EJ22 power loss and misfire troubleshooting

[BigPapaV]

Hello, first post here after reading a lot.  Please forgive the length, I feel a little history is important for troubleshooting, and it may help document my path for the next person. 

I've been working on resolving two related issues on my 1986 Vanagon swapped with 1992 Legacy EJ22.

Initial situation: Van failed smog at 15 mph, but just barely passing at 25 mph.  Engine would experience mild misfire at idle (500 rpm).  Engine would lose power, as in cut out and sometimes die during two general scenarios: A. while accelerating from stop and during great changes over approx 2000 rpm; B. at constant speed on flat ground randomly, bumps in the road didn't matter (tested on freeway). 

Initial set-up: 
MAF - AUTECS (22680AA160F, original to engine?) 
O2 sensor - AUTECS(?), original to engine? 
ECU - AUTECS (22611AA931F from 1992 Legacy), not sure if original to engine 
Fuel injectors - gray tops
Fuel rails - black stickers, Left 17523AA050, Right 17522AA090

Idle hang - I've read this is common. 
Engine idles at 500 rpm affording to tacho. On start-up idle shoots up to 1500 rpm and hangs for a couple minutes.  Starting to drive it'll still hang for another 5 minutes or so until I've been able to get past the stop signs and onto a side road going 30 mph or so.

Research pointed to many possible problems that led to troubleshooting and some fixes:

1. Vacuum leaks? - all intake manifold small OD hoses renewed (sump vent, IACV inlet, etc). PCV valve replaced. 

2. Spark plugs bad? - Replaced recently with NGK set, but pulled all plugs again, all plugs looked almost like brand new, very little discoloration. Thinking lean cylinder condition? 

3. O2 sensor bad? - removed and checked voltage w/ propane torch, reacted like Haynes said it should. 

4. Ignition coil pack bad? - couldn't open the connector to check per Haynes, and I didn't want to destroy it since the pigtails seem short for repairing with a new connector type. Tabled this because it doesn't sound like a cylinder is ever cutting out. 

Test Drive 1 - short drive around the city 

Test Drive 1 Conclusion - no change. 

5. MAF connector bad? - checked if bad connector first, engine would die when red wire jiggled, so connector was depinned, no more dying from jiggling.  Misfire at idle still present. 

6. MAF bad? - haven't cleaned yet, but wary of stories that they get ruined from cleaner sprays.  Checked air filter, mildly dusty on the outside. Bought spare Denso MAF and Denso O2 Sensor to replace as set in case I need to go there.  Tried "unplugging" MAF by removing red wire (connector depinned) while engine running and engine died.  Not sure if this is any real indication of MAF being good or bad? 

7. Oil level wrong? - changed oil + oil filter, and good thing I did because the sump was only about half full 😅

Test Drive 2 - drove over an hour on streets including flat ground and hilly.  Misfire at idle still present, power loss events got worse and worse toward end of drive.  Near end of drive found positive battery cable loose.  Tightened up, power loss event less frequent but still happening over 2000 rpm. Arrived home.  

I'm pretty sure the battery cable loosened up during the drive. 

Test Drive 2 Conclusion - Overall no change, except less ticking from valve lifters. 

I then figured on focusing on possible Fuel System issues. 

8. Water in fuel? - didn't do any checks / changes since I figured probability was low, fuel tank was ~1/2 full, maybe 4 months old

9. Fuel filter bad? - replaced fuel filter, looked original to van 

10. Dirty injectors? - Found out EJ22 ECU needs to be matched to injectors, and that the red tops perform better in general.  Also found out that it seems gray top injectors weren't installed on 1992 Legacies, could be wrong here.  

Bought a set of red top injectors (for 1992 EJ22 from eBay). Sent both sets of gray and red top injectors for cleaning before installing the red tops. Test report indicated that both sets had very poor spray pattern and flow before cleaning.  Reinstalled red top injectors. Also cleaned the two ground points under the air filter housing. 

Test Drive 3 - Engine started up to 1500 rpm as before, came down to idle at 500 rpm fairly quickly (improvement).  Test drive around city to warm up, then uphill to load the engine.  

Engine didn't lose power at acceleration from stop anymore, unless revved over 3500 rpm like for gear shift.  Engine also didn't bog down and lose power suddenly going uphill.  

Test Drive 3 Conclusion - Much better, all in all very satisfied with the improvement, but not 100% yet. 

11. Unrelated, coolant temp gauge not working - coolant gauge on the dash still didn't work after refurbing the cluster earlier - I found the wire in the engine compartment and pushed it onto the "sender" nub. 

#11 Result - the gauge works!  But now the engine misfires constantly at idle, worse than before, so much that it will die within 20 seconds of starting unless I increase throttle. WTF? LOL. Hmmm maybe voltage leak-off to ground somewhere, and this is when I decided to post this.   See video:

Moving forward, I think the best bet is to check ground / voltage drops on that coolant temp wire, and the harness bundle it goes into.  And also check all the ground connections to the engine. 

Question 1 - has anybody had similar experiences to above that had worked through it and came to a solution for their set-up? 

Question 2 - anybody have a simple diagram or description of all the engine and electronics grounds, or ideas for upgrading the grounding of the engine?  I seem to recall some saying there could be between 4 to 6 ground points in the engine bay(?)

Question 3 - anybody experienced misfiring due to a bad ground or other wiring issue similar to above? 

Question 4 - anybody know if there's a ground wire in this bundle from which the coolant temp sender wire pops out of? 

Many thanks for taking the time to read! 

 

-V