I did the HEI test on all coils and I'm getting a strong spark from all of them. My ignition harness appears to me to be wired correctly, I don't think I have any issues because of that. I have some new, much shorter Magnecor spark wires on the way and I've also picked up a noid light to test all the ECU signals to the injectors (hoping to test this weekend). We'll see if that turns anything up and go from there.

Team as far as ignition harness goes, yes I have the 4 signal wires to ECU and ive tested continuity of all of them directly to the coil connector. Also have 12V switched from ECU to relay box. These 5 wires from ECU 'Y' into the ignition harness with the 4 power wires coming from the Rywire relay/fuse box then head to the coils. The low amp ground from this harness is grounded to the chassis and not all the way back to the ECU. The relay box also gets 12V power directly from battery as per the Rywire instructions. Pin E on all of my coils is tested and receives 12V switched. Pin A as mentioned above gets continuity to the ECU trigger pins. Pin D is now independently wired directly to battery negative. The harness has a ground ring terminal for each housing which I have to assume has coil pins B & C both wired to it together since I'm seeing continuity between Pin B and both the Block and the chassis. Just realizing I didn't do any real continuity testing for Pin C, Likely I should but as mentioned, HEI testing the coils while cranking the engine gave a strong consistent spark while cranking a cold engine.

 

B and C are specifically not supposed to be wired together as per the diagram. Supposedly there are reasons for doing that wrt there being two separate circuits or windings in the coil that makes the IGN-1A unique and as powerful as it is (I can’t recall specifically which it is at the moment). But again it’s been argued and done differently by various people on here. That’s not the recommended way. Idling is not going to result in the same loading or simulation as peak torque or peak rpm.

I seem to recall you had a coil or maybe two fail previously as well? Might have been somebody else, but my memory is it was early in your original startup several years ago. Not sure I’d be assuming anything at this point given all you’ve been through.
.

 

Correct, I had both my leading coils fry very early on at different times about a year apart from each other. Both were replaced with no other changes and have been fine since though. I'll have to do more testing with Pin B..... I agree there is lots of discussion on this so it often challenging to siphon through it all. I've read that Pin B doesn't matter if it's connected to Pin C as long as its grounded to same spot as the ECU sensor grounds (which I believe is the block in my case). But at this point it's something to consider, if my noid light test seems to check out and my new plug wires don't change anything, then it's likely I'll need to revisit this.

 

My $0.02 on this subject as a semi-retired EE... It's not that the IGN-1A is special or really all that inherently different from any other inductive "smart" coil, the 3 ground pins are necessary because the coils have essentially 3 electrically isolated circuits, each of which requires an electrical return path.

The B pin provides a return path for the TTL logic trigger signal from the ECU. Since the ECU's coil TTL trigger output circuits are powered off of the ECU's +5VDC regulated sensor supply, they need to ground to the ECU's sensor/signal common ground pin. This ensures a clean TTL signal to trigger the coils, so you should never share this ground with either of the other two coil grounds (pins C & D)

The D pin (Battery ground, high current) provides the return for the primary side of the coil - an inductive coil is basically a transformer, so the circuits of the primary and secondary transformer windings are electrically isolated. The primary side of the circuit charges the primary coil winding (hence the high-power qualifier), based on the dwell settings, and when the TTL trigger signal commands it to "let er rip", the primary field collapses, inducing current in the secondary coil windings, which ultimately discharges to ground thru the spark plug.

The C pin provides a return path for the secondary winding/high voltage discharge side of the coil circuit - and closes the loop to the point that the spark plug grounds to.

 

Thanks, I’m only a バカME and an aging, forgetful one at that. It had all been discussed on the forum before in previous threads, maybe by Lance himself(?). I should stick to moment of inertia and other such mindful topics instead.
.

 

I agree keeping the IGN1A coil's pin B (signal ground) separate from the other two high-power grounds is probably the most important thing. Twisted shielded wire for the coil signals is a good idea, to avoid picking up interference. Imagine if the front trailing coil gets triggered by noise when the front leading trailing coil fires, that would be like zero trailing split which is generally considered unsafe in boost. The front trailing coil getting triggered by the rear trailing coil might be even worse, although I don't have a good guess which parts might get damaged from a trailing coil firing 180 degrees too early.


