meerkat8701

Man that sounds really rough. Having been forced to start from scratch a couple times I fully understand your situation. As long as you have your health and youth you can get back on top of things and start anew. What doesn't kill you can only make you stronger. I genuinely hope everything works out for you man!

/Anthony

fc pro am

iTrader: (8)

Best of luck, I've been there before myself, it's not cool at all and I wouldn't wish this on anybody. Last year I decided I too needed change and decided to give my life to God. Failed relationships, loss of jobs, not seeing my son had me in a deep hopeless place. Religion might not be the go to option for some people, but I quickly learned the love that comes from community.

Trying to get out of situations all on your own is a huge task. I learned to serve others first, put my personal situation to the side and focus on loving others. I pray that you are doing ok and can find positive supporting environment that will listen to you and uplift.

Akagis_white_comet

iTrader: (4)

I apologise for the brevity of this reply as my car, compy & belongings are in storage. Typing this at library with left hand only as my right pinky is broken (boxer fracture). Very difficult to type, but im getting food stamps and hopefully disability too. Filed 4 it last wk. Currently at a shelter for now.

I'll keep in touch via here, facebook & gofundme as much as I safely can. Any help is much appreciated.

thanks,
Travis

Akagis_white_comet

iTrader: (4)

After 3 weeks in a splint, my right pinky finger is healed enough to type with. Not perfect, but useable nonetheless.

I'm still trying to get my business off the ground, with more ideas naturally coming to me as time progresses. I'm currently working on a killer upgrade for the FC's overhead console. Can't give any details till I crack the electronics portion of it, but it will be frickin' sweet!

Something odd has been happening recently. My youtube channel has received attention from Tennessee roughly every week, along with someone calling my Google Voice number (which only a select few have). Both occurred on the same day as well, 9/21/2014. The call had its number deliberately blocked, was late (11:27pm) and no message was left. Anyone that knows my current situation and knows me knows not to call with a blocked number, especially so late, and even fewer people know my Google Voice number. Even telemarketers don't call that late, especially on a SUNDAY NIGHT!

Has to be Trula calling, after she watched my youtube channel.

Akagis_white_comet

iTrader: (4)

The past 2 days have been quite the adventure to say the least. It could be a missing scene from Planes, Trains and Automobiles

After hearing that some would-be thief got into my car by cutting the strap holding the driver's door shut and couldn't find the battery (a tip to all whom relocate theirs, do it SMARTLY!), it was time to get the car to a better place.

That was right after selling my washer & dryer to a friend for $150 and giving her a hand with getting them set up, despite a ridiculously late start on such that ended up biting me in the rear. Car fired up nice and easy, then went around the block and TWO TIRES came off of their rims . I had broken my cardinal rule about checking tires (use the gauge, don't eyeball it) and this is what happened. So we got the car back to Charles' house very carefully after airing one up to 50psi, it leaking the whole way back. After a hard night sleeping in the passenger seat, the car went to a nearby tire shop that took care of both tires for $20. So we're good to go right? To quote Frylock "Oh hell no"

Put some gas in the tank and she stalled and kept randomly stalling like the plugs just quit firing (ridiculously flooded, beyond stupid flooded). Again and again, it kept happening. After 0.7 miles and four times that the car fell on its face, Charles called his roadside assistance who sent out A&A Towing with the destination being Defined Autoworks. The tow truck driver dropped the car 2.5 miles away, claiming that the CVS Pharmacy on Broad St was the right address which is clearly wasn't, then left and went RIGHT PAST the shop!

So after a few more FoFs and a push-start with Charles' AE92 Corolla, the car coughed, putted and sputtered right into Logan's driveway at 5pm yesterday. It'll be getting the Turbo's Y-pipe modified and linked into the intercooler with another section which will fit perfectly and fell into my hands a couple years ago by accident. Next weekend, we shall have boost!

