| Author | Message | ||
     Timebandit |
Today I got the Low Fuel Lamp and the Check Engine Lamp on a cold start with a full fuel tank. Ambient temp = 52*F. I have been getting the LFL without the CEL for quite some time, at ambient temps from 40 to 70*F. I've always wondered why the LFL lamp will be on when I start the bike when it wasn't on when I shut the bike off. Today things were different. I got the LFL and the CEL at the same time. AT=52. The history for my bike is that the LFL comes on all the time, without the CEL, even at temps down to 40-45*F. Today the AT=52 and I got the LFL as expected, but the CEL came on for the first time. I let the bike idle in the driveway for a few minutes, then both lamps went out. I just pulled diagnostic codes. Only one code is set: "b 1005 LOW FUEL SYS ERR" SW 6.1 CAL 1.0 I've seen the threads that refer to fuel pump issues and Instrument Cluster issues, and I'm not sure which (if either) is the root of the problem. Some threads suggest that SW 6.2 solves the problem. Can anyone fill me in? I'd like to have all the ducks lined up before I file a warranty claim. Thanks! | ||
| Zac4mac |
I had an e-mail conversation with an anonymous Buell engineer a couple of years ago about this issue. I found my issue was caused by the thermistor sensor drifting out of its resistance range. He was unaware of that failure mode but told me of another caused by faulty firmware. End backstory. Two failure modes with different solutions - 1) LFL/CEL at startup, both go out around 151 dF(OpTemp) - IC's firmware continues to poll sensors after shutdown - replace IC with current firmware(6.2 for 08/09, 7.1 for 2010). 2) LFL/CEL come on around 151 dF - sensor drift, mine drifted high. Spec is 800-1400 Ohms, mine read 2200 Ohms at 40 dF - either replace pump/sensor ass'y(proven) or tap a parallel resistance to trim back in to spec(unproven as yet). Zack | ||
| Timebandit |
Thanks for the help, Zac. I wasn't aware that there were two failures modes. It looks like I had been experiencing a what you're calling Failure Mode 1: I'd perform a cold start with a non-full / non-empty tank and get the LFL whenever the ambient temp was "cool." For me, that's been the current weather pattern of 45-55 degrees. Starting the bike in the same conditions with a full tank would cause LFL+CEL lamps to come on and set DTC b 1005. The lights would go out after the system came up to temp. Today I had a similar occurrence: Tank was not topped off, so on a cold start I only got the LFL without CEL. AT=45*F. Rode 0.2 miles to refuel. Topped off with 3.86 gallons, which tells me there was 5.3 - 3.9 = 1.4 gallons in the tank when the LFL came on by itself. It shouldn't come on unless fuel level < 0.8 gallon. I re-fueled and restarted the bike at the gas station. Engine was still cold. The only difference is that the fuel was now topped off. On restart: LFL+CEL. It looks like I get the LFL whenever AT<55*F and CT<151*F, regardless of fuel level. The CEL only comes on when CT<151*F, when AT<55*F, AND the tank is topped off. To confirm this, today I rode 55 miles and stopped to refuel with 1.7 gallons of fuel. Immediately after refueling, I started the bike to move it from the pump to a parking space. CT remained hot, and I got no lights on the restart. I moved the bike to a parking space and went inside to hit the bathroom and came back out 3 minutes later. AT=55*F. After 3 minutes at AT=55*F, I restarted the bike: LFL+CEL+Coolant Cold. After a minute the system was back up to temp and the lights went off. My problem is that the AT<55*F and CT<150*F causes the LFL light to come on, regardless of fuel level. LFL+CEL comes on whenever the above conditions are satisfied AND the tank is topped off. It's funny that the CEL only comes on when the tank is topped off AND the coolant temp is cold. This has to be a software glitch. Looking at your description of Failure Mode 2, I'm thinking this could be these could actually be the exact same failure mode. 151*F is the temp at which the coolant temp reading switches over to COLD. It could be that whenever CT <=150*F the COLD condition is satisfied, and the LFL+CEL lights are erroneously triggered. The difference between me (Mode 1) and you (Mode 2) seems to be is that I haven't gotten the CT to drop below 151*F because my AT is too high. You can do that if you ride in colder weather. Of course, I checked the DTC again today -- no new codes. Only one in the list: "b 1005". (Message edited by timebandit on November 20, 2011) | ||
| Timebandit |
Trying to be a bit more concise, here are my cold start observations: AT<55*F and CT<151*F and tank not Full ==> LFL AT<55*F and CT<151*F and tank Full ==> LFL+CEL+DTC "b 1005" It would be interesting to see if I could ride in cold weather to lower CT<151*F and induce Failure Mode 2. My current AT=45*F isn't cold enough to do get the CT that low, no matter how I ride. (Message edited by timebandit on November 20, 2011) | ||
| Avc8130 |
