honda_luvr_2000

I have a 97 Civic DX hatch. Engine harness & ECU converted over to OBD1. I did this because I had a 93 Si that was well rotted out, and picked up the hatch for cheap because it needed a motor. I had the D16Z6 engine, trans, engine harness, ECU and converted wiring from the 93 in the 97 hatch with no running problems at all for a year. I drove the car two weeks prior to putting in the CR-VTEC build with still no problems.

Now, with the CR-VTEC all in, it has a poor idle (kind of rough with popping from the exhaust), and after it begins to warm up it will frequently stall out. After driving for a little bit the check engine light (CEL) comes on and sets DTC 43-2 fuel supply system too lean. I have had several people check it out, and give some things to check, but so far no luck.

Here's the nuts&bolts...
First the build:
-OBD0 B16A1 cylinder head and intake manifold
-valve clearance (lash) is in spec
-98 Accord 4cylinder throttle body
-98 CR-V stock block and injectors
-CR-VTEC conversion kit
-OBD1 computer (P28 D16Z6) and engine harness
-OBD1 DOHC distributor

What's going on:
It has a rough idle, and once it's begun to warm up (130ish & up degrees) it will stall frequently at idle. When driving it seems to run fine, but after so long will set DTC 43-2 Fuel Supply System Too Lean. Watching the DATA list with a tester, the O2 sensor will drop to 0.00-0.04V and just stay there not long after starting the car, maybe within 45 seconds. When it stalls it's like someone just turned the key off, very abrupt with no sputtering or choking before, but it will start right back up (not like a bad ignition coil, ignitor, or main relay issue)

What I've checked/done:
-checked fuel pressure- 41 PSI with vacuum hose off fuel pressure regulator
-verified fuel pressure was not dropping off and still steady when dies
-tried fuel enrichment with propane torch in intake tube- no change to idle or O2 reading.
-pinched fuel return line to increase fuel pressure, up to 60 PSI and no change in running or the O2 reading.
-verified ignition timing is right
-verified cam/crank timing is right
-spark is ok
-checked spark plugs- ok
-checked compression- 165 PSI with no large varience between cylinders
-put in another O2 sensor- same result
-put original O2 in another car- verified it is reading correctly
-put original O2 back in
-checked for short to grounds on O2 sensor wires- none
-checked for high resistance or opens in O2 sensor wires- none
-swapped another known good ECU (P06) in- still same result
-checked MAP voltage while idling at it's best near 800 RPM- fluctuates from 0.86-0.92V
-checked for vacuum leaks by spraying brake wash all around intake manifold- didn't find any
-found that if I pull the MAP sensor connector off and start the car it will run rough for 1-2 seconds, throw the code for the MAP sensor, and then go into limp mode- idles smooth at 1200 RPM & O2 reads 0.96-0.99V, but when driving it bogs real bad from a stop (too rich)
-cleared DTC & swapped known good MAP sensor on- still same result (poor idle & lean)
-checked injector resistance- 10.9 Ohms same as D16Z6 injectors (spec 10-13 Ohms)
-checked injector flow rates & flow rates of D16Z6 injectors- same (240)
-I know that the injectors are okay (I had the B20 engine running before I tore it down)
-checked resistance of all 3 distributor sensors (CKP, TDC, CYL) all in spec 350-355 Ohms (spec 350-700 Ohms)
-adjusting idle screw seems to have no real effect on engine's idle speed (even fully closed it still runs)
-pulled plug off IAC valve and stalled
-tried setting base idle by the service manual, no help
-swapped on 98 CR-V throttle body & P30 intake manifold, no change
-have changed MAP, IAT, TPS, IAC all no change
-watching ECT sensor it rises normal (doesn't jump all over or skip ranges)
-both IAT & ECT sensors read within 2 degrees of each other at cold start-up
-noticed when checking for vacuum leaks by spraying brake wash (starting fluid) around intake I got some mist into the throttle body & the idle smoothed out and O2 reading came up to near 0.9V, then idle went back to "normal" (poor) and O2 reading back to 0.00V
-noticed IAC command is at 400ish mA while running, didn't have another OBD1 car to compare it to
-noticed injector duration is at 2.60ish while running and when is dies
-VTEC solenoid is off while running, and when there isn't a DTC set it does activate VTEC at correct RPM
-had a friend hook up a P30 with Hondata S300, and tried an LS base map & GSR base map, ran no different
-tried messing with the fuel maps & turning off sensors, but no real help.

