Project OldTree: The 12 Days of Rotormas
07-23-13, 02:54 PM
#376
Watch the video first, made on the way out
then view the datalog for the return trip, attached here
Due to the car misbehaving as shown in the video, and not wanting to restart a couple blocks away from the gas station, I turned back early. while making another log. Stupidly, I forgot to save the first one before starting the second log.
As seen in both the video and log, it is running quite rich, then jumps to lean for some reason, then back again. All of this was done under 4000rpm using only the primary injectors which never came close to an unsafe duty cycle. All 3 load parameters, Delta, Fuel & Ignition are also logged too.
I've seen this strange behavior once before on the first log in neutral, but it disappeared on logs 2, 3 & 4. Any ideas on why it's acting like this?
many thanks,
Travis
then view the datalog for the return trip, attached here
Due to the car misbehaving as shown in the video, and not wanting to restart a couple blocks away from the gas station, I turned back early. while making another log. Stupidly, I forgot to save the first one before starting the second log.
As seen in both the video and log, it is running quite rich, then jumps to lean for some reason, then back again. All of this was done under 4000rpm using only the primary injectors which never came close to an unsafe duty cycle. All 3 load parameters, Delta, Fuel & Ignition are also logged too.
I've seen this strange behavior once before on the first log in neutral, but it disappeared on logs 2, 3 & 4. Any ideas on why it's acting like this?
many thanks,
Travis
07-24-13, 11:40 AM
#377
Moderator
iTrader: (3)
Join Date: Mar 2001
Location: https://www2.mazda.com/en/100th/
Posts: 30,764
Received 2,559 Likes
on
1,820 Posts
i have to wonder if its just low on gas 
, mine did almost the same thing, putting 5 gal in it fixed it, the surprise was that the gas gauge works! (my car is built from floor sweepings)
your car actually runs pretty nicely, its 1000% ahead of mine, the E6K was a really rough starting place!
that being said just looking at the logs, it runs nice, but its really rich, 8% throttle, 1500 rpm it should be @14.7:1, and not 11.6 like it is.
, mine did almost the same thing, putting 5 gal in it fixed it, the surprise was that the gas gauge works! (my car is built from floor sweepings)your car actually runs pretty nicely, its 1000% ahead of mine, the E6K was a really rough starting place!
that being said just looking at the logs, it runs nice, but its really rich, 8% throttle, 1500 rpm it should be @14.7:1, and not 11.6 like it is.
07-26-13, 12:46 AM
#378
Rough Tuning: A guide for annoying neighbors at 1:30AM with your Wankel
After a lengthy discussion with Stevesimon and J9FD3S and some old fashioned searching on top, I have an answer to the car's Lean Spike Behavior.
The short version is that its Ludicrous Rich mixture (11:1 at idle!) was putting in so much fuel that it was accumulating over time, making the plugs not fire (not enough air in mixture) every so often. When this fuel-laden mixture hit the O2 sensor, all it could see was the air, so it appeared ridiculously lean due to the fuel still being in liquid form. The further the car got from home, the more frequent this behavior became untill it finally flooded out near the end of the video.
So with this awesome knowledge fresh in mind, I've spent all day tuning and am making good progress. Thanks to J9FD3S, his clever injector time formula and a bit of my own reasoning on linearization (go further into vacuum, not towards boost), I now have a working prototype map. It's still not perfect due to a miscalculation of Manifold Pressure at WOT on my currently N/A car (thought it was 27 inHg, actual is ~20), but the result is that the car starts with no throttle input now, idles at ~12.5:1 @750rpm on her own and holds a suitable AFR through 3000rpm and 18.7 inHg.
At 3000rpm, the car wants to stumble & shake considerably. AFRs are in the low 12s/high 11s, so I would immediately suspect this. Primary Injector DC peaked at 14.5% @ 3100rpm, so they're nowhere near unsafe DCs and could logically be leaned out a bit more. Secondaries are still intentionally silent as the primaries are currently proving to be more than able to do the job on their own.
Included is a short 20 second log of the essential 22 parameters during my most recent tuning session and a copy of my current map, named in J9's honor. Enjoy seeing what I see and feel free to give constructive criticism, comments and suggestions.
One thing I'd like to announce is that as I'm getting the rough tuning done, I'm writing a tuning guide for rotary owners about to make the jump to a standalone. It is geared toward bringing the gap between the First Start on a downloaded base map and the First Drive in order to get the car safely running & drivable to a dyno for proper in-depth tuning. Items such as the Lean Spike issue are covered in sufficient detail but in terms easily comprehensible by the average person seeking their first standalone. One such tidbit is that increasing injection time in a cell in the fuel map results in your AFR decreasing. I forgot this a couple times, so I'm putting it here as a reminder too.
The short version is that its Ludicrous Rich mixture (11:1 at idle!) was putting in so much fuel that it was accumulating over time, making the plugs not fire (not enough air in mixture) every so often. When this fuel-laden mixture hit the O2 sensor, all it could see was the air, so it appeared ridiculously lean due to the fuel still being in liquid form. The further the car got from home, the more frequent this behavior became untill it finally flooded out near the end of the video.
So with this awesome knowledge fresh in mind, I've spent all day tuning and am making good progress. Thanks to J9FD3S, his clever injector time formula and a bit of my own reasoning on linearization (go further into vacuum, not towards boost), I now have a working prototype map. It's still not perfect due to a miscalculation of Manifold Pressure at WOT on my currently N/A car (thought it was 27 inHg, actual is ~20), but the result is that the car starts with no throttle input now, idles at ~12.5:1 @750rpm on her own and holds a suitable AFR through 3000rpm and 18.7 inHg.
At 3000rpm, the car wants to stumble & shake considerably. AFRs are in the low 12s/high 11s, so I would immediately suspect this. Primary Injector DC peaked at 14.5% @ 3100rpm, so they're nowhere near unsafe DCs and could logically be leaned out a bit more. Secondaries are still intentionally silent as the primaries are currently proving to be more than able to do the job on their own.
Included is a short 20 second log of the essential 22 parameters during my most recent tuning session and a copy of my current map, named in J9's honor. Enjoy seeing what I see and feel free to give constructive criticism, comments and suggestions.
One thing I'd like to announce is that as I'm getting the rough tuning done, I'm writing a tuning guide for rotary owners about to make the jump to a standalone. It is geared toward bringing the gap between the First Start on a downloaded base map and the First Drive in order to get the car safely running & drivable to a dyno for proper in-depth tuning. Items such as the Lean Spike issue are covered in sufficient detail but in terms easily comprehensible by the average person seeking their first standalone. One such tidbit is that increasing injection time in a cell in the fuel map results in your AFR decreasing. I forgot this a couple times, so I'm putting it here as a reminder too.
07-30-13, 11:49 AM
#379
Moderator
iTrader: (3)
Join Date: Mar 2001
Location: https://www2.mazda.com/en/100th/
Posts: 30,764
Received 2,559 Likes
on
1,820 Posts
try this map.
changes from your map are:
o2 target map is set to factory like values, for reference, it is actually turned off
changed oil metering map, to one that looks more like the stock map, from SAE paper. more RPM and load points, slightly more oiling
changed the timing vs coolant and timing vs air intake to factory values, from Henrik.
cleaned up the timing map a little, it had some stray values here and there, timing is retarded over ~10psi, added 750rpm row.
changed the injection angle to factory, from S4 and Rx8 training manual.
changed the priming map under 65c to factory setting.
cleaned up fuel map a tad, lots of odd stray values, added 750rpm row, and -9psi row (i changed mine to PSI, i hate KPA, its lame). from datalog cleaned up the areas around idle, and 3000rpm, it should be more consistent.
over 10psi, fuel injection is just set to 85% duty@7000rpm.
changes from your map are:
o2 target map is set to factory like values, for reference, it is actually turned off
changed oil metering map, to one that looks more like the stock map, from SAE paper. more RPM and load points, slightly more oiling
changed the timing vs coolant and timing vs air intake to factory values, from Henrik.
cleaned up the timing map a little, it had some stray values here and there, timing is retarded over ~10psi, added 750rpm row.
changed the injection angle to factory, from S4 and Rx8 training manual.
changed the priming map under 65c to factory setting.
cleaned up fuel map a tad, lots of odd stray values, added 750rpm row, and -9psi row (i changed mine to PSI, i hate KPA, its lame). from datalog cleaned up the areas around idle, and 3000rpm, it should be more consistent.
over 10psi, fuel injection is just set to 85% duty@7000rpm.
