Honda Odyssey Forum banner

Dead battery, then all hell breaks loose. Misfires and crank sensors oh my.

15K views 65 replies 12 participants last post by  matty.g  
#1 ·
2013 Ody, 130k miles

Kids left door open this morning 10 minutes max. Cranks but won't turn over. Come back an hour later and it starts.....barely. Drive to get gas, get home and it picks up a P0339 for crank position sensor circuit intermittent then P0302 and P0306 for misfires on 2 & 6. I get a new battery, does not fix it. I change the crank sensor, does not fix it. Swapping plug and coil on cylinder from 6 to 5 and the misfire does not follow it.

So now what?

Is the crankshaft sensor learning procedure a real thing? Is it really only doable by a dealership?

What else am I needing to look at?
 
#3 ·
Yeah, cleared and gone through a couple times. And to your point, I don't have a Honda specific scanner. But I'll say the crank sensor is stored in permanent but doesn't keep coming back, rather the 2 and 6 misfire repeatedly return. Based on my understanding of how permanent codes work though that crank sensor can't go away because I won't be able to satisfy the driving conditions for it to clear naturally and I don't have an HDS.

The Ody service manual tells me to start going after continuity on the fuel injectors next. I guess that's where I'm off to now.
 
#4 ·
Am I wildly missing something here on fuel injector troubleshooting? Checking continuity and resistance across any part of the fuel injector system requires the intake manifold to be removed? And to swap injectors to identify if the fault moves requires you to swap them, reinstall the whole manifold then give it the ol' college try?
 
#5 ·
All of this sounds to me like you had a battery near end of life. That's it. Everything else is just stuff happening when it would be better left alone.

I would have bought a new battery, charged it up with a battery charger, reset codes, see what comes back, be extra cautious about believing the codes until they repeatedly come back after resetting, and then I'd learn more about not using run-on sentences, but no time for that. :ROFLMAO:

A low voltage can cause all kinds of temporary errors, but I don't think it will actually start breaking stuff in your car.
 
#6 ·
All of this sounds to me like you had a battery near end of life. That's it. Everything else is just stuff happening when it would be better left alone.

I would have bought a new battery, charged it up with a battery charger, reset codes, see what comes back, be extra cautious about believing the codes until they repeatedly come back after resetting, and then I'd learn more about not using run-on sentences, but no time for that. :ROFLMAO:

A low voltage can cause all kinds of temporary errors, but I don't think it will actually start breaking stuff in your car.
Lol, fair enough.

These codes have come back with the new battery now more than 4 times after being cleared. And the new battery is sitting at 12.6V off and getting 14.2V from the alternator. I'm happy to keep clearing and trying but is there reason to think they'll just go away?
 
#7 ·
change the crank sensor
Honda OEM, or aftermarket?

Could this be VCM damage?
Pull and inspect rear plugs. . . for oil.

Muzzle that thing.
 
#9 ·
Aftermarket, but like I said, the crank code does not keep coming back. I think it's stored in permanent because I haven't satisfied the conditions to clear it. I've only pulled front plugs for simplicity, I'll check rears.

im wondering if something didnt snack on the wiring harness under the hood,

have you had a good look under there?
I just changed 2 fuel injectors for funsies to see if the misfire moved. Spoiler alert, it did not. I checked those injector harnesses and they looked fine.

I'm with oldskewel that this is related to the battery, but I'm now up to clearing the codes 7 times and the misfires return each time. Can we check in line fuel pressure?
 
#10 ·
It was already mentioned, but if you used an aftermarket crank sensor I'd take it off and put the original back on. If it wasn't bad then you will likely induce other problems that can be caused by aftermarket electrical parts like a crank sensor. Same with coils. Even if the original was bad, the aftermarket won't be any better and will cause issues.

Put everything back the way it was and start over with a fresh charged battery. If you used genuine Honda parts then you can likely trust them. Otherwise, you likely installed parts that are worse than what you removed. Seen it too many times.

