legendre

Howdy folks, first post. I'll get right to the reason I had to finally reg for this forum..

Wife's car, 1998 CR-V AWD, 150K, very clean shape overall and well maintained, 1.5yr on valve job / timing belts / water pump.

All seemed fine, and then it was about three days from the first report of trouble to total loss of driveability. Starts fine when cold, idles fine. As it progressively warms, ability to open the throttle falls off. By the time it's fully warm (15 min idle), 1/4 throttle will stumble the motor, possibly to a stall. 1/2 throttle will stall it outright. If forcibly stalled in this way, the engine can be difficult to restart - which becomes easier, as it cools down.

If I try to drive it when warm, it's sorta-kinda possible if I don't mind wrecking the clutch (feathering it heavily with the motor at high RPM), but otherwise it's essentially useless.. yes, it's the worst "hesitation" ever. More like a gigantic hole in the gas pedal.

I have: checked compression, valve lash & timing, replaced fuel filter and spark plugs, measured fuel pressure, tested ignition timing & quality, tested the TPS output voltage, and a few other things I'm probably forgetting.

Crazily enough, the CEL is NOT on..! (Or it wasn't, until I pulled the conn. for the TPS - which stalled the engine and set a code which I've yet to reset).

How much fun is that? So where should I go with this next?

(BTW - I'm a retired motorcycle tech, mostly familiar with Brit, German and Italian machines of the 1950s-1980s vintage, but have worked on a lot of 'Big Four' Japanese bikes. If you need a word of advice, let me know and I'll meet you down in the bikes section.. )

Jjbaehr

I know you said you checked the valve clearance, are you sure it's right? This sounds like tight exhaust valves to me. These crvs are plagued with this problem. I adjust them on the loose side, I'd rather a little noise over a drivability issue any day. Also honda has a recall on the ignition switch for a stalling concern with this particular crv. Sounds unrelated but wouldn't hurt to have them check to see if it has been done before. I'm pretty sure they are covered for life.

Jjbaehr

It could also be a partially clogged cat converter, disconnect the exhaust before the cat and see if she runs any better

legendre

Yes. Checked, double-checked and triple-checked.

For what it's worth, the valve arrangement on this engine is very similar to many motorcycles. I am completely at-home and comfortable, both measuring and adjusting this type of setup; I've literally done hundreds of these.

Yes they are. Note that this one had a valve job about 1.5 yrs ago.. don't ask, it's one of the most irritating things I've ever had to deal with on a "modern" engine. Idiot previous owner let the valves go for 130K.. and we got it just as the valves were starting to go out. I had no idea that an engine of this vintage would have mechanical valve lifters.. so I got to pull the head and replace valves.

That I haven't heard. How does the ign. switch cause this problem? Does it trouble the ignition, or the fuel, computer or what??

legendre

Plenty of flow (blows your hand right off) can be felt at the pipe, and it doesn't act up when it's cold - just as it warms.

Suppose it's not outside the realm of possibility.

Oh, I also forgot to mention.. I suspected bad / watery fuel, so I added two bottles of yellow HEET to the tank (it was at 1/4 full), with no result. Then I topped off the tank with fresh 88 octane - again, no change at all.

Fuel quality seems OK. ETA: Or at least it's not throwing any red flags.

Jjbaehr

Watery fuel is a good possibility, although I would expect a misfire CEL by now. I have seen distributors do some crazy things in the past, the sensors inside are rather hard to troubleshoot, I don't like to guess when diagnosing vehicles but honda distributors are sort of an exception to that rule, it wouldn't hurt to swap a know good one just to see if it's the culprit. As far as the ignition switch goes the contacts wear out and it literally just turns the car off randomly. Easy way to see if it's bad is to just jiggle the keys in the ignition while it's running and see if it turns off. Again it's doubtful but it wouldn't hurt to have honda check to see if it's been done and just make an appointment to have it fixed. So you don't have future stalling problems.

Jjbaehr

And you measured fuel pressure when it's acting up correct? Not at cold start?

PhilStubbs

Sounds like a fuel pump on its way out. Pinch the return line after the regulator with some vise grips and see if it gets better. I know that sounds like troubleshooting the regulator, but Honda regulators almost never fail. I would have said never, but then I know one if you would come in here and be like "my sister's boyfriend's cousin's best friend's dad worked with a guy who's regulator went bad".

legendre

I didn't have the right tool for it, so a friend made an adapter fitting for me that replaces the banjo bolt on the fuel filter and includes a 3/8" NPT female thread for attaching a pressure gauge - that seems to be how the 'pro' kits work.

Pressure is solid at 30 psi, hot or cold, and it jumps immediately to 45 psi when you open the throttle and the regulator pulls in. This also kind-of rules out the fuel pump, because if the pump were going bad, I'd not see consistent fuel pressure readings - if the pump flow drops off, so does the pressure - right?

I mean seriously, the pressure is rock-solid as I'm holding the throt half-open and the motor is simply heading for a stall.. the reading doesn't even flicker. Ditto for the ignition signal.. you can watch any plug you like (via a timing light), and it will hit bright and on-time right down to the moment the engine stops turning.

You folks sincerely think this is still fuel pump related?? Heck, I'd love to see a pump fix it, but I have a funny feeling that I'd be changing out a perfectly decent pump.

PhilStubbs

I mentioned pump not knowing you checked the pressure. That does rule out the pump.

Have you tried a fresh set of plugs?