07-30-13, 01:03 PM
#380
Coolant Leaks solved, Vacuum leak found and solved
I was feeling sick of coolant leaks and decided that my ghetto-fabbed coolant return line arrangement had to go. Loosened the drain valve and nothing came out...it was completely gunked up. Pulled it out of the radiator and drained the coolant. This is an important lesson to keep your cooling system clean and free of foreign substances. With the coolant drained and the valve de-gunked, re-taped with teflon tape and reinstalled, the heater return hose got pulled and the dremel came out to play. Lopped off the end of the hardline so it is closer to parallel to the strut tower sheetmetal instead of protruding into the Wastegate Actuator. This allow the 10AN hose to run on top of the frame rail, far enough away from the engine to not cause a vibration/heat failure. Route it to the return tee and we're all set. Best of all, it is RTS-ish. RTS is short for Return To Stock. In this case, it just needs some regular 5/8" heater hose and a 90 degree elbow. The original hose could also conceivably be used if spliced slightly longer.
And the deed is done. All that needed removed was roughly 3/4", but the difficult approach made my cut a bit angled with the longest part of the removed section being 1". This was accomplished by carefully threading the power cord for my dremel through front to back and using a #420 cutting disc to get a guide slot. With that in place, I extracted the dremel and did the rest from above using the rotary file cutter attachment. After the hardline was deburred, some 10AN hose went on there and the remnant elbow was placed on the return tee akin to the original 13B setup with a pex splicer mating the two pieces together. A pipe clamp attaches the hose to the front vacuum line guide and creates about 1/4" clearance from the actuator, so all is happy! If I spent more time on it, I'm sure I could square up the cut hardline end, but it'll do for now.
Before Pic
Removed Section Pic, with a normal Bosch-type relay for size reference
After Mod Pic (Black & blue)
While draining the coolant into my awesomely useful clear drain pan (a 15qt Sterilite box), I saw quite a bit of floating debris and even more in the bottom. An old but clean-ish tshirt acting as a fine screen fixed that, so the car has properly uncontaminated coolant now.
Unfortunately, these efforts did not cure my coolant leak. Further investigation led to drips on the hose clamp above the tee fitting, meaning that either the section attached to the lower pump housing neck (stock FC radiator hose) or the tee fitting are the culprit. A quick exam showed considerable rust on the tee, leading me to believe the upper hose section was the suspect. Pulled all of it apart and found noticeable scoring on the inside from the tee. Although I could not see any cracks or holes, I suspected a couple tiny pinholes and the scoring on the inside was troubling enough to warrant replacement with a spare one along with the rusted tee fitting.
Now I know what you're gonna say...this is hard! And normally I would agree, except I've got a world-reknown hotrod shop a couple miles away, a hardware store with an excellent plumbing/electrical department and the ability MacGyver solutions that is second to none. So late last night after a chat with Moldypoo, I got into the plumbing zone and found what I need to link 1.75", 1.5 and 0.625" hoses together at a common point with materials suitable for use in the 3rd Level of Hell...
Enter the #51159 radiator hose drain adapter from Jegs. On one side, a 1.75" hose fitting, on the other is a 1.5" and right in the middle is a 1/4" NPT threaded port. Now comes the fun part: maneuvering a 5/8" hose into there. Turns out it took a 1/4" NPT close hex nipple, 1/4" NPT tee (as an elbow), another 1" nipple, 1/2 to 1/4" NPT adapter and a 5/8" hose barb. While a bit Rube Goldberg, this arrangement eliminates the original heater hose entirely (no more 20+ year old rubber hose in a vital area!) and the benefit to the tee was simple: I needed to make a sharp bend and have 3 of them leftover from the early vacuum experiments. The unused port will get a drain valve to aid with air bleeding & bottomside maintenance. Coincidentally, the valve was free with the drain adapter.
Now, back to the matter at hand. After seeing the little blue target dancing around in a circle on the fuel map in ECU Manager,I grabbed the carb cleaner and sprayed around the intake gaskets. Vacuum leak, rear edge, upper half of Throttle Body-UIM junction has reappeared. Taking a look at the pictures on the previous page spelled it out: RTV failure at the thinnest part of the gasket. As one can see, the FD gasket is much thinner than the outer edges of the upper runners than the Cosmo's UIM & throttle body castings, only covering about 1/3 of the area in question. What happened is that while the RTV held at 15 inHg, 20 inHg sucked it right in thanks to Bernoillu's Principle. Due to the small area, Bernoulli's Principle magnified the pressure exponentially to where it pulled enough of the RTV right off at the point where the gasket/flange area difference was greatest to create a leak. While better RTV such as Permatex's "The Right Stuff" may work, trying to band-aid the problem will just make it re-occur in a less opportune location such as while driving somewhere. After waiting all day on a call back from Ray Crowe whom was trying to get the proper gasket from Mazdaspeed for me, Cody Johnson said "Why not make the gasket". The light bulb went on in my head, and 30 minutes later, I had one made from scratch using the throttle body as a template. It's slightly different from the Cosmo one in that I made the BAC opening two symmetrically placed round holes instead of the smiley-face, but it works and I have no more vacuum leaks! Cost $0.00
Now, back to tuning. With the car stumbling considerably, idling low and behaving strangely after doing the aforementioned cooling system work despite zero changes to the ECU, I'm re-examining all of the maps & logs. Plugs were a bit fouled (understandable), so I wiped off all the gunk but couldn't clean the center electrode at the moment and reinstalled them. Car behaved identically. Also tried reloading the working map, no behavioral changes...
At the moment, J9FD3S has tweaked a map for me (the "J10" revision) and it takes considerable throttle effort to start, plenty of backfiring and shaking in doing so. Runs lean, idles low (600-650rpm) despite messing with the idle adjustment screw on the TB, stalled on her own. Log included
And the deed is done. All that needed removed was roughly 3/4", but the difficult approach made my cut a bit angled with the longest part of the removed section being 1". This was accomplished by carefully threading the power cord for my dremel through front to back and using a #420 cutting disc to get a guide slot. With that in place, I extracted the dremel and did the rest from above using the rotary file cutter attachment. After the hardline was deburred, some 10AN hose went on there and the remnant elbow was placed on the return tee akin to the original 13B setup with a pex splicer mating the two pieces together. A pipe clamp attaches the hose to the front vacuum line guide and creates about 1/4" clearance from the actuator, so all is happy! If I spent more time on it, I'm sure I could square up the cut hardline end, but it'll do for now.
Before Pic
Removed Section Pic, with a normal Bosch-type relay for size reference
After Mod Pic (Black & blue)
While draining the coolant into my awesomely useful clear drain pan (a 15qt Sterilite box), I saw quite a bit of floating debris and even more in the bottom. An old but clean-ish tshirt acting as a fine screen fixed that, so the car has properly uncontaminated coolant now.
Unfortunately, these efforts did not cure my coolant leak. Further investigation led to drips on the hose clamp above the tee fitting, meaning that either the section attached to the lower pump housing neck (stock FC radiator hose) or the tee fitting are the culprit. A quick exam showed considerable rust on the tee, leading me to believe the upper hose section was the suspect. Pulled all of it apart and found noticeable scoring on the inside from the tee. Although I could not see any cracks or holes, I suspected a couple tiny pinholes and the scoring on the inside was troubling enough to warrant replacement with a spare one along with the rusted tee fitting.
Now I know what you're gonna say...this is hard! And normally I would agree, except I've got a world-reknown hotrod shop a couple miles away, a hardware store with an excellent plumbing/electrical department and the ability MacGyver solutions that is second to none. So late last night after a chat with Moldypoo, I got into the plumbing zone and found what I need to link 1.75", 1.5 and 0.625" hoses together at a common point with materials suitable for use in the 3rd Level of Hell...
Enter the #51159 radiator hose drain adapter from Jegs. On one side, a 1.75" hose fitting, on the other is a 1.5" and right in the middle is a 1/4" NPT threaded port. Now comes the fun part: maneuvering a 5/8" hose into there. Turns out it took a 1/4" NPT close hex nipple, 1/4" NPT tee (as an elbow), another 1" nipple, 1/2 to 1/4" NPT adapter and a 5/8" hose barb. While a bit Rube Goldberg, this arrangement eliminates the original heater hose entirely (no more 20+ year old rubber hose in a vital area!) and the benefit to the tee was simple: I needed to make a sharp bend and have 3 of them leftover from the early vacuum experiments. The unused port will get a drain valve to aid with air bleeding & bottomside maintenance. Coincidentally, the valve was free with the drain adapter.