I just started having similar problems with my 09 CR. The BIGGEST problem I have right now is no warranty so I am not sure what to do. Both the IC and the Fuel Pump are quite expensive to replace just because the CEL/LFL comes on. Any suggestions for an affordable fix? ac | ||
| Timebandit |
I see three options, ranging from the cheapest to the most expensive: 1. Cheapest: Ignore the lamps until you warm up the engine. (Ignore DTC.) 2. Next-cheapest: Hack the sender's voltage output. (It's not clear whether the amount of drift is temperature compensated. Compensating at one temperature might screw things up at other temps. I'd hate to fix the problem of a premature LFL at cold temps by eliminating an appropriate LFL at higher temps.) 3. Expensive: Replace IC and/or fuel pump assembly. (I really think that this is an IC issue. What they really need to add is a map of resistance vs. temperature.) | ||
| Zac4mac |
The main difference in the 2 failure modes is #1 is generated from historic data and #2 is real-time. If you have a mode 2 failure, LFL/CEL after reaching OpTemp, a trimpot should work fine. The thermistor is linear, so you "shift the curve" back to where it's supposed to be. A 1/4 watt should do, there's virtually no current. Otherwise Timebandit has the options sussed. Zack | ||
| Lrholy06 |
Hey Zack, I was looking through Mouser's website for a trimpot that would work and got a little worried that I might pick the wrong one. Just typing in trimpot alone you get nearly 300 results. Are we looking for a SMD resistance trimmer or a through hole one? Do these need to be RoHS compliant? I dont imagine that it matters much for what we are using it for but I would like to try it. (Message edited by lrholy06 on November 21, 2011) | ||
| Timebandit |
My sensor was working properly at "high" temperatures, and I have recently experienced what looks like sensor drift at "low" temperatures with the recent weather change. My problem is that until we have a warm-up, I can't confirm whether or not the situation returns to normal with increasing temperature, or the drift remains when the temperatures rise. Not having the second data set makes it impossible for me to be certain that adding offset wouldn't result in proper calibration at one temp and miscalibration at another. If that's the case, then the net result would be trading-off a Type 1 Error for a Type 2 Error. Obviously, that wouldn't be good. I need more sample data to really know the answer. I'd be really interested in knowing how it works out for you guys who try trimming the sensor Z. Please keep us posted. In the interim, I'm going to go for the expensive fix because my bike is still under warranty. Thanks for your help! It gives me lots of helpful information to present to the dealer. | ||
| Zac4mac |
@Lrholy06 - I would use a "thru-hole" myself, it will be larger and easier to use. RoHS is only important if you want to go into production... no big deal to a "hobbyist" although components that are NOT compliant are hard to find anymore. I still use 63/37 lead-tin solder on my bench, I dislike the lead-free junk. @ Timebandit - just because you don't get a CEL/LFL doesn't mean the sensor is OK, just that you're still in the "accepted" resistance to avoid a CEL. e.g. say at 65 dF you had 900 Ohms when new but the thermistor drifts high. Now you might see 1100 Ohms at 65 dF. Still in the "accepted range" of 800-1400 but when you go cooler, now you're "out of range". I expect the resistance range for the thermistor will remain close to the original 600 Ohm sweep, just dislocated. If so, the trimpot only shifts it from a 1600-2200 Ohm sweep back to an 800-1400 Ohm sweep. Only way to check is subject the sensor to a similar temp sweep used for the spec of 800-1400 and measure. That would require a Tenney(environmental chamber) big enough for the bike, or pull the pump/sensor and measure it an a smaller unit. Worst case scenario, you readjust the trimpot occasionally to keep proper function and display. Zack | ||
| Timebandit |
Update: I took the bike to the dealer for warranty service. HD has decided that I'm going to get a new instrument cluster that will have to be loaded with SW 6.2. I was told that it's not possible to reprogram the old IC, which has SW 6.1, with SW 6.2. It sounds like you get one shot at burning the SW into the IC, and any reprogramming has to be done on a fresh IC with blank PROMs. Parts are ordered today. We'll see how long it takes to get them. Thanks for the help in identifying the nature of my problem. It felt pretty good going into the dealer knowing what to expect. (Message edited by TimeBandit on November 29, 2011) | ||
| Njdevils1990 |
I just started having this issue on my 09 cr about another ago. It seems that the CEL&LFL come on (at the same time each time) about 3-4 minutes after the bike has been started, usually in temps under 55, 60 at highest. It then will go off after about a minute or two of riding. I have also noticed that the fuel pump does not "prime" it's self when I initially turn on the key and kill switch like it should. I will take a video and post what I mean. I also have the same code b 1005. In the electrical diag. book it says either a short/open circuit from the IC to the fuel pump connector or the IC is bad or the fuel pump is bad. I think it's the fuel pump but I'm not 100% as its only occurred in the colder temps and I have not properly diagnosed it as I dot really know how to use a multi meter correctly lol. | ||
