Hello All,
Got a project car that needs to go back to being a daily driver. 1994 LX, F22B2 at 129K miles. We overhauled the engine (essentially a short block rebuild), but during that time the 1/4 tank of gas went bad. Initial lightoff after the overhaul went well, but then would start no more. After trying multiple things, I gave up and had the car towed to Honda Repair. They confirmed bad gas as the cause, and drained the tank, added a few gallons back in, and off we go.

The car ran for a few days, then would not start. We re-drained the tank at home, added 2 gallons plus a few ounces of fuel dryer (BG product), changed the fuel filter, and eventually it started back up. We immediately filled the tank (BP, not Sheetz) and added the rest of the dryer and treatment. Car ran great for two days, and then back to not starting.

I suspect that, even after draining (from the drain plug), there's still some 'fuel' left in the bottom that can't get out. after driving, that stuff swirls up enough to get picked up by the fuel pump, which promptly goes to the fuel rail. After accumulating and sitting overnight, "badness happens" in the fuel rail and the previously running car goes back to sitting very still in the back yard.

Questions- How can you get 100% of the contents out of the tank? Do I have to drop it and tip it (I sure hope not)? Has anyone tried to wash it or rinse it out after getting it empty?
And the filter- Does it also need changing? If it's only for solids, not liquids (like a coffee filter)- and my problem is liquid (either water or broken down gas), then I'm thinking it should be okay.
Appreciate any past experience or recommendations. Thanks!

 

usually theres a drain bolt

 

yes, I used it, but the drain bolt is not at the hard bottom of the tank. There was still some amount left in there after draining from the bottom plug.

 

You do not have a "bad gas" issue......time to start testing other possible issues.

 

I left out the process of elimination:
-Compression is good at 135-137psi in all four cylinders
-Spark test is good in all four cylinders; plugs clean and properly gapped
-Fuel pressure is good; 42 psi static (no vacuum) and 35 psi running
-timing is dead on at -15 degrees,
all of the above verified by the Honda tech, who diagnosed as bad fuel.
The car ran for a day or two both times after draining the tank.
I also confirmed bad fuel with what I pulled out of the tank when I drained it. The last little bit was almost the color of iced tea, and what I put in a clear container produced a 'slug' the size of a silver dollar that stratified at the at the bottom of the container after settling.

If the problem is not what's already (and still) in the tank, PLEASE, tell me what else it might be!

 

another question- I intend to blow back (compressed air) whatever is in the fuel line, from the fuel rack connection to back into the tank. Will the fuel pump free-wheel in reverse?

 

which other thread are we referring to ?

 

Ok, drop the tank and clean it

 

So, here's what I did:
1. Disconnected battery. Opened the access plate in the trunk, and removed the TLI (three of the five nuts would not fit a 10mm socket, and one had nearly disintegrated, btw)
2. Hand pumped the tank empty. Used a flexible rod on the suction hose to hold it into the low point of the tank. I'm pretty sure I got everything out. The tank looked pretty clean, and so did the first 8 gallons I removed. The last 1/2 gallon was a little darker, but no debris and no fluid separation.
3. Disconnected the fuel line to fuel rack, and loosened the nut on the bottom of the fuel filter. This drained the fuel rack, fuel filter, and fuel line in between.
4. Reinstalled/reconnected everything except the fuel line to the fuel rack, and added 2 gal of new gas.
5. Turned the key to 'run' several times, enough to purge the fuel line with about 6 ounces of fuel into a cup. The fuel that came out looked pretty clean and clear.
6. Reconnected the fuel line to the fuel rack, and loosened the service port bolt on the fuel rack. Switched the key on twice more to get any air out of the fuel rack.
7. Tightened everything up and pressurized the fuel system once more for a leak check, and then began cranking the car.

After three 10-second cranks, a cylinder started to fire- then two, then the engine started. We drove down to the same gas station where we got the 2 gallons, and added eight more gallons (but not full). Drove around neighborhood for at least half an hour. Stopped and restarted the engine multiple times, no issues. Ambient temperature, by the way, was ~55 F at that point.
Six hours later, the car would not start at all. Ambient temp now down to ~35 F.

And I don't think this is a temperature change issue. Apparently there's something that gets 'fixed' temporarily by purging the fuel system that gets 'unfixed' by sitting still for a while. It definitely used to be bad gas, but now... I'm at a loss.
Dirty fuel injectors? Demonic possession?
Any ideas or suggestions would be GREATLY appreciated.

