock system can supply all three of these signals but can they supply these signals to 4 separate coils?

I was thinking of putting them in parallel in such that the same signal would be divided between the two and operated similar to that of the coils we normally have. It sounds like a logical idea, the only other problem is the ground that is required - I have yet to find the ground for the coils on the stock system as the new ones actually have a pin location for the ground whereas ours have only two pin/bolt locations that have the +12V constant and trigger. Should I connect Cylinders 1 and 4 together at ground and then 2 and 4 at ground as well? Or would I be able to ground them all together on the same cable/ground point?

 

[polarmoment]

Now I'm kinda confused, does anyone know how the two coils within the coil pack are separated during the dissipation of energy? The new coils are separated into individual coils allowing each plug to have it's own "coil" but they, the dumb coils, require a +12V constant, trigger and ground. I know that our stock system can supply all three of these signals but can they supply these signals to 4 separate coils?

I was thinking of putting them in parallel in such that the same signal would be divided between the two and operated similar to that of the coils we normally have. It sounds like a logical idea, the only other problem is the ground that is required - I have yet to find the ground for the coils on the stock system as the new ones actually have a pin location for the ground whereas ours have only two pin/bolt locations that have the +12V constant and trigger. Should I connect Cylinders 1 and 4 together at ground and then 2 and 4 at ground as well? Or would I be able to ground them all together on the same cable/ground point?

1. The "ground" for the stock coil IS the trigger. They are supplied constant 12v and the ECU signals the power transistor to ground the coil to charge it, then release the ground to spark. The stock power transistor is the limiting factor when it comes to spark energy, not the coil. The 'ground' they're saying to ground to the head is the ground for the secondary winding. In the stock coils, there is no need for such a ground because the secondary's poles are connected to both plugs in the pair.

2. No, the stock system can't provide the proper signals to 4 dumb coils because either you'll run into current limiting when run in parallel or charge time issues if run in series. It could probably do it for the smart coils (which aren't really smart coils, but that's another discussion), but you better have dwell control in your EMS so you don't burn them up.

3. I think you would be well-served to examine some wiring diagrams for the stock ignition system as well as some reading on how inductance and resistance work in a coil to create a spark and really make sure you understand how all this works before you dive into this project. There's a lot of marketing hype in those links that rely on selling numbers to people who don't understand any of what they're saying.

Any coil can put out that kind of energy if driven properly.

 

[Moonlight GSX]

1. The "ground" for the stock coil IS the trigger. They are supplied constant 12v and the ECU signals the power transistor to ground the coil to charge it, then release the ground to spark. The stock power transistor is the limiting factor when it comes to spark energy, not the coil. The 'ground' they're saying to ground to the head is the ground for the secondary winding. In the stock coils, there is no need for such a ground because the secondary's poles are connected to both plugs in the pair.

2. No, the stock system can't provide the proper signals to 4 dumb coils because either you'll run into current limiting when run in parallel or charge time issues if run in series. It could probably do it for the smart coils (which aren't really smart coils, but that's another discussion), but you better have dwell control in your EMS so you don't burn them up.

3. I think you would be well-served to examine some wiring diagrams for the stock ignition system as well as some reading on how inductance and resistance work in a coil to create a spark and really make sure you understand how all this works before you dive into this project. There's a lot of marketing hype in those links that rely on selling numbers to people who don't understand any of what they're saying.

Any coil can put out that kind of energy if driven properly.

I think you may have misunderstood what I meant... the four inductors won't actually be in "parallel." What I meant by this is that they all would receive their own 12V and ECU trigger (yes the ECU is the trigger - the ground is only there to actually create the charge on the secondary winding), 1 and 4 would receive their trigger from pin A10 (ECU then to power transistor) and 2 and 3 would receive their trigger from A23. The last and final signal is from C58 which is the Engine Ignition Signal (I'm still researching that - but have a relative idea).