The only other ideas I have would require using a scanner to watch the live feed from the ecu. If you had a cel that would help a lot. At this point without getting my own hands on the car anything I suggested would be just a guess.

For the hell of it, try what I mentioned and see what it does. I have seen good fuel pressure and the pump still be the problem even though it makes absolutely no sense.

legendre

FYI, I +also+ think this is ultimately fuel-related - but the issue is on the delivery side, rather than the supply side.

That is, I suspect that the injectors are getting fed a crap signal.. for one reason or another. Any thoughts on that one?

legendre

Plugs are fresh NGKs; you can see the complete laundry list in my first post.

Yes, I like that idea a lot. How is it done? Can I use a general-purpose oscilloscope?

Jjbaehr

Injector pulse is Also controlled by pickups in the distributed.

Jjbaehr

Distributor*

PhilStubbs

I'm saying an obd2 scanner. You can see a live feed of all of the engine sensors.

legendre

Ah, OK.. nope, don't have one at my disposal.

There is one thing I still need to confirm, though. While I was able to get the correct low and high outputs (0.5V + 4.5V) from the TPS, I wasn't able to confirm that there were no drop-outs or noisy/jumpy points across the range.

Could be the TPS is dropping out (right in the range where it would be worn the most - low-mid range, where you cruise).

legendre

I wanted to check into this. So I'm looking at p. 4-15 of the service manual, which has the ignition wiring diagram.

In addition to receiving 1 input from the ECM (which is basically the 'FIRE COIL' signal), the dist. also produces 4 output signals. But I'm not entirely sure what they are used for (in the ECM) or how they actually supply the data:

TDC: TDC sense
CKP: Crankshaft position
CYP: Cylinder position
TACH: Tach signal.. comes off the coil, so one pulse per spark, I'd guess.

They seem kind of redundant (for instance, if you +know+ the crankshaft position via CKP, don't you also know when #1 is at TDC, or when #3 is at BDC or anything else??). Can anyone elaborate on how they are used in the control scheme?

Jjbaehr

I honestly have no clue, silly question but have you unplugged your o2 sensors and taken it for a ride? Maybe you have a lazy o2 sensor causing the fuel trim to be out of whack?!?! Just a thought. Throwing out whatever comes to mind

legendre

Just re-checked it with the help of another set of hands.

TPS voltage is perfect.. 0.5v closed, 4.5v at WOT and perfectly smooth up & down the range.. Just like new.

legendre

At this point, any idea is worth considering.

So yeah, I've read that unplugging the O2 (the one on the exh. manifold? is there another?) will cause the system to run open-loop.

Though I'm not really sure what that ultimately means, as in, which sensors does it ignore - and how does it fudge or guess which values to use in-place of data inputs?

So just pull the plug on the manifold O2 and give it a try?

And BIG thanks to you, and everyone else who is taking a shot at this stupid deal. I think it's apparent that I've kinda run out of options.. and it's not like this is my first rodeo, either.

If this thing just had a carb (heck, even four of them!) on it, it would be running last month. Frustrating..

(ETA: Is it OK to pull the wires to the injectors, and replace the TB with a side-draft Walboro =P)

Jjbaehr

Yea the one on the manifold, the second one is after the converter, that one just sees if the converter is doing its job properly, it's an easy test to rule it out, and yes it will stay in open loop causing it to run rich. I would do that, then take a fuel sample and look for water or excessive alcohol content, you can get an alcohol tester at the auto parts store,then I would Swap distributors. If you rule all those things out then I'm truly stumped.

legendre

Ok, that sounds like the same situation as on my Nissan (which is a V6, so it has two exh. manifolds and 4 sensors). First 2 are in the manifolds, and are IN the ECU loop. Second set of 2 are after the converters, OUT of ECU loop, and just check to make sure that everything else is in-order. The latter will set codes if there is trouble, but they play no role in engine management.

Did you read my comments about fuel, early in the thread? Guessing I had water in the fuel, I added 2 bottles of HEET (methanol, yellow) to 1/2 tank. Made no diff, so then I topped the tank with fresh fuel.. again, zero diff.

I do have what amounts to a spare dist.

Jjbaehr

I do remember reading that, two bottles of dry gas will not fix problems with gas like this. This was pumped from a 2012 crv I fixed at the dealership a week ago. Filling a glass bottle and actually examining it is what I meant.

PhilStubbs

open loop means the ecu ignores the o2 sensor and runs on its pre-programmed map. the o2 sensor constantly adjusts idle and part throttle for 14.7:1 afr and somewhere around 75%(dont remember the exact number) throttle the ecu ignores it and runs on the pre-programmed map. its possible that its your problem, but im not convinced. you can try unplugging it and see. it is in the exhaust before the catalytic converter. unplugging the one after the cat would do nothing. it only monitors emissions and has no effect on drivability.

the sensors in the distributor each have a function. tach signal is like you said, one pulse per cylinder fire. ckp tells the ecu which cylinder is in need of fuel(the ecu pulses the injectors one at a time like spark plugs). cyp tells the ecu where the pistons are to fire the plugs at the proper timing and the tdc sensor basically keeps all that in check by letting the ecu know each time the crank spins once to start the whole process over again.

have you tried pinching the return line yet? its free and takes 30 seconds to do. it will rule out the pump, then i can move on with my order of troubleshooting

Jjbaehr

This was literally 60/40 water to gas, I swear she filled her tank with a garden hose, she actually made it 3 miles before she experienced any issues! And yes it still ran, barely tho! Pumped straight from the full station