53 replies to this topic

[Ocabrera1966]

Well, the mystery deepens!

 

I went to the shop and saw the work being done first hand. The sock filter looked ok, and when air was blown through the hose backwards from where it connects to the fuel pump it flowed through. There was a little tiny (1" x 3/8") plastic coffee strainer looking thing inside the hose that looked clearly past its prime. That was removed, and the fuel strainer will be bypassed while an external FRAM filter will be placed before the pump. There is some debris in the tank, but nothing major. That will be flushed out and the tank cleaned.

 

Not sure if any of those things is the cause of the problem as the car started up fine for the shop after the tow truck dropped the car off. They drove it a bit and noticed the same hesitation at 3k that I get, but the car did not shut off. Maybe having the car towed caused the fuel to slosh around enough to loosen whatever is impeding the flow??? 

 

I don't know. This little car is kicking my butt at every turn.

[Ocabrera1966]

Got the car back today. Fuel tank has been cleaned out, filter is new, pump is new, strainer has been bypassed and an external filter fitted in-line before the pump. So far, it hasn't stalled for lack of fuel. But, it bogs down big time at about 3000 RPM, until you push the pedal down and get past it. Idle is all over the place too. 

 

Just refuses to run right. 

 

I was thinking of going the devac route, but I am very close to giving up having spent a ton of $$$ throwing parts at the thing. 

[greasemonkeyreborne 5x1g's]

thnx for the follow up on the fuel tank and strainers. i bought an 87si complete in 2012 w 129k. i couldn’t run it due to damage. i finally brought it home last week. about two gallons left in tank. i pulled the harness off the main relay and jumped the fuel pump. pumped the fuel out by disconnecting the return hose near the intake manifold. noticed really good return flow u til it got low on gas. then flow greatly reduced. i think that was point it got crap sucked in lscreens.

honda says 7.8 oz should come out in ten seconds. i got 4. assumption is clogged screens, or fuel pump mechanism failing???

car fired right up w fresh fuel and other maintenance i did to the fuel rail and injectors.

cars not drivable due to damage, but i’m sure it’s staved for fuel at higher rpms

i’m going to do what u all recommended and and drop the tank and inspect for trash. pull screen in tank , etc...

i really want to clear the lines and run the fuel via a bucket and see if the return flow gets to where it needs to be. if not, then it will prove to me the pump is trashed. currently, the pump sounds great. nice quiet hummm

good luck w the 3k thing. keep us posted

d

[Ocabrera1966]

Well, I took it out tonight after dark (no AC, and this is Florida). Idle bounced up and down a few times and the the PGMFI light came on (code 1). Afterwards, idle was relatively smooth, and there was no bog at 3k rpm. In fact, it was actually fun to drive.

Car has a new O2 sensor, doesn’t mean it’s not faulty or the wire back to the ecu isn’t bad. So what other than fuel to the injectors does the ecu control that is out of the circuit when it goes into closed loop mode?

[Ocabrera1966]

Did some research and I think I answered my question regarding what the ecu is using to control the fuel mix. Based on what I’ve read, could be one of the following:

1. O2 sensor. Mine is new, but it’s been welded into the collector pipe just below the exhaust manifold. This shouldn’t affect the idle, but replacing it with a four wire (heated) sensor, is an easy and inexpensive test/fix.
2. The temp sensor under the distributor. I have a new one sitting in a box, and that’s an easy fix as well.
3. The MAP sensor. Expensive, but easy to replace.

I can’t think of anything else the ECU can bypass, and since the engine runs fine in closed loop mode, I think it’s likely to be one of these things.

Thoughts?

[greasemonkeyreborne 5x1g's]

does it stumble at 3k parked, in neutral, and just pressing gas pedal? if so. this could aide in troubleshooting in the driveway.

consider retesting ohms on tps several times. maybe random dead spot like someone mentioned. but it happens randomly.



how many miles on car. are wire harnesses in original configuration. wondering if p.o. did some rigging

i once test drove a crx si ten years ago w a bad stumble. i was fumbling w the wires under the hood and happened to grab the dizzy pick up wires ( connector on top of dizzy). and got the stumble to repeat. it was a damaged hall effect pick up outta the dizzy.

i know that’s reaching here. but try wiggling wires while idling. maybe 3k hits a particular harmonic frequency via vibration. wiggling might reproduce

i’ve been blessed w smooth running si’s for the past 25 years. sorry i can’t be of any help. i think you might be on to something the rest of us may experience as all of our cars age

[Ocabrera1966]

In the driveway, I can rev past 3k even with the initial high, then surging idle. Seems to be an issue only under load.I'm going to test the MAP sensor tonight and replace the temp sensor (although that looks a bit difficult to get to). The thing runs beautifully in closed loop mode. If I can't figure out what the cause of the problem is, I would prefer it stay in closed loop mode. Is there any way to force it into staying in closed loop mode? 

[greasemonkeyreborne 5x1g's]

i’m sorry but this is beyond what i’ve ever had to get into. it’s sounding more like a control situation and not a loose wire. i’d guess ur on the best path at the moment. i assume the three items list3d in prior posts r the likely cuprits in running a smooth open loop?

another goofy situation i heard someone have was a loose alt belt causing volt loss when punching throttle.

my advice isn’t going to help here...

[Ocabrera1966]

MAP sensor tests fine. Temp sensor has been replaced. I tested the connections back to the ecu, and those are fine too. Sadly, the problem persists.

