Tech is stumped. What causes this misfire?

I suspect the plug wires or cap have a crack in them, letting moisture in and creating a path for the high voltage to leak to ground rather than going down the wire to the spark plug. Cracks like that can be hard to find, and may only open up when the engine is warm. Fortunately, they are not that expensive and easy to change.

Bill the Engineer

 

Yes, I think you should remove the spark igniters and inspect the spark plug wells for water. You may need to use a mirror and a trouble light to illuminate the spark plug wells.

If you see moisture in the spark plug wells, then you will know the problem is water leaking onto the engine valve cover and then getting into the spark plug wells, causing misfire.

In that case, blow out the spark plug wells with compressed air. Make sure the rubber O-rings on the spark igniters are good (they should be if the igniters were just replaced.)

Then look at the alignment of the hood to the cowl tray. You may need to add a supplemental weather strip to keep water from leaking down the cowl tray onto the engine.

An alternate possibility is a problem with the wiring harness connector leading to the misfiring igniter. Inspect the connector carefully to make sure there are no faults which would allow moisture into the low voltage connections. Make sure the connectors are locking to the igniter wiring harness sockets.

Most modern gasoline engines, including the Prius, have a spark igniter sitting on top of each spark plug. This allows more precise control over ignition timing and avoids the issues you cited. There is no distributor cap and no high voltage wires - that is an obsolete design which disappeared ~20 years ago.

 

Somewhere along the way you did change the plugs, right ??
Although rare, it is possible to have a extra spark path across a defective or damaged plug insulator.

 

I hope you installed the correct NGK or Denso iridium plugs.

 

Have you told them about how you can make it stop the mis-fire/running rough ?
And did they "hear" you when you said that ?

I still think that the problem going away if you turn it off and back on should be a BIG clue........for somebody.

BUT......with a 13 year old car, I'm not sure finding the cause is a prudent thing to do. Likely costing you a LOT more money than the problem is worth. Simply re-start it whenever you notice the problem..........or just drive it. Didn't you say that it goes away on it's own in a minute or two ??

 

I've just re-read this thread twice over looking for where I missed that, and I haven't spotted it. Is it possible that was from somebody else's thread?

Yes, those are signs of serious misfiring and meant to be taken seriously. One of the predictable casualties of sustained misfiring is eventually the catalytic converter (whatever fuel isn't burned in the misfiring engine ends up down there and melts the converter insides into a block of gunk) and that is not an inexpensive part.
(The block of gunk then increases back pressure, which doesn't help the engine run any better, so it might be worth checking just to make sure that isn't something that's already happened and complicating your diagnosis. I don't know if there are good noninvasive ways of measuring back pressure, through an O2 sensor port, maybe?)

There are a classic three basics to check for misfiring. The engine needs fuel, it needs compression, it needs spark. It's really a matter of focusing and not letting up on those areas until the problem is found. It can be very helpful to write down (maybe even post here) everything tested and the result of each test.

As far as I can see to sum up, so far you have:

• replaced all spark plugs
• replaced one fuel injector
• replaced fuel tank, canister, and lines (and presumably the soggy fuel in the old tank)

Is that right so far? So, then:

• What are the (quantitative) results of a compression or leakdown test? (Conventional compression test requires Techstream to command the engine to crank for the test; a leakdown tester can be used by anybody with a source of compressed air.)
• What spark tests have been tried, other than replacing the plugs? An easy test, while the engine is misfiring, is to unplug one igniter at a time to see if the misfiring gets worse. It gets worse every time you stop a good spark. If it doesn't get worse, you've found a bad one.
• At least, that used to be a simple test. In a modern car the ECM looks for an "IGF" confirmation signal from each igniter, so if you unplug one it will see what you did there, Probably it will only log codes (P1300, P1305, P1310, P1315 for cyls 1, 2, 3, 4 respectively) and you can even check for those with a code reader afterward to confirm that it saw what you did.
  
• Also, you can remove plugs from the engine, leave them plugged into their igniters, ground their metal shells to the engine and watch them for sparks while someone cranks the engine. (Just long enough to see sparks, and mind the fuel-laden mist puffing up out of the spark plug holes. Or unplug the injectors for the duration.)
• What is the fuel pressure? It should be ok, as the pump is in the tank and the tank is new. Testing it is a lousy job: getting access to the connection where the service manual wants you to test it makes injector replacement look easy, and requires you to hire an 8 year old with triple-jointed wrists. If you have your old injector from cylinder 1, I wonder whether a machinist could make a pressure test fitting out of the top, O-ring end of it, that could be used just by sticking it in the fuel rail in place of a real injector. If the rail is unbolted, then such a fitting (and the other injectors) would need to be securely wired to the rail to avoid being popped out by the fuel pressure.
• Injector problems can cause misfiring either because all the injectors are bad, or because there's a wide difference in how bad they are (the ECM only gets O2 readings for the engine as a whole, so it has some ability to adjust for the overall condition of the injectors, but no way to compensate for differences between them.) If you had a problem with injector 1 and it was the only bad one, replacing it could have improved things; in a different case, it could have made things worse to have one now that's brand new and others aren't.
  
• I've collected test data on two PriusChat members' fuel injectors, and so far am exactly 1-for-2 in having them turn out to be the problem (got a coin to flip?). The details are here and here and in the threads leading up to those. In astrolink's case the injectors were the problem (as clearly shown by the problem going away with replacement, and the poor test results coming back for the old ones before restoration) and astrolink was happy. In CClay's case, they were not the problem (as shown both by the engine still misfiring after replacement, and the good test results coming back for his old ones), and he was disappointed. I do still have CClay's old set, restored to the condition shown in the test results, if your diagnosis leads that direction.