Something that doesn't get discussed often is the rectangular capacitor labeled 'condenser' which is mounted near the factory igniter (page B-1a of the factory wiring diagram). If people are removing that when swapping to aftermarket coils, it might be wise to try adding it back so it can help keep the 12V supply to the coils stable. The capacitor should be physically near the 12V supply (pin E on the IGN1A coils) and the battery ground (pin D on the IGN1A coils), if possible.

 

Well seems as though this Rywire harness was a waste of money on my part. What I think I'll do then is snip pin B from each of the coil connectors and re-wire those together and run a single ground to a sensor ground in my ECU. When we say this we mean physically connecting the ground from Pin B of coils directly to one of my sensor ground wires going to my ECU? Does it matter which one? My ECU pinout (Adaptronic Emod013) has multiple sensor ground pins, but doesn't specify which particular sensors are grounded to each of those pins. If I was to do this without advice I'd say Pin 2H is the likely candidate as connector 2 has all 4 of my injector driver outputs on it. Any failings to that logic? TIA 3M would be another candidate as I know the only sensor grounded there is my oil pressure sender.

 

Different situation for me as this is in the Rx8 chassis. Correct me if I'm wrong but I don't believe there was a capacitor for the factory coils.

 

Diving into this a bit more, would there not be obvious signs if my plugs were firing at the wrong times? Timing light checks out for all 4 coils (at idle) and would that explain fouled plugs? In my mind I'd expect the opposite, overly hot/lean looking plugs?

 

yes, there is a condenser on the factory RX8 ignition system. I’m not sure if your Rywire harness plugged into the factory wiring or not, but seems likely not. If it did plug into it then the condenser will be functioning, at least within the factory parameters it was designed for. It’s mounted on the top forward LH side of the engine:



.

.

.

edit: it probably isn’t operating since the IGN-1A coils are powered direct to the +12V battery connection. Just wasn’t thinking it through.
.

Also, any pin labeled SGND should work

 

Spark noise gets pretty deep into electrical theory, I have an EE degree but I'm not an expert at magnetics and fields. I have some practical understanding of it, from helping work on automotive control systems similar to the Adaptronic.

From what I understand, the spark energy required to jump the gap and will be lower at light load (like idle) and higher at high load (full throttle, high boost). So if you had a way to measure the amount of magnetic interference trying to mess with your signal wires, the interference should be the weakest and least likely to affect things at idle or free-revving when you are checking timing with the light. The interference should be the strongest and most likely to affect signals when the engine is at full boost and high RPM when most sensible people would not stick their heads near a car's engine bay to watch a timing light.

On piston engines, those guys can sometimes see weird burn marks on cylinder walls or unexpected bearing wear at certain crank angles that might be evidence of combustion happening dozens or even hundreds of degrees before the engine should be making power. Firing the spark 10 or 30 or 60 degrees too early might also melt or burn things in ways that look different than simply trying to make too much power on weak internals. The piston guys also have the benefit of a huge community gathering and sharing data. I'm not an engine guy, but I occasionally talk to engine guys and hear about interesting stuff like that. I'm not sure if there's a rotary equivalent. I've seen some pictures of rotors that have dents in the faces, I suspect that could be from early combustion but I don't have a good way to prove that. I wish I had a better answer as related to rotaries, sorry.

I agree that pins labeled SGND (sensor ground) should work for the IGN1A pin B. The goal is that the low-current devices (like pin B of the IGN1A coils, or the MAP sensor ground) have a wire going straight to the ECU which is not shared with any high-current devices (like pins C or D of the IGN1A coils, or the ground wires for the cooling fans).

 

i have heard of a shop that set the ecu up on the engine to run it, and then they setup another (Motec) to log the first one. it is an advanced move, but when you're an engine shop that has a dyno, i guess you'd do stuff like that.

 

Thank you all for the shared info, I appreciate the discussion on this.

Team appreciate you sharing that info on the condenser. I do have this as a remnant in my main engine harness still. Can see I have it grounded to housing 1 (photo below) but as you mentioned above, the only thing wired to ECU for my coils is 12V switched and the 4 trigger signal wires so this condenser is doing nothing. Based on the factory coil wiring diagram you provided there is the signal wire from ECU, then a common ground for all 4 coils followed by an additional ground from all 4 coils going to this condenser. Would it be worth while scoping out which pin that condenser is on the main harness and splicing Pin B (sensor ground) from my AEM coils to this same pin? In my head this is likely not the case as this condenser is grounded to the same location as Pin C of my coils and I should just stick to grounding to a sensor ground pin as labelled in the ECU.