In your first post you say "it cranks but won't turn over." Let's get our wording on the same page. "Crank" and "turn over" refer to the starter "turning over" the engine. So, your engine "cranked" and "turned over" but would not "start." Now that we're on the same page with that we can go further. Did it crank slower than normal when this happened?
 
#11 · (Edited)
It was already mentioned, but if you used an aftermarket crank sensor I'd take it off and put the original back on. If it wasn't bad then you will likely induce other problems that can be caused by aftermarket electrical parts like a crank sensor. Same with coils. Even if the original was bad, the aftermarket won't be any better and will cause issues.

Put everything back the way it was and start over with a fresh charged battery. If you used genuine Honda parts then you can likely trust them. Otherwise, you likely installed parts that are worse than what you removed. Seen it too many times.

In your first post you say "it cranks but won't turn over." Let's get our wording on the same page. "Crank" and "turn over" refer to the starter "turning over" the engine. So, your engine "cranked" and "turned over" but would not "start." Now that we're on the same page with that we can go further. Did it crank slower than normal when this happened?
Noted on the nuance of my wording. I'll be better going forward. FWIW, with the new battery it starts smooth and easy every single time. The misfires don't set until about 60 seconds in. And I did put the original crankshaft sensor back in.

However, I went through the Honda service manual troubleshooting and actually arrived at a bad PGI-FI main relay. I was actually able to switch that relay with an identical part number and it opened up a host of new DTC troublecodes and the misfire didn't set. So I feel good about the relay.

Problem now is, I can't find what amperage it is? It's part number 7002 61 2523 and is a 4 pin relay. I found an Autozone #20580 but not sure how to confirm it's identical?
 
#13 ·
I was just reading your post and wondering how in the world a PGMFI relay would cause a misfire. I go back to my first suggestion. Put everything back how it was and start over. You're making things worse by just firing the parts cannon at this. I highly doubt the low battery caused the crank sensor to fail and even if it did, replacing it with an aftermarket leaves much to question. It's just going to dig your hole deeper.

After you're done putting the original crank sensor back in, you might even disconnect the battery and short the cables together to discharge all the capacitors in all the modules, then hook the battery up again and see what happens.
 
#14 ·
I was just reading your post and wondering how in the world a PGMFI relay would cause a misfire. I go back to my first suggestion. Put everything back how it was and start over. You're making things worse by just firing the parts cannon at this. I highly doubt the low battery caused the crank sensor to fail and even if it did, replacing it with an aftermarket leaves much to question. It's just going to dig your hole deeper.

After you're done putting the original crank sensor back in, you might even disconnect the battery and short the cables together to discharge all the capacitors in all the modules, then hook the battery up again and see what happens.
Just to clarify, everything (including crank sensor) has been back to original since 730 this morning. There's no parts cannon going on here, just following Honda troubleshooting guide.

I'll short the cables when I get back home. I've pretty much been through the entire Honda service manual at this point and will accept all suggestions.
 
#15 ·
Short the cables (remove them from the battery first!) and then reconnect. Start it up, let it run, and lets start from the beginning. Get the exact codes that are left and describe the exact symptoms. What kind of scan tool do you have?
 
#17 ·
Radio presets are stored in non-volatile memory, they don't get erased. You shouldn't have to do it for more than 30 seconds.
 
owns 2006 Honda Odyssey EX
#19 ·
Alright. Did the short. Screenshot with the permanents is pre-starting and the other is after a minute of just being on.

Also, the question was asked. My app is OBD Fusion and my adaptor is a Veapeak from Amazon. I have a ton of experience with this combination on the BMW's that I flip, but this is new to me with the Honda.
Image

Image
 
#66 ·
my 2010 pilot was doing this till i done the valve asjustment. the usa manual says to do it every 75-100k miles and europ modil manuals say every 20k miles. when i done mine it was clear off what it should have been and got rid of the multiple missfires but i still had 1 bad coil pack that the miss fire would follow. replaced the coil and still #3 missfire. replaced all spark plug no missfire. just today 8/29/25 started a 1 2 3 missfire going down the highway. popped a code for oil pressure solenoid. read some pages and the main cause was a weak battery so off to get a new one.

i would say to have your rockers adjusted/valve adjusment done.
 