| Njdevils1990 |
Here's my video I just took it sounds like the pump isn't too happy can I get some feedback on this? What do you guys think? I'm taking it in to Brian's hd/Buell for diagnostic work to see if they can narrow it down to a bad pump or IC. Video http://i70.photobucket.com/albums/i118/Nirvana1221 1990/13b08c53.mp4 | ||
| Timebandit |
Update: New IC fixes problem. Took 14 days to get fixed. (Message edited by timebandit on December 09, 2011) | ||
| Zac4mac |
@Njdevils - go into diagnostic mode and check your fuel pressure, that's the best way to see it's condition. Fuel pressure should be 380-415 kPa when running and steady at around 500-517 kPa at key-on engine-off kill switch-on. From my own experience(I'm on my 3rd pump) the static test will show issues first. @ Timebandit - I believe it was Erik who told me they wanted a programming port on the IC but HD said too much $$, so the IC has to be opened up and re-flashed at the manufacturer. Zack | ||
| Two_seasons |
One of the better threads in the 1125 section. Updates are important. Thanks. | ||
| Timebandit |
> "I believe it was Erik who told me they wanted a programming port on the IC but HD said too much $$, so the IC has to be opened up and re-flashed at the manufacturer." By Erik, do you mean Erik as in Erik Buell? Man, I know I'm sticking my neck out to say this, but if Erik wanted a programming port on the IC, that says that even though he's a great mechanical engineer, he's a really piss-poor software engineer. There's no reason for a re-programming port on an instrument cluster, unless it is your intent to ship the instrument cluster with software that you know to be defective, and postpone worrying about fixing your bugs until later on. For Pete's sake, this isn't a piece of free personal computer software we're talking about, where people will accept a piece of crap because it's free -- this is an instrument for a motor vehicle. Motor vehicle instruments aren't supposed to need programming updates to work right. They're supposed to work right when you buy them. Instead of asking for a programming port, a better approach would be to get the software programming done right the first time, so that no future updates are necessary. It's not like this is all that complicated. This is an ongoing problem with poorly-skilled computer software programmers, who act as if it's OK to ship something that doesn't work now and fix it later. Unfortunately, that poor excuse for a work ethic, and the poor work output that results from it has become so commonplace in the computer software industry that people are now beginning to accepting it as the norm. Now this crappy design standard and the acceptance of mediocrity is finding its way into motor vehicles. That's just not acceptable. Today, most software developers' standards are unacceptably low. In their mindset, it's OK to ship crappy, bug-ridden software to meet an early release date, and worry about fixing the defects later on. The RIGHT way to do it is to do it right the first time. If you do that, there's no reason to even consider needing a programming port on an instrument cluster. Maybe I expect too much -- I design medical instruments, where you have to have your shit together or people will die if you screw up. It has to be right the first time. Period. It would be nice if the guys who design instrument clusters for motor vehicles took their jobs as seriously. | ||
| Timebandit |
> "I just started having similar problems with my 09 CR. The BIGGEST problem I have right now is no warranty so I am not sure what to do. Both the IC and the Fuel Pump are quite expensive to replace just because the CEL/LFL comes on. Any suggestions for an affordable fix?" The out-the-door price for my IC replacement was just under $300, or $275 for the cluster plus tax. That's what my dealer was paid by H-D. I expect that if a customer took their bike in for an IC replacement, there would probably be a labor charge for the IC replacement. If the IC comes with the software update already installed, then you could probably DIY the replacement for about $300. | ||
| Two_seasons |
Timebandit, you better not mention your BadWeb screen name when interviewing with Microsoft...  jk | ||
| Zac4mac |
@Timebandit - yes, that's the "Erik" I meant. Being able to flash the IC in the field would be a good thing. As it stands, the ECM can be done that way; mine got at least 4 different maps flashed over the first 2 years. BTW - all the re-flashes were "no-charge" and took 15 minutes apiece. I used to work in R&D at a medical device company. FDA testing would take a lot of time and effort, and still didn't ALWAYS expose any inherent flaws... Early released product got heavy forensic scrutiny when failures in the field occurred. I don't miss the FDA or HR one bit. I built prototypes under a microscope for 10 years, now I fix welders and cranes. With the more complex medical equipment like electro-surgical generators, there were Ethernet connections for remote diagnostics and to flash updates to the software. As far as a simple/cheap fix for LFL/CEL, have you tried a trimpot yet? That's as cheap and simple a fix as I can imagine... Zack | ||
