TitaniumCoder47

1998 Honda Civic EX Coupe

I've been wrestling with this symptom for the past few weeks, and it takes all of these variables for it to happen.

When my car sits for an hour or more...
1. I start it up... RPM goes to 1500 or 2000 and begins to idle down as it should
2. RPM hits about 1000 and then drops rapidly like car wants to stall
3. ECU kicks up RPMs to 1500 or 2000 to prevent stall
This repeats about 6 or 7 times. After that the car is fine and eventually settles on about 700 RPMs.

Now, change it up a bit.

Again, my car has sat for an hour or so...
1. I start it up and immediately start to drive
2. Car accelerates but as soon as RPM hits about 2000, I lose power and it drops fast
3. I play with the gas a bit, spurts of power between losing power and making like it wants to stall
4. After about 3 minutes of this, car drives fine; no further issues driving or stopping at lights or acceleration

For the longest time I was not even getting a check engine light. (more on that in a sec) So in my efforts to nail the issue, I have done/checked the following:
- Swapped out the fuel pump with another one I know is working
- Cleaned the spark plugs (NGK) and checked resistance with my multimeter
- In process of pulling the plugs, one of my wires broke, so I now have a new set of NGK wires
- Swapped out the ignition coil (I had the distributor off to check it for oil leakage maybe causing a short, and I had a brand new OEM coil so I put it in regardless)
- Ran Seafoam through the throttle body (which had a bit of carbon on the throttle plate itself)
- Replaced the fuel filter (just last night)
- Checked and adjusted the timing

In addition to the above, I also have the following "new" parts (replaced in past 1-3 years):
- Spark plugs (NGK Iridium)
- Complete distributor
- Fuel pump
- Fuel pressure valve
- Timing belt and water pump
- Thermostat (OEM)
- And... a rebuilt head about 3 years ago (due to a hair line crack in the original one caused by overheating)

Yeah, I know. I practically have a new car with all the parts I've replaced at this point, which is why I would really like to fix this problem and get as much mileage out of my Honda as possible. But it obviously won't make sense to put another big item into it. I'm hoping maybe I just need to replace an O2 sensor or a fuel injector or something?

Tonight a check engine light FINALLY came on - P0171. I've read up on that which leads me to lean towards a sensor or fuel injector since I've already eliminated a few of the possibilities.

Thank you for any advice

Jimi Hondrix

Disconnect negative terminal from negative battery post. Remove IACV and throttle body..clean both thoroughly with Throttle Body Spray (not carb cleaner) and a toothbrush. Do ECU reset afterwards

TitaniumCoder47

I have a bit of Seafoam left which I used for a throttle body clean. Will that work for this cleaning by toothbrush?

Jimi Hondrix

Should be okay to use..you can call Seafoam tech support and ask them, Concern is to avoid damaging sensors. Throttle body spray is around $5 :https://shop.advanceautoparts.com/p/...Fc-KswodaesIcw

TitaniumCoder47

Thanks. I went ahead yesterday and took off the throttle and plate, fuel rod, and injectors. I cleaned everything with the remainder of the Seafoam spray I had used for the throttle body a week or so ago. I cleaned the injector filters as well. Then I did the ECU reset. While the car does drive smoother and seems to have more pep, the core symptom still remains. This morning, for example, I warmed it up for about 2 minutes and then drove off. I didn't have any issues. I drove to church and let it sit for an hour and a half. When I left church, I just started the car and drove off. Around 600 feet up the road, while accelerating, the RPMs were about 2000 or so and I lost power. I eased off on the gas and just kept it where it was, not trying to accelerate but rather just maintain speed. About 200 feet later, it was fine and I drove all the way home with no issues (incl. stopping at about 3 stop lights with RPMs at 700 and no issues accelerating). If I let the car sit for another hour or so and try to drive off, it will do the same thing. At about 2000 RPMs, it will lose power until I keep it even for a bit. Then it will be fine. I'm really at a loss as to what this could be!

TitaniumCoder47

I have some new information.