The only difference between these "dumb" coils and our coils is that the two coils we have (1/4 receive same trigger and 2/3 receive same trigger) have been divided into 4 separate coils with an external ground whereas ours is an internally capacitive ground. In theory the concept makes sense it's just how to wire it. Now, I have made a diagram of the modified setup that would utilize the four coils (will post once I get home from leave), the setup makes sense the only thing that is still confusing is the capacitive ground that may or may not be needed, the trigger signal is accounted for from the ECU and power transistor, and the 12V constant is also accounted for. All the required signals for the coils will still come from the stock 3 pin that is located on the intake manifold, the only thing that isn't on that 3 pin is the external ground that can be easily made by just grounding those pin locations to the head to complete the circuit for the secondary windings in the individual coils.

 

[95g-s-x]

I would assume that the 1/4 and 2/3 coils would need to be run in series just like a COP setup

 

[Moonlight GSX]

I would assume that the 1/4 and 2/3 coils would need to be run in series just like a COP setup

The 12V constant would in theory be in series, but you can't daisy chain as the properties of inductive coils in series or parallel will have effects on the current and voltage drop across them. But in the mindset of putting them in "series" as they get their own feed from the main 12V (from switch) and trigger (from ECU) then this would, again in theory, eliminate the chance of any voltage or current drop across any of them.

 

[95g-s-x]

But can the power transistor handle the load of triggering bolt coils from the single channel?

I looked at the "smart" coils. I don't see why you couldn't use these and eliminate the power transistor all together. The Miatas use a VERY similar transistor (mitsu made, and appears to use same connector) and convert to 1zz coils eliminating the transistor.

Here is the link. Look at the pic for the 1.6l

Upgrading to Coil on Plugs (all years) COP writeup - Miata Turbo Forum - Home to the Turbo Miata

I actually just compared the pinouts on the miata transistor to ours. Their connector is numbered backward from ours (ours 1-8 left to right, theirs 8-1 left to right). If you re-label their pins to match ours, the only difference is pin 4 and 5 on the miata. Pin 4 on the maita gives feedback to ecu....assuming it is a tach signal. pin 5 gives the cluster its tach signal (we dont have a wire at pin 5) Their new pin2 triggers the cyl 2/3 coil and their new pin7 triggers the 1/4 coil

 

[Moonlight GSX]

But can the power transistor handle the load of triggering bolt coils from the single channel?

I looked at the "smart" coils. I don't see why you couldn't use these and eliminate the power transistor all together. The Miatas use a VERY similar transistor (mitsu made, and appears to use same connector) and convert to 1zz coils eliminating the transistor.

Here is the link. Look at the pic for the 1.6l

Upgrading to Coil on Plugs (all years) COP writeup - Miata Turbo Forum - Home to the Turbo Miata

I actually just compared the pinouts on the miata transistor to ours. Their connector is numbered backward from ours (ours 1-8 left to right, theirs 8-1 left to right). If you re-label their pins to match ours, the only difference is pin 4 and 5 on the miata. Pin 4 on the maita gives feedback to ecu....assuming it is a tach signal. pin 5 gives the cluster its tach signal (we dont have a wire at pin 5) Their new pin2 triggers the cyl 2/3 coil and their new pin7 triggers the 1/4 coil

Interesting, didn't know that about the Miata.

Although I have read in our forums that the P/T can go bad from time to time and I could see it being an issue for when I decide to rock 4 separate coils. But in all honesty, as the P/T is only a "switch" I would think that it would be okay to handle the load for 4 separate ones as it'll still receive only 0-5V at the trigger and the other 12V constant as well. But you're idea to use the "smart" coils makes sense too, and just now I think I see where you're coming from on using the smart coil with the Miata?

The only difference between the smart coil and the schematic for the COP on the Miata is that there is a necessary ground that goes to the head for the coil - I take all the other signals are supplied by the stock system? Correct me if I'm wrong but that's what it seems like... If so please be right!

