Hard start when not stone cold or up to temp (lean closed throttle mixture)

[TheOriginalEasyrider]

This sub-forum is pretty dead so hoping there is a fuel injection wizard on TC 96 motors here.

2010 FXSTC. Problem has existed since very early in this bike's life. I'd say since day one but can't honestly confirm that. The only change made early on was switching from oem to NGK iridium plugs. I can't remember what I did yesterday much less 13 years ago. There was possibly a set of Champion RA8HC plugs before the NGKs. I really don't think this is a plug issue but I'll go back to the bone stock plug as a test if I have to.

Problem is that the bike starts great when stone cold or fully up to temperature. If it's somewhere in the middle and especially on the cold side (like a two mile ride to the gas station) the engine mixture acts like it is way lean and starving for fuel. If you are old enough to remember having a kickstarter, if you didn't kick the panhead over three times with the ignition off the kickstart would launch you over the handlebars. That's what's happening here except that the engine turns over briefly, backfires through the throttle body and the starter jacking gear makes this horrible noise. Usually starts up (with persistent pressure on the starter button) after that.

Side note: I have not checked the fuel pressure, fuel filter or any of that mostly because this has been going on forever.

I have a V&H FP3 management tool installed now but of course it had this behavior back when the bike was bone stock. I thought maybe fooling with the low RPM, closed throttle cell might fix this but so far, no luck. There's probably a temperature range where these symptoms can be reliably reproduced. The engine is pretty cold blooded. I have a Jagg oil cooler on it but have it covered unless I am going on a long ride on a very hot day. For a typical short ride the oil temp gets up around 190F. Even on a very hot day it might only get to ~ 220F.

This is obviously (at least to me) a problem with a too lean mixture but only in a very narrow range of circumstances. The usual detonation and white smoke coming out of the throttle body. Other than that the bike starts and runs flawlessly. 20W-50 Mobil 1 full synthetic. Battery is fine. Plugs properly gaped and look fine. ~23k mostly gentle miles.

Various changes to the FI map for VE and likely suspects cells made no difference.

I don't see how this could be a timing problem and there is no way I know of to check the timing anyway. If it's a timing problem then it's in the ECM or whatever is telling the ECM what the engine temp is. I've not seen anyone anywhere complaining about this so I am inclined to think that the ECM and the engine temp sensor is probably designed and working correctly.

Thoroughly cleaned the throttle body, air filter and IAC sensor. The engine temperature sensor must be working otherwise the bike wouldn't run as well as it does, right? The FP3 app does display a lot of information but I don't really know what to look for other than the temp. As I said, there may be a small temperature band where this is most likely to happen but that really wouldn't tell me much as far as what I should be looking at to actually fix this.

I've been screwing around with this for years so I would be suitably impressed if someone was to just throw out the winning solution. What I am NOT interested in doing is shotgunning parts based on hunches and wild-assed guesses so please spare me.

What I'm hoping for are things I can check/inspect/measure or otherwise eliminate as possibilities or confirm as the cause. Whoever nails this problem cause will get my undying adoration.

 

[joel]

Personal I would ohm test the crank and temp sensors cold, at the time of hard start, and also at fully warmed up when it starts normally.
Crank sensor out of range can cause timing issues, hard / no start symptoms.
We have been dealing with a older Mazda b2600 with extended crank no start at my shop, temp sensor ohm tested in range, almost matching a new sensor, decided to install the new one anyway. Starts right away at all Temps now.
Just my thoughts, both sensors can cause your symptoms.

 

[TheOriginalEasyrider]

Thanks Joel. Your suggestion makes sense although doing those tests would be a royal pain especially since you said that the bad part read pretty much the same as the good part. At this point I never know when I will have a hard start except as previously stated. Sometime it starts just fine after the two mile ride to the gas station. Other times not. Anywhere between ambient temp and ~ 190F might exhibit these symptoms. Probably closer to ambient and 140F. Perhaps lower.

Just asking but if there is a problem with these sensors wouldn't they throw a code? And why do you think that a sensor that had resistance readings matching the new part caused the problem? I'm not doubting your results. Just wondering how a sensor giving appropriate readings would the the source of failure. And especially a no start.