So this morning I let the car warm up for about 8 minutes before driving to work. No issues.

About 40 minutes ago, I plugged in my CEL reader and started the car. RPMs were about 1500 and starting to idle down. But instead of waiting, I immediately drove off across the parking lots behind the office complexes. Sure enough, the car lost power like it wanted to die. I cycled through the values, and the only one that jumped out at me was SHRTFT1. It was a whopping -21 the entire time the car had no power. Twice during this time (which lasted about 2 minutes), the car would get power like it was a person under water, surfacing to gasp at air before plunging below again. This cycle happened twice, then the car had full power with no issues on the 40 minute drive home. I kept my eye on the road and on the SHRTFT1 and only saw it drop as low as -8 one time. Usually it was ranging between 0.0 (when I let my foot off the gas) to mid 30s.

I've done a bit of research, and it appears the SHRTFT1 is directly related to the O2 sensors and the injectors. On the 40 minute drive home, I could not really pick up a correlation between the number and whether I was accelerating or sitting at a stop light. The only direct correlation was when I let off the gas to coast--it would zero out.

I'll do some more research tonight, but if anyone has an idea, I would love to hear it!

Jimi Hondrix

Recreate then conditions that will cause the low power condition and then disconnect the primary o2 sensor harness and see if the engine reacts differently. Disconnecting the o2 sensor will cause the sensor to run in an open loop, run rich and throw a CEL. Reconnect the primary o2 sensor regardless if it solves the problem and then purchase a new one.

If you want to check injectors..remove them from cylinder head and spray them into a bucket while cranking engine. Note the spray pattern, amount and timing of spray for each one. Also, inspect if they are leaking from any part of the injector including o-rings. lightly lube o-ring with synthetic grease (ex: Napa Syl-Glide) before reinstalling the injectors.

Jimi Hondrix

OP: I forwarded this entire thread to a much more knowledgeable person than myself and here is his reply:
(copied and pasted) " I'm expecting it to read like decel fuel cut 2000-1200 RPM, but it doesn't quite match, so I don't want to yell that out without more data....LOL. Maybe open up the PCM and look for toasty spot or leaking capacitors?

Thinking about fuel cut:If so.... that should never happen until the TPS voltage says the throttle is closed (with RPM still high) and it should not enter decel fuel cut in a stone cold engine (I want to say it enables fuel cut about 130F or higher but I don't know the exact temp for sure).

....so does the OP have a TPS problem? Low coolant level problem? Coolant flow blocked to the IAC valve? (If IAC incorporates thermal valve, poor coolant flow can result in fast idle staying on too long or too high, incorrect RPM for the temp the ECT indicates to the PCM) Temp sensor problem? Thermostat problem? IAC stuck open? Vacuum leak? (cruise hose came off manifold? Something else?)

I'd use a scanner that shows TPS as a voltage value instead of %, (or use a voltmeter) I expect 0.49v closed (Accurate adjustment is critical on Hondas TPS)
Watch TPS while driving, see if it shows anywhere near a closed voltage (or below) when he's driving and it acts up, while the throttle isn't really closed.'.

mk378

You can test TPS with the engine not running. Press the gas pedal down slowly and watch for the reading consistent increasing with no jumps or "dead spots".

Fuel trims outside of +- 10% are a problem. A large negative means the ECU is thinking too much gas is entering the engine, maybe that is actually the case, maybe it is a bad sensor reading. Fuel pressure should be checked.

civicservice

You get the most out of fuel trim readings when measured against another parameter - usually engine speed. What are your long/short fuel trims:

1. at idle
2. under load (~3500 rpm)

TitaniumCoder47

Thanks! I tested the TPS tonight, and it showed a steady climb and steady drop as I changed the throttle position. At least I can rule that out!

Beginning of the week, I pulled off the front O2 sensor (easy to get to) and examined it. Lots of carbon or whatever caked to it. I cleaned it up a bit and set it aside. Then this evening I jacked up the car and crawled underneath to locate the rear O2 sensor. Unfortunately, it's NOT easy to get to and would require taking off the catalytic converter. I decided to go for it and check that out too, but the nuts/bolts on the cat are so rusted that WD40 and a breaker bar just about stripped the heads. So I bailed on that for the time being.