A big thank you to Charles for going way beyond the extra mile in getting my car to Logan. I owe you big man! Also, I want to thank Logan for taking the car on with next to no advanced notice, doing the work quite inexpensively ($70), scheduling it to work optimally with my schedule and for providing last-second directions to the shop as my internal GPS had failed and the tow truck had left us high and dry. Last, I want to thank my buddy Todd for spotting me his place this weekend and volunteering to go up to the shop next weekend to retrieve Project OldTree upon completion.

Akagis_white_comet

iTrader: (4)

It's time for another update to my loyal viewers.

The car is back home now in a friend's garage next to his in-progress SRT4 Conversion that I'll be wiring up later this week. I'd like to give a big thank-you to Logan at Defined Autoworks for spotting me 3 new spark plugs as mine were beyond usable from chronic flooding and doing some basic tuning on my Haltech to make the car not foul plugs anymore and giving the car a jumpstart to get it going. I also want to thank my bud Jerry for lending a hand in getting the car back home for the next bit of work.

At the moment, Project OldTree is in the middle of a BIG electrical cleanup. I was pretty dissatisfied at my previous electrical center in how it simply added onto the stock fuse box while not being up to my standards for appearance. In short, wires were added and not contained properly, as were additional fuses and relays. While the VW setup works quite well as it was, it left a bit to be desired in terms of aesthetics.

Now the car is getting an overhaul for that corner, from a very surprising source, one whom I've openly criticized often for taking shortcuts in engineering. Pictures will follow in my next update, including before & after the turbo Y-pipe mods.

celbii

iTrader: (2)

Been following this build for quite a while now, glad you were able to hang onto it despite everything that has happened. Hope you got a housing/job situation figured out. Good luck man!

Akagis_white_comet

iTrader: (4)

Thanks Celbii. The one lesson my RX7 has taught me is that there's ALWAYS a solution to every problem if you look at it correctly. And that brings me to what I was hinting about before:

Y-pipe is modified, thanks to Logan at Defined Autoworks


There's always some compromise with any car modification, and one thing that annoys the **** out of me is poorly planned and/or executed electrical modifications. Sometimes one has to get creative when adding a standalone, such as with the VW setup I was using since 2010 as seen above

But what I didn't like about it was the lack of a cover on the fuses & relays. Things get dirty on a street car, and relays/fuses are one area you don't want to meet dirt. So I've been hunting for a better answer and found one in the usual clever way.


A Close-up, showing the small notch in the radiator core support to clear the cover. Small price to pay, but well worth it.


And here's the money shot, what's inside the box:


ALL of the stock relays are gone now. The stock fuse box is no more as well. It's one-stop shopping for EVERYTHING electrical. Car now has a dedicated 120A fuse just for the alternator, to fix the annoying (and not particularly safe) piggybacking that Mazda did from the factory. Also has a dedicated Key-ON relay triggered by the GREEN DIAGNOSTIC CONNECTOR. Also added a Starter relay which permanently fixed the click-click-start issue. The A/C relays are in there too, along with one just for the wideband controller. The only thing not inside the box is the stock Main Relay and the Fan relays, which I've outsourced to a Volvo 2-speed controller. Best of all, there's a ****-ton of room leftover for future goodies. Still got 3 relay sockets open, 10 more fusible links and 9 fuse sockets free.

Not too shabby for a prototype eh?

EDIT: I made an error in the previous wiring & fabrication efforts. Well, this is somewhat of an error in the S4 FSM that was giving me fits over the last week. F-65 & F-66 (Lockup & 4-3 Kickdown relays) are CONSTANT 12V+, NOT IGNITION SWITCHED 12V+. Using either of these for the ignition-switched 12v source will result in the car not shutting off when the ignition is shut off. I found this out the hard way and since then, have reverted to using the Green Diagnostic Connector for Ignition Switched 12v+.

jjwalker

Akagis, what the hell is all that over on the passenger side where the stock air box would be??

