Ignition Fail on Dyno!!! Help Please.
 

I couldn't get a full dyno run in today because the ignition kept falling off before the rev limiter was supposed to kick in. Rev limiter was initially set to 7000rpm, but the ignition was cutting out consistently at a much lower rpm. Changing the knock window seemed to drop the rpm this occurred, but it remained consistent after that at 5800rpm exactly - WTF? I'm running 3.0.3u firmware. Please look over my msq and tell me if there's another setting that could be doing this. Yes, I know the rev limiter is set to 5750rpm now. It was not while attempting to tune.

Power was about 200HP at 5800rpm and torque about 180ish just before ignition fell off.

Details: 96miata with 99 bottom and top end, 0.040in cometic head gasket, RC550cc injectors in sequential, DIYPNP, 3.0.3u, boost at 9.5psi or about 165kPa, new NGK 6 series spark plugs gapped to .028", blue plug wires with less than 5k miles, no issues other rpm, seems to be software limit related. AFRs about 11 and advance about 13deg when issue occurs. Let me know if any other info is relevant.

have a log of the event?

Will it freerev through 5800 if you aren't in boost? I.E. while stopped in neutral, can you rev it slowly past 5800rpm?

Here is a representative run where it seems to lean out briefly at about the 5800rpm mark, which might also be uncombusted fuel from missing spark. 14143.906 is the point I'm referencing.

Haven't tried free reving it - good thought. Too late now but I'll try tomorrow. I also noticed that the mapDOT goes a little bit crazy and map drops about 5kPa abruptly, but not sure if that's normal. Could this be as simple as my line from the manifold to the MS? Will check that line also.

While we're at it. Any explanation for the seemingly false "knock" I get intermittently. I'm not sure exactly the window for this particular run, but likely about 4000rpm to 5500rpm.

Alright. Doesn't happen when free revving - not in boost. What's that tell us? Does that make it appear to be a boost specific or load specific problem? BTW, I did a quick pull on the street after free revving with the limiter still set at 7000rpm and it did the same crap at 5800rpm in boost. I won't believe its knock cause my timing is so damn low up top. I wouldn't think it would be the spark getting blown out cause my plugs are gapped to .028. A buddy of mine suggested plug wires as he had a similar problem with new bosch wires on his turbo colt. Again, mine are NGK blues with 5k miles. What about coil packs? I've heard that the failures would be more random. Hoping to hear something back from the community soon. Thanks

Uh well are you sure that knock is false? How much timing are you running at what psi at 4000-5500RPM?

If your spark cut-out is load specific (in boost), then that means your spark is weak.

Can be:

plugs
wires
coils
dwell

or some wacky ass MS shiz

Do you have a dial-type timing light? Might be worthwhile to make sure you are actually sparking when you think you are. 13 degrees BTDC is plenty conservative . . . assuming it's right.

Nothing else really stands out. There are some wierd spikes in you PW plot, but not in the area of interest. All parameters other than AFR are smooth and continuous from surrounding values. It definitely looks like a misfire and excess O2.

I'd also recommend logging ignition advance.

13*???????????

Damn that is absolutely super weaksauce. You can run more timing than that.

You SHOULD run more timing than that.

That's a good point. If you confirm your timing is correct, just remember that the higher the cylinder pressure, the harder it is to jump that spark gap. So, overly delayed timing "could" generate a misfire if you're on the edge.

To put this in perspective, at 35psi, from 4500-5500 I run 14* This is super duper retard but 35psi on my turbo is WAYYYYYY more power than 9psi on GT2x. Also WAYYYYYYYY more catastophoploey detonation potential.

At 20psi I run 17*

and this is on same 93 octane as you with much more airflow and power. No WI.

also, never ever upload "CurrentTune.msq" unless you want to make us all crazy. File>Save As that junks.

Dead Fael, [feyl]

13° at 10psi is a bit low, but not the issue, if he were to run 17° at 20psi, his motor would blow. Please don't compare your built low CR motor to a stock High CR motor.

Love Brain.


SCCAAAAAX, where are you souring the map line from? Do you have MAPdot enabled?

Hmm? I ran more timing than that on my FM basemap for 11:1 uberstock 03 factory pistons.

I agree that the timing is not the cause of the case of "runlikecraps"

Has it run to red line in the past?

OK, to address some of the above and some additional info.

I didn't think about the CurrentTune issue. Haven't run into it myself yet. I'll be sure to rename prior to loading in the future.