I went ahead and put the front O2 sensor back on and started up the car. After not even 2 minutes, it idled down to about 1200RPM and then dropped like a dying duck. I had my CEL reader plugged up, so I grabbed it and checked. Sure enough, the SHRTFT1 value was dropping at the same time to a whopping -27 as the car struggled to stay running. Then the ECU would increase the RPMs to about 1000, and the SHRTFT1 would be 0. Then after a couple seconds, it would drop again. This happened about 3 times, and I turned the car off.

I'm will probably have to gamble with some money and parts at this point. But maybe I can still increase my odds? Should I replace just the front O2 sensor first? Or should I replace both and the cat at the same time? (someone said when a cat goes bad, it can cause really poor engine performance) Or would a bad injector do this, so I should just focus on testing those first?

TitaniumCoder47

Well, I have some more information...

I pulled off the injectors again. Then one at a time, I connected the negative lead to the battery. Then I readied the positive lead but did not connect it to the positive terminal. I then rigged up a hose over the inlet and filled the hose with about 4 inches worth of gas. To simulate pressure, I blew over the opposite end while touching the positive lead to the positive terminal. In every case, a single STREAM of gas would emit. No spray like seen on YouTube. Rather, a single thin stream of gas. I tried both with and without the tiny filter in the injector.

Perhaps I was not doing something right? Or perhaps I have bad injectors? Otherwise, I don't know why I would get a single thin stream of gas instead of spray.

Jimi Hondrix

To test pull the injectors from the engine, reconnect harness wire and spray them into a bucket while cranking the engine. Watch for spray pattern, amount, timing and leaks.

Similar to the video below but use a bucket to catch spray. Can also disconnect the distributor harness to stop spark (added safety):

TitaniumCoder47

Well, this forum has been awesome! Thank you everyone for the excellent help in diagnosing my Honda's symptom. The final outcome is that I replaced the front O2 sensor and all four injectors, and it's like I have a new car. I have tested it over three days, and the symptom is completely gone. Additionally, my Honda is a lot more responsive in driving. The engine sounds a tad quieter too, though not sure how that is possible. For any interested, here's the cost breakdown to get to this point:

44.78 - NGK (8034) HE76 Premium Spark Plug Wire Set from Amazon (only replaced because one wire broke when I was trying to get to the plugs)
21.39 - Duralast FF696DL fuel filter from AutoZone*
30.35 - Denso 234-4099 Oxygen Sensor from Amazon**
72.66 - Genuine Honda Set of 4 Fuel Injectors (reman w/ 10yr warranty) from excellent eBay seller
Total $169.18
*Been a couple years since I replaced, so I count this as general maintenance
**Probably didn't need to replace, but was going this route prior to overwhelming evidence the injectors needed to be replaced

Additionally, I bought a timing light and ECU reader, which I consider investments, not supplies/parts, since I will use them many times over.
26.69 - INNOVA 3551 Inductive Timing Light (Amazon)
34.99 - ANCEL AD310 Classic Enhanced Universal OBD II Scanner Car Engine Fault Code Reader CAN Diagnostic Scan Tool - Black (Amazon)

Thank you everyone again for the help in diagnosing my car! It is back on the road as my daily driver, and I hope to get a few more years at least out of it!

(Since I replaced two things at once, I don't really know which or both was the problem. However, based on the injector tests, I'm like 95% sure they were the original injectors and had reached the end of their lifespan. Nineteen years wouldn't be too bad for injectors, I would think!)

Jimi Hondrix

OP..thanks for the update and it's always nice when a thread comes to a conclusion..especially a positive one. You did right by purchasing an OEM primary o2 sensor (BTW great price) and injectors.The timing light and scan tool also a very good investment. Here's o2 info: What the Home Mechanic Needs to Know about O2 Sensors