Akagis_white_comet

iTrader: (4)

Turbo Control Solenoids for the twins. Based on a Simplified Sequential mod I made for the 20B Cosmo, then upgraded a few times for serviceability and reliability. 50ft of vacuum line there

Akagis_white_comet

iTrader: (4)

It took awhile to pin down the problem, but I solved my no-boost issue. Turns out it was a combination of issues, a few being due to me not reading the Haltech manual sufficiently years ago. And then the Main Relay started acting up by not disconnecting when the key was shut off every so often. It turns out that my previous statement about F-65 and F-66 being constant 12v+ was incorrect after all. They are indeed Switched 12v+.

So after a bit of minor rewiring to flip the solenoids' polarity around to match the Haltech (Signal is GROUND!) by using the long-deprecated Air By-pass Relay connector for Switched 12v+, and resetting the wastegate to trigger at 10psi instead of 28inhg, I fired the car up and took it around the block at 9pm last night. Got 5psi of boost and a nice big backfire in 1st gear, repeated in 2nd gear as well. In my distraction, of holding my phone up to see the boost gauge (backlight wasn't working), I forgot to get a datalog of the drive.

Car is currently set up in Sequential Mode. I still have to correct the Turbo control relay's wiring

Akagis_white_comet

iTrader: (4)

Sorry I've not been updating the thread as much as I did before. Inbetween wrenching, I've been working on my site as much as possible to get it ready for launch. It's almost done, I just need to get the paypal stuff set up to work right and get some better pics here & there.

So I went on Google Play and I found this: RX-7 Club app

Now if only the rotary engine live wallpaper was still there...

Akagis_white_comet

iTrader: (4)

Got a new door a few days ago, installed it yesterday and it fit perfectly. Was a local guy I found through a friend, same color and completely straight & true. Swapped the harness, then bolted it up with very little effort. Came from a Sport model, so i had to swap the window regulator too. Latched & locked with no fuss too. See the pics for how messed up the B-pillar is, while leaving the important stuff such as the latch bar completely untouched.

Oh, more good news. My website is finished now. Link is in my profile. Time for sleep now, I work too much lol

Akagis_white_comet

iTrader: (4)

It's that time again, time for another update to Project OldTree. I know I've been slacking and I apologize, but I've been wanting to save up several small items for a suitable update to the thread. Between the following, a friend getting divorced and lending a hand rebuilding that friend's now ex-wife's former Neon better than stock, I've been pretty busy. So with that prefaced, let's jump in.

First up is that I have a new fuel pump, scored for dirt cheap. There's nothing wrong with the FD pump I've been using since 2008, but I found a legitamite Walbro 255 GSS315 for $20 from a very reputable source we all know. Small price to pay for insurance against insufficient fuel pressure under boost. It was a bit worrisome when it suddenly conked out, but Kevin Landers took care of it because he's awesome like that. Next, I finally got motivated and reassembled the interior, with one important upgrade...MUSIC! Stereo is installed for good now, despite breaking the little "light box" for the cigarette lighter due to the stock wiring having insufficient slack for sane installation. Seriously Mazda??? So when I get a chance to do so, I'll be fabbing a new one out out of steel and extending the associated wiring by a couple inches. Okay, sidequest over. Back to the real upgrade: Over the winter, I put together a kit to eliminate all of the fir tree-type trim clips permanently because I absolutely despise how difficult they are to remove. Now every single panel bolts in with regular metric hardware. No more damaged, broken or destroyed rare interior panels because of those polymer evergreen bastards. 10AE guys know this all too well.