 

UPDATE: not the injectors. I had them bench tested and cleaned by CruzinPerformance.
Any occurrences of heat soak/boiling gas occurring in fifth gen Hondas? I've seen that sixth gen (2000) doing that, complete with Honda TSB (00-024). I seem to have the same symptoms.

 

Glad you got all of that said "bad gas" out of there......but I don't understand how you can say "It definitely used to be bad gas, but now... I'm at a loss." and still have the same issue

You are dead set on "bad gas", I tried a while back so I'm not sure what I can say to get you off that topic and get to actually testing something when the car acts up. Good luck.

 

Maybe I'm not being clear, but you're missing that word WAS in that pull quote. I am still not sure what the problem is now- which is kind of the point of posting here.
Here again, is what I've tested that's good (or scroll up to the post on 17 Feb): Compression, spark, timing, fuel pressure. And all verified separately by a Honda tech.

Add to the above the fact that when the fuel system was drained and refilled, the car would start and run for a short period of time. Three iterations for proof. But the issue is not fixed.
I don't know what to test next. Is there a test kit for signal (pulse width, etc) to injector? Igniter? ECM? FITV?
Any helpful suggestions greatly appreciated. ["Go test 'something'," "just buy a new car", or "go F yourself"- those don't count.]

 

Look, an issue that shows up every now and then can be hard to trace.....let the car sit for a while as you play around with the bad gas that is still left in in after everything you've done to remove it....lets start to think about this point for just a minute.

And if you want to go there....I have a car - the 89 LXi in my sig line - that I put about 1/4 to 1/2 a tank of gas into it twice a year.

If you want to "Buck up butter cup" I said on Feb 18th to DROP the tank and clean it-----could the sock filter on the pump have "gunk/dirt" built up on it, or still in the tank in such a way that after a bit of running it gets so clogged that there is not enough fuel pressure to run the car....oh wait, a fuel pressure test when it wouldn't start might show that, sorry my bad your "Honda tech" tested that so it can't be the issue.

You are going to have to test things when it will not start (compression and timing don't count in this part).

When it won't start - what do the tests show .

Or is this gas SO bad that even if you spray starting fluid/carb cleaning into the intake - open the buttery fly on the TB and all of that - that is still won't start? Wait, the plugs must me soaked in bad gas cause when you pulled them when the issue was happening they were soaked in bad gas.

You can think I'm an A hole, that is fine I don't care. At the end of the day I HAVE tried to get you away from this bad gas stuff and get you to test for things WHEN the car acts up.

Have a good day my friend

 

Thanks poorman. Let me be crystal clear this time- IT"S NOT BAD GAS ANYMORE, and it hasn't been for a while, but SOMETHING is still wrong.
And, this is not an intermittent problem. The car is occasionally running, but not starting most of the time.
(and hey, we're all A-holes at some time or another, but you're trying to help and I do greatly appreciate that!)

Plugs: are pretty new (Sep 16), and gapped about a month ago. They're clean, and really haven't run enough not to be.
After failed starts, they are moist (thin coating of gas, just enough to make them shine) but not dripping (i.e. not flooded). For the amount of cranking that goes on before pulling them, I would almost expect them to be flooded, but they never are.

Fuel pump/ sock: In all that draining, I never pulled any significant particulate matter out of the tank. just a small piece or two. And, fuel pressure (I measured, then local Honda shop did too) 43 psig is middle of the range. When I pulled the injectors, I purged the fuel line (by running the pump) into a cup several times. It's delivering about 1-2 ounces per second, which should be enough (that's ~40 gallons/hour). Whatever the problem is, I don't think its in the pump sock, the pump, filter, or the lines.

I've got the injectors back now, and I'll look at voltages on the connectors before I hook them up. I will try starter fluid again tonight too and update.

 

BTW and if anyone's interested: I ran the numbers from the injector test sheet. An injector (IPT 4008) wide open (unpulsed) will deliver 71 ml in 60 seconds. Convert from metric, times four injectors, and that's 4.5 gallons/hour. Which checks with mileage: at 60MPH, a car getting 20 MPG will burn 3 gallons an hour. (and that's with pulsed injectors, obviously, and return fuel etc.)

So in my case, the fuel pump/sock/lines/filter has WAY more CAPACITY (volume flow) than what's required for the engine at full speed and load. So fuel flow rate (at least to the rail) is not the problem.

 

At the start, not sure of the order, you said something about "rebuilding" the engine or bottom end, whatever. By chance is the main ground (G101) on the drivers end of the plenum loose or have :gunk: built up on it. This ground is the "main ground" for the ECU so if loose well you will get all kinds of issues....no start, dies when running, ect. Again not sure if the intake/plenum was messed with during the work but something worth checking to be sure it is attached, clean and tight. I and others have posted pictures of it several times so it won't be hard to find a pic of its location.

