ICE ignition
 

Hey all. I got my 2010 R3 this past May and it wasn't the strongest engine. However, after a track day it started misfiring and lost more pull. It sometimes even stalls at stop lights.

I went to ICE ignition and Michael offered me their new LSX coils. He said they would be great for the RX8. He says they're about twice as powerful as GM coils, and even a little better than the IGN-1A. I expect the test results shortly.

I just got the coils and wires today :mdrmed: I got the ebay harness, and new set of plugs off amazon. I hope to get it all together this long weekend. I need to set dwell soon.

"About twice as powerful as GM coils". Seems legit.

You'll need to find a way to mount them. Do they provide a spec sheet like dwell time, charge time, discharge time? I'm sure they're more than suitable, but would like to see numbers. Keep in mind that our coils need to be able to fire every 5-6ms, which is more than most engines ever do.

I'd like to see the A Little Better Than 1GN-1A Study.

You do realize many of newer 8 owners have never seen an 8 track?
:)

Our old Chevelle (bought new) had a eight track.

I didn't realise they look similar to the MSD blaster coils :/ It's a new product for ICE and I've requested some clarification along with the test results. He's not claiming multiple sparks. How thick is the MSD iron? The ICE coil weighs 430g. Power is directly related to its mass.

Loki, I see your point about dwell. 9000 rpm is 6.67ms per revolution. Dwell for these coils is 5.8ms @ 14V (10,344 rpm), 4.8ms @ 16V (12,500 rpm) with their voltage booster.

edit: Michael just replied to me (it's currently 9am in Oz). The MSD blaster coil is apparently a copy of the stock GM coil. Its iron is 11 mm thick. The ICE iron is 21 mm thick; almost twice the mass (and obviously not identical). I'll post the test results as soon as I get them from Michael.

in on this upcoming shitshow

We need more reputable members' input.

Charles, that is basically porn. Thank you. Thank you ;)

From Michael K at ICE:

"Ok, some figures for you.

At 10 amps primary current (which is where the dwell table I gave you runs our coil) they are neck and neck, 144 vs 142m/j in the IGN1’s favour. Our coil can run forever at this amperage and not get hot. Can’t say about the IGN1, but should be the same.

At 12 amps primary current, it is 160 vs 172m/j in our favour.

At 14 amps primary current, it is 172 vs 180m/j in our favour again.

The IGN1 suffers more as we push harder and would have to run hotter at the higher settings.

Will have results at 16 amps primary current in the next couple of days. I’ll let you know"

I also learned yesterday that D580 coils use a different connector than D585 coils...as well as the harness I bought. Same pin out though...

If they claim it's the best then they should become a vendor and sell them here.

Good morning.
As you may remember, I have your kit and recently installed an SBG kit more or less to see how they worked as a anecdotal type of review for my, and others here as well, benefit.
I know very little about the technical aspects of the ignition systems as you just detailed, am N.A., and can only give impressions from my butt dyno.:)
I have kept your kit in case I was not satisfied with the SBG kit.

As when I installed your kit, I noticed better performance, but, I also realized, just as when I did yours, I also installed new plugs, cleaned the MAF sensor, installed a new air filter, and reset the NVRAM. (20 Brake Stomp).
So any increase in perceived performance is to be expected, imo.

I really can't say there is a difference between them based on that, and other than the heavier wiring harness (which might be better, Idk) and the different mounting bracket which is aesthetically pleasing to my eye, I don't know if the increased cost of the SBG can be rationalized for a stock application.

I also am aware of another owner who experienced a failure of one of the IGN-1A coils on the kit and switched to your kit.

Do you believe this statistic you quoted affects the performance of an RX8 with a stock setup?

I always appreciate your feedback, and your availability has always been a very positive aspect of doing business with you.

I wondered if the 40% duty cycle you referenced could actually be detrimental in N.A. application, or could that be more of a concern for those tuning?

I'm sure you understand most of this BC, but for the uninitiated...

Duty cycle is the on vs off time for any given piece of electronic equiptment. A 100% duty cycle capability would mean that a part could remain in the On state indefinitely without degrading, a 50% duty cycle rating would mean that for the time spent On there would have to be an equal amount of time spent Off (a 50/50 split). For a coil On time would be the total dwell time, with Off time being the "rest" time between the coil discharging and the coil beginning to charge again.

For most electronics (coils in this case) exceeding the maximum duty cycle will cause an exponential increase in the temperature in the junction of the BJT. Junctions of transistors are annoyingly fragile, and thermal runaway sets in almost instantly, so that increased temperature (caused by too much power) can lead to increasing temperature (thermal runaway) which leads to immediate failure.


Here's two examples. Lets say all of these engines are running at 9000rpm for simplicity.

9000r/min = 150 r/s which gives us (1/150)s/r = 6.667ms/r

Since we're talking about duty cycles for ignition coils the ratio that we're looking at is dwell time/total time, and we're going to look at that 40% number and a theoretical 100%.

In a conventional piston engine (not using wasted spark ignition) each cylinder fires once for every two revolutions. At 9000rpm we get our 6.667ms/r * 2r = 13.334ms between every ignition. To get the duty cycle we just have to do Required Dwell/13.334ms, to reach a maximum duty cycle of 40% we do 13.334ms * 0.40 = 5.33ms Dwell timeto avoid overheating the coil. At a 100% duty cycle the maximum possible dwell time is the full 13.334ms.