As far as I can tell, something seems to be telling the computer to cut fuel even though the O2 is indicating it needs to add fuel.
I have not had the chance to go through the wiring testing all the sensors & grounds for excess resistance, shorts to power, shorts to ground, shorts to each other, or voltage drops. That's my next step. I've been going crazy over this. Any ideas or directions to go would be much appreciated. Thank you for your time.

-Mike

honda_luvr_2000

any ideas anyone?

honda_luvr_2000

A little update:

I had the chance to get some readings off of a 94 Civic EX for comparison...
-once warmed up (190F) idled at about 750 RPM
-fuel injector duration was at 2.15-2.25ms at idle
-IAC command was 305mA

So those two readings are different than mine. I realize the displacement is different, but the ECU is the same and even the LS & GSR fuel maps didn't help. I'll have to check mine again later at the same temp to see where it tries to get.

I'm really hoping someone will be able to think of something I've missed. I don't want to be the only one posting in my own thread, but so far, here's another.

honda_luvr_2000

Seriously... anybody got an idea??? I'm stumped.

$hawNPeeZyy

Try using an ecu that is from another "cr-vtec" motor that runs correctly. Remember the b20 block/vtec head is not an oem engine and therefore requires an ecu that is chipped and tuned accordingly to the engine and its modifications.

honda_luvr_2000

Thank you for at least giving some input & an idea... However I got it fixed Saturday night, looks like at about the same time you put in that post.

SO for anyone who cares to know:
I put the Z6 injectors in, presto, runs great. I swear the B20 ran with the injectors I had in originally. So I don't know if there is a difference in the ECU operation of OBD1 & OBD2 injectors or if there was just some thing wrong with them. I'm going to look further into that. I'll probably put them in my 00 Civic EX and see what happens. If it goes lean too then I know they are messed up, but if not then it must have been a control issue. Either way, it's running beautifully and drives like a dream. Had it at an autocross yesterday and had a blast.

$hawNPeeZyy

Good to know that you finally got everything sorted out. Im working on a b20/v project myself so let me know if you have anymore problems and I would be happy to provide any input

mouab18c1

iTrader: (5)

Running obd2 injectors on an obd1 ecu will not effect anything as long as they are stock injectors. All stock injectors are 240cc. You need a wideband to indicate if you are really running lean or rich.

honda_luvr_2000

Not true on both accounts.
As for injector flow rates... https://honda-tech.com/forums/uk-europe-regional-129/honda-acura-oem-fuel-injector-flow-rates-2628134/ and http://www.massivespeedforums.com/20...or-flow-rates/ When you say "all" or "none" or "always" or "never" you're usually not correct.
As for seeing lean or rich... watching the reading of the O2 sensor with a scan tool on a line graph will also indicate lean or rich (that's how I did, and that's without a wideband). When my O2 voltage reading went to 0.00-0.04 volts and stayed there, that's really running lean. Aside from setting DTC 43-2 (43) "fuel supply system too lean" telling me that it was running lean.

mouab18c1

iTrader: (5)

A wideband is more correct since it is displaying it in your face. Injector flow rates my be different but if your running a stock b20v, then getting stock 240cc injectors would be your best route. And of course a tune. A frankenstein motor with no tune is the same as a time ticking bomb.

mouab18c1

iTrader: (5)

And I meant all stock b series Injectors are 240cc

honda_luvr_2000

Oh okay.

As for Wideband, yes it is a wider range of reading that it puts out for finer tuning, but as far as reading steady lean/rich, the factory O2 is able to do that very well. But yes, it is very helpful to have a wideband for tuning purposes.

Yes I agree that not tuning a frank motor or any heavily modified engine is not a good idea at all.

B20vtec_eg6

another thing to consider running a p28
the fuel map isnt meant to run for the b16, or the b20

i had my old ls running a b16 map, and it ran like crap, bogging out and what not cause it was dumping too much fuel
and it might be running pretty lean cause im sure the b20vtec would run with more fuel than the d16z6

i had a friend who ran a stock p28 on a stock gsr, he ended up blowin his motor by washing out his cylinders, it ran too lean, which heats up the cylinders, and causes the pistons to expand, then your walls are toast

try getting your ecu chipped, and at the bare minimum find someones dyno'd base map for a b20vtec
its better than nothing, or a p28 basemap