And that brings me to the next item on the list of things that have busied me over the winter. I am translating the Eunos Cosmo's Wiring Diagrams to better understand how Mazda put the 20B-REW together. In doing so, I found out that the N390 AND NF01 ECU's do NOT have a 5v Sensor VREF source in them. It is supplied from the EC-AT Control Unit. Also, while dabbling around in the diagrams, I cracked how the Cosmo's ignition system works. The N390 (13B-REW) is basically the same as the FC's ignition system, but with the coils separate from the igniters. On a related note, the FC's UNKNOWN FOURTH IGNITION SIGNAL WIRE mystery has been solved. It is for diagnostic feedback and tells the ECU that the Trailing coil/igniter is working properly. In other words, it's an early form of a self-diagnostic system. The 13B Cosmo has the exact same setup on it.

Also, I'm doing to do some rumor dispelling right here. The 20B's NF01 ignition system is a bit more complicated in that it has a feedback wire on BOTH leading and trailing coils As for their basic operation, it's just a beefed up version of the FC's trailing system on both leading & trailing coils. When #1 fires, the ECU tells the igniter to "advance" to #2 so it will be ready to fire when the ECU gives it the signal. After #2 fires, by grounding the Trigger wire, the ECU tells the igniter to "advance" to #3 via the 3rd signal wire. Once #3 fires, the system automatically resets to enable #1 to fire again. So with a little thought, I built a system to decode this setup and 'translate' it into a Direct Fire Ignition System for the Cosmo.

Turns out the tachometer is run by the Trailing coil (ala FC style), so I did some digging and found that the 13B & 20B Cosmo have different tachometers to account for the different number of pulses generated by the trailing igniter. So I got into the zone and made a tachometer driver circuit to go hand-in-hand with the Direct Fire Conversion kits. Oh, and the Wiring Diagrams are now in both English and Japanese, and COMPLETELY SEARCHABLE by device name, connector name or even wire color and pin number...

Another little side quest I did was cracking open the Main Relay as it has repeatedly stayed engaged with the ignition off. What's strange is that it looked new inside. No signs of corrosion, pitting on the contacts, burn marks, absolutely no evidence of malfunctioning that I could readily discern. While in that area of the car, I took out the stock self-tapping screw, drilled out the hole and gave it permanent M4x0.7 threads just like in the hatch. Add a 1/8" fender washer with one side cut off and it fits the 5v bus bar nice and tidy, making it so the new Main Relays can be mounted side by side in the existing M6x1.0 threaded holes. After dissecting the Main Relay, finding it to be seemingly in good shape and replacing it with two separate 40A relays for reliability/replaceability, the car still keeps the relays energized. So it's looking like my ignition switch is the culprit in this case. For being nearly 30 years old, I'm not too surprized at this development, but it really sucks because replacing it is a major chore that I'm not looking forward to. This would also explain the infamous click-click-start issue that seems so prevalent with FCs, in that it produces the same mode of failure that the window switches do. With the Main Relay/Ignition Switch matter finally identified, I moved up front and finally mounted the Volvo Fan Control Relay. With some careful filing, it hooked right onto the fuse box's original mount.

On a related note regarding electrical reliability and sustainability, I finally got a new headlight motor for $25 shipped from Game of the Wangan here. Arrived a day early, so I wired it up to the bench supply & meter for a little datalogging. Turns out it draws 5 Amps in both directions when running properly. I would suspect the motors that operate the power windows, sunroof and rear wiper draw a similar amount of current. Seems kind of odd that Mazda would fuse these circuits for 3x their proper draw, but there's bound to be a reason that I just can't figure out at this second in time. As soon as I went to swap it into the car, I found that it was a DRIVER's side motor, not the passenger side one that I purchased. Thankfully, the person whom I bought it from is a fine upstanding member of the Rotary community and gave me a prompt refund for the entire amount, and said to keep the motor on top of that as he had no use for it anymore. So with that in mind, I'm going to crack both motors open and do some R&D on an upgrade for them.

