Thats great news about the breaker. 200 amps is much more appropriately sized for your 0 gauge cable. Eaton Bussman has great stuff, that is what I used for all of the electrical gear on my car.

 

I'm glad you told me that you didn't have issues with the breaker popping while trying to start, I was thinking it was normal and just something I'd have to live with lol. This one seems to be much more quality than the old one, that's for sure.

 

I ended up having a cam sync issue while trying to fire up my car on the Fueltech ecu recently. I was cranking it endlessly and it was seeming to be so close to catching. I was really worried about cooking the starter, but never had issues with the breaker. Ended up putting some filters on the cam/crank sensors through the Fueltech software and the thing fired right up.

 

Glad OP fixed this! I've had similar happen from:

• Spark blowout. It was so bad the car would glitch out and lose sync. Chased a sync issue for a while but eventually an ignition system upgrade fixed it!
• Trigger settings. Had to switch from going low to going high to fix a weird intermittent sync issue.

Just in case anyone troubleshooting in the future finds this! So glad to see a thread like this where OP was able to find the fix. **** like this is what makes forums great.

 

Totally agree. Thank you both for posting up your experiences too. This is the kind of info that needs to be posted to help others.

 

Correction, my issue is not fixed. I took the car out and did a couple more pulls tonight and I got a CEL for 'trigger reference error' again and on the second pull, it killed the engine.

I keep forgetting to set up the new knock sensor that I installed a few months ago, so maybe that's the issue. But even if it isn't, it obviously still needs to be done. So between that, building a new harness, and getting a new tune from scratch, I'm not sure what else to try. Actually one thing I was thinking about is maybe changing the TPS. We checked it months ago, but maybe there's an anomaly at the top end that we didn't see.

 

Another thing I don't think I've mentioned that is kinda odd; one of the OBDII apps on my stereo will open immediately when the problem occurs, even when I don't have any adapter connected. Once again, I'm not sure if it's a feature or a bug. It seems more like a bug though.

Also, my flex fuel sensor might be going bad. I'm saying this more so because the temp reads 192F when the car is cold and then drops as I drive it. The ethanol reading seems like it's OK. It does read 1% when the key is to accessory but it always reads 8 or 9 when it's running. But it makes me wonder if it's affecting this too.

 

A few updates.

The flex fuel sensor just needed to be set up in NSP, and that's good to go. The knock sensor was set up at the same time, so that isn't the issue either.

I built a new harness, running the CAS wire as far as possible from the coils just in case there is any interference, and it's not near the alternator either. I also replaced the Hall Effect CAS, and none of these changes have fixed my issue. I guess my next step is trying a new tune, like C Ludwig suggested. Maybe first I'll check the spark plugs to make sure they're not too worn down, or replace the coil packs in case there are some bad ones, like fireindc suggested. I upgraded to IGN-1As but that was about 11 years ago. I've already replaced one coil when I noticed it was bulging a couple years ago.

I haven't pushed it to WOT throttle yet since these changes, but my tuner says you can see it in the logs, trigger tooth count is all over the place instead of a flat line. So I've just been taking logs of regular driving instead of doing WOT pulls.

 

I came across this thread while researching causes for misfires and CAS errors.
Plugs were ruled out, as NGK R7420-9 are fairly new and not fouled.
The harness didn't seem a likely culprit, as it is a 5 year old Rywire harness, and the car is autox only. The only drive time is has seen is some breakin, autocross and dyno.

Now, if the CAS were easy to get to, I would have taken a close look just to be sure. But it is buried behind a custom catch can behind the alternator, and that is a lot of disassembly I would like to avoid.

Next most likely was the ground connections, and that did yield good results.

Most existing grounds were disconnected, the surface of the mounting point sanded down to bare metal, the lugs cleaned, and re-mounted.
The ground from the upper intake to the firewall was removed, and replaced with one from the firewall to the engine using 4 AWG cable.
A 4 AWG lead now connects the battery directly to the engine.
The ground from IGN 1A coils, the battery and the firewall are all connected within 2 inches of each other on the engine.
The negative power lead from the coils is connected directly to the battery.

At an event a few days ago, I experienced no misfires.
Checking the Haltech logs, there were no trigger errors. Previously there had been quite a few.