#22 ·
The problem can be with scan tools and these newer vehicles. Generally there are two types of codes. Pending codes are codes that the computer has seen one time but require it to see a second time before turning on the MIL. These codes will show on a scan tool as "pending." Permanent codes are usually what come up when the computer has seen the problem twice and turns on the light. It then sets a "permanent" code. This is called two trip detection. There are also codes that will go straight to permanent without going to pending first. These are continuously monitored systems and codes. If you unplug the temp sensor it will immediately set a permanent code and turn on the MIL. If you have an evap leak it will set a code as pending and wait until it sees the problem again on the next drive cycle before going to permanent and setting a code.

However, on vehicles 2013 (or so) and newer, even if you clear the codes the codes are still stored in a permanent memory and it takes something like 50 or 80 drive cycles without seeing the problem in order to clear even after clearing codes and repairing the problem. These phone based scan tools are good but they have their quirks so trying to figure out what it's telling us. If those codes are in "permanent" status but there's no check engine light then I'd say it's the 2013 or newer issue. If there's still a check engine light on then the permanent codes mean they are active and turning on the MIL and in need of repair.
 
#24 ·
OK. Let's go with that for now.
While it's running and actively misfiring, unplug the coil packs one by one and see how the engine reacts. If both #2 and #6 are indeed misfiring then unplugging their coils will have no affect on how the engine runs. Unplugging other coils you should hear a noticeable RPM drop. We need to verify which cylinder(s) are misfiring.
 
#26 ·
So, definitely #2 and #6 are misfiring. That's good data. Next we need to figure out what's missing. It sounds like it's not the coil, the plug, or the injector since you swapped all those and had no change. Here's the next step I'd try. Unbolt the coil on #6. Start the engine. Slowly lift up on the coil and see if you can hear the coil firing. As you lift it up you should be able to hear it snapping in there as the spark jumps from the coil to the plug. You can also, remove it completely and use an incandescent test light connected to ground and let the spark jump from the coil to the tip of your test light. It won't hurt your test light. We need to know if there is actually spark occuring in those cylinders. Try not to let the coil fire without having a path to ground for the spark. That can damage the coils quickly.
 
#28 ·
Do you get a good strong spark? It should jump out of that coil a good 1/2 inch or so.

Next would be to determine if the injectors are actually being fired. That's a little more difficult. Need some way of getting a test light on the connector. Not sure what you have for tools.
 
#29 ·
Yeah, it was good spark. I was certainly a 1/2" away and it was consistent and bright.

That's going to be tough, I have a regular test light. No oscilloscope. I can confirm the injectors had 12V with power on, but obviously that's a big jump to "are they firing?"

What tool would allow you to get under the intake plenum while installed?
 
#32 ·
That doesn't sound very safe to me. You swapped injectors with no change, correct? So it doesn't seem like an injector flow problem to me. We just need to see if the control signal is making it to the injector from the PCM.
 
#31 ·
A T-pin, backprobe, or wire piercing tool are the best things. You need some way to get on the control wire from the PCM to see if the PCM is firing the injector. The PCM will be providing a ground so you'd want a test light connected to battery positive and then probe the control wire with the other end and see if the light blinks.
 
#35 ·
Without looking at a wiring diagram, most of the time, the 12v power wire to each injector will be the same color on all the injectors, or all on one bank. The control wire will be different on each injector. So, whichever wire is not the same color on each injector is the control wire from the PCM.
 
#36 ·
In many cases, when the PCM sees misfires it will shut down injectors as a strategy to save the catalytic converter. So, it's possible that the injector may fire for a few moments then quit. If you see that, the misfire may be caused by something else and the injector stopping may be normal. If that happens, clear codes, start the vehicle, and try again. See if that's repeatable.