Edit:
A: 0-5V ECU Trigger (A10 for 1/4 and A23 for 2/3)
B: Ref Ground (original ground ref on power transistor?)
C: Ground to Cyl Head (self explanatory)
D: Battery Ground (directly to battery or just somewhere on chassis?)
E: +12V Constant (switched by ignition - always hot)

Is this correct? The only other thing is where does C58 (Engine Ignition Signal - tachometer signal) go? Rather how would that be linked?

 

[95g-s-x]

Interesting, didn't know that about the Miata.

Although I have read in our forums that the P/T can go bad from time to time and I could see it being an issue for when I decide to rock 4 separate coils. But in all honesty, as the P/T is only a "switch" I would think that it would be okay to handle the load for 4 separate ones as it'll still receive only 0-5V at the trigger and the other 12V constant as well. But you're idea to use the "smart" coils makes sense too, and just now I think I see where you're coming from on using the smart coil with the Miata?

The only difference between the smart coil and the schematic for the COP on the Miata is that there is a necessary ground that goes to the head for the coil - I take all the other signals are supplied by the stock system? Correct me if I'm wrong but that's what it seems like... If so please be right!

Edit:
A: 0-5V ECU Trigger (A10 for 1/4 and A23 for 2/3)
B: Ref Ground (original ground ref on power transistor?)
C: Ground to Cyl Head (self explanatory)
D: Battery Ground (directly to battery or just somewhere on chassis?)
E: +12V Constant (switched by ignition - always hot)

Is this correct? The only other thing is where does C58 (Engine Ignition Signal - tachometer signal) go? Rather how would that be linked?

If you look closely at the miata diagram, the ground pin for the coils are grounded to head AND all tied together and then grounded at our pin 3, which is a chassis ground. So I would wire it like this

A: Ecu Triggers (PTU pin 7 for 1/4 & PTU pin 7 for 2/3)
B: PTU pin 3 (ground)
C: Ground to head
D: Battery Ground ( sould be able to chassis ground, but battery neg if breakup occurs)
E: PTU pin 6

Only thing I'm not sure of is the tach....looking at the diagram, you may actually tie all of the AEM coil pin B together and run that to PTU pin 4

 

[Moonlight GSX]

If you look closely at the miata diagram, the ground pin for the coils are grounded to head AND all tied together and then grounded at our pin 3, which is a chassis ground. So I would wire it like this

A: Ecu Triggers (PTU pin 7 for 1/4 & PTU pin 7 for 2/3)
B: PTU pin 3 (ground)
C: Ground to head
D: Battery Ground ( sould be able to chassis ground, but battery neg if breakup occurs)
E: PTU pin 6

Only thing I'm not sure of is the tach....looking at the diagram, you may actually tie all of the AEM coil pin B together and run that to PTU pin 4

Wouldn't pin A be ECU Triggers PTU pin 7 for 1/4 and then pin 2 for 2/3? Other than that it makes a lot of sense. The ground to head won't actually be part of the harness will it? At least that's what I'm seeing, most likely I could wire that in the bundle but just have a whire divert off to the head for it's grounding point (so I don't have a "free" wire). But other than that I thank you for the help! I'll have to see how well this works.

 

[95g-s-x]

Wouldn't pin A be ECU Triggers PTU pin 7 for 1/4 and then pin 2 for 2/3? Other than that it makes a lot of sense. The ground to head won't actually be part of the harness will it? At least that's what I'm seeing, most likely I could wire that in the bundle but just have a whire divert off to the head for it's grounding point (so I don't have a "free" wire). But other than that I thank you for the help! I'll have to see how well this works.

yes, your right on pin a.....had a brain fart. let me know how it works....i am curious about the tach.'

The ground to head, you could tie them all together and ground to one point, or go individually, really up to you.

 

[polarmoment]

The only difference between these "dumb" coils and our coils is that the two coils we have (1/4 receive same trigger and 2/3 receive same trigger) have been divided into 4 separate coils with an external ground whereas ours is an internally capacitive ground.