Additionally, the starter has never spun as fast as it does now. The FD starts on the first try

I guess this is just a reminder that the whole system needs to work.

And I am going to get out the boroscope and check that CAS connector.

 

So are you pretty certain that it's fixed now? It's a good sign that it performed well recently. Was your trigger tooth count all over the place before? Were you feeling the misfires during normal driving or did it feel normal like mine and you just knew of the misfires based on the logs?

One of my local buddies said in the past he's put all his issues into ChatGPT and was able to fix them, so I figured I'd give it a shot. After going back and forth for about an hour, Chat determined that the most likely cause is the coils being old enough that they're allowing EMI through the sensor ground wiring. It suggested swapping the coils, and I bought some Wednesday night from SBG. It also suggested disconnecting the sensor ground wiring from the coils and instead running that ground from the coil to the engine block, which I will not be doing lol. The coils are already grounded according to the instructions, with B to sensor ground, C grounded to their respective housings, and D grounded directly to the battery. And power coming from the ignition relay which pulls directly from the battery through a fuse.

I haven't checked the spark plugs yet, but I probably will. I have 4ga grounds from the chassis to two places on the UIM, and one place on the block near the original ground from the starter harness to the block (actually I may have removed that one, I don't remember). I have the battery grounded to the transmission with a 0ga cable and battery power through a 0ga directly to the starter. Maybe I'll move one of the UIM grounds directly to the block near the coil grounds, but I haven't had any issues with them before.

I've had an RX-8 2kw starter on the car for several years now so it already cranks stupid fast lol. One of the best mods. My buddy and I built my engine in 2014 and it still fires up fast hot or cold even though the compression numbers are in the 80s. But I had those numbers about 8 years ago when I tested it, so I don't fully trust compression tests anymore.

One other possible cause I was thinking was maybe the bearings are worn out to the point that it's moving the trigger wheel far enough on WOT pulls that it's causing the issue, but it's still moving enough during normal driving to cause the trigger tooth count errors. Chat suggested end play might be the problem too. The gap specs fine at around .9mm, but that's with the engine off of course, so I guess it's still possible. The engine is about 12 years old afterall.

Side note, I like your sig. I've told people that it's my own personal roller coaster.

 

I don't do any "normal" driving in this car, other than staging in and out of grid. It gets trailered to events.

The misfires were definitely noticeable, I could feel and hear them, 1 or 2 per event.

Logs showed a lot of trigger errors, don't know about tooth count, I will check on that.

Definitely fixed now; no misfires that I could hear or feel, and the trigger log is flat - no events.

I also did the RX-8 starter mod awhile back. It is just so nice to start on the first crank.

The starting wiring you mention seems a good idea. I used 4 gauge, which seems to be performing well.
​​​​​​​
Re the Sig: yeah, my favorite quote from a passenger.

 

Interesting. Mine either feels fine but still with tooth count issues, or loses everything at WOT. I've never felt it miss. Good to hear yours is fixed though.

I probably went a little overboard with 0ga lol, but at the time I was chasing starting issues. I had the blue connector jumped but it wasn't enough, so running a 10ga wire all the way from the blue connector to the starter fixed it. I already had the 0ga installed so I left it.

I also just rediscovered another issue where my ECU is getting power when just connecting the battery, and it seems to be sending power through B11 and A26 to the relays rather than receiving power from the relays like it's supposed to. I thought I had it fixed but I guess not, and it makes me wonder if that's part of the issue.

 

What do you mean by it's sending power to the relays? If you pull relays, and probe the terminal at the relay that goes to ECU pin B11, you see +12V on that terminal? B11 is switched power into the ECU, and also is connected to the crank and cam +12V wires on the "4 core" in a typical Haltech harness. Is it being "backfed" by one of those? A26 is from injection relay to ECU and injectors. If you pull the relay and probe the terminal at the relay that goes to ECU pin A26 are you seeing +12V at that terminal? Is this an "all Haltech" setup that uses their fuse/relay block and wiring or is it a custom harness that is wired to the FD fuse and relay block equivalent of what would be provided by the Haltech stuff per their wiring diagram? You said before that you're losing trigger and WBO2 right? Trigger (crank position sensor) is fed +12V by B11 and the WB1 typically gets power from A26 if you're wiring per Haltech diagrams. If you have some weirdness in your wiring that is causing you to see voltage when you shouldn't on those pins, it could cause issues with the things being powered by them through noise, ground loops/potential differences. A wiring diagram of your harness would be helpful in diagnosing it. You need to verify the only +12V "in" on B11 or A26 is coming from the appropriate relays in the relay/fuse block and power isn't coming "in" from somewhere else that is separately powering something that typically uses B11 or A26.