Now, back to the matter at hand. After seeing the little blue target dancing around in a circle on the fuel map in ECU Manager,I grabbed the carb cleaner and sprayed around the intake gaskets. Vacuum leak, rear edge, upper half of Throttle Body-UIM junction has reappeared. Taking a look at the pictures on the previous page spelled it out: RTV failure at the thinnest part of the gasket. As one can see, the FD gasket is much thinner than the outer edges of the upper runners than the Cosmo's UIM & throttle body castings, only covering about 1/3 of the area in question. What happened is that while the RTV held at 15 inHg, 20 inHg sucked it right in thanks to Bernoillu's Principle. Due to the small area, Bernoulli's Principle magnified the pressure exponentially to where it pulled enough of the RTV right off at the point where the gasket/flange area difference was greatest to create a leak. While better RTV such as Permatex's "The Right Stuff" may work, trying to band-aid the problem will just make it re-occur in a less opportune location such as while driving somewhere. After waiting all day on a call back from Ray Crowe whom was trying to get the proper gasket from Mazdaspeed for me, Cody Johnson said "Why not make the gasket". The light bulb went on in my head, and 30 minutes later, I had one made from scratch using the throttle body as a template. It's slightly different from the Cosmo one in that I made the BAC opening two symmetrically placed round holes instead of the smiley-face, but it works and I have no more vacuum leaks! Cost $0.00
Now, back to tuning. With the car stumbling considerably, idling low and behaving strangely after doing the aforementioned cooling system work despite zero changes to the ECU, I'm re-examining all of the maps & logs. Plugs were a bit fouled (understandable), so I wiped off all the gunk but couldn't clean the center electrode at the moment and reinstalled them. Car behaved identically. Also tried reloading the working map, no behavioral changes...
At the moment, J9FD3S has tweaked a map for me (the "J10" revision) and it takes considerable throttle effort to start, plenty of backfiring and shaking in doing so. Runs lean, idles low (600-650rpm) despite messing with the idle adjustment screw on the TB, stalled on her own. Log included
Last edited by Akagis_white_comet; 07-30-13 at 02:42 PM. Reason: fixed first picture
07-30-13, 01:32 PM
#381
Moderator
iTrader: (3)
Join Date: Mar 2001
Location: https://www2.mazda.com/en/100th/
Posts: 30,764
Received 2,559 Likes
on
1,820 Posts
At the moment, J9FD3S has tweaked a map for me (the "J10" revision) and it takes considerable throttle effort to start, plenty of backfiring and shaking in doing so. Runs lean, idles low (600-650rpm) despite messing with the idle adjustment screw on the TB, stalled on her own. Log included
07-30-13, 04:06 PM
#382
Moderator
iTrader: (3)
Join Date: Mar 2001
Location: https://www2.mazda.com/en/100th/
Posts: 30,764
Received 2,559 Likes
on
1,820 Posts
The directors of the firm hired to continue the credits after the other people had been sacked, wish it to be known that they have just been sacked. The credits have been completed in an entirely different style at great expense and at the last minute.
ok, i made the coolant temp correction richer, and tweaked the coolant and air temp correction map, it was pulling 2 degrees of timing, oops.
i also bumped the idle speed to 1100 when cold, and back up to 850 when warm.
the log kind of looks like it was running like poop
ok, i made the coolant temp correction richer, and tweaked the coolant and air temp correction map, it was pulling 2 degrees of timing, oops.
i also bumped the idle speed to 1100 when cold, and back up to 850 when warm.
the log kind of looks like it was running like poop
07-30-13, 06:27 PM
#383
Backfires and sets off alarms
3 more logs and I set off some guy's car alarm because the car backfired
This is still on the J10 map, not the J10-1 one as I've yet to try it.
So my computer decided to delete the tuning guide I was painstakingly working on, along with my tuning behavior notes and all previous logs. Fortunately, all the logs are backed up on the laptop and I was able to recover some of the notes. The guide is gone though
Logs contain lots of weak voltage, ****-poor running, low idle and backfiring loud enough to set off an alarm on a vehicle parked 20ft away...the four major food groups!
Battery is on the charger was it was reading 12.0 in the Haltech during these logs. Just a hunch, but insufficient voltage seems seems a likely culprit for such behavior as it can't keep a relay on...
Enjoy!
This is still on the J10 map, not the J10-1 one as I've yet to try it.
So my computer decided to delete the tuning guide I was painstakingly working on, along with my tuning behavior notes and all previous logs. Fortunately, all the logs are backed up on the laptop and I was able to recover some of the notes. The guide is gone though
Logs contain lots of weak voltage, ****-poor running, low idle and backfiring loud enough to set off an alarm on a vehicle parked 20ft away...the four major food groups!
Battery is on the charger was it was reading 12.0 in the Haltech during these logs. Just a hunch, but insufficient voltage seems seems a likely culprit for such behavior as it can't keep a relay on...
Enjoy!
07-30-13, 09:31 PM
#384
Well, time to check all that wiring!
Although my car has a 13B, it had problems long ago where sometimes it would run perfect, other times it couldn't start, or would start poorly and run like crap. Turns out it was a bad (leading) coil. And man oh man, did it take me forever to figure that one out.
Sooooo.... check the coils too. You never know if they are all firing properly. Do you have others that you can swap in?
What about CAS? I've heard that the 20B (or was it 13BREW??) were known to have issues with the CAS's.
I don't really know anything anymore, just shooting in the dark.
Although my car has a 13B, it had problems long ago where sometimes it would run perfect, other times it couldn't start, or would start poorly and run like crap. Turns out it was a bad (leading) coil. And man oh man, did it take me forever to figure that one out.
Sooooo.... check the coils too. You never know if they are all firing properly. Do you have others that you can swap in?
What about CAS? I've heard that the 20B (or was it 13BREW??) were known to have issues with the CAS's.
I don't really know anything anymore, just shooting in the dark.
07-31-13, 12:20 AM
#385
Idle Tune 2 used the Holy Hand Grenade...It's Super Effective!
With a freshly charged battery, I tried the J10.1 map, it was okay but didn't have the right 'feel' to it and seemed to search for a happy idle point in 4 different cells. So, what I did was...
Thou shalt take out thy Holy Pin and counteth to Three. Three shall be the number of counting, and the number of counting shall be Three. Once you have reached Three, being the Third Number, thou shall lob thy Holy Hand Grenade of Antioch toward thy foe, who shall snuff it.
So I reverted to my Idle Tune 2 map and gave it a crank. She started with no effort and idled like melted butter at 730rpm at the limit of the idle adjust screw on the throttle body (dipped lower near the middle of the adjuster). I believe the car has said that this is how it wants to idle., so here are the pertinent details
12.5-13.0 AFR (12.7 average)
15.5 inHg
730rpm plus or minus 15rpm (within FSM spec of 750rpm +50 or -50)
~3.3% Injector DC
~2.75 ms Injection Time (+ or - 0.005)
10.6 degrees Ignition Advance
I did notice in the log that when the idle dropped to 700ish, the AFR was up near 13
Both the map & a log are attached.
Thou shalt take out thy Holy Pin and counteth to Three. Three shall be the number of counting, and the number of counting shall be Three. Once you have reached Three, being the Third Number, thou shall lob thy Holy Hand Grenade of Antioch toward thy foe, who shall snuff it.
So I reverted to my Idle Tune 2 map and gave it a crank. She started with no effort and idled like melted butter at 730rpm at the limit of the idle adjust screw on the throttle body (dipped lower near the middle of the adjuster). I believe the car has said that this is how it wants to idle., so here are the pertinent details
12.5-13.0 AFR (12.7 average)
15.5 inHg
730rpm plus or minus 15rpm (within FSM spec of 750rpm +50 or -50)
~3.3% Injector DC
~2.75 ms Injection Time (+ or - 0.005)
10.6 degrees Ignition Advance
I did notice in the log that when the idle dropped to 700ish, the AFR was up near 13
Both the map & a log are attached.
07-31-13, 05:53 AM
#386
Junior Member
Join Date: Jul 2013
Location: MD
Posts: 1
Likes: 0
Received 0 Likes
on
0 Posts
1986 RX7 rear seats HELP
Hey guys I am new to the RX7 world. I just inherited a 1986 RX7 from my father. He loved the car so I decided to buy it after he died and finish it up for me and my boys. I would like to know if anyone can help me out with finding a rear seat. if anyone has one they want to part with or where to get one. I know they are little but My boys would like to be able to go out in the car with me.
07-31-13, 10:55 AM
#387
Moderator
iTrader: (3)
Join Date: Mar 2001
Location: https://www2.mazda.com/en/100th/
Posts: 30,764
Received 2,559 Likes
on
1,820 Posts
the voltage/spark plug thing makes me wonder if we couldn't try again?
j10.2 i looked at the perfect idle map, and made the idle cell the same.
j10.2 i looked at the perfect idle map, and made the idle cell the same.
08-01-13, 03:23 PM
#388
Making progress
After adding too much oil from misreading the dipstick 
, the car was running quite funky even on the IT2 map with a low 600rpm idle, struggling to hold it, lean spiking/backfiring from running too rich, and FLATLINE OIL PRESSURE. Above 1000rpm is when the gauge would respond.
Now I know what everyone things: stock gauges are junk. Well, it's what I have to work with right now.
Drained 3 quarts from the pan and she's back to almost normal now Still idles a hair low (685 rpm give or take), but everything else is back to what it should be. Just for S&G, I reinstated the 8000rpm rev limiter and let her wind out to check the AFRs through the entire range while logging it. Looks good, though a bit rich in the higher RPMS. AFR valleyed at 10.2@ 7200rpm. From there to 7600, she leaned out slightly to 10.6.