I've thought for years that there is something on my Harley that doesn't know what the engine temperature is and of course the temp sensor is the most likely suspect. But if that sensor were bad the bike wouldn't run and it would throw a code.

My FP3 app does read a variety of sensors including engine temp. I've been too lazy to do this but maybe any time I start the engine when it's in the "likely to have a problem" range I should have a look to see what the engine temp is. If the reading is reasonable I would think that would pretty much eliminate the temp sensor as the cause and maybe it's a bug in the ECM when it doesn't react sanely to the data it is being given. But if this was an ECM problem I would expect to see a lot of complaints about that which I don't.

As adverse as I am to shotgunning in parts I did look into what was involved in changing the temp sensor and the process looked to be non-trivial. The CPS looks easy enough to change but I think it would be quite a stretch to rationalize how that sensor could be causing this problem.

Also, this is NOT an extended crank/no start problem. Cranks fine and fires immediately. Backfires through the intake immediately. Retarding the timing a tad in that situation might help. Also turning the engine over with the ignition off to prime the cylinder could help. Unfortunately I see no way to do either on my FI TC.

I very much appreciate your feedback and you could very well be right about the temp sensor being sketchy. But the bike starts and runs fine all day long so this would not be a typical failure mode for a simple rheostat electrical device.

I'll give another look at how to remove that sensor and do bench testing on it. THANKS!

A note for clarity: When I refer to temperature I mean oil temp not engine temp. It's a button push to check the oil temp on the dipstick.

 

[TheOriginalEasyrider]

Looked around for info on the engine temperature sensor and I apparently had that one confused with something else (Maybe the MAP sensor). Doesn't seem all that difficult to change... $50 +/-. I only found one video where he actually changed it and reconnecting the cable looks like it's a bear. Of course he didn't remove the horn like the service manual says to. Certainly doable even at my limited ability.

Every video I found with a bad engine temperature sensor threw a code and the symptoms was stalling when the engine was up to operating temperature. I am getting no codes and the bike runs great all day long except as noted.

In one of the videos he referenced a table in the service manual that showed the various resistance readings that should be there for specific temperatures. I have a Clymer manual and there is no such table for my 2010 FXSTC. I'd also have to fashion a cable to take the readings and all of that would be easiest if done with the sensor removed from the engine. So except for replacing a potentially good sensor with one that might introduce new problems there is certainly merit in swapping the sensor out as a test. I'd do it in a heartbeat if it was throwing a code but....

If I get annoyed enough and no one else chimes in with any bright ideas I may go forward with your suggestion. I have long thought that the problem had something to do with your suggestion. But this sensor is a really primitive rheostat so while it's possible that it gets stupid in some narrow temp range that seems to me to really be a reach. Definitely the prime suspect as you say though.

Perhaps looking at what the FP3 app has to say when trying to start when the engine is warm and not cold or hot might shed some light on this. If it's reporting some clearly bogus temp in that instance then you will definitely be right in your diagnosis.

 

[dolt]

Deleted; double post.

 

[dolt]

Pull the sensor and test it. A faulty temp sensor will not always throw a code. True that stalling is a common symptom but so is hard starts, erratic idle, etc. If you can record data with the FP3 and it records engine temps; check that out as well after the next outing; if the ETS is faulty, the data should show it. I have read so many posts from guys chasing similar problems that check for intake leaks, clean/adjust the IAC, re-flash the tune, etc., etc. chasing these issues only to take the advice of the more "seasoned" owners that tell them from the jump to replace the ETS but the most common response from the afflilcted is "it didn't throw a code". Now, I am not saying that the ETS is the problem but at least it should be eliminated.

I told you the Clymer manual falls short of the OEM service manual. You really should get the OEM manual for your bike; best tool in the tool box. They are available on line and not expensive. Sometimes you can find the hard copies on Ebay for very reasonable prices as well. I am sure Thunderogue or another forum member would be happy to post the section with the sheet you saw in the video. That page is not in my OEM manual for the '05 FXTD but it probably can be found in the touring model manual.