And with that, I found another ignition switch for cheap too It's a S5 switch (FC01, no airbag version), but it's almost a drop-in swap for the S4 (FB01) one. The only difference is the Key Reminder connector (F-21 on S4, C-10 on S5). The wires are identical and go to the same locations for the same functions, it just has a different style of connector. A big thank you to Game of the Wangan for turning an error into awesome solutions for several issues all at once. We need more guys like you here!!!


To summarize:
Trim clip elimination kit
Walbro fuel pump
13B & 20B Cosmo Engine Control System translated
N390 & NF01 Ignition systems decoded & translated into Direct Fire for D585 coils
FC Coil unknown signal wire mystery solved (Feedback for diagnostics)
FC/Cosmo Tach driver adapter circuit designed

On the books for next time:
Upgraded S5 radio trim light box

Aaron Cake

iTrader: (1)

Just want to say that I've tried looking at your website but the left side menu is maddening in how it pops open but if you slightly move the mouse it's gone again. There needs to be a timer in there of a few hundred MS after OnMouseOut or whatever you are using to debounce it.


I hate fir trees. Swapped a bunch of them with nutserts years ago, and will be planning all nutserts on the Cosmo.

What type of signal does it produce?

Drive the tach with the tach output on the Haltech or other standalone.

Or just connect it (in an FC application) to the leading coil.

Akagis_white_comet

iTrader: (4)

Thank you for the feedback on my site Aaron. I'm not sure of the exact issue you are experiencing with it, as neither my laptop, desktop or several computers at the local library have displayed this specific issue. Just to be on the safe side, you can click the menu items and it will take you to the main page for the vehicle in question, complete with buttons that cover all items in the 'jumpy' menu.

Care to buy a trim clip kit from me? I include an install tool with every kit and a few extra inserts just in case something goes wrong.

As for the tachometer, not all members can afford a standalone, yet still desire an ignition upgrade. Or in some cases such as with the JC Cosmo, a standalone isn't possible (YET) due to the Palmnet ECAT-ECU-HVAC-ETC communications. Also, when transplanting a 20B into a FC/FD with the JC Cosmo coils, it sends 50% more pulses to the tachometer, making it read high (3000 instead of 2000rpm). My converter box fixes this too.

As for correctly Standlone'ing a Eunos Cosmo, that's a project for the future once I've finished translating the shop manual.

Aaron Cake

iTrader: (1)

IE11 on Windows 7. Not sure what to say beyond that really. I just sort of gave up because the menus were so spastic. Really I'm not a fan of popout menus on a website anyway. Ever tried to use them on a touch screen? Brutal in many cases. But generally the issue can be fixed by setting a timeout before the menu disappears after the OnMouseOut. I haven't looked at the code on yours but most menus are a fairly standard script.

I'm not sure why one would need an ignition upgrade but isn't making enough power to warrant a standalone. Maybe there is a case I'm not seeing.

Standalone is easily possible on a Cosmo. Just make a parallel installation, keeping the stock ECU but controlling the injectors, coils and other engine bits (fans, etc.) with the standalone. Very commonly done on other vehicles like the RX-8.

Any decent standalone has configurable tach pulse settings. For the exact purpose of matching the stock tach to the wrong engine.

Not trying to rain on your parade of course, as I realise my post seems rather negative. Just trying to understand the need for converters and such.

Akagis_white_comet

iTrader: (4)

Aaron, your points are perfectly valid as always, and yields a much more informed community. Hardly negative in my book.

My whole site was made in Firefox and behaves properly in Windows and Linux OSes, as well as on my Android 4.0 phone's browser. Perhaps it's a bit of funkiness with IE? Wouldn't be the first time Microsoft screwed something up there As far as touch screen browsing goes, it appears to work fine with my phone's browser. Anyway, for those whom aren't too keen on pop-out menus, the site works just the same by clicking on the car's main link on the sidebar.