Thanks again Scott for your input, appreciate hearing from you. When I tore my previous engine down there was no signs of detonation or damage to any of the rotors, even down to seal groove clearances there were in excellent shape. The biggest thing noticed when I tore the last engine down was just the amount of carbon build-up, another indicator of incomplete combustion. Looking back at all that I've learned since tearing that engine down, I think it's feasible to assume that the amount of carbon likely played a factor in the very abnormal apex seal wear I found on Rotor 2.

 

On the subject of grounding, and as can be seen in the photo above/below:

I have my battery in the trunk. Negative terminal is grounded to chassis in the trunk. Have another ground wire from neg terminal that runs up and connects to a mounting point on the trans. I then have a grounding strap from the trans mounting point to my chassis/subframe near the passenger motor mount. Up top I then have the main engine harness grounds on each of the two rotor housings along with my Rywire coil sub-harness (Pin C for sure and likely Pin B as well) grounded to the same spot on the housings (can be seen in the photo in previous post). I have a ground strap that goes between each of the two rotor housing grounding points then a grounding strap from the rear housing that drops under the car and grounds to the chassis/subframe at the same point where the strap is from the trans/battery to the subframe.

 

Tuned up to 11psi last night (link for video below). Seems to be pulling well but EGT's are a bit of a concern. There was a lot going on so I can't say exactly when I saw this but my gauge peak shows rear at 988C (1810F) and front at 915C (1680F). I'd say most time while I was watching during pulls I was seeing peaks around the 900-940C (1650F - 1740F). I've known my exhaust manifold may be a bit of a flawed design, we'll see if we can offset the rear/front temps with some more tuning. My exhaust is not helping either, I have all 3" exhaust but the flanges on the HKS catback and the BHR midpipe have 2.5" flanges on them; ideally I'd like to get these all cut off and replaced with 3" v-bands.

 

probably not any different than running a 3” exhaust flange gasket with only a 2.5” thru-hole like some people on here do for low boost control.
.

 

That was my initial thoughts as well, I know adding restriction is a common thing for guys to do on the IWG turbo's if they are experiencing boost creep. I've know this would be a choke point in my build, but figured it should be good enough for my initial goals, especially for only running 10-12psi of boost. I still think replacing the flanges with 3" vbands is the way to go but I also think EGT's can be addressed much more via tuning. This was initial tuning session, and I had some minor issues at the start of the session which delayed getting past some idle and free rev tune clean up.... which in turn made us run out of time at the end of the session once I did start boosting. This is nowhere near a race ready tune, but enough of a tune for me to at least stop driving the car like a granny. Need to check on a few thing here not before continuing again. Developed some sort of rattle that I can hear/feel around the trans tunnel, also after returning home I'm certain I can hear an exhaust leak; so hopefully the rattle and exhaust leak are the same culprit and I just need to give everything a good re-tighten. Will likely give the engine another smoke test for vacuum/boost leaks now that I have some miles and heat cycling through it.

 

i think you should finish the tune first, but when i put the Big Dumb Exhaust on my car, it did drop the EGT's, so that is a thing.

its pretty easy to put an old pressure/boost gauge into the o2 sensor fitting and see what the backpressure is.

 

I've seen this vicariously through a friend as well, he has a similar setup with a 9180EFR iwg. He swapped out his 3" exhaust for a true 3.5" and saw a significant drop in EGT's and a noteworthy drop in EMAP (I don't have the actual numbers available to me so this is more word of mouth from a trusted source)..... but he's also running 28psi on E85.

Interesting idea, I actually have a second O2 bung not being used right now on my downpipe, my Adaptronic wideband is threaded into it right now but I have it disabled and am feeding the ECU my AFR's from my AEM UEGO via serial cable instead. The question though, most people are measuring EMAP (ie pre-turbine); this wouldn't be a true emap/imap ratio, so what baseline numbers would I compare that exhaust backpressure to? Secondly, if I was basing my judgement off front EGT's alone I'd say everything is A-OK; it really just seems to be my rear EGT's getting hotter than desirable.


I agree though, I still think this can be addressed via tuning and I should wait to pass final verdict until that is addressed. Was just simply stating my observations during my first boosted pulls on this new engine and new tune.


I did get under the car and found the rattle that I have developed is my midpipe contacting a drive shaft tunnel brace. I've tried to address it temporarily for now but I may need to remove a section of my exhaust and bend a hanger to gain a bit more clearance properly. This was likely always close clearnace but I did smack a good pothole on a recent drive and there was a huge scrape down the rear portion of my exhaust piping, so I'm fairly confident this has caused the exhaust to sit slightly lower and start making contact with the brace.