 

Yep, that's exactly what's happening, but I figured out where it's coming from. With this harness build and 2500 install, I made a fuel pump harness that goes to three relays in the back. Those three wires going into the ECU (A24, A7, and A8) are getting voltage from the opposite side of the relay (85) because I initially wired it backwards and had the power going straight through the diode into 86 and on to the ECU, then that power is being transferred to B11 and A26 and sent to pin 87 of the ECU and injector relays. I rewired it so the power is going into 86 and should be blocked by the diode, but it didn't work because the relays are no longer blocking power, if they ever did. So I need to find a relay that will definitely block that power.

On top of that, my two radiator fan relay wires for relays 2-4 (A31 and A32) are feeding power into the ECU as well. I swapped the front harness three years ago and I've had some weird power issues ever since, but I was able to get the car operating normally so I didn't think much of it. When I installed the new ECU and engine harness this winter, I tried to go back to the stock fuel system and it didn't work, so that's even more proof that it's messed up.

My TMS4 chirps when I connect the battery, and that's caused by the fuel pump relay wires. I also have "residual voltage" of around 2V on pin 87 of the ACC relay, which feeds Vehicle CAN, flex fuel, boost control solenoid, EVAP, and IAC. When I disconnected each of those connectors in the engine bay (aside from EVAP and IAC), the voltage would increase, or at least I think it was, which I found odd. But once I removed power from the fan relays with the 60a fuse, the residual voltage went away along with the power going to A31 and A32.

So the fuel pump relays and front harness are causing all my issues. Hopefully it's causing the trigger issues as well lol. I'm not using the Haltech relay box, but I have mine wired exactly like theirs. My tuner says it's just trigger issues. When I've looked at the logs, it looked like I was losing more things, but I'm not the most knowledgeable on that, so it could just be trigger issues.

 

The diode across pins 85 and 86 in a relay is for flyback suppression and it will not control current flow across the coil, in any case you would need current flow to the ECU on a relay trigger wire since the ECU is just grounding the coil through its transistors. It is correct to power pin 86 and trigger the relay via 85 going to pin A24, and so is seeing +12V at the ECU on any pins that trigger relays. Seeing power at A26 or B11 from another source when whichever relay or switch is feeding your ECU/harness to injectors and such power is disabled is what isn't right, you don't want multiple current sources on any circuit because not only does it allow for odd things to happen due to differing voltage drops on those circuits and possibly propagate noise, but it compromises fuse protection.
So, ensure that A26 and B11 only see +12V when injection relay and ECU relay (haltech terms) respectively are enabled, these relays triggered when the ECU grounds pin A25 when power is seen on A13, which sees +12V from your ignition switch as in stock FD wiring.

As before, a diagram is nice to have (IMO, having done wiring and built many ECUs, mandatory) for these sorts of things. Stuff like seeing voltage on pin 87 of some relay that should be powering multiple things is odd and would be pretty much impossible for anyone to figure out without being there or seeing a diagram.

 

IMHO, I would still look at your crank angle sensors. I have had an ignition problems with misfires for over one year with 6 melted IGN1 coils, many fouled plugs and poor idle. My crank angle sensors were stock Denso's and tested good on resistance testing. I have changed many coilsand even went to a RMagic system and fried the ignitor too. I was troubleshooting everything. Finally, I bought an automotive 2 channel oscilloscope and found the one CAS for cam G was bad despite normal resistance. It was randomly dropping signal on cam G and showing variable wave heights. I then replaced them with some from Atkins, but unfortunately, they were new knockoffs and both were having intermittent poor signal. I then bought two new OEM's from Ray Crowe and one was good, but the other was still having problems. I even messed with the gap many times to try to get the bad one to work. No go. I finally bought another new one from Ray and both work perfectly on the oscilloscope now. My remaining problem is the power spikes from the melting coils screwed up many of my electrical components. It even fried my electrical load control unit which made my alternator stay at high output all the time (15V or even a little more), which hurt my battery too. These small handheld oscilloscopes are not high-end high bandwidth units, but for $85 with a signal generator too, it found problems and I likely would have never figured out what was wrong without it.
Mike