As for the electronics I've designed, I'll try to put it as best I can. Not all are as godly as you are Aaron . For things like adapting different ratios and tires, a signal converter is VERY useful. Even more useful if it includes Electronic to Mechanical output conversion onboard and at a price that makes it available to anyone, even us broke S4 owners

The Tach adapter has a simple purpose: save an output on a standalone for something more useful, in the case os a retrofit. On a stock engine, it behaves the same as the tach output from the FC's trailing coil. Kind of necessary when running D585s with a stock ECU in a FC. As for why one would do so, it's been proven many a time amongs RX-8 owners that the D585s are fantastic and they're easily packaged wherever the owner desires. Since I mounted mine directly on the shortblock, it left the entire corner behind the headlight open for my end-all, be-all fuse/relay box. Flexibility and packaging is the key

As far as the JC Cosmo and a Standalone is concerned, wouldn't the two ECUs' signals conflict with each other? IIRC, the JC Cosmo uses a Palmnet system for in-car communications and the ECU's end of that network is calculated based on how much fuel it is putting out (Injector DC), Throttle position (TPS), and so on. For example, at idle (750rpm), let's say that the stock 550 injectors are at 2ms pulse width with the NF01 ECU. But with the Haltech/Megasquirt/etc, the user tunes it down to 1.5ms. Which reading does the Palmnet system use?

It's a difficult problem, even moreso since all the documentation is still in Japanese.

Okay, now back to the fun stuff. Since the last update, I've put about 45 miles on the car to get acquainted with its behaviors and quirks. In the past 48 hours, she has...
Fouled six BUR9EQ spark plugs
Drained her battery twice
Had a bus bar fail from corrosion (38 OHMS, WTF???)
Stalled a bunch, then fell right on her face on startup
Had the Taurus fan on high drag down the electrical system to 12v while running
Stalled some more
Refuse to go into boost
Backfire a lot!
You get the idea

So after a bunch of fretting about, I got some fresh BUR7EQ plugs for the Leading holes last night, ordered some new BUR9EQ Trailing plugs for arrival tomorrow afternoon, bypassed the bus bar in question by tacking all of its terminals on one stud. Soon as I did that, the fan issue vanished into thin air and she stays at a nice, solid 14.0v with the Taurus fan on high

With no further tuning, she is quite smooth in vacuum when given about 1/3 or less throttle. As soon as she crosses into boost, she backfires and wants to flood right at 1psi when the secondaries come online.

On an interesting note, it turns out my insurance agent at State Farm is into RX-7s also. She wanted one right out of high school in 1986

unek87

Tried going to your link for your website and I guess safari is having problems . Always try to support forum members that produce goods for my rides .

Aaron Cake

iTrader: (1)

A 2nd set of sensors is installed to run the standalone, and then the standalone is either triggered in parallel with the stock VR sensors or another wheel added. Standalone controls the coils and injectors. Stock coils and injectors are disconnected. Or if the stock ECU freaks, then the injectors are replaced with resistors. Since the Cosmo is well before OBD2 I can't imagine this would cause any issues. So as far as any other systems on the car are concerned, the stock ECU is still in charge. So gauges and such should work just as they did.

Needs tuning.

Akagis_white_comet

iTrader: (4)

Well, it looks like fate has thrown a monkey wrench into the works YET AGAIN, with the details here: https://www.rx7club.com/haltech-foru.../#post11908925

ObliqueFD & I did some tuning and all went really well. Car was picking up power, running smoother and happier untill we stopped for the night. The next day, it turned to total dog crap. Searching idle, stumbling like an irishman at 3am, stalling when put into gear and given a LIGHT load with feather light throttle.

It appears heat-induced, as the behavior gets worse the warmer the engine gets. Took a wild guess and replaced all the vacuum caps, finding every single one cracked from heat failure. No changes, so ObliqueFD & I did some tuning on it today to see if we could iron out the cyclic behavior. No dice there.

This was a working, stable map that fired up easily with little fuss, and responded well to throttle for being only half tuned.