Next, can we assume most of the tests, spark AT the plugs, fuel pressure was tested and verified WHILE the car was acting up (no wanting to start)?

Checking for "pulse" at the injectors - sounds like you might be getting the pulse as the plugs are moist/wet. A noid light or a good voltmeter can check for this....just remember they are always hot and the ECU sends the ground signal to "fire/open" them......or just open the butteryfly on the TB and give it a good shot of carb cleaner/starting fluid and see what happens.

Again check all of this when the darn thing is acting up/not starting,

I don't want to jump to a bad ECU yet - I hate to spend money without testing everything out first.

 

OK, injectors are in and fuel line is hooked up (I purged the fuel line into a cup before connecting, and fuel is nearly clear- very clean).
-Checked the ground at the back of the IM. (I had also added an additional ground wire there on a different bolt earlier). They both were tight.
-Before reattaching the fuel rail, I put a little compressed air down the return line to the tank- it is not obstructed. (just an easy check while disconnected).
-All four injectors are getting battery volts on the red/black wires. ->[resistor, PGM FI relay, and all the wires in between are getting power]
-Installed a pressure gauge in the service port, and pressurized the fuel system by turning the key on a couple times. (and bled off the air in the fuel rail). Fuel pressure solid at 40psig+.
I cranked it over 5-6 times for 10 seconds each crank. No start.
That was about all the battery would take- it's on the charger now. I'll try starter fluid after dinner.

 

Removed intake duct and gave it starter fluid. No start.
Next I removed one plug at a time, and checked for spark with plug held against the coolant outlet (on the engine block). All four are sparking. I also held it on the valve cover over the spark hole for #2-4, and got a nice little POOF of fire. Everything I need for engine start is there, right?

On a whim, I broke out the timing light and hooked it up to the #4 wire while checking it against the coolant outlet. The plug sparks, and the light lit up. And here's the weird part. When I put the plug back in and hooked it up, I reattached the timing light and cranked it over. The timing light did not come on. Same for al the cylinders.

Am I right on this? The spark plug is sparking when held outside against the block, but NOT sparking when it's in the cylinder. What could cause that??

 

Maybe the spark plug wire is arcing and it only does it in the normal in-the-cylinder position. At night/in the dark run the engine and look for sparks. You can mist some water on it to make it more obvious. Where there is spark there is a problem.

 

I rechecked spark with an actual spark tester- and I had no indication of spark. So I swapped out the coils, and the engine instantly started on the first attempt. I tried multiple times, and it started every time on the first touch of the key. Apparently, the coil was just strong enough to give a spark on spark plugs held against the block, but too weak to give a spark in the pressurized cylinder (where it really matters). We'll run for a few days to make sure of the fix, but it is looking very good tonight!
A couple observations for posterity:
1. There was a debate on this blog a while back on the use of spark testers (like this one: OTC 6589 ) vs. using a spark plug held on the side of the block. That debate has been settled (for me, anyway). It is possible to get a 'false positive' using spark plugs this way with a weak coil. [and as a side note, a false positive on spark checks can drive you crazy!!] I would have found the weak coil a lot sooner had I used an actual checker. I think the 'color of spark' business is also bunk, as I had a nice, bright blue/white spark using spark plugs, but I had zero spark in the cylinder.
2. There are some things worth the dealer markup, and the ignition coil is probably one of them. The bad coil was bought new from Autozone six months ago. And for three of those months, the engine was in a thousand pieces in my garage! I'm not saying there are no good aftermarket coils out there, but I'm 0 for 1 with them.

I'll follow up in a week or so, hopefully to confirm the fix. Thanks all, and esp. poorman212 and TomCat39, for the help!

 

Here is to hoping this issue is resolved and you can worry about things like actually driving the car

 

Just to follow up- car is fixed and stayed that way. It now starts faster than my '09 Accord :-) !

 

Just something to add to the topic of not being able to fully empty the tank with the drain plug; I can confirm that there is about 2 gallons or so that don't drain out fully when the car is sitting level and the drain plug is pulled out on my 1992 Prelude. For future reference, having gunk (or in my case, rust) built up in the bottom of the tank can cause problems for the fuel pump that can really only be resolved by removing the tank, tilting the tank and letting it all out through the drain hole.

 

Good ole Auto Zone lifetime warranty parts! They guarantee a lifetime of heart ache and parts swapping!

 

Did you wind up going with Denso coils ?