TPS, Injection time & Duty Cycle traced after Manifold Pressure well. DC peaked at 76.7%, so we're still in the safe zone. Could be leaned out a bit overall, but is a good starting place for a street tune.
As the coolant still had some trapped air pockets from rebuilding the bottom radiator hose/heater hose junction, they began to work themselves out by making the Coolant Temperature spike up to 220, receiving an instant Key-off from me as the fan wasn't behaving properly due to the switch not being immersed in coolant as one needs to be to operate correctly.
Map & Log attached. I forgot to rename the log to match the map's name before uploading, so there you have it.
, the car was running quite funky even on the IT2 map with a low 600rpm idle, struggling to hold it, lean spiking/backfiring from running too rich, and FLATLINE OIL PRESSURE. Above 1000rpm is when the gauge would respond.Now I know what everyone things: stock gauges are junk. Well, it's what I have to work with right now.
Drained 3 quarts from the pan and she's back to almost normal now Still idles a hair low (685 rpm give or take), but everything else is back to what it should be. Just for S&G, I reinstated the 8000rpm rev limiter and let her wind out to check the AFRs through the entire range while logging it. Looks good, though a bit rich in the higher RPMS. AFR valleyed at 10.2@ 7200rpm. From there to 7600, she leaned out slightly to 10.6.
TPS, Injection time & Duty Cycle traced after Manifold Pressure well. DC peaked at 76.7%, so we're still in the safe zone. Could be leaned out a bit overall, but is a good starting place for a street tune.
As the coolant still had some trapped air pockets from rebuilding the bottom radiator hose/heater hose junction, they began to work themselves out by making the Coolant Temperature spike up to 220, receiving an instant Key-off from me as the fan wasn't behaving properly due to the switch not being immersed in coolant as one needs to be to operate correctly.
Map & Log attached. I forgot to rename the log to match the map's name before uploading, so there you have it.
08-02-13, 01:40 PM
#389
Test Drive 2: The Duel of AFRs
Last night was spent tuning the fuel map to level it out to 12.5-13.5 AFRs up to 6000rpm. Each step got a new name, ala IT2.1, IT2.2 and so on through IT2.5.
Just got back a much closer gas station, while getting video and datalogs of both to and from. My commentary isn't too great after having been up all night, but it does line things up between visible/audible behavior and clues/details found in the datalog.
Map & logs attached.
To Gas Station, corresponds with "To Gas" log:
From Gas Station 1 (Stalled)
From Gas Station 2 (Restart)
The reason for the split return video is because I didn't know how long it would take to get the car going again. You can see where they line up by playing it side-by-side with the "From Gas" datalog. The click I mention is the circuit breaker for the battery tripping (not sure why, see log for clues). I've not edited these at all, so what you see and hear is what I saw, heard and attempted to give good commentary on.
Car stalled & didn't want to restart when I got back home. Based on the consistently rich AFRs, my money's on the plugs being a bit fouled. After a nap, datalog analysis.
Just got back a much closer gas station, while getting video and datalogs of both to and from. My commentary isn't too great after having been up all night, but it does line things up between visible/audible behavior and clues/details found in the datalog.
Map & logs attached.
To Gas Station, corresponds with "To Gas" log:
From Gas Station 1 (Stalled)
From Gas Station 2 (Restart)
The reason for the split return video is because I didn't know how long it would take to get the car going again. You can see where they line up by playing it side-by-side with the "From Gas" datalog. The click I mention is the circuit breaker for the battery tripping (not sure why, see log for clues). I've not edited these at all, so what you see and hear is what I saw, heard and attempted to give good commentary on.
Car stalled & didn't want to restart when I got back home. Based on the consistently rich AFRs, my money's on the plugs being a bit fouled. After a nap, datalog analysis.
08-07-13, 08:29 PM
#390
Houston, we have a problem...
Checking the datalog from Test Drive 2 showed me that the alternator's voltage regulator was going south as its output wasn't consistent with engine RPM. A few times, I saw the voltmeter needle bouncing off of 15v like a tach bouncing off the rev limiter. This behavior was witnessed at 1500rpm, so couple it with the other inconsistencies and the voltage regulator is the issue.
Yesterday after work, I went to the junkyard with one of my coworkers and snagged a new alternator for $10, a tire changing jack for $2.50 and a GM knock box (#16065711, cross references to the one mentioned in the MegaSquirt manual). Knock box was from a Chevy Astro and within easy reach, mounted right beside the Map sensor & EGR solenoid. As for the jack, I knew that a 20B conversion adds some weight to the car, so its jack needed to be appropriately uprated as well. My first choice was a 99+ Cougar jack as it has an integrated handle like my old mystery jack did and its donor car weighs just shy of 3000lbs. Couldn't find one of them, so I kept hunting around...and found the perfect one in the trunk of an Infini J30, Nissan's 3500lb answer to the Eunos Cosmo. Somehow, I had a feeling that the right one would come from a fullsize Japanese car all along...
It just happens to fit perfectly in the stock FC location, has the correct type of saddle and its handle (the ? hook kind) fits in the FC's tool bag in the spare tire well. It's rated for 1100kg (2420 lbs), more than enough to do the job no matter what, which was the goal.
They only had one 92-95 Taurus and it was a 3.0L, so I pulled it and it tested bad at the local Autozone. Took it back to the junkyard, then took another look around for the other Ford 3.8's. All I could find was a lonely old 3.8L Windstar buried in the back of the yard. Pulled its alternator, got it tested which it failed due to Autozone's pigtail being damaged. However, in a rare turn of events, the gentleman whom did the test could tell by ear that it was working properly. So I gave him the junkyard pigtail so they could fix their tester and went over to Advance Auto whom validated the Autozone fellow's suspicions that the alternator was indeed fine.
The new alternator met the wire wheel to de-gunk and shine it up a bit before install. One thing I've done differently this time around is add a second circuit breaker for the alternator. I had originally planned on this and bought a second breaker, but never needed it untill now as the alternator's output stayed within the proper range. This oversight had the unfortunate effect of disconnecting the battery during Test Drive 2, showing that a runaway taurus alternator can put out over 150 amps, even when you factor in resistance from the ~15ft of battery cables. The new setup will have the alternator go through its own 150A breaker under the hood before meeting with the 12v bus bar. Should the car ever experience another runaway alternator, the front breaker will trip sooner than the rear one did, disconnecting the alternator output cable from everything, then the battery warning light will come on to show that the car is running off the battery. Failure Mode: Apollo 13
A strange little quirk I noticed is that the voltage regulators have different part numbers. F1DU-10-316-AA is the regulator on the old alternator, while F4ZU-10-316-AC is the new one. It may not be relevant that the old one is from a Taurus while the new one is from a Windstar (both were 3.8's, all original Motorcraft parts too), but everything else is identical as far as I can tell. Part numbers don't cross-reference to each other, as the F4ZU is a Load Response Controller type (response time 2.5 seconds) while the F1DU is not. Looks like it just needed waking up like the old one, but works just as well as the old one did on my 13B. Revved it up to 7500 a couple times to see its response and got a constant 14.6v, so we're back in business.
So I couldn't resist hooking up the GM Knock Box to play around with it a bit. Turns out that it can work on either 12v or 5v, but 12v operation requires a Zener Diode to trim the voltage down to 5v.
Also, the Haltech can't tell the difference in voltage as both states are too high. 12v operation yields a drop to ~8v during a knock event, while 5v input power turns into an output of ~4.8v during knock.
This won't do, the difference between good and knocking needs to be much further apart. So it's time to build a voltage divider into the circuit. Since I've already got a convenient 5v bus bar, I omitted the diode and pulled yet another item stashed away since my R/C car days, a pair of resistors & bracket from a Traxxas Rustler (circa 1997). As I had purchased an ESC following its assembly, these sat in the parts box for the next 15 years. A couple short M3x0.5 screws & nuts bolted everything snugly together, then the bracket's mounting hole was opened up to M6 and attached to the strut tower along with the knock box. The bracket also acts as an aluminum heatsink for the resistors too
With the resistors & knock box mocked up using the bench supply, I simulated knock by applying 5v to the sensor wire (blue). A quick tap yielded 4.8v while tapping it much faster showed as low as 3.6v on the multimeter. Last, as the two remaining OE Mazda sensor connectors appear to be 0.110" which I don't have, all 3 were replaced with common bullet connectors for simplicity.
So today I made a couple changes to the map file and did another datalog, saved as IT2.6 and attached below: Closed Loop O2 Control is enabled now as follows:
30 second post-start delay, 500-3500rpm, 1000rpm idle, 200 degree/50% throttle cutoff, Target AFR for 0-1000rpm/14.6-23.3 inHg set to 13.7, above 1000 set to 14.0 (2.25v)
AVI #5 (Orange/Green wire) is now the input for the GM Knock Control Box set to pull 10 degrees whenever its output drops below 4.8v. This didn't work too well as its voltage stayed at 4.15 during the datalog
It was running heavily rich in the 11s at 730rpm, partly due to the timing being pulled prematurely and consistently rich/stumbly when revved to 7250 with a cloud of blue smoke resulting from the exhaust. O2 Control didn't work as intended for a reason I've yet to discover (insufficient adjustment range?). On a good note, the new alternator stayed solid in the 13-14v range.