Yeah, they've been divided into 4 separate coils, each with their own resistance and inductance. If you wire 1 & 4, 2&3 in parallel per channel of the power transistor, you will hit the current limit. If you run them in series, you will increase the charge time of the circuit.

The power transistor is still DRIVING the dumb coils. It's still sinking ALL the current in the circuit. It's not just providing a signal to the coil, it's doing all the work. They're calling the smart coils "smart" because they have an internal IGBT driver (in reality, smart coils also have built-in dwell, but like I said, that's another discussion). The dumb coils do not have an internal driver and must be charged by the stock power transistor. That is why you will run out of spark energy.

The OE coils will provide more energy than these if you try to run them on the stock power transistor.

If you were to pick up a second power transistor and run one of the cylinder pairs on it, it would work, but you essentially need a channel of a power transistor for each coil if you want to get full spark energy out of them.

 

[Moonlight GSX]

Yeah, they've been divided into 4 separate coils, each with their own resistance and inductance. If you wire 1 & 4, 2&3 in parallel per channel of the power transistor, you will hit the current limit. If you run them in series, you will increase the charge time of the circuit.

The power transistor is still DRIVING the dumb coils. It's still sinking ALL the current in the circuit. It's not just providing a signal to the coil, it's doing all the work. They're calling the smart coils "smart" because they have an internal IGBT driver (in reality, smart coils also have built-in dwell, but like I said, that's another discussion). The dumb coils do not have an internal driver and must be charged by the stock power transistor. That is why you will run out of spark energy.

The OE coils will provide more energy than these if you try to run them on the stock power transistor.

If you were to pick up a second power transistor and run one of the cylinder pairs on it, it would work, but you essentially need a channel of a power transistor for each coil if you want to get full spark energy out of them.

The second transistor was also a thought. But as for running them in series or parallel the actual power going to them WON'T be one coil to another, it'll be each individual coil receiving their own power. In a way think of it being one coil in it's own individual circuit is what I was saying not having them all, in theory, on the same circuit. But I do understand what you mean, trust me I do.

There is one more thing that has come up, on our normal schematics there are two capacitors: one is at the coils themselves and the other is at a mid-point on the +12V switch line. With that being said for the grounds on pin C do I need to put a capacitor on them? And should I have to worry about the capacitor that is already (should be) in place that is off the +12V line? Just making sure I've got everything in order before I change this system out... it'll be awesome (hopefully) when it's finished.

 

[Moonlight GSX]

Here are some schematics I've come up with that would be the build for the individual coils. One of the pictures, with the smudges, is the setup that would run the "dumb" coils and power transistor, the other two are the wiring harness and over all schematic for the "smart" coils.

The one with the power transistor may need a second transistor as stated before but I'd have to figure out how to wire that to work properly... a whole lot of splicing.

 

[polarmoment]

Your dumb coil schematic is exactly what I said wouldn't work. You're trying to drive the coils in parallel. What happens is you get about 3A of charge current through each coil, since the transistor limits current to 6A total per channel. That's why you'd need a second transistor.

The top and bottom diagrams should work with smart coils, but you can't run dumb coils in parallel.

 

[Moonlight GSX]

Would this work? Not that 'd rather go the smart coil route... but another option none the less.

 

[polarmoment]

I *think* it's ok. I'm on a 7 year old laptop and can't see the diagram all that well, but it looks like what I was talking about. The idea is that each channel of each power transistor is only driving a single coil. Like I said earlier though too, the power transistor limits coil current to about 6A. Spark energy = ½L*I² and the transistor is limiting 'I' to 6A- same with stock coils. The only way you'll get more energy out of the ignition at that point is if the coils are of higher inductance.