Brain - boost is sourced from the back top of the manifold (99manifold on 99 motor in a 96). It is bridged to feed the boost guage on the interior side. It is possible some small amount of boost could leak somewhere in that line. MAPdot is not used for AE if that's the question. It's straight TPSdot, but I'm at 100% throttle the whole time, so AE won't matter on the top end. How else would I "enable MapDOT?"

I was actually a couple steps ahead of you guys in thinking my base timing might have been lost somewhere along the way and between having the motor in and out of the car. Just checked with a timing light and the timing is running right where I've got it in SW between 14 and 15deg at idle - I had the laptop sitting in the engine bay while checking timing.

I also went ahead and purchased another set of NGK6's - I'm running the V-Power BCPR6E-11 gaped to 0.028" as indicated several times. They're not the standard 6E-11s, but I should be OK with these, right?

I also went ahead and purchased another set of spark plug wires - NGK Blues just in case. Wires and plugs will get installed tomorrow night when I pick the wires up. Will let you know how that goes.

Just to verify, did you guys have a chance to look at my dwell settings? I don't think this is the issue because my battery voltage stays between 13.4 and 13.8 very consistently and I believe I've never touched these values from the base map. A confirmation from one or more of you would be good.

So, I called up flyin' miata to see what there thought was. Coils - there guess was coils. My wires don't really have enough miles on them to likely be the issue. Apparently this is extremely common. So much so that they keep a supply of coils for the '94 to '00 cars on-hand at all times. Mother F###! I hate paying more than $100 for maintenance items, but I guess I'm nickel and diming myself in spark plugs and wires. Not to mention the additional tuning time on the dyno. They stated only way to tell was to swap out a known good set of coils. The bench resistance check will tell you that the coils are good, but not the ignitors.

They also recommended to check the resistance of the wires. I have a really shitty multi-meter and measured 360 Ohm, 330Ohm, 299 Ohm, and 278 Ohm front to back. Please don't read into the actual values because my multi-meter is likely well out of cal and I took these relatively hot. However, I was happy to see the values in the same range. I'm thinking likely not the plug wires at this point. Also likely not the relatively new plugs.\

DAMN COILS!

Coils

or

DWELL

someone please look at his dwell values. Yes improper dwell will cause high rpm boost-only misfires.

Shoot me a pm. I might have coils.

resistance on wires should be 16k ohm per every 3.2ft.

anyways, increase your tpsDOT to 75%/s.

on your log where you said the event happened is exactly where you surpassed your 50%/s threshold.





you should really move your map line to off the fpr, the 4th runner is really noisy.

Brain,

Is that TPSdot in the AE settings? That's only thing that makes sense.

Shueind, not interested in used coils, but thanks for the offer. Don't plan to worry about this again in the future. I go all new on 99% of my stuff.

Well if you are going to drop money on new coils, I do not suggest going with OEM coils. A new coil pack is $200-$230 from finishlineperformance.com, as opposed you can do a LS2 coils swap for about that price and have a much better spark output. For reference my OEM 94 coils would not do good above 9.5psi on the dyno. I would start getting spark blow out.

Also where in Dumfries do you live? I went to FPHS years ago.

yes, in AE wizerd

$140 Club Price at Mazda - Werd

It's the mother f'ing coils. I changed the plugs and wires out, moved the map line to the middle of the mani and made the TPSdot change Brain suggested. All with no results.

Damn - about $50 wasted on parts that didn't need to be replaced.

Problem is that the coils ship from Cali and I live in VA. AutoX Saturday, so I need to find a temp replacement if I want to go. Missed the entire season last year and looking forward to whooping up on some BMWs.

Thanks for the help guys and I'll be sure to post back after checking with another set of coils.

I live right near Walmart at 234 and rt 1, but moving out of Dumpfries...

Can anyone backup claims to stock coils not being sufficient for this level of boost - 9.5psi. I don't want to waste $140 for less than adequate solution. Is the LS coil swap batch also? If not, what changes will be necessary in terms of wiring and msq? Is the idea to use used coils? Aren't the wires like $125 on their own? FM sells an LS3 kit for like $700. F that.

I'd heard that the 01-04 COPs work well with moderate boost. Is this a good option? How would I use a timing light with no plug wires? Seems it would be cheaper. Do I have the hardware on the MS side to support COPs? That would be 4 sequential signals instead of 2 batch, right? What other changes and what wiring suggested - like special shielding, guage, etc.

The stock coils tend of vary on how well they work. Some people have good luck with them and they work fine for 15psi. Other people such as I have them crap out at a low level.