Alas, I have some good news. I have found that my target AFR for running N/A should be 14:1. Idle was pretty close before enabling O2 control, so it's gonna need more work all the way up to 8000rpm & 20 inHg. AFR info here: https://www.rx7club.com/2nd-generati...g-afrs-878432/
Yesterday after work, I went to the junkyard with one of my coworkers and snagged a new alternator for $10, a tire changing jack for $2.50 and a GM knock box (#16065711, cross references to the one mentioned in the MegaSquirt manual). Knock box was from a Chevy Astro and within easy reach, mounted right beside the Map sensor & EGR solenoid. As for the jack, I knew that a 20B conversion adds some weight to the car, so its jack needed to be appropriately uprated as well. My first choice was a 99+ Cougar jack as it has an integrated handle like my old mystery jack did and its donor car weighs just shy of 3000lbs. Couldn't find one of them, so I kept hunting around...and found the perfect one in the trunk of an Infini J30, Nissan's 3500lb answer to the Eunos Cosmo. Somehow, I had a feeling that the right one would come from a fullsize Japanese car all along...
It just happens to fit perfectly in the stock FC location, has the correct type of saddle and its handle (the ? hook kind) fits in the FC's tool bag in the spare tire well. It's rated for 1100kg (2420 lbs), more than enough to do the job no matter what, which was the goal.
They only had one 92-95 Taurus and it was a 3.0L, so I pulled it and it tested bad at the local Autozone. Took it back to the junkyard, then took another look around for the other Ford 3.8's. All I could find was a lonely old 3.8L Windstar buried in the back of the yard. Pulled its alternator, got it tested which it failed due to Autozone's pigtail being damaged. However, in a rare turn of events, the gentleman whom did the test could tell by ear that it was working properly. So I gave him the junkyard pigtail so they could fix their tester and went over to Advance Auto whom validated the Autozone fellow's suspicions that the alternator was indeed fine.
The new alternator met the wire wheel to de-gunk and shine it up a bit before install. One thing I've done differently this time around is add a second circuit breaker for the alternator. I had originally planned on this and bought a second breaker, but never needed it untill now as the alternator's output stayed within the proper range. This oversight had the unfortunate effect of disconnecting the battery during Test Drive 2, showing that a runaway taurus alternator can put out over 150 amps, even when you factor in resistance from the ~15ft of battery cables. The new setup will have the alternator go through its own 150A breaker under the hood before meeting with the 12v bus bar. Should the car ever experience another runaway alternator, the front breaker will trip sooner than the rear one did, disconnecting the alternator output cable from everything, then the battery warning light will come on to show that the car is running off the battery. Failure Mode: Apollo 13
A strange little quirk I noticed is that the voltage regulators have different part numbers. F1DU-10-316-AA is the regulator on the old alternator, while F4ZU-10-316-AC is the new one. It may not be relevant that the old one is from a Taurus while the new one is from a Windstar (both were 3.8's, all original Motorcraft parts too), but everything else is identical as far as I can tell. Part numbers don't cross-reference to each other, as the F4ZU is a Load Response Controller type (response time 2.5 seconds) while the F1DU is not. Looks like it just needed waking up like the old one, but works just as well as the old one did on my 13B. Revved it up to 7500 a couple times to see its response and got a constant 14.6v, so we're back in business.
So I couldn't resist hooking up the GM Knock Box to play around with it a bit. Turns out that it can work on either 12v or 5v, but 12v operation requires a Zener Diode to trim the voltage down to 5v.
Also, the Haltech can't tell the difference in voltage as both states are too high. 12v operation yields a drop to ~8v during a knock event, while 5v input power turns into an output of ~4.8v during knock.
This won't do, the difference between good and knocking needs to be much further apart. So it's time to build a voltage divider into the circuit. Since I've already got a convenient 5v bus bar, I omitted the diode and pulled yet another item stashed away since my R/C car days, a pair of resistors & bracket from a Traxxas Rustler (circa 1997). As I had purchased an ESC following its assembly, these sat in the parts box for the next 15 years. A couple short M3x0.5 screws & nuts bolted everything snugly together, then the bracket's mounting hole was opened up to M6 and attached to the strut tower along with the knock box. The bracket also acts as an aluminum heatsink for the resistors too
With the resistors & knock box mocked up using the bench supply, I simulated knock by applying 5v to the sensor wire (blue). A quick tap yielded 4.8v while tapping it much faster showed as low as 3.6v on the multimeter. Last, as the two remaining OE Mazda sensor connectors appear to be 0.110" which I don't have, all 3 were replaced with common bullet connectors for simplicity.
So today I made a couple changes to the map file and did another datalog, saved as IT2.6 and attached below: Closed Loop O2 Control is enabled now as follows:
30 second post-start delay, 500-3500rpm, 1000rpm idle, 200 degree/50% throttle cutoff, Target AFR for 0-1000rpm/14.6-23.3 inHg set to 13.7, above 1000 set to 14.0 (2.25v)
AVI #5 (Orange/Green wire) is now the input for the GM Knock Control Box set to pull 10 degrees whenever its output drops below 4.8v. This didn't work too well as its voltage stayed at 4.15 during the datalog
It was running heavily rich in the 11s at 730rpm, partly due to the timing being pulled prematurely and consistently rich/stumbly when revved to 7250 with a cloud of blue smoke resulting from the exhaust. O2 Control didn't work as intended for a reason I've yet to discover (insufficient adjustment range?). On a good note, the new alternator stayed solid in the 13-14v range.
Alas, I have some good news. I have found that my target AFR for running N/A should be 14:1. Idle was pretty close before enabling O2 control, so it's gonna need more work all the way up to 8000rpm & 20 inHg. AFR info here: https://www.rx7club.com/2nd-generati...g-afrs-878432/
08-20-13, 12:41 PM
#392
The ties that bind us
Thank you for the praise driftsequence. Turns out I forgot one simple thing in regard to adapting the GM ESC Box that the MegaSquirt article didn't mention directly: its starting voltage when not knocking.
The MS assumes that the neutral point input voltage is 5.0v and the GM ESC box's output voltage during knock is lower than that (user-defined Knock Threshold voltage). However, the Haltech PS1K/PS2K have their neutral point at 2.5v to facilitate the use of their Trim Control **** in both directions (advance and retard) as it operates exactly like a TPS. It's a simple 5v potentiometer that adds/subtracts ignition timing linearly based on a user-defined maximum amount in ECU Manager (10 degrees by default IIRC). In this case, the specified AVI seeing 0.00v would tell the ECU to pull timing by the full 10 degrees while the full 5v would mean advancing timing by the full 10 degrees. Seeing 1.25v or 3.75v would mean pulling/advancing timing by 5 degrees respectively.
My last experiment on the car yielded a neutral point voltage of 4.0v using both resistors in series with the ESC Box's output, which won't do. In the previously-noted test which yielded 4.15, I saw in the log that ignition timing was advanced by 17 degrees. Did the math and found it was close enough (1.65/2.5=67% of total advance, or 13.2 degrees). Since my current map is quite conservative on ignition timing and not in boost yet, nothing bad happened from this artificial advance.
After re-examining the article, it dawned on me that they were using a 4.7 zener diode as a voltage divider (12v source power) to bleed off all voltage above 4.7v. It then hit me to do the same. Rewired the resistors as a voltage divider and got 2.5v when standing alone but 0.01v when joined up with the ESC box due to the resistors consuming 10 watts each. So it went back to the drawing board to make a new divider the ultimately consisted of eight 470 ohm 1/4 watt resistors in series and the output wire to the ECU between resistor #1 and 2. Add it back in and the circuit's output when not knocking is 2.5v on the dot! Major thanks to my brother Dan for soldering all of it together for me.
Now, dinner, some hot glue to keep everything absolutely solid, heat shrink and some padding on all sides so it stays put inside the aluminum bracket.
In recap:
Megasquirt thinks 5.0v on a selected input means no knock, Haltech says it is 2.5v
Use several resistors as a voltage divider to make the GM ESC box output 2.5v when not knocking, less when knocking. At 5.0v output, knock shows up as 4.8 or less, so 2.4v is the knock threshold for retarding the timing by 0.8 degrees.
Had to cheat a bit on the throttle body to up the idle to 750-800rpm by using a longer bolt for the low-end throttle stop. TPS still reports 0%, but a longer bolt let me put the Idle Adjust Screw closer to the middle where it needs to be. Per advice from C. Ludwig, it turns out there is a bug in ECU Manager v1.11 regarding Closed Loop O2 Correction as the Max Increase and Max Decrease values are backwards. When I put in a Max Increase of 20%, it pulled 20% of the fuel as soon as O2 Correction kicked in and stalled the engine. Killed off O2 Correction for the time being and saved the settings/map as FT1.0.