I also got onto every one of my turbine/exhaust manifold flange bolts; 2 of these were definitely "loose" gave them another 1/4 to 1/2 turn each. Other than that my EGT probe sealing caps seem to like to back themselves off slightly over time, both were loose, one took another 1/2 turn, the other I managed a good 1.5 - 2 turns on it. Have not been able to get out for another drive yet. I'd also like to do a smoke test through my intake and see if I've developed any vacuum/boost leaks between engine install and now (I smoke tested the engine on the engine stand before installing to check UIM, LIM, and injector ports). I comp test likely isn't a bad idea either but as it stands, I don't suspect any compression issues.

 

i wanted to start mine with pre turbo pressure, and then do post turbo, and that would let you see where you are on the turbine map.

but i found out if you block off the EGR port, then the EGR port is blocked off, so i punted and did post turbo pressure.
mine was 12psi, so i put a hollow cat in it, which dropped it to 7psi, and then switched to the Big Dumb Exhaust. pressure dropped, but i'm still tuning so i haven't checked it again.
its a FEED exhaust, so i'm expecting exhaust pressure to be negative

or in summary, 7psi is too much, although car was silent, and boost control was really easy. FEED is maybe not enough, boost control went out the window, and its not silent anymore!

 

it likely has more to do with the front manifold runner being straight in while the rear runner is significantly longer with tight bends in 15° pie segments (6 pies per 90°), along with the runners being 1.5” Sch. 10 pipe (1.68” ID).






When you have only that many pie pieces it creates a bit of a ridge at every joint. Which then reduces the effective flow diameter to be smaller than the actual pipe/tube diameter.

I’m in the middle of having custom pie pieces cut at 10 pieces per 90 (9° segments) for that reason. It makes for a smoother bend, but higher cost and effort due to the number of welds required.



.


however, unless it needs to corkscrew in as on the manifold pic above, or have a tighter than 1D radius, then more often than not a smooth mandrel bend is better.

Not meaning to cast fault on this though, because on an RX8 chassis fitting the EFR B2 frame turbos in a low mount position is a real booger that presently requires a custom manifold. The reality of it is what it is.

Some people have modifed their chassis to have the similar Turbosource cast manifold fit in a similar fashion, but it has this same high rear rotor EGT issue. Even though the bends are smooth, the rear runner bend into the turbo flange on the cast manifold is exceptionally fubar. It could have been much better and it baffles me why they did it like that. It’s really almost square rather than radiused, yet the space is there to have a much better radius and it somehow was completely overlooked. The inside radius on the opposing engine flange end isn’t much better.



.

in that regard, the custom manifold is much better on the rear runner inside radius at both ends, yet they end up pretty close to equal between the good and not so good of each one. Which levels the playing field between them.
.

 

Yowzer those are some high EGTs... Is that on ethanol? Highest I've gotten the car to on some hard 13psi street 160+ mph pulls is a tad over 1600 degrees f with 93 and meth injection.

 

No this is on straight 93 pump fuel. I'm currently waiting for the exhaust shop I want to use to free up some floor space where I'll have all my exhaust flanges cut off and replaced with true 3" V-band. Once that is done I have some brand new ID1050X's to replace my current ID1000 primary's with. We'll see where that gets me..... if the rear EGTS are still sky high after that then the only thing I can focus on will be my turbo manifold design. My front EGTs currently are slightly higher than what you're indicating at 1650F ish; so that seems perfectly acceptable to me especially with no water/methanol.

My car is currently parked for winter (has been since mid Nov).

 

Straight gas makes sense but still seems scary high. I have some 1050x primaries they do me well. What afr are you running at that boost and where is the wideband sensor?

Can you magically fit a cast ewg manifold on your setup? They are perfectly mirrored

​​​​

 

No unfortunately the chassis just doesn't have the room for low mounted like that and even runners, at least not anything that is mass produced, the closest cast I can get to mine would the Turblown IWG, but that's not a perfect fit either and still favours heat to the rear runner as it is not even length either. I have a local friend who uses it in his Rx8 swap. AFR is around 11.1 at 10psi and the wideband is about 24-30" downstream of my turbine outlet.

I'm also almost certain my ID1000s are having some sort of issue; hence the new once to swap in, so we'll see where that gets me too, I seem to be absolutly dumping fuel on decel, so who knows what that's messing with under load...