The primary inductance on these coils is 4.8mH. The primary inductance on the stock coils is ~3.5mH. So at the same charge current, you'll net about 23mJ more spark energy over the stock coils. If you want to get some real energy, find a set of 4 honda igniters (which hit current limiting around 8A) and run those. The extra 2A of charge current will almost DOUBLE the spark energy (from 86mJ to 153mJ). You'll need a way to convert the ECU's TTL signal to the 12v rising edge trigger that the honda igniters use though. If honda parts aren't your forte, the bosch unit on the earlier 1.8t motors was a 4channel igniter that charged to 8-9A.

That's the thing AEM isn't really addressing with the dumb coils- it's not the coil that's the problem, it's the driver. They don't tell you that you need to use a higher energy igniter if you want more spark energy. In fact, these coils pale in comparison to standard, dirt cheap OEM GM DIS coils (6mH primary, stock DIS module charges to 7A). You'd make more spark energy on the OEM coils with a bosch igniter than you would with those AEM coils on a pair of stock power transistors.

Just a minor point too, the output of the power transistor is not TTL. TTL is a 0-5v logic signal (TTL stands for 'transistor-transistor logic'). The output of the power transistor is.. well, a power switch.

 

[Moonlight GSX]

I *think* it's ok. I'm on a 7 year old laptop and can't see the diagram all that well, but it looks like what I was talking about. The idea is that each channel of each power transistor is only driving a single coil. Like I said earlier though too, the power transistor limits coil current to about 6A. Spark energy = ½L*I² and the transistor is limiting 'I' to 6A- same with stock coils. The only way you'll get more energy out of the ignition at that point is if the coils are of higher inductance.

The primary inductance on these coils is 4.8mH. The primary inductance on the stock coils is ~3.5mH. So at the same charge current, you'll net about 23mJ more spark energy over the stock coils. If you want to get some real energy, find a set of 4 honda igniters (which hit current limiting around 8A) and run those. The extra 2A of charge current will almost DOUBLE the spark energy (from 86mJ to 153mJ). You'll need a way to convert the ECU's TTL signal to the 12v rising edge trigger that the honda igniters use though. If honda parts aren't your forte, the bosch unit on the earlier 1.8t motors was a 4channel igniter that charged to 8-9A.

That's the thing AEM isn't really addressing with the dumb coils- it's not the coil that's the problem, it's the driver. They don't tell you that you need to use a higher energy igniter if you want more spark energy. In fact, these coils pale in comparison to standard, dirt cheap OEM GM DIS coils (6mH primary, stock DIS module charges to 7A). You'd make more spark energy on the OEM coils with a bosch igniter than you would with those AEM coils on a pair of stock power transistors.

Just a minor point too, the output of the power transistor is not TTL. TTL is a 0-5v logic signal (TTL stands for 'transistor-transistor logic'). The output of the power transistor is.. well, a power switch.

Now all of that actually made sense, as for the TTL logic I know that 0-5V is a typical TTL signal and all it's doing is telling the PT to "open and close" causing the secondary coil to charge to ground (correct me if I'm wrong). But which would you say would be cheaper? I may have to do more research on shipping cost but would you have a rough estimate?

Seems on the Bosch 4 channel I'd have to split the logic signal of each pair into separate signals, I guess that wouldn't be too bad but the cost of one of those new is outrageous. Unless I'm looking at the wrong one? Found a VW one... I take that's the one you're talking about? I may go the route of the custom wiring harness and use the smart coils due to removing the power transistor all together - one less thing to cause a problem. Rather than wiring - P/T - wiring and then coils it'll just be wiring - wiring - coils. Figured I'd have less of a headache than dealing with them, God forbid one goes out.

Bosch P/N: 0 227 100 211

 

[polarmoment]

Cheapest route to upgrade the ignition would be to exhange the P/T with the bosch igniter (find a used one, they're about $45 tops) and stay with the stock coils. It would put out more energy than an MSD.

When you change the coils, the only thing you're really changing is the inductance of the circuit. But if you look at the equation for spark energy, the charge current has an exponential effect on spark energy, while the inductance has a much smaller, linear effect.

