JohnBradley

iTrader: (30)

88590015. That would work well since I have more than a few Evos that I can test this on at this time. Thanks

bradrs

Before you are TOO sure about the constant charge time, try to measure it.

I was told by a very knowledgeable source on the 2G DSM that there was a dwell map that was mapped via battery voltage. But when I actually measured it, I found that the dwell did drop as RPMs climbed.

I also monitored bat. voltage and charge current at the same time. Bat. voltage climbed a little at some points, but did not follow the same curve as the dwell drop off. And the charge current absolutely did drop off.

Logically, they would map the dwell time vs. bat voltage so that you had a constant charging current. That is not what I saw.

It would also make sense to have a minimum off time to allow for a long spark, and to allow for the coils to fully discharge(VERY important if you don't have any way to actually monitor coil charging current).

I would highly recommend measuring dwell using a oscilloscope directly on the signal wire going to the coil, before assuming it is what you think it is. And be careful about the coil charging. If it doesn't fully discharge before you start dwell again, your coil current will go through the roof, and the coil could overheat quickly.

Mattjin

I am actually not a beginner and I do know how to measure the charge time with a scope, which is why I said 99%. It is more likely not a dwell but a fixed charge amount. Dwell is more based on old style ignitions (points) with a charge amount based on crank angle, but this ignition is more based on a fixed charge time that is mapped via battery voltage (this part I quickly tested). At the moment having adjustable charge time is not high on my priority list but next time I am under the bonnet with the scope out I will be doing the necessary measuring.

Actually, more likely than a dead time, I more recently believe that the second map is for use during accel enrichment. Some ecu's map the dead time but personally I would have thought this would not need to be mapped at a factory level. Both maps are in the code next to each other so they are linked, and I once remember reading in the factory manual that the charge time is altered under acceleration. The bigger thing I am trying to find is the edge control for the signal, if indeed it is selectable. This would allow open adaption of just about any CDI system there is.

Anyway stay tuned, I will complete the testing within the next couple of weeks.

MalibuJack

iTrader: (5)

FWIW The stock coils with an HKS DLI2 worked fine for me until I started making more power. I tried several different COP combinations and setups, all of which had problems. I went back to the stock/HKS combo for the past few years until I picked up a sparktech COP setup.. Its basically using an M&W Box and Chrysler 300m coils. It works really well and doesn't give me any drivability problems.

The problem with some ignitions is their REALLY sensitive to plug temp and gap, and can take some trial and error. In my case, I went back to a stock heat range, and slightly tighter than stock gap. Previously I was running NGK plugs 2 steps colder, and a tight gap at .25

The one thing I found was a compatibility problem with the UTEC and M&W ignition box, this is definitely related to the settings your looking for in the ECU, however removing the UTEC solved the problem for me.

The short answer is there are definitely maps for coil configuration, charge time (dwell), and a trigger offset (ms to wait before starting to charge, aka dead time) however none of these have been tested well or documented. I recall also seeing a trigger output voltage as that has to be regulated to keep from burning out the ignitors on the stock coils. This is also relative to battery voltage as stated in a previous post.

Most of this data is either not discussed, or fairly new. I've been aware of it for awhile but didn't notice that their not in most of the ECU definitions as I don't use it on my car as the Sparktech COP setup works fine with the stock trigger settings, and the dwell setting is not relevent for CDI, only the trigger dead time.

bradrs

The map I was pointed to on the 2G was for battery voltage as well. And I was told that there wasn't any other adjustment for RPM. So I made a modified EPROM with identical table values(so that it should be constant time no matter the voltage), and then ran the tests. It still dropped off with RPM. I don't have enough insight into the factory code to know what the cause of it is, I could just tell that it wasn't behaving as it should, according to the people who have reverse engineered much of that code. If your results show that the RPM is irrelevant on the Evo, that is good to hear. It would make it much easier to create good maps for different coils and ignitors. I am just pointing out that if you haven't tested with respect to RPM, you might want to look at that while you are looking at the battery voltage test.

I couldn't tell from your description. Did you actually test at different RPM while doing that battery voltage test?



Aren't there plenty of good CDIs that have that capability? The companies that don't have this capability are that way because they aren't willing to respond to the market. Even if you add support in for them, they still would lack harnesses. And there are companies on the market that already make plug and play CDIs for the Evo, that will work as is.

If you can get the dwell stuff figured out, that would allow more options for inductive ignitions. And they can improve the charging time, even with the stock coils. Getting the stock hardware working better would probably be more interesting to most people.

Mattjin

Verified today. I have the coil charge time map. I had the scope out on the bench for work on another car so I thought why not do a quick test. Tonight sometime I will post it for 8859. Actually, it might be better if I post it in the New Maps thread so its easier to find for others. The dead time I didnt test though, and in fact there are 3 maps related to the charge time. That will require some high rpm work with the scope connected and at the moment its not that important a map for anyone.

*** Posted ***

bradrs

Verified meaning that map #1 was just coil charge time with respect to battery voltage, and flatlining it produced a constant charge time across all RPM?

That is very good info.

Mattjin

You know what they say about looking a gift horse...

The map is there and responds to battery voltage. I have no idea what it does much above idle because I spent all of about 2 mins with some quick map changes to verify it has an effect before I returned to the customers car I was working on. I know it works, now someone else can finish off the map data while I get to work looking for other more useful map information. It might trim with RPM, it might be a dwell map and not a charge time, but I know for certain that if you do something like knock 50% off that map it will knock 50% off the signal to the coils.