P.S. Even though my CAS are stock variable reluctance (analog) sensors, the oscilloscope will evaluate Hall sensors too.

 

It looks like the Haltech has a built in oscilloscope. It would be worth while taking a look at the outputs for the two crank angle sensors. I found this overview on Haltech's website to guide through setup: https://support.haltech.com/portal/e...u-oscilloscope

I had to do this on my car during initial start up on my Fueltech system. I had a little bit of electrical noise on the two channels and I was able to apply some digital filtering and get the sensors to provide clean data. The built in oscilloscope helped diagnosed my issue.

 

Oh yeah, I forgot about the wiring diagram. I drew a rough draft at work and I was going to draw a nice one online but maybe this one will be good enough for now. This is the way it's wired now, before I had 85 and 86 swapped. I bought new relays and I checked them today. They still show the same resistance on both sides, but I did notice that some of the relays are giving different readings from each other, and one of the old ones is reading no resistance at all. The best one is at about 99 ohms, and most of them are around 97 or 95, with one in the 50s and one in the 20s, and the one at 0. The only wires I'm not showing on this are the 5V, signal ground, input, and resistor being used for sending fuel level to the ECU, but those wires don't seem to be an issue.


Maybe I'm misunderstanding, it's SUPPOSED to be sending 12V all the way to the ECU with just the battery connected? Because that's what it's doing. I'm not sure if they've all done that in the past, I never checked them because I didn't have these issues. So if that's the case, even if I fix the coolant relay wiring, it will still send 12V to the ECU on those wires as well? Those trigger wires see 12V even if the relays are removed, so I know there is another problem there. The ECU itself is what's passing the voltage to the ECU and Injector relays through A26 and B11. If I disconnect the ECU, that issue goes away, along with the TMS4 chirping. And with the 1500 swapped in, I don't see voltage on A26 at least, I don't remember about the ECU relay. So either way I'll be sending the ECU back to get repaired during winter.

I did change some things Friday night. I completely removed the coolant fan control module and wiring harness from the recall in 93 and 94. I had removed the module itself years ago but the harness was still in place until Friday night. And I only put one fuel pump relay back in instead of all three since I'm only using one pump currently. Anyway, the TMS4 module doesn't chirp anymore when I connect the battery, but it does after I shut the car down, although not as fast, it will give a chirp about every three seconds.

I did get the CEL for trigger reference error Saturday morning after only starting the car and idling, so that's a bit of a change as well.

I replaced the crank angle sensor and still have the issue, so unless the new one is bad, I don't believe that's it. Although as many problems as you had with yours, maybe that IS it lol. Holy ****. I'm using HE, not sure if they're any more reliable or not. I've never heard of so many issues with the stock VR sensors. I'll take a look with the oscilloscope though, or have my tuner do it since I have no idea how lol. And I'm still planning on swapping the coils as soon as I get them from SBG, should be this week or next hopefully. Thanks for the input!

Yeah I'll take a look at that. I'm really glad I don't have to buy one, I'd never figure out how to work that lol. Thanks!

 

EDIT: NOW my post decides to show up after I posted it again lol.

 

If you search for an electrical diagram of how relays work, there is an electromagnetic coil between pins 86 and 85; it's normal to measure 12V at pin 85 when the ECU is not pulling it down to ground.
Example diagram here, it doesn't show an ECU so you have to imagine the ECU output pin is acting as a switch that is open or connecting pin 85 to ground. https://www.hagerty.com/media/mainte...relays-part-3/

The relay trigger outputs from the ECU are probably not designed to see constant battery power on the other side of the relay coil. Try moving the relay coil power (the 'Fuse' in the lower-left of your diagram) from the battery to the same switched 12V source that turns on the ECU. That may not solve the problem, but it should avoid the ECU (and whatever else) getting powered when the key is off and might help avoid draining the battery as quickly when the car is off. The relay coils shouldn't draw much power, usually less than 1 amp each, so you shouldn't need a huge wire to send switched 12V power to the fuse.