The solution for the longest time has been to go with Toyota COPS as replacements. They are relatively cheap and easy to do. You can buy almost PNP setups for $300-$400 from TSE. Or you can do it yourself for $75-$150 depending on how good you are at finding used coils and such. The benefits of COPS were that the fit perfectly into the spark plug holes, no longer need spark plug wires, they output a tach signal, and can be ran in wasted spark of sequential.

Now there is a movement to running LS style coils. FM has their kit which is expensive as you have seen. A big part of that cost is that they buy new OEM coils from GM which are costly. You can do it for a good bit cheaper if you are ok with DIY. The bracket from FM is $45 which is not horribly priced. Their custom spark plug wires are Magnecore and cost $89 with a lifetime warranty on them. You can order the wires from FM or get them straight from Magnecore for around the same price. There have also been several users who have used stock spark plug wires fine with LS coils. With the FM bracket there are 2 types of LS coils you can use. You can either use LS3 coils which is what they sell you, or you can run some LS2 coils. Buying the coils online from ebay is much much cheaper. They generally run $15-$25 depending on which ones you get. The pigtails for the coils can also be picked up for between $25-$40 depending on where you get them. Downfall of LS coils is that they do not output a tach signal so your EMS has to do that. Since you have a 96 miata your MS does this already so it is not an issue. Also so far everyone has been running the coils in sequential mode, but I do not see why you could not run wasted spark.

As for the comparison which is better between Toyota Cops and LS coils, it varies. The LS coils definitely have a much stronger spark then the COPS. The upfront costs of the LS coils can be a bit higher if you go with FM bracket and wires. On the other hand you can get brand new LS coils for about what you pay for used COPS. The pigtails come out to about the same cost.

I know for my first LS2 coil setup I spent about $125 total to get it up and running. That was buying 4 new coils from ebay, pigtails, making my own bracket to hold them, and then reusing my stock spark plug wires. I am now going with the FM bracket because it is much cleaner. I am also switching to a coil that will fit their bracket. I am planning on reusing my stock spark plug wires for the time being, but will buy the FM ones in a bit once I have some more money.

For the LS coil setup, I would prolly get the FM bracket and buy used coils online. Which one "fit" as you identified your current LS coils do not. Also, in sequential are you guys using coils designed for two cyclinders just with one wire, or does each ls coil feed a single cylinder? I don't see how they could also be used in batch otherwise. With the LS coils, is it necessary to run any additional wiring back to the MS, or can the stock harness wires be tapped? What's the wiring look like? I/Os on the board or pins on the connector would be helpful.

What exact changes are necessary in msq? I know dwell settings will change depending on which coils I end up with identically. Are typical dwell curves available online for LS coils from the manufacturer, or do I need to trust a forum map?

I am busy at work today and don't have access to all my info so I will post more later tonight.

The coils that fit in the FM bracket can be bought new here for $23.xx each.

For wiring the Megamanual has a ton of info on the coils. They have all the info you needs for Dwell settings and curves. James one of the coders for MS runs LS coils in his drag car. So there is plenty of documentation.

As for full sequential or wasted-spark, each coil has its own built in ignitor and coil. So I have each ones "signal" wire wired up to my MS3X. I see no reason why you could not pair the signal wires for 1/4 and 2/3 and have the coils run wasted spark. I just have never done it.

As for wiring it is not harder to do then COPS. I can help you with that if you need it.

Yeah, I think this is the way I'd like to go. I'm going to go ahead and purchase the FM bracket and coils tonight prolly. Interested to hear more about how wiring works for sequential - no fing around with batch if I'm going to do this thing. If I could reuse the stock harness, that would be great. Sounds like I'll be down at least two signal wires and maybe corresponding grounds? Is this something I've got to worry about either generating or being susceptible to noise?

Ah, In batch I would just tie the signals to the coils together - gotcha now. I was like - each coil only has 1 wire - WTF.

What ecu you are as that will determine how hard it is to run full sequential.

Here are the FM Big Spark Kit install instructions.

Diypnp

You would have to do some research and see if your Diypnp could support running sequential spark and if not what needs to be done to make it support sequential. Then it would just be running 2 additional wires back.

Looks like DIYPNP supports the two additional signals on WLED and ALED I believe. Would be great if someone could confirm. Also if someone could confirm if ALED and WLED require jumpers internal or if its just a software setting on outputs. Imagine I need a jumper from ALED and WLED to my DB connector and to wire in at my terminal boards. It appears 100 Ohm resistors will also be necessary at R1 and R2. Anyone that has done this on the DIY?