Now, in a rotary engine (or a piston engine with a wasted spark ignition system, technically) the revolution time stays the same, but the coils are each required to fire once per revolution. That leaves the total time at 6.667ms, meaning that to reach a maximum duty cycle of 40% we do 6.667ms * 0.40 = 2.67ms Dwell time to avoid overheating the coil. At a 100% duty cycle the maximum possible dwell time is the full 6.667ms.


To prevent the coil from overheating there are two types of protection systems built into some coils. The first is an auto-discharge system, used in coils like the D585 (I run them in my car), which discharges the coil after a certain amount of charge is stored in a capacitor. The second is power limiting (I think most coils have this) that effectively stops charging after a certain amount of charge is stored, I'm assuming that's what the IGN coils have built in. Each has their downside, an auto discharge can cause a high power spark at a bad time, while a power limiter can cause a weak spark if the dwell time is set too high.

In the RX8 on the factory S1 tuning, coils are set to dwell for about ~3.5ms, which is already well beyond the 2.67ms maximum of our unnamed coil with a 40% maximum duty cycle, but since it's protected by power limiting all it will do is limit the size of our spark. On the same factory tuning we would need a coil witha minimum duty cycle of ~3.5ms/6.667ms = ~50%.

So yes, a coil with a 40% duty cycle isn't suitable for any Rx8 application (boosted or not) if the RPM range extends above (1rev)/((3.5ms/0.40)(1s/1000ms)(1min/60s)) = 6857rpm, because at or above 3.5ms Dwell we run into either the power limiting or we overheat and destroy the coil outright.

Thanks guys.

So Charles, in plain talk, you think the IGN-1A coil does not fire as efficiently at higher RPM as the GM coils you use.
Am I correct?

Legot, I am still learning about these things, and because of certain influences beyond my control, at times I have had trouble grasping things.
(Sometimes the little hamster falls asleep in the wheel :))
If I understand you correctly, you are saying that you believe the 40% duty IS detrimental.

The reason I'm asking so specifically is that my 8 is still running great, but I have noticed at the upper end of the RPM range a type of momentary stumble or plateau that I had assumed may be because I forgot to clean the ESS when I installed the SBG kit, or perhaps some other reason I'm not aware of.

If there is a real possibility that the SBG coils are responsible, I would seriously consider re-installing the BHR kit without doing anything else, to see if that stumble goes away.
It would be a PITA to do it for no other reason, but I am willing to do it for the greater good.:)

Yes, the math says yes. I've never used either kit so I'm not going to comment on quality, but if a piece of hardware isn't capable of doing what you tell it to do, it probably won't work as well as it needs to.

Great. I did a lot of reading before I bought it, but I don't recall that bit of info.
I might have read it somewhere, but it didn't register if I did.

Charles, you're right about coils putting out "enough" energy. I don't know about the Renesis, but Yamamoto has a chart that shows performance on a 1970s Wankel plateauing around 120 mJ of spark energy. The better the combustion system, the less energy you need. An OEM is going to use a good enough system to meet emissions etc. Under normal circumstances, I don't expect much gain with an upgraded coil. And like you said, "aftermarket" doesn't guarantee better performance, or reliability over stock.

A little back story, I didn't set out to look for coils for my RX8. I'm working on a mountain motor, and I wanted a killer ignition. I was going to use ICE's race coils, as they put out 150 mJ with their 7 amp street system, and can put out 200-300 mJ. Over several months of working out the details, Michael told me about his new LS1 coils and so I went with that. I bought my car in May, and I asked him how they'd do with the rotary, and he said they would be good.

He also assured me that the wires don't put out much RFI. Aside from the EMU issue, I don't want to hear static on the radio.

Perhaps Michael's number mainly show how robust the coils are. It seems the numbers that really matter are optimal dwell time, max duty cycle, and how much actual power the coils put out at 9000 rpm.

I continued the discussion here because I believe it's relevant the the ICE thread.
If they believe them to be superior to the Merc coil, and if my issue disappears when or if I reinstall the BHR kit, then I think that would be relevant to this thread and beneficial to the entire forum, which is one of the reasons why I bought another kit I didn't need.

It wouldn't be scientific, but it would be enough to convince me.

DC is not inferior to AC in regards to being transmitted long distances along naturally resistive wires. In fact, HVDC links are well established practice nowadays. The difficulty occurs if you are using a low voltage instead of a high voltage. If the same wire is used in both cases then if you increase the voltage by a factor of 10 then you reduce the resistive losses by a factor of 100. Changing the voltage up and down is quite straight forward with AC - just uses a transformer. With DC this is very difficult, and in the early days just not economically practical.

I would say that an ignition coil emits an alternating voltage - the coil output swinging positive during the dwell period, and negative when the primary current is switched off. The actual current that flows in the plug is more of a pulsed DC - nothing flows in it during the dwell period when the coil swings positive. What's more, GM LS coils have a diode in line with the secondary winding where it goes to ground that blocks any positive output voltage, so this further guarantees the plug gets a pulsed DC current.

I don't think you understand the relationship between current and voltage well enough to be making assumptions. What you're saying really doesn't make any sense.

Edit: Just so I don't seem like too much of a jerk, if we were operating in DC for ignition we wouldn't be seeing the AC waveform, we'd be seeing a fairly smooth ramping of DC voltage. That would be because a DC system would be using a DC-DC boost converter rather than primary and secondary coils (which physically can't be used to step up DC).

WhatI am saying is, the coil puts out voltage in both directions but because that voltage is assymetrical the plug only conducts current in one of those directions, hence my assertion that the coil puts out an AC voltage but a DC current flows on the secondary side. The plug is acting as a crude rectifier.