Although the O2 sensor is new, it’s position has been relocated further down the exhaust path than the original. So it’s possible it’s not reaching the right temperature. A four-wire O2 sensor might help here as they are have a heater circuit provision.

A new fuel pressure regulator is on its way. It’s a cheap part and easy to replace. The car’s idle gets out of whack with the fuel pressure regulator disconnected (while plugging the manifold port, of course). My understand of this is that disconnecting the FPR from vacuum will result in max pressure. Seems odd that the idle would become erratic with it at full pressure.

Not sure if I sure go the DEVAC route and eliminate as many sensors, dashpots, and valves as possible. I’d be happy to run it in closed loop all the time and disconnect the PGMFI light in the gauge cluster.

[1985 CRX SI]

I would try another set of injectors. Maybe one injector is dumping more fuel that the others and causing the O2 reading to be off and the ECU is cutting fuel to all to the injectors correct the imbalance. Injectors work best if they are a balanced set.

[Ocabrera1966]

I had a leaking injector when I first got the car, I bought an entire set of flow tested and balanced injectors. All the o rings are new as well.

Slowly narrowing it down by process of elimination. Throttle body is coming off to be cleaned and I’ll check the IACV while it’s off. There’s an IACV fix posted elsewhere that I want to try since the initial symptom on startup is the high/then surging idle.

On the plus side, I drove it early this morning and got a ton of compliments from little kids to older folks. Even a couple of dudes, one in a slammed 2000’s civic Si and the other in a WRX. That was at a gas station, so the second the I turned the car back on, the issues resumed until the PGMFI came on.

[greasemonkeyreborne 5x1g's]

glad ur getting the “love” on the car. they’re so old now and u never see them driving around

u posted on an older thread. if i understand u correct, once the engine light comes on, the car smoothes out at 3k? if so, then computer is putting ur setting at default. i hate to throw this in the mix as another part, put can u get a hold of another ecu? i’ve heard some people comment goofey things happen and a new ecu fixed it

i also saw in ur garage this car has 325k on it? is engine/harnesses/sensors original from what u can tell?

[Ocabrera1966]

It says something when kids pull their faces out of their phones and tablets, roll down windows and point at the car. Especially here where I live, where supercars and big dollar luxury cars are not uncommon. I had a red 1986 Si purchased new in 1986. It had white Enkei wheel, and the same "Mugen Power" decal on the back, so the car is pretty special to me, if most on the road have never seen one.

 

Engine/harnesses/sensors seem to be original, except for those I've replaced. I have a spare throttle body that's been cleaned and has a known working TPS. If the TB i the car now is a full of grunge as the one I just cleaned, that could possibly be the cause of the problem. It is missing the IACV, so the old one will need to be transplanted from the TB in the car now. At the very least, it is the cause of the high idle at startup.

 

I have a strange feeling the issue lies with the TB/IACV/TPS. Code 1 on the ECU is O2 or fuel problem. The 2k idle on startup points to something in the TB/IACV/TPS throwing the ECU out of whack. Really, the ECU only uses TP, MAP, and O2, so if ELECTRICALLY the sensors are working, it's possible something mechanical and/or vacuum related is interfering. 

 

Thanks for all the input guys. I'll keep you posted.

[Ocabrera1966]

IACV is out of the car. Looks like this is the cause of the high 2k idle, as the black cover may have screwed itself out over the years. That, or the poppet wasnt rising high enough to close off the flow of air as the thermal wax heats up and pushes the poppet up. Looks like the poppet is fully closed when there is a 3mm gap from the top of the poppet to the top of the black disk that covers it. I measured this with a depth gauge, and tested it by heating the IACV in a pot of boiling water to get the poppet to close. I can adjust it when its back on the car, but I have to wait until I get replacement o-rings which should be here by Friday.

However, I dont think thats going to resolve the bog at 3000 RPM and the code 1 on the ecu. Something producing an electrical signal is causing the ecu to go into closed loop mode. One solution may be to do a devac. This would eliminate the numerous sensors inside the black box on the drivers side of the car. Chances are those sensors are no longer being made and finding used replacements would be a crapshoot at best. But, eliminating all those sensors would leave TPS, MAP, TEMP, and crank angle (not sure if Im leaving anything out), Ive replaced and/tested the O2, TPS, TEMP, and MAP sensors, so not much left.

Oh, a replacement ECU is also on its way.

I would gladly pay for someone to diagnose the thing and say oh, its this ______, and it will cost $$$ to fix it. Otherwise, its trial and error. By the way, arent ANY Honda mechanics still alive out there who worked on these cars back in the day???

Edited by Ocabrera1966, 23 October 2018 - 10:39 PM.

[greasemonkeyreborne 5x1g's]

you’ll get plenty of free advice here. my feeling is it’s an age/high mileage gremlin. helping t solve this just helps the rest of us in the future.

several years ago an rpm facebook began. apparently most of the members r there? someone mentioned it’s quiet there as well.


IACV....

two of the 4 orings r available from honda. the top lid and the two mounting to the air intake.

there’s another on the idle set screw. that could only be bought as a set. it’s been out of stock for years. the other is the oring around the black plastic screw in base u removed to get to the plunger. that was never available

i’m sure the the other orings can be found. i was a plastics engineer. designed seals. coolant seals will be epdm, fuel seals will be buna-n (old version), modern, better choice is viton


i always pull the iacv off any engine i find in the yards for spares. i pulled one off this 220k crx last week. ive always wanted to match those orings. if the ones on ur order don’t work, then we can work together on a group project to get rings