Over the last few days, 8/11-8/16, I got some help from Colten Palm on Ohio Rotaries for some road tuning. With him driving, the car is coming along well just be going up and down a nice long stretch of road by my house in 2nd and 3rd gear. As of right now, car is now tuned to 4250rpm with a mostly steady 13.0:1 AFR and the map is saved as FT1.1. Had some fan issues due to the redundant Ford Taurus fan circuit breakers (used as a short extension to make wiring easier) which kept tripping after a few seconds, so they're gone now. Even in its current N/A form and with an ACT Prolite flywheel, it is deceptively powerful and smooth. Let off the throttle and she purrs like a tiger cub.
The other issue we've encountered is that the 150A breaker that the alternator runs through keeps tripping. Voltage doesn't spike up to indicate overcharging, so my money is that heat soaking is making it trip artificially early. During a no-load temperature test using a hair dryer & thermometer, the breaker didn't trip at all, even at 190 degrees Fahrenheit. At such an elevated temperature, its load capacity is probably reduced enough to trip under the alternator's output though. Made a heat shield for it but saw little difference, so it may just be a defective breaker.
Ironically, the fan issue and the Alternator Breaker sidelined the car at the same house twice a few blocks away from home. The owner of said house there is an oldschool gearhead who is restoring a 73 Corvette with a 454 big block and has much appreciation for rotary engines as a result of a buddy of his owning a FD back in the late 90s. When I told him that my car is just like a FD, but with 50% more displacement, his jaw dropped...
The past couple days have been an emotional roller coaster. My girlfriend of the past 2.5 years, Trula, broke up with me quite reluctantly on 8/18 at 10:30PM and it has really messed with my drive to get anything done. Out of proper respect, I won't go into details publicly but I've sequestered myself away for a week or so to figure out what I want and need. At the moment, I am doing okay but I know it comes in waves so this week I am writing my thoughts down as they come to see what 'sticks'. I have an idea of what the answer is and how to get there. Please, no discussion about this or its effects.
With all of that said, the car is done to Stage 1: N/A drivable. And untill today...
You might be missing some of the benefits that stereo can provide!
The MS assumes that the neutral point input voltage is 5.0v and the GM ESC box's output voltage during knock is lower than that (user-defined Knock Threshold voltage). However, the Haltech PS1K/PS2K have their neutral point at 2.5v to facilitate the use of their Trim Control **** in both directions (advance and retard) as it operates exactly like a TPS. It's a simple 5v potentiometer that adds/subtracts ignition timing linearly based on a user-defined maximum amount in ECU Manager (10 degrees by default IIRC). In this case, the specified AVI seeing 0.00v would tell the ECU to pull timing by the full 10 degrees while the full 5v would mean advancing timing by the full 10 degrees. Seeing 1.25v or 3.75v would mean pulling/advancing timing by 5 degrees respectively.
My last experiment on the car yielded a neutral point voltage of 4.0v using both resistors in series with the ESC Box's output, which won't do. In the previously-noted test which yielded 4.15, I saw in the log that ignition timing was advanced by 17 degrees. Did the math and found it was close enough (1.65/2.5=67% of total advance, or 13.2 degrees). Since my current map is quite conservative on ignition timing and not in boost yet, nothing bad happened from this artificial advance.
After re-examining the article, it dawned on me that they were using a 4.7 zener diode as a voltage divider (12v source power) to bleed off all voltage above 4.7v. It then hit me to do the same. Rewired the resistors as a voltage divider and got 2.5v when standing alone but 0.01v when joined up with the ESC box due to the resistors consuming 10 watts each. So it went back to the drawing board to make a new divider the ultimately consisted of eight 470 ohm 1/4 watt resistors in series and the output wire to the ECU between resistor #1 and 2. Add it back in and the circuit's output when not knocking is 2.5v on the dot! Major thanks to my brother Dan for soldering all of it together for me.
Now, dinner, some hot glue to keep everything absolutely solid, heat shrink and some padding on all sides so it stays put inside the aluminum bracket.
In recap:
Megasquirt thinks 5.0v on a selected input means no knock, Haltech says it is 2.5v
Use several resistors as a voltage divider to make the GM ESC box output 2.5v when not knocking, less when knocking. At 5.0v output, knock shows up as 4.8 or less, so 2.4v is the knock threshold for retarding the timing by 0.8 degrees.
Had to cheat a bit on the throttle body to up the idle to 750-800rpm by using a longer bolt for the low-end throttle stop. TPS still reports 0%, but a longer bolt let me put the Idle Adjust Screw closer to the middle where it needs to be. Per advice from C. Ludwig, it turns out there is a bug in ECU Manager v1.11 regarding Closed Loop O2 Correction as the Max Increase and Max Decrease values are backwards. When I put in a Max Increase of 20%, it pulled 20% of the fuel as soon as O2 Correction kicked in and stalled the engine. Killed off O2 Correction for the time being and saved the settings/map as FT1.0.
Over the last few days, 8/11-8/16, I got some help from Colten Palm on Ohio Rotaries for some road tuning. With him driving, the car is coming along well just be going up and down a nice long stretch of road by my house in 2nd and 3rd gear. As of right now, car is now tuned to 4250rpm with a mostly steady 13.0:1 AFR and the map is saved as FT1.1. Had some fan issues due to the redundant Ford Taurus fan circuit breakers (used as a short extension to make wiring easier) which kept tripping after a few seconds, so they're gone now. Even in its current N/A form and with an ACT Prolite flywheel, it is deceptively powerful and smooth. Let off the throttle and she purrs like a tiger cub.
The other issue we've encountered is that the 150A breaker that the alternator runs through keeps tripping. Voltage doesn't spike up to indicate overcharging, so my money is that heat soaking is making it trip artificially early. During a no-load temperature test using a hair dryer & thermometer, the breaker didn't trip at all, even at 190 degrees Fahrenheit. At such an elevated temperature, its load capacity is probably reduced enough to trip under the alternator's output though. Made a heat shield for it but saw little difference, so it may just be a defective breaker.
Ironically, the fan issue and the Alternator Breaker sidelined the car at the same house twice a few blocks away from home. The owner of said house there is an oldschool gearhead who is restoring a 73 Corvette with a 454 big block and has much appreciation for rotary engines as a result of a buddy of his owning a FD back in the late 90s. When I told him that my car is just like a FD, but with 50% more displacement, his jaw dropped...
The past couple days have been an emotional roller coaster. My girlfriend of the past 2.5 years, Trula, broke up with me quite reluctantly on 8/18 at 10:30PM and it has really messed with my drive to get anything done. Out of proper respect, I won't go into details publicly but I've sequestered myself away for a week or so to figure out what I want and need. At the moment, I am doing okay but I know it comes in waves so this week I am writing my thoughts down as they come to see what 'sticks'. I have an idea of what the answer is and how to get there. Please, no discussion about this or its effects.
With all of that said, the car is done to Stage 1: N/A drivable. And untill today...
You might be missing some of the benefits that stereo can provide!
08-20-13, 03:27 PM
#393
It's funny I just took my car out for a ride last night and had my center cluster disassembled...I forgot what its like just having the radio turned on a few clicks so its barely audible.
Glad to see the knock control is coming along. What does the engine bay look like with all of your wiring all over the place? haha. I now appreciate the care that is put into our engine bay. Most things are easily accessible and dont require much disassembly to get to other parts.
Glad to see the knock control is coming along. What does the engine bay look like with all of your wiring all over the place? haha. I now appreciate the care that is put into our engine bay. Most things are easily accessible and dont require much disassembly to get to other parts.
09-08-13, 08:00 PM
#394
She isn't leaking, she marks her territory!
I apologize for the lack of updates as of recent, these past days have been VERY difficult on me and we all know why. This past friday (8/30), I spent the day with a good friend whom I had lost touch with over the past couple years. Prior to her arrival, I made a video message to Trula to express all of my thoughts & feelings in a clear, visible, unmistakable manner. Uploaded to youtube, set as private (link required to see it) and gathered a few opinions from people closest to me about its content. It was unanimous and one friend felt strongly enough about it, his exact words being "She NEEDS to see this, NOW!", that he found her info on his own and sent it to her without me knowing untill after he did so. Not exactly cool, but with good intentions so I'm not mad about it. Also, as fate may have it, it turns out a quite famous someone has been watching the whole situation unfold pretty closely despite being exceedingly busy. I can't name any names, but he's always been someone I looked up to since I was 10 years old. If you know me, you know who it is and why I have the specific ringtone and message tones on my phone.
Now, back to the car. No stereo update this time. After a worry-free drive to the store and back, the alternator pulley nut came loose as I was going to drive to the hardware store...