The coil itself- as long as it's more than a few mH of inductance and low enough resistance to charge quickly- is largely irrelevant. It's the circuit that charges the coil that makes the most difference in spark energy in an inductive setup. AEM would prefer you didn't know this.

 

[Moonlight GSX]

Cheapest route to upgrade the ignition would be to exhange the P/T with the bosch igniter (find a used one, they're about $45 tops) and stay with the stock coils. It would put out more energy than an MSD.

When you change the coils, the only thing you're really changing is the inductance of the circuit. But if you look at the equation for spark energy, the charge current has an exponential effect on spark energy, while the inductance has a much smaller, linear effect.

The coil itself- as long as it's more than a few mH of inductance and low enough resistance to charge quickly- is largely irrelevant. It's the circuit that charges the coil that makes the most difference in spark energy in an inductive setup. AEM would prefer you didn't know this.

So what's the difference between what you just said and COP setups? I'm not knocking it but how come everyone goes for COP setups when they can just change the P/T and up the output of the stock coils rather than change out a whole system and pay $$$? I dunno, just seems logical to go the cheap route you know?

I'll still have to make a wiring harness or at least modify mine won't I? But due to our coils being paired I'd have to tie the signals together rather than run four individual lines yes? Here is the wiring diagram for the Bosch Ignitor, It would have to be paired differently yes? No lie, I like the idea but I want to understand it before I head down that path, I already understand the other two, but this third one has me lost (well just a little - concept makes sense).

 

[polarmoment]

People go COP because it looks pretty, because other more modern car have it so "logically it must be better," etc. Most people don't fully understand how ignition systems work, and they therefore make decisions based on marketing and "what so and so is doing" rather than based on the facts of what their setup needs.

COP coils have very low inductance. The proper way to drive COP coils is to ramp dwell based on engine load. Short dwell for low load conditions, then ramp the dwell up for higher boost. This keeps them from overheating under normal operating conditions. There are only a handful of engine management systems that can do this, unfortunately.

For what it's worth, GM's 1000hp Buick indycar engines ran standard road car ignition coils. I used to have a picture I took with my phone of the Renault F1 engine in the back of Gerhard Berger's B197 Benetton, and you can see the standard Denso bike coil sticks in place.

You wouldn't have to tie any signals together. Just run it on only 2 channels. Leave 2 channels of the igniter empty. I suppose you could tie them together for a bit better on-resistance of the drive transistors. IGBTs have notoriously high on-resistance and tend to drop as much as 2V across them (so you're charging the coil with 11 volts instead of 13). They work great in CDI applications, but can be less effective in an inductive setup unless the dwell time is modified to take that into account.

 

[eclipsegsx1736]

If you wanted to run these types of coils, the "smart" coils with the built-in igniters would be the way to go. You can eliminate the factory igniter/PTU and wire these directly to the 0-5v falling-edge trigger signal coming out of the ECU.

If you eliminate the factory PTU, you'll lose your tach signal, but luckily our low-res crank signal happens to have the same 0-5v square wave and 2 PPR signal that the stock tach needs so you can jumper the crank signal to the tach input and it works fine.

However, you still run into the issue of proper coil dwell control. They will *probably* work just fine on the stock ECU with the factory dwell time, but they won't be optimised without proper adjustment of dwell time to suit these particular coils. I say they'll *probably* work fine because the stock ECU is already set up for waste-spark, so it cuts the dwell time down low enough at high RPM to prevent running the coils at too high of a duty cycle.

If you really want a good ignition system, get a Series II AEM EMS that has 4 individual coil drivers so you can run sequential ignition. These AEM "smart" coils in sequential with proper dwell control have been run well over 1000hp on 4 cylinder setups.

Beau

 

[Moonlight GSX]

If you wanted to run these types of coils, the "smart" coils with the built-in igniters would be the way to go. You can eliminate the factory igniter/PTU and wire these directly to the 0-5v falling-edge trigger signal coming out of the ECU.

