Boy your pretty lucky to have two techs come down to figure this out for you. I still have yet to figure out my customers car 100% and its on adaptronic. I been tossing aem and adaptronic around myself.
Hope you get it all figured out.

 

Will do, thanks!

 

Very thankful they are coming, primary reason I went with Adaptronic

Thanks!

 

have you tried shifting the ecu vac nipple source yet ???

 

I took off the uim yesterday and it appears that all the nipples on the back are full time vaccuum.

However, I will try shifting the source next time I run the engine

 

Small Update:

Set the correct Dead Time Correction for the fuel injectors in the ECU, replaced Wideband, and

Swapped around the primary injectors and coil packs between middle and rear rotors and confirmed visual spark from grounded plugs externally.

Car started alot easier, AFR gauge quicker to respond, and no more backfiring at idle. However, vaccuum was reading low and still felt like 1 rotor was down.

At this point, I'm suspecting this whole issue is a bad leading coil pack. Strong enough to pass the timing light and visual spark test but weak enough to fail when running.

Tonight I'm going to use a Spark Stress Tester with adjustable range to test the coil output from each coil

 

take some pics of where you draw your vac sources from

if the ECU map sensor line is lower than the injectors or teed with a faulty fuel reg then it may have some fluid in it affecting your readings

if it is hooked to a purge air source it will also read low as it it taking its source in front of the throttle body

at least two of the nipples on the phelonic spacer are purge air source
.. for the OMP spider, and for the purge / PCV valve switching from the crankcase ventilation and charcoal canister

 

There is not a spacer on a 20b, 13b-re or 13b-REW.

 

All four of the nipples on the rear of the UIM go directly the primary runner portion of the plenum, both vacuum and pressure.

 

Good clarification

I'm using only those nipples on the rear of the UIM. The brake booster line is going into a fitting that I tapped into the rear as well.

 

do you have a non return valve in that brake booster line ? is it facing the correct direction ? have you tried pinching it closed ?

 

Good question. What is under all the "aluminum foil", there is a 1 way valve in the factory hose at the booster. It does not effect idle or how the car runs, but you will pressurize the booster when you see boost, which makes for a fun time stopping.

Btw: tuner tip of the day. The S5 n/a throttle cable is only 2 ft long, makes for much cleaner install.

 

Checked cold compression again with full charge on battery, ~95-100psi all faces

Check valve is correct, foil has also been removed

I checked all of the coils last night with a "Spark Tester". Its essentially a spark plug lead that plugs into the wire, then has an adjustable gap inside a tube with two points and a ground wire. It forces the coil to max out.


Results:
All the coils fired strongly, except center trailing. Center Trailing would arc from external of the coil to the bracket, leaking horribly. This only occured with the spark tester installed.

I purchased a brand new coil and positioned it in the Center Leading, and moved the previous good leading to center trailing.

Next:

Started the car and checked timing with light on all wires. Each fired consistently except trailing center (original moved up from leading) and trailing rear. These were showing sporadic in/out lighting pattern on the timing light. Triple checked the light connection and plug wires, even turned it off and retested with spark test - good. Restart car and light misfires on center and rear trailings.

Consistent with the misfiring light is the misfire in the sound of the motor. I didn't think trailing coil misfire could cause rough idle or rev, but it seems to be the case. Car can hold somewhat stable idle but if I rev, it will backfire and sound rough.

I'm going to replace all original coils with brand new coils today. May be possible that some of these used coils fire ok at cranking/low rpm on tester, but not up to speed.

Apart from the misfire, car was still only pulling -12inch vaccuum at 1100rpm. Map sensor matches boost gauge, map sensor reads 60kpa (100-60 = 40kpa, 40kpa = 12inch/hg)

Easy to start cold, very difficult if not impossible to start when over 160*

 

To add to this, plugs were brand new, and I could not hear, see, or feel any vaccuum leaks anywhere around the LIM/UIM. Sprayed carb cleaner around every area I could think of that may leak.

 

On your difficult hard hot start issue. Are you also showing more than normal battery voltage drop? I have the same issue as you when cold the engine starts perfect. When she gets warm, the starter seem to struggle turning her over. It's much weaker. I was thiking of doing Rx8 starter upgrade, but I'm not gonna waste the money right now since I will be converting to the t56 in a couple months. Then I'll just have to use the 20b starter.

 

try things with all the trailing spark disconnected entirely

it is a long straw but it may be an indication that the + coil wiring cannot support the multiple coils hung off it
im expecting each coil is drawing at least 7 amps .. and 6 x 7 = 42 amps

 

you could also try moving the mixture trim down a ( lot ) so that you have some ( more ) realistic numbers for injector time

i am expecting 1.8 ms is going to run shitloads better than 4.? ms.. which would flood even with 550 injectors at idle

 

Guys,

Replaced all the coils with brand new duralast d581's. Lifetime warrenty in case we blow one.