Pulley is trashed, same with the new taurus alternator recently acquired. Due to my head being a mess for reasons needing no further mention, I destroyed my good previous alternator like an idiot. Never knew cast aluminum shatters like glass when dropped/thrown from 4ft. Thankfully, Marcus from ZN Performance has spotted me a spare S5 alternator to get me going in the meantime. Same wiring as the Taurus Alt, so it was easy to put in aside from the spacer being a tad tight. Went for a test drive and it was a steady 14.0v without the fan kicking on yet. With the fan on low, it holds a steady 13.0v.
Went up to Quaker Steak & Lube for the friday night meet on the 23rd, and got a surprising amount of attention. Even when parked next to a 8.4l Viper SRT-10, people were coming over to see my car instead because it was so different than the regulars' Corvettes, JZA80 Supras, G35s and so on. My car was the only rotary there, although a turbo miata came in later and parked next to my car. Overall, massive positive vibes there and it's exactly what I needed to gain some much-needed mental clarity.
This was the first highway cruise at 60-70mph and the car stayed at a solid 210-215 degrees. Unfortunately, the Subaru fan switch kicking on at 203 meant the alternator was under an extra unnecessary load the whole way there. Thankfully, my battery is beefy enough to take the strain and fill in the gaps where necessary, including post-drive cooldown. On the way back, temperature started to climb about a mile from home, hitting 230 degrees when I sidelined the car for an extra cooldown break in Jegs' parking lot as a Panda AE86 went by, probably on his way to deliver tofu somewhere (it has the decal). After getting back home, I cracked open the JC Cosmo FSM, delved into it and found that its coolant gauge calibration chart. The middle of the gauge is where the needle stays untill 235 degrees, then starts climing again. So based on this, the 210-215 degrees on the way up to the meet is just about right where it wants to be.
I have a coolant leak somewhere near the front of the car and suspect it is the top radiator hose at the copper elbow nearest the radiator. My hunch is that it only starts leaking once the system is pressurized, as there was only a slight amount of coolant on the elbow's underside when I checked it today before my buddy Syed came by to requisition my assistance on his 2001 Accord's O2 sensor troubles. After checking every fuse & circuit in the car, it stumped me that it would say the O2 sensor heater circuit was kaput but I didn't have the wiring diagrams to do a pinout check at the connector to be certain. Did an impromptu oil change on the Accord, finding both the filter & drain plug to be excessively overtorqued by the dealership during its previous oil change. Had to hammer a screwdriver through the filter to get it loose and the aluminum crush washer was forcibly contoured to the plug's shaft while the hex head showed signs of impact wrench use. Some careful hammer work, prying and a 17mm socket got the damaged crush washer off to be replaced by a spare dowty seal. Also, Autozone gave him the wrong filter as it wouldn't thread on. Fortunately, I had a spare one from an oil change special last month at Advance Auto Parts and it worked like a charm. My advice was to just say you have a RX7 from now on and get a cap-style filter wrench.
So after that ordeal, I needed some more positive vibes and gave him a ride in the RX7. I swear that she gets faster every time she's driven. With only about 1/2" of throttle in 2nd, 20mph turned into 50+ suprizingly quick and with a smooth, constant force pressing you into the seat. When we got back to the house, he only said three words: I want one!
Today, 9/8/2013, I tackled the coolant leak with a couple of 1.5" Polypropylene insert elbows from Lowes, specifically Genova #350715. At $1 each, it was worth a try. Unlike the loosely fitting 1.25" copper elbow joiners from before, these are sized to join 1.5" polypropylene tubing with hose clamps and fit nice and tight onto the radiator hose segments. Did the homework and found out that the material is good for up to 275 degrees Fahrenheit before it starts to soften. As my engine stays at 210-215 normally and only peaked to 230 because it was leaking coolant under the previous incarnation of the top hose, this should work pretty well untill I can do something better. Just to play it safe, I added some Redline Water Wetter to help offset the lost coolant that was replaced with water. It should help a decent bit.
Next update will have pics.
Now, back to the car. No stereo update this time. After a worry-free drive to the store and back, the alternator pulley nut came loose as I was going to drive to the hardware store...
Pulley is trashed, same with the new taurus alternator recently acquired. Due to my head being a mess for reasons needing no further mention, I destroyed my good previous alternator like an idiot. Never knew cast aluminum shatters like glass when dropped/thrown from 4ft. Thankfully, Marcus from ZN Performance has spotted me a spare S5 alternator to get me going in the meantime. Same wiring as the Taurus Alt, so it was easy to put in aside from the spacer being a tad tight. Went for a test drive and it was a steady 14.0v without the fan kicking on yet. With the fan on low, it holds a steady 13.0v.
Went up to Quaker Steak & Lube for the friday night meet on the 23rd, and got a surprising amount of attention. Even when parked next to a 8.4l Viper SRT-10, people were coming over to see my car instead because it was so different than the regulars' Corvettes, JZA80 Supras, G35s and so on. My car was the only rotary there, although a turbo miata came in later and parked next to my car. Overall, massive positive vibes there and it's exactly what I needed to gain some much-needed mental clarity.
This was the first highway cruise at 60-70mph and the car stayed at a solid 210-215 degrees. Unfortunately, the Subaru fan switch kicking on at 203 meant the alternator was under an extra unnecessary load the whole way there. Thankfully, my battery is beefy enough to take the strain and fill in the gaps where necessary, including post-drive cooldown. On the way back, temperature started to climb about a mile from home, hitting 230 degrees when I sidelined the car for an extra cooldown break in Jegs' parking lot as a Panda AE86 went by, probably on his way to deliver tofu somewhere (it has the decal). After getting back home, I cracked open the JC Cosmo FSM, delved into it and found that its coolant gauge calibration chart. The middle of the gauge is where the needle stays untill 235 degrees, then starts climing again. So based on this, the 210-215 degrees on the way up to the meet is just about right where it wants to be.
I have a coolant leak somewhere near the front of the car and suspect it is the top radiator hose at the copper elbow nearest the radiator. My hunch is that it only starts leaking once the system is pressurized, as there was only a slight amount of coolant on the elbow's underside when I checked it today before my buddy Syed came by to requisition my assistance on his 2001 Accord's O2 sensor troubles. After checking every fuse & circuit in the car, it stumped me that it would say the O2 sensor heater circuit was kaput but I didn't have the wiring diagrams to do a pinout check at the connector to be certain. Did an impromptu oil change on the Accord, finding both the filter & drain plug to be excessively overtorqued by the dealership during its previous oil change. Had to hammer a screwdriver through the filter to get it loose and the aluminum crush washer was forcibly contoured to the plug's shaft while the hex head showed signs of impact wrench use. Some careful hammer work, prying and a 17mm socket got the damaged crush washer off to be replaced by a spare dowty seal. Also, Autozone gave him the wrong filter as it wouldn't thread on. Fortunately, I had a spare one from an oil change special last month at Advance Auto Parts and it worked like a charm. My advice was to just say you have a RX7 from now on and get a cap-style filter wrench.
So after that ordeal, I needed some more positive vibes and gave him a ride in the RX7. I swear that she gets faster every time she's driven. With only about 1/2" of throttle in 2nd, 20mph turned into 50+ suprizingly quick and with a smooth, constant force pressing you into the seat. When we got back to the house, he only said three words: I want one!
Today, 9/8/2013, I tackled the coolant leak with a couple of 1.5" Polypropylene insert elbows from Lowes, specifically Genova #350715. At $1 each, it was worth a try. Unlike the loosely fitting 1.25" copper elbow joiners from before, these are sized to join 1.5" polypropylene tubing with hose clamps and fit nice and tight onto the radiator hose segments. Did the homework and found out that the material is good for up to 275 degrees Fahrenheit before it starts to soften. As my engine stays at 210-215 normally and only peaked to 230 because it was leaking coolant under the previous incarnation of the top hose, this should work pretty well untill I can do something better. Just to play it safe, I added some Redline Water Wetter to help offset the lost coolant that was replaced with water. It should help a decent bit.
Next update will have pics.
09-17-13, 05:19 PM
#398
Well, my cheap fix for the upper radiator hose failed after 10 miles. It was noticeably wet at the lower elbow and there was a MASSIVE leak somewhere near the turbos. It was leaking at the rear of the passenger side fender, and a considerable amount at that. I'm not sure if it was on the turbos as there wasn't any steam, or if the leak was in another location such as the Heater Core to hardline hose (probably the original from 1986...), due to lack of evidence. All I know is that it was hemorrhaging liquid coolant at the rear bottom edge of the fender and I could not find any visible drip-paths.
Combine this with the oil pan and turbo return leaks and I'm ready to take it to a shop as I'm completely sick of fluid leaks now. No, I'm not going soft, I've just had enough of dealing with leaks for the past 8 months...
Combine this with the oil pan and turbo return leaks and I'm ready to take it to a shop as I'm completely sick of fluid leaks now. No, I'm not going soft, I've just had enough of dealing with leaks for the past 8 months...