How is the hall-effect crank sensor getting power? Is it sharing a length of wire with the wideband or any other 12V device? How is the hall-effect crank sensor getting grounded? Is its ground getting shared with any other devices?

 

I was thinking the diode was supposed to block that voltage across. Good to know.

One of my local buddies suggesting moving that wire to 12V switched too and it didn't make sense to me at the time, but now I get it, and I feel like an idiot for not seeing that before lol. That should fix that part of the issue, since it will only be sending that voltage for the couple seconds I have key on before starting. Thanks!

I usually have my battery disconnected when the car isn't on, but this would still be better. I figure I'll use 20ga for that wire.

I'm using a Haltech flying lead harness, so it's getting power directly from 87 of the ECU relay, which goes to CAN, and also B11 which has been getting voltage due to these power issues, although it's around 2V and not 12V like A26 has been seeing. And it's grounded to signal ground, which is all tied together so it's tied with every other sensor and anything else (IGN-1A coils) using sensor ground.

 

The relay coil power might not matter when the ECU is powered on, but with some luck it might help to stop the weird voltage thing. Many ECU output circuits have a flyback diode, when something like a relay or boost solenoid turns off there will be a high voltage spike when the magnetic field collapses. When that voltage at the output pin goes above the battery voltage, the internal flyback diode provides a path for the voltage spike to get dissipated to the ECU's 12V power supply pins (and maybe even back to the vehicle's battery) to avoid damaging the MOSFET / transistor in the output circuit. When you have the coil of the relay connected to the battery but the ECU's 12V power supply pins are not receiving power, that flyback circuit can send a small amount of current out to the ECU's 12V power supply pins. To be safe, it's a good idea to check that the crank sensor 12V supply looks good after re-wiring the relay coil power to switched ignition power.


Another thing that might help; if the car can run well without the flex fuel sensor I would try disconnecting it. Flex fuel sensors can sometimes add noise to the ground wire, some people advise to connect the flex sensor to the chassis/battery ground and not let it share any wires with the ECU's sensor ground or MAP sensor ground or crank sensor ground.

 

Yeah it doesn't seem to, because both the fuel pumps and radiator fans seem to operate normally. I had wondered if this might be part of the problem with the trigger issues though. Thanks for the explanation! That makes a lot of sense now. And it brought up another idea for troubleshooting. I'm going to check voltage on B11 again with both connectors disconnected, and then with A connected, and then again after changing that relay coil power wire.

Well dammit lol. I hadn't heard that about flex fuel sensors. Maybe I'll be building another harness haha. I'm not sure how it will run, I guess it should be OK since I'm not using ethanol yet and it's also not tuned on ethanol yet. I don't think the tune is dependent on it yet. Thanks for the tip!

 

Before rewiring the fuel pump relay, I checked voltage on B11 again and got no voltage with neither connector connected, and also with A connected. I did have 12V on A25 now though, and I don't think I did before. But removing the ECU relays got rid of that voltage, so it's passed somewhere in there I guess, even though it's wired exactly like the Haltech relay box.

I ran the new wire from X-05 B/W to the fuel pump relays and primed the fuel system with the switch I installed to ground the pump relays, and they came on. Then I noticed I didn't have the ECU connected (which I thought was a problem at first but shouldn't be), and I don't believe I had the key on, which should have made priming impossible since it wouldn't get voltage to the relays, so I'll need to check into that just to rule out something else wrong with it.

At one point early into troubleshooting, I noticed the TMS4 chirped again when connecting the battery, but it didn't do it again after that other than one small chirp when connecting the battery once or twice. I'm going to drive the car tomorrow and take a log to see if the trigger issue has been affected by the changes.

So as of right now, A25, A31, and A32 are the only wires getting 12V to the ECU.

 

I took some Oscope logs tonight, but I can still see the problem just by glancing at the screen. I should be getting my new coils early next week.