Car started very quickly cold. The idle was perfectly smooth, sounded entirely different. Didn't have the timing light tonight, but visual external spark test showed blue/purple spark compared to orange.

Had to shut it down after a few mins as it was too late and loud.

Observations

-12-13inch vaccuum at 1600rpm
-13-14 at 2k rpm

If I drop down to idle around 1000rpm is sounds very fluffy and weak, vac drops down to -11

Revving from 1000rpm to 2000rpm is smoother but shows slow response and backfires. Over 2k is cleaner.

AFR is consistently around 12.5-13.0

I will check out the MS values in a little while. After setting the injector lag times, I was able to pull out a lot of fuel from previous settings.

Timing was 0*tdc at 1000-1250rpm, and 10* at 1500-2000rpm. Split is at 15*

I don't want to jump to conclusions, but I believe the engine has low intial compression and needs to break in from these RA seals on original housings.

The cranking speed is lightning fast, as I have a 0 gauge wire direct to starter. Its definitely enough speed. Hot starting has to be compression or tuning/other variables.

Tomorrow ill warm it up and let it idle for a good while.

 

I haven't caught up with the entire thread yet, but wow what a saga. It's nice that you're documenting all this.

I suspect the ignition coils got hurt when the Microtech was not configured to run them correctly. I've seen this before with GM LS-type coils, in extreme cases it's possible to produce visible smoke from an ignition coil if you send it the opposite polarity signal for long enough.

As you're starting to learn, starting the engine can be pretty easy when the air & fuel & compression & spark are close enough. For the hot start, I'd recommend first double-checking that the ECU is still getting the proper signal from the CAS; the VR / magnetic sensors inside those units can be affected by heat and cranking speed, so it's possible to have a signal that is strong enough to trigger the ECU when cold but not strong enough to trigger the ECU when warm.

If the ECU is seeing the CAS signals and trying to fire coils and injectors, tuning hot start is actually pretty easy compared to other aspects of standalone ECU calibration.
1. With your foot all the way off the throttle, decrease the cranking fuel all the way down to zero (in case the engine was flooded from the last attempt).
2. Crank the engine for a few seconds, and if it fires with zero fuel you know the engine was far too rich the last time it cranked or idled.
3. If the engine doesn't fire after 2-3 seconds, stop cranking it, increase the cranking fuel by a small amount... I'm not familiar with cranking fuel tuning with Adaptronic but try to watch the fuel pulse on the software, try not to increase by more than 0.5ms - 1.0ms per attempt.
4. It's unlikely the engine is going to start during the first few attempts, so repeat steps 2-3 adding cranking fuel in small increments until the engine starts. Be sure to let the starter rest between attempts, try not to crank the engine for more than 5-10 seconds at a time, and give it time to cool down between attempts. If the software shows fuel pressure, try to make sure the engine has consistent fuel pressure during each attempt.
5. Keep an eye on the fuel injector pulsewidths. Cranking fuel is probably not going to be more than 10x larger than idle pulsewidths. For instance with E85 and tiny primary injectors, my car's cold idle pulsewidths might be ~6-8ms, and cranking pulsewidths about 60-100ms. Your cranking fuel will likely be below 50ms with big injectors and gasoline, especially when warm. If you get to big pulsewidths and the engine doesn't start, something else is wrong... increase airflow by opening the idle solenoid, or giving it throttle and repeat... also double-check spark is present and correct with a timing light. You shouldn't need different ignition timing for cold start vs hot start.



Cold start is significantly more difficult, because the engine is only truly cold once or twice per day. Luckily it sounds like you've got that working OK.



Personally I would ditch the CAS for a crank-mounted trigger, either the FD setup or the FFE kit. Don't want to bore you with the details but there are so many downsides to the CAS it's not funny.

Good luck,
-s-

 

I hear elliot and andy fugured it out. Filll us in

 

It turns out that the 20b has two firing order versions
1-3-2 and 1-2-3. This one was programmed for 1-3-2, changing it to 1-2-3 fixed everything. Car sounds totally different and very fast throttle response

Thankyou everyone for all the help. Car will be dyno tuned in a few weeks.

 

^ nice. glad you figured it out.

 

LOL. How do you know which 20B fires in what order?

 

I was pretty certain you had a timing issue. Glad you got it sorted.

All the Cosmo 20B engines are 1-2-3. Maybe the 13G or some of the custom short crank engines are 1-3-2?

 

Thats exactly what i thought. The 13g's had 132 I thought. Maybe the aftermarkets as well.
Glad to see its up and running and was a simple fix