Also, thanks for the link to big spark instructions - very helpful. However, where do I come across said wiring harness used in the FM kit? It seems they pulled from a GM setup. Also, anybody know what the connector type is to interface with the stock coil plug? I'd like to keep everything as neat as possible.

Finally, dwell curves for the Ebay coils I just purchased at the link above.

Is it also necessary to jumper this way for this particular sequential setup outputing on ALED and WLED. I would imagine that ALD is ALED and WLD is WLED. It's not extremely clear.:

http://www.diyautotune.com/diypnp/do...sted_spark.jpg

"The picture above (attached) shows it jumpered for four separate spark outputs. You'd set it up this way for 4cyl sequential ignition, or for 8-cyl wasted spark (using two 4-tower wasted spark coil packs). All outupts, S1, S2, S3, and S4 would be used. These are the outputs to the ignition coils." DIYAutotune

Is it still necessary to activate and configure ALED and WLED in software outputs? I would imagine not, but not 100% sure if there is a relay that must be activated or anything else electrical I'm not thinking of. It seems for sure that R1 and R2 will need 100 Ohm resistors. I'll run the outputs for 3 and 4 to the DB connector. Grounding for the coils will be fed back through common ground on the main harness and sheild should not be necessary for this boolean signal.

It will not longer be wasted spark. Does that make it "Single Coil"? Does it stay "Falling Edge" and "Going High"? All other Ignition Options screen settings appear to stay the same.

Under More Ignition Options screen still looking for some dwell settings. Also some direction for cranking and not simply base dwell would be appreciated. Values against voltage would also be appreciated. Perhaps I'm not using the right search keywords, but not finding a whole lot on use of these LS coils with Megasquirt. Much less on LS sequential coils on Miata. And nothing on LS sequential coils miata diypnp. Opportunity for a writeup during and after? Wiring diagram shouldn't be terribly difficult.

Buy ls coil pigtails on ebay

Cut factory connectors and splice ls connectors on.

Prefer to butcher as little as possible on the harness. Anyone know where I can find the mating connector for wiring harness at back of coils?

I have never found a connector that looks like the one on the stock coil pack. To get one I literally cut off the stock connectors off a dead set of stock coil pack to make my set pnp. If you are going full sequential you would still have to run an additional 2 wires.

Ah, I'll butcher my crappy coils then. I realize I'll need wires for #s 3 and 4.

The correct dwell settings appear to be listed here(also attached): http://www.megamanual.com/seq/coils.htm#ls2

Would the correct settings in % be as follows and can anyone verify that all these dwell values make sense for the ls2 coils?:

Voltage Percentage
8.0 144
10 116
12 100
14 91
16 84

Another thing to watch out for is the coils wiring. The FM kit uses LS3 coils they claim. I do not know if those use the LS1 or LS2 wiring pattern.

Wiring pattern for LS1 and LS2 appear to be the same, right? Are the capacitors something not already contained in the MS spark circuitry? Do I need to add them in.

The wiring pattern is reversed between the 2. Look closer at the orientation of the small catch to hold on the pigtail.

My FM bracket came in today. I will be pictures of it and my coils in a bit.

The brackets are actually very simple. Two end pieces with rods to hold all the coils in place. For someone who can do good custom fab work, they would be very easy to replicate. For me who can not fab, they were worth every penny. I should get everything wired up and installed sometime this weekend.
https://www.miataturbo.net/attachmen...ine=1332856183
https://www.miataturbo.net/attachmen...ine=1332856183
https://www.miataturbo.net/attachmen...ine=1332856183
https://www.miataturbo.net/attachmen...ine=1332856183

With the coils on it just barely fits with the coolant reroute.
https://www.miataturbo.net/attachmen...ine=1332856183

I'd rather have seq. fuel before seq. spark.

and you run out of outputs really fast on a diypnp.

I have seq. fuel now. Not sure about Shueind.

So, nobody has answered my questions about setup in software or the issue with capacitors and other wiring on physical setup.

Brain, do you have seq spark on your setup?

Yes. But it don't have diypnp

I am running an MS3X so I have sequential fuel and spark on my car. I honestly do not know about what software settings have to be changed or what mods need to be done to the DIYPNP to make it work with sequential spark.

IIRC all you gotta do is move whatever's on ALED and WLED to the spark outputs.

PT6 and PT7 are already used for the fuel drivers.

So that leaves you with PA0 for the fan control...I think that's it, so if you dont need any other outputs (like boost or vics or whatnot) you can make it happen.

But don't quote me on it.