 

[joel]

If you get on Facebook look for the group "Harley PDF Manual Trade & Share."
Join and look at the files section, I downloaded the repair manual for my bike from there for

 

[TheOriginalEasyrider]

Ummmm ... I never said I didn't have the MoCo service manual. I have it in hard copy and in PDF format. What I said was that I find the Clymer manuals to be better written, better illustrated and easier to follow. Before slamming me you might think to check to see if that info is actually in the OEM service manual available online. It isn't. Or if it is they have it very well hidden. I did find a few ETS tables online but no idea if they are applicable to my year/model. If you want to repeatedly hit me over the head insisting that I buy something you think I should have, just send me a check and I'll happily make those purchases.

Apparently HD no longer publishes the table listing resistance readings for various temperatures. Doing this sort of test appeals to the engineer in me but in reality, if I have to pull the sensor anyway I might as well just put in a new one and see if that fixes my problem.

There's no way I can see to test the temperature sensor beyond verifying that the resistance changes with the measured temp. This obviously works. Whether it gives some bogus reading at a particular temp remains to be seen. If so, this would be very weird behavior for this type of device (based on my many decades of experience in the electronics industry). Not saying that this scenario is impossible... just that it would be a very unique failure mode.

I found a youtube video where a guy did extensive testing of a defective ETS. It read correctly at a variety of temperatures but as the sensor cooled down the resistance readings were jumping all over the place. I'm not sure of the significance here since a high resistance reading would indicate a cold engine and my bike acts like it thinks the engine is hotter than it really is. For this type of failure the resistance would have to be in the < 400 ohms or so range. I'm guessing the correct ETS resistance for the temps where I'm having a problem should be around 1k ohm.

Monitoring the FP3 app readings before I start the motor each time should shed some light on what's happening. If the FP3 has a data logging feature I don't see it. But as a practical matter all that's important is what the ETS is reporting to the ECM at the precise moment I am trying to start the engine.

Of course another possibility (and perhaps a more likely scenario) is that the ETS is working correctly but is the wrong part for my particular motor.

Curiously, I just set up the sensors and the FP3 app reports that the battery voltage is 11.4v when in fact, the voltage is 12.9v. I hate it when a test instrument reports a bogus value.

 

[dolt]

OK apologies on the manual comment and callling "slamming" is pretty harsh; I did "wink" after the comment. Addtionall, in fairness, I did not know you were not in possesion of the OEM manual, you only commented on the Clymer manual. I so you have to let me off the hook on that one. I am happy to learn that you do have the OEM manual. If you ever get into serious drive train issues, I think you will find the OEM manual pretty handy.

JMHO but I suspect the resistance readings are pretty much the same from model to model and even if they aren't any table of readings you use, assuming the ETS is faulty, should show an anomaly. Sure, youj can replace the ETS without testing but a simple test, that works, won't cost $50. I don't agree that the only thing that is important is what the ETS is reporting to the ECM on start up. During dyno testing that I was observing, the tuner reflashed the ECM with some changes to the tune and suddenly the bike was running much worse with the flash. Mystified, the tuner looked at the recorded data and saw that the ETS had flatlined at 70*. ETS was replaced and the tuner was able to finish the tune.

I wasn't sure whether the FP3 collected data or not. I do know that the current iteration has a "Sensor Data" mode but not sure what the mode entails.

The PN for the ETS for your bike is 32466-99 which tells me that it is the correct part for your bike; has been used since '99 for all EFI models. Same PN for allTC touring and softail models since 1999, so you can set that thought aside.

I have never asked but has this hard start problem always presented? In stock triim before the FP3?

Anyway, apologies if I have offended; no disrespect or offense was intended in this or my previous posts as it is just not my nature. Good luck, I hope you find a solution.

 

[joel]

I know nothing about what or how to see data from FP3 tuners, do you use a diagnostic cable hooked to the laptop?
If so can you see graphs of the data from current or recorded data?
You can often see glitches of a sensor as temperature changes, engine vibration can be duplicated while using a screwdriver handle tapping on a sensor.
I need to run down to my shop so I'll bring my scanner back home hook up to my bike and see what I can do with it on sensor graphing,
I don't remember but have you confirmed all ECM related grounds are removed and cleaned.