10-26-13, 07:38 AM
#399
So things have settled down a bit. Upper radiator hose is fixed with a couple silicone elbows, two hose joiners from Jegs and a leftover section from the original S4 top hose. No more leaks here!!
One coolant leak appeared at the rear of the engine bay, somehow ending up at the passenger fender behind the wheel. Did some checking around and found the heater core supply hose wasn't on the firewall fitting enough. So it dripped onto the vacuum line running right below it. Shoved it back on further and gave it a proper hose clamp and it stopped leaking as well.
With this said, there's only one possible source for a leak. Coolant exits to the overflow bottle as it should once the engine is warmed up. But after it cools down, it doesn't pull it back in. A pinhole in the overflow bottle tube is the primary suspect for it, so it's getting replaced with fresh hose. And with that said, that should be all of the coolant leaks. But just to be certain, it'll get pressure tested anyway.
Now onto something else. My buddy Jeff at work, a fellow car guy with an impressive yet unfinished 87 Camaro, is lending me his garage to deal with the oil pan & turbo oil drain leaks. As soon as I popped the hood, his jaw dropped. As I am writing this now, the car is sitting in his garage, awaiting some wrench time today & tomorrow to get it solved once and for all. As soon as we get the pan & drains sorted out, it's back to work on his daily driver camry's engine swap (hole in block). Fortunately, mine will be easy to pull thanks to everything being labeled, intelligently organized and accessible by normal sized human hands. Yay for smart!
In the meantime, I've been biking to & from my job, 5 miles one way. It's exactly what I needed, same with where I work too. As I'm fond of saying, "a voice of sanity in an insane world". Plus, I'm making enough to take care of everything I need to do, have plenty leftover for all the fun stuff and I actually look forward to going to work now. It's people that make all the difference. So I just want to give a big thank you to Jeff, Bridgette, Justin, Josh, Alex, Stacey, Darryl, Mike and Dante at work for being so cool to hang out with everyday and seeing the good in people. The world needs more people like y'all.
With all of this said, and once the oil issues are resolved, the car will be finished, minus A/C, by Christmas. Next paycheck will see more awesomeness, so stay tuned.
One coolant leak appeared at the rear of the engine bay, somehow ending up at the passenger fender behind the wheel. Did some checking around and found the heater core supply hose wasn't on the firewall fitting enough. So it dripped onto the vacuum line running right below it. Shoved it back on further and gave it a proper hose clamp and it stopped leaking as well.
With this said, there's only one possible source for a leak. Coolant exits to the overflow bottle as it should once the engine is warmed up. But after it cools down, it doesn't pull it back in. A pinhole in the overflow bottle tube is the primary suspect for it, so it's getting replaced with fresh hose. And with that said, that should be all of the coolant leaks. But just to be certain, it'll get pressure tested anyway.
Now onto something else. My buddy Jeff at work, a fellow car guy with an impressive yet unfinished 87 Camaro, is lending me his garage to deal with the oil pan & turbo oil drain leaks. As soon as I popped the hood, his jaw dropped. As I am writing this now, the car is sitting in his garage, awaiting some wrench time today & tomorrow to get it solved once and for all. As soon as we get the pan & drains sorted out, it's back to work on his daily driver camry's engine swap (hole in block). Fortunately, mine will be easy to pull thanks to everything being labeled, intelligently organized and accessible by normal sized human hands. Yay for smart!
In the meantime, I've been biking to & from my job, 5 miles one way. It's exactly what I needed, same with where I work too. As I'm fond of saying, "a voice of sanity in an insane world". Plus, I'm making enough to take care of everything I need to do, have plenty leftover for all the fun stuff and I actually look forward to going to work now. It's people that make all the difference. So I just want to give a big thank you to Jeff, Bridgette, Justin, Josh, Alex, Stacey, Darryl, Mike and Dante at work for being so cool to hang out with everyday and seeing the good in people. The world needs more people like y'all.
With all of this said, and once the oil issues are resolved, the car will be finished, minus A/C, by Christmas. Next paycheck will see more awesomeness, so stay tuned.
11-11-13, 10:35 AM
#400
After a few weekends, the car is being reassembled now. One thing I need to mention is:
NEVER ATTEMPT TO MOUNT A 20B WITHOUT THE PROPER ENGINE STAND ADAPTER.
My engine went on the stand and flipped over to re-seal the oil pan without much fuss, but turning it back right-side up was a different story. The top bolt hole on the thick middle iron broke off, then one of the bolts securing the "arm" to the Engine Stand broke in half, dropping the engine on the ground. The engine landed on the Turbo Control Actuator and passenger side engine mount, bending the TCA rod and bracket a bit. Miraculously, there was no other damage that I am aware of and no one was in the line of fire either.
I am not sure if my thick middle iron had a weak casting in this location, it was another kind of unforeseeable failure or a combination of the two, but the stands's "arms" were secure and the bolts used on the iron were threaded in fully. Strangely enough, one the SAE bolts between the "arms" and engine stand head failed by breaking while both of the metric bolts used for the iron to "arm" connection survived. Bottom one was bent about 25 degrees and is obviously not reusable in the bellhousing. Either way, bolts are cheap enough to replace.
Based on this experience, I don't think that it is safe to have a 20B on an engine stand without the correct stand adapter, especially if it is a fully dressed 20B-REW. This is good reason enough to invest in a stand adapter from Pineapple Racing as it spreads the load over four bolt holes instead of just two. The bolt hole I mentioned is sheared off nearly flush, but perhaps a talented TIG welder could fix it when it's time to rebuild the engine. Will get pictures on the next update. I was VERY lucky on this failure and it could have turned out MUCH worse. Chances are that if you are reading this, you won't be as fortunate.
After the engine stand crisis, I found that the front turbo oil return tube was only on finger tight at the Front Cover. Just to make certain, I replaced its gasket along with the other two oil return gaskets on the shortblock and gave them a thin coat of Ultra Black RTV on both sides for redundancy.
With the turbos back on, my buddy Jeff & I tackled the Turbo Control Actuator. I'm surprised to say that the rod straightened up acceptably using only a vise and the bracket only needed a little torch & hammer work to go back on like stock.
After several late nights, engine & transmission are back in the car now with the correct Orange 3.909 speedometer gear. Later today, the plan is to finish up the remaining parts (driveshaft, exhaust, wiring, vacuum lines, etc) and get the car back to drivable by the end of the week. Big thanks to Jeff & Bridgette for lending a hand last night when dropping the engine back in and mounting the transmission. Y'all kick ***!
NEVER ATTEMPT TO MOUNT A 20B WITHOUT THE PROPER ENGINE STAND ADAPTER.
My engine went on the stand and flipped over to re-seal the oil pan without much fuss, but turning it back right-side up was a different story. The top bolt hole on the thick middle iron broke off, then one of the bolts securing the "arm" to the Engine Stand broke in half, dropping the engine on the ground. The engine landed on the Turbo Control Actuator and passenger side engine mount, bending the TCA rod and bracket a bit. Miraculously, there was no other damage that I am aware of and no one was in the line of fire either.
I am not sure if my thick middle iron had a weak casting in this location, it was another kind of unforeseeable failure or a combination of the two, but the stands's "arms" were secure and the bolts used on the iron were threaded in fully. Strangely enough, one the SAE bolts between the "arms" and engine stand head failed by breaking while both of the metric bolts used for the iron to "arm" connection survived. Bottom one was bent about 25 degrees and is obviously not reusable in the bellhousing. Either way, bolts are cheap enough to replace.
Based on this experience, I don't think that it is safe to have a 20B on an engine stand without the correct stand adapter, especially if it is a fully dressed 20B-REW. This is good reason enough to invest in a stand adapter from Pineapple Racing as it spreads the load over four bolt holes instead of just two. The bolt hole I mentioned is sheared off nearly flush, but perhaps a talented TIG welder could fix it when it's time to rebuild the engine. Will get pictures on the next update. I was VERY lucky on this failure and it could have turned out MUCH worse. Chances are that if you are reading this, you won't be as fortunate.
After the engine stand crisis, I found that the front turbo oil return tube was only on finger tight at the Front Cover. Just to make certain, I replaced its gasket along with the other two oil return gaskets on the shortblock and gave them a thin coat of Ultra Black RTV on both sides for redundancy.
With the turbos back on, my buddy Jeff & I tackled the Turbo Control Actuator. I'm surprised to say that the rod straightened up acceptably using only a vise and the bracket only needed a little torch & hammer work to go back on like stock.
After several late nights, engine & transmission are back in the car now with the correct Orange 3.909 speedometer gear. Later today, the plan is to finish up the remaining parts (driveshaft, exhaust, wiring, vacuum lines, etc) and get the car back to drivable by the end of the week. Big thanks to Jeff & Bridgette for lending a hand last night when dropping the engine back in and mounting the transmission. Y'all kick ***!