The 01+ VVT coils are good over 20psi.

I have some in my garage if you want.

I think I get it now. Output port settings are only enabled when you want to use the ecu logic to control them as with Knock In on PA0. That's not the case here. Spark output is only through hardware beyond when ALED and WLED are enabled for spark outputs 3 and 4 when single coil is selected instead of wasted spark. ALED and WLED are empty on my board - 99% sure anyways. Only other software settings that matter (other than single coil) are the dwell settings, which I posted previously for the ls2 coils.

Quick logic check. The coil currently feeding cylinders 1 and 4 holds the connector that should be used for cylinder 1, correct. Same thing with coil plug for 2 and 3 feeding the sequential for 2?

Just want to be sure. I'll label them now in that case.

OK, got all my stuff today. I'd post pics but it's identical so far to Shuiend's setup.

I labeled the harness plug going to Cylinders 1 and 4 as "1" and the harness plug going to cylinders 2 and 3 as "2". Then, I removed the old coil pack and temporarily put the new setup in place. I verified that the stock plug wires will fit this setup if you use 1 to 4 from right to left, which is exactly opposite what FM recommends in their directions (of no consequence).

Now, onto wiring. Each stock wiring harness plug has only got 3 wires, which must be (+), signal and ground. I'm wondering how to handle ground. I believe it is correct to have the signal ground go back to the ECU, but should the coil ground be run back with the signal ground, or should I tie all 4 coil grounds together and ground to the engine? I'm worried that the LS2 coils might carry a bit too much current back to the ECU. Actually, doesn't most/all of the current go into spark? WTF is there a 4th wire ground for on the LS coils?

Signal ground.

Power ground

Tie each set of grouds togrther. Four and four.

Versions of 3.1.x before 3.2.1 of the ms2/extra firmware had a sequential fuel bug that caused stumbled and misfires. Not that?

G

I have 3.0.3s. WTF. I hope that wasn't always the problem. Oh well, got all the equipment for sequential ignition now...

I'm currently tuning a force fed 94 and I'm experiencing spark mis-fire at about 4psi and above at around 5200 rpm.

This is using MSII+extra. Plug gap is .025 and the spark plug wires are good. Thanks for the tip on checking the firmware, I will have to do that next.

My question for you is, is it pretty common for the 94-97 coil packs to not handle boost very well? Is there any drop in upgrade that could be purchased to keep this MSPNP truly PNP? I could change the setting and wiring for Toyota Cops but then we can't switch back to stock ecu on the fly... Actually, I need to verify that his MS box has enough outputs for sequential spark and fuel.

Thanks!

Grant,

How much power is your car making at the wheel? Just curious, b/c the car I'm working with has a Jackson Racing M45 kit.

My Toyota cops run off the igniter in batch and are drop in plug and play.

G

My car is 210whp at 5800rpm and 9.5psi before the spark/fuel whatever takes a dump. Should be about 250 to the wheels max with a good tune. Yes, it should be possible to drop in another set of coils in batch mode. Actually, if I can't get the sequential to work on ALED and WLED for outputs C and D, I'll put it back in batch by simply changing settings in SW and running IGN1 to Cyl 4 and IGN2 to Cyl 3 on the wires I've already run to the board from those coils. You coul always strip back the stock connectors and put in an intermediary connector for each to swap out the pigtails without pulling the harness on the coils, injectors, CAM sensor, etc.

OK, someone please clarify,

Here: http://www.diyautotune.com/diypnp/do...n_control.html
under "Basic Logic Level Output", which this seems to be with LS coils, it states that Spark A is IGN1, SparkB is IGN2, Spark C runs to WLED and Spark D runs to ALED.

Some other sources have said:
Spark A is IGN1
Spark B is Cyl3
Spark C is IGN2
Spark D is Cyl4

I've also seen one that starts with Cyl4.

These appear to be contradictory. Currently, I have this wired:
IGN1 to Cyl1
IGN2 to Cyl2
WLED to Cyl3
ALED to Cyl4

Ignition set to "Coil on Plug" in software. R1 and R2 now have 100 Ohm resistors. WLED and ALED wired to Cyl3 and Cyl4 signal in. Dwell settings imported for LS2 coils.

Does all of this appear to be correct, or do IGN2, WLED and ALED need to be swapped around to match firing order?

Abcd = 1342.

use output test mode and just make sure the order is correct. it needs to match the firing order of 1342.


where:

spark A = cylinder 1
spark B = cylinder 3
spark C = cylinder 4
spark D = cylinder 2



it appears you need to swap c and d