-Chap

 

Back pressure might be a worthwhile test, just to rule out the earlier misfiring possibly damaging the catalytic converter in a way that's now confounding your troubleshooting efforts.

As a test for water in the fuel, some can be pumped into a glass. If you get some stiff wire and wind it around the four injectors and the fuel rail so they won't pop out of the rail under pressure, you can lift the rail up so the injector bottoms clear the engine (best to use some compressed air first to blow away any grot around the holes where they sit). You could stick a hose of suitable diameter over the bottom of an injector (no. 1 would be easiest), run it into a glass, enable the fuel pump and trigger the injector using a 9 volt battery across its terminals. (9 V is enough to open the injector, and less likely to overheat it than if you fed it 12 V for longer than the brief pulses it is built for).

Then you'll know what you're squirting into the glass is exactly what's getting squirted into the engine. If there is water in it, after you let the glass stand for a time, it will clearly stratify with the water at the bottom. Like you, I don't think it sounds likely you'll find any.

Roughly the same setup, using a graduated cylinder and a stopwatch instead of an ordinary glass, is how you could compare the flow rates of the four injectors. As you can see by checking the lab report links I supplied above, that's basically what the bench tests are: what volume of fuel flows in so many seconds at a known fixed pressure, under controlled conditions. A new or fully restored set looks to be matched within 1% or so. It isn't really about the "quality" of the injectors - they might all be of fine quality, but there could still be flow differences between a brand new one and ones that have been in long service. Cleaners you pour in the tank might make some difference, but astrolink's results showed it wasn't very significant (an ImportTuners article in 2011 reported similar results, link in astrolink's thread somewhere). The bench restoration uses stronger solvents and techniques that aren't available while on the car.

The dealership might have equipment to measure the injector flow rates. Sometimes it is helpful to ask them to give you the quantitative test results as opposed to just saying they did the test and it was ok. (In some cases, that helps remind them to really do the test.)

You can treat the "after restoration" section of either of the reports I linked as indicating about what the results for a good set should look like (possibly adjusting if tested at a different pressure or with a different viscosity test fluid than Rich used).

I don't know if you have any wrenchy tendencies yourself or just have the mechanics do everything, but I should mention in case you might think of getting at the injectors yourself that great care is needed when inserting the upper O-ring ends into the fuel rail. I didn't stress that enough with CClay and he ended up learning some of it the hard way. Most importantly, if you don't succeed at full insertion using your fingers, know that cranking down the fuel rail bolts will not finish the job for you - it will tear the O ring, bend or crush the injector instead.

I made a how-to video for him that I haven't posted anywhere yet (some of the audio was crummy and I haven't had time to re-do that track) but I can send it to you if it would help. It covers the 'enable fuel pump' and 'trigger injectors with 9 volt' business I referred to above.

-Chap

 

I would first check fuel trim with computer. Fuel trim numbers at idle and under load will tell you if it could be vacuum leak or they may hint to fuel delivery problems. If fuel trim numbers appear to be ok then I would swap two igniters together. If it still misfires in same cylinder I would check spark plugs. If they appear ok I would just swap two and maybe do the compression test at the same time. If its ok and it still misfires from same cylinder I would check the wiring harness more. And only after that I would take fuel rail out to test injectors or just replace that one injector or just swap two injectors to see if misfiring cylinder finally changes. Taking the injectors out will cost at least new injector bottom O-rings (If something breaks it's lot more) and testing them takes almost as much time as all other tests.

 

[edit: apparently that was someone else who turned his off and the problem went away. Sorry for the confusion.]

OK.....and my memory isn't too good sometimes......but didn't you say that shutting it OFF and then re-starting makes the problem go away.
Or was that someone else ??

If not, I agree that running it while a noticeable misfire is occurring is not a good thing to do.
Maybe it is time to consider a trade-in.

 

But if there was some serious contamination in the gas, normal use might not get rid of all of it all.
A good dose of a multi-purpose fuel additive might be a good idea.
Also I would be covering the hole in the filler neck with some metal duct tape.......just because. It won't last forever but should rule out that as a cause. With difficult cases like this you should ALWAYS fix everything that you know is not right because Mr. Murphy often will bite you in the butt if you don't.

And just to be clear here.................you have a 13 year old car with presumably a LOT of miles on it.
Just exactly why are you continuing to pour money into it instead of using the money toward a replacement ???

 

I've had two bouts of misfires in the past year on my 2001, and in both cases fixed the problem by adding water remover to the fuel. Both times it started with a heavy rain. I went through a lot of the checks discussed here the first time it happened. More recently after 6 months it happened again and this time I added the water remover without even opening the hood and it resolved itself almost immediately. I should also mention that I've been living with the occasional evap codes for a year or more, just clearing them when they occur.​

My working theory is that water is somehow getting into the tank in really wet weather. I don't know how, but since I have an antidote in the form of a case of fuel additive I'm not losing much sleep over it. A lot of what you've described sounds familiar to me. I wouldn't rule out contaminated fuel...

 

And what criteria are you going to use to deem it "scarp" ??
I think you passed that point long ago.
Maybe you could use the car itself for a down payment on a newer one ?
Of course, if you can't make payments then you are in a nasty bind.

 

someone said cover that hole in the filler neck. that's the first thing you need to do tomorrow morning. may not fix anything but it sure won't hurt and may be the whole problem. after all, how you think that water got in there in the first place? diy if you can.