If you eliminate the factory PTU, you'll lose your tach signal, but luckily our low-res crank signal happens to have the same 0-5v square wave and 2 PPR signal that the stock tach needs so you can jumper the crank signal to the tach input and it works fine.

However, you still run into the issue of proper coil dwell control. They will *probably* work just fine on the stock ECU with the factory dwell time, but they won't be optimised without proper adjustment of dwell time to suit these particular coils. I say they'll *probably* work fine because the stock ECU is already set up for waste-spark, so it cuts the dwell time down low enough at high RPM to prevent running the coils at too high of a duty cycle.

If you really want a good ignition system, get a Series II AEM EMS that has 4 individual coil drivers so you can run sequential ignition. These AEM "smart" coils in sequential with proper dwell control have been run well over 1000hp on 4 cylinder setups.

Beau

Here's the thing about the tach input, I'm switching my CAS/CPS sensors to the 1G black top CAS, so that I may, in the future, either advance or retard timing with adjustable gears. As you know the stock sensor won't fit under there unless I run stock cam gears. But that's a different topic.

Between what you just posted and what polarmount posted both are logical and I would like to get rid of a the factor of the PTU/Ignitor but then again getting a AEM EMS isn't cheap. So here's my question to both of you:

What would YOU do in this situation? I'm sure polarmount (taking a wild guess) would rock the Bosch 4 Chan Ignitor and call it a day as it is much cheaper but doesn't rid me of the "ignitor" portion. As for eclipsegsx1736... I'm not sure what you would run so I'm curious as to see where you stand on this as well.

Overall I have read that the AEM EMS is an awesome piece of work, but hard to work on and DSMLink V3 is much cheaper and NOW (not 3 years ago) more capable of being awesome - but in the wiki doesn't show that it can control dwell time. I'm wondering if they will add that in someday or should I transfer to the AEM? I have V3 right now, it's not on my mod listing because it's not installed (not even in the same country actually).

Let me know what you guys think...

Edit:
You also mentioned the 4 individual coil drivers, would this mean that I would wire it to four separate inputs on the ECU or am I looking at the wiring diagram wrong?

Re-Edit:
Just found my own answer the pin locations on the AEM module would be: P10 (coil 1), P23 (coil 2), P52 (coil 3), P62 (coil 4). From what this looks like I'll have to re-wire part of my harness to achieve this?

 

[polarmoment]

I would probably just run smart COPs. The 350Z/altima coils have done over 1kwhp, as have honda K series, but those are way too long for the DSM valve cover. Toyota 1ZZ/2ZZ/1NZ/2AZ would also work. Toyota coils also have an IG-confirm signal that might work with the DSM tach signal.

I actually designed a circuit that splits the IGT signal from the Honda ECU on my MR2 into 4 separate channels. The intent was to move to COP, and the breadboard version worked perfectly, but I really had no reason to do it, so it's still in my work box. For that setup, I had both 350Z coils and R6 coils with separate igniters. Neither was 'better' than the other, but the R6 coils fit better in my valve cover so that's what I would have done.

 

[Moonlight GSX]

I would probably just run smart COPs. The 350Z/altima coils have done over 1kwhp, as have honda K series, but those are way too long for the DSM valve cover. Toyota 1ZZ/2ZZ/1NZ/2AZ would also work. Toyota coils also have an IG-confirm signal that might work with the DSM tach signal.

I actually designed a circuit that splits the IGT signal from the Honda ECU on my MR2 into 4 separate channels. The intent was to move to COP, and the breadboard version worked perfectly, but I really had no reason to do it, so it's still in my work box. For that setup, I had both 350Z coils and R6 coils with separate igniters. Neither was 'better' than the other, but the R6 coils fit better in my valve cover so that's what I would have done.

But even with those would I still have to run some type of external amplifier? Such as MSD or Dynateck?