Map and logs of stalling:
https://www.rx7club.com/attachments/...-map-copy-.xml

Aaron, could you please take a look at this when you have a chance to? I'm not sure what the issue is, and I'm not sure exactly how to interpret the logs properly to deduce the cause correctly. It is a compressed file, so change the extension to .ZIP and it will unzip to produce a Haltech map and logs for ECU Manager

Aaron Cake

iTrader: (1)

I'll have to look at it another day. I don't have the Haltech software on this computer and the download is crawling at 20K....2 hours remaining. Screw that.

Akagis_white_comet

iTrader: (4)

How the hell are you only getting 20k??? I'm pretty sure this isn't 1998 and we're not all on dial-up anymore. Then again, I thought downloading Zelda mp3s on Napster at 5k then was pretty awesome Thanks for checking out the map & logs.

We re-examined the car yesterday and found that my TPS was wired incorrectly, due to misinformation from the HITman's pinout which is floating around the net. So after cracking open the Cosmo wiring diagram and translating wire colors, then correlating it with two other sources, it is wired correctly now. With the TPS mounted on the throttle body, the proper wiring is as follows

Top pin: 5v
2nd pin: Narrow range signal (Pin 3F on N390/NF01 ECU)
3rd pin: Ground
Bottom pin: Full range signal (Pin 3G on N390/NF01 ECU) This signal also goes to the Cosmo's ECAT Unit too, most likely for kickdown control.

So we popped it back on, re-calibrated it and fired it up again. Car acted fine when cold, but got more and more glitchy as it warmed up, with the TPS voltage values jittering more and time went on. Result is that the car couldn't tell what the throttle was doing and kept dumping/pulling fuel to compensate, hence the 'looping' idle when viewed in ECU Manager.

So, TPS is suspect, and on a whim, we tried cleaning it per the MR2 owners club guide. It worked SLIGHTLY better, but still jittered too much to be stable. Something else I saw was that it wasn't behaving linearly anymore too. Definitely time for a new one, on the way now, along with a BAC valve too.

Akagis_white_comet

iTrader: (4)

The stumbling/drifting lean behavior has finally been pinned down. After replacing the TPS due to jittery behavior and dead spots, then redoing EVERY ground which contributed to the TPS behaving erratically. Ground Bus Bar was showing 50+ ohms of resistance and all of the 3/8" ring terminals were corroding from within (and showing unacceptable resistance as a result), so they all were replaced. The new bus bar has five M6x1.0 studs, is hand fabricated entirely from 304 Stainless Steel and is a drop-in replacement for the no-longer-zinc-plated GM one it replaced.

With that done, the new TPS recalibrated in ECU Manager, she's still running not well. Vacuum is low, noticeable vibration, searching idle. Thought it might be the ignition, so we checked the whole system one piece at a time. Middle rotor leading coil was sketchy, so we swapped in a spare and the car behaved identically. So a compression test was in order and that spelled it out.

Significantly Low Compression on Middle Rotor, Front & Rear were OK but not great
Using the RotaryResurrection method, she produced even bounces at around 20psi on front & rear, while 15psi on Middle. Overall were 70 front, 60 middle and 70 rear respectively. But oddly enough, it's showing that all the apex & side seals are intact while the engine runs like it's got no compression on the middle rotor. But the plus side is that since nothing has actually blown, the rotor & housings should be fine

j9fd3s

iTrader: (3)

before you go and rip the engine apart, i would do some more tuning. we had an Rx8 come into the shop, and it was obviously running on 1 rotor. after a compression test, one of the rotors was in the mid 5's, when the spec is about 8. so we called Mazda, and they had us replace the entire ignition system. since it was their money, we did.

car fired up, we ran it for a while, and retested compression, and it was in the mid 8's on both rotors.

moral of the story is that the car has to run right to have good compression, and as such i wouldn't be too quick to rebuild if it doesn't run right and compression is even.

magic 8 ball would say the ecu needs tuning