| Timebandit |
Zac, maybe you forgot this -- I haven't had any need to try the trimpot (Fix #2) because I went with Fix # 3 -- new Instrument Cluster (IC). Now that I have the new IC, I have no more problems with in appropriate idiot lamps on a cold start. The root of the LFL + CEL problem has been confirmed to be a software defect. New software fixes the problem. The only hard part about getting the fix is that the instrument cluster cannot be programmed in situ. You have to buy a new one. Although the new IC is not the cheapest way to approach the problem, I would recommend it for anyone who wants to expend the minimal amount of brain time to obtain a guaranteed result, which also happens to be the definitive solution to the problem. For someone who values their time more than the $300 cost of the new IC, this is the definitive fix for the problem. The trimpot idea is a really good one. I can see where it would be particularly appealing to guys who are out of warranty and don't want to spend the $$$ for an out-of-warranty IC replacement. There's another potential fix that we haven't talked about: see next post. | ||
| Timebandit |
> "I see three options, ranging from the cheapest to the most expensive: 1. Cheapest: Ignore the lamps until you warm up the engine. (Ignore DTC.) 2. Next-cheapest: Hack the sender's voltage output. (It's not clear whether the amount of drift is temperature compensated. Compensating at one temperature might screw things up at other temps. I'd hate to fix the problem of a premature LFL at cold temps by eliminating an appropriate LFL at higher temps.) 3. Expensive: Replace IC and/or fuel pump assembly. (I really think that this is an IC issue. What they really need to add is a map of resistance vs. temperature.)" To that list, I should probably add another option: 4. Really Labor Intensive: DIY re-program the Instrument Cluster. As Zac mentioned, the IC doesn't have a reprogramming port. Reprogramming of the IC would require disassembly. To do this, someone with a little hardware/software knowledge would need to disassemble the cluster and reprogram it on the benchtop. The easiest way to do this would be to diff the code between an old/new instrument cluster, but it would also be do-able if you only had an old IC, a good knowledge of hardware programming, and the willingness to spend the time reverse engineering the solution. Hopefully there are PROM or EEPROM inside that could be replaced/updated with a software patch. This method would be a bit of work. It would only be worthwhile to someone who already knows how to do this sort of thing and enjoys tackling a niche market problem. If there are enough interested people with the LFL + CEL problem, there could be a niche market for selling an IC rebuild service that makes the problem go away. | ||
| Njdevils1990 |
Just checked pressures in the diagnostic mode Key on (engine off): goes up to 500-505kpa and then drops as it is only priming the system Engine idling: ranges from 390-405kpa So I'm thinking it's an instrument cluster issue. | ||
| Froggy |
Or you know, you may have an idea for some additional functionality in the future like say, a gear indicator or voltage display. Personally I would love the ability to reconfigure it, maybe so I can ride in diagnostic mode without disconnecting the VSS, or maybe so I can see instant fuel consumption at the same time as something else like the MAP reading. | ||
| Njdevils1990 |
So figured out the pattern on my bike with code b 1005. If the coolant temp is higher than 151 degrees and air temp is lower than 55 I get a cel and low fuel light. But if the air temp is higher than 55 the cel and low fuel go off. Only happens when the bike is warmed up and seems to go off also with the air temp being above 55. Any thoughts? | ||
| Timebandit |
> "So figured out the pattern on my bike with code b 1005. If the coolant temp is higher than 151 degrees and air temp is lower than 55 I get a cel and low fuel light. But if the air temp is higher than 55 the cel and low fuel go off. Only happens when the bike is warmed up and seems to go off also with the air temp being above 55. Any thoughts?" I have some thoughts that are on-topic. First, I'd like to ask you to tell us your IC software number and calibration number. If we're going to figure out how the IC software behaves, it's important to correlate the malfunctioning behavior with the software revision/calibration numbers. Second, Your observations seem quite similar to mine; the wrong lights come on when the ambient air temp is cold (below 55*F), and they don't come on when the ambient air temp is warm (above 55*F). The difference seems to be that I experience the problem when performing a cold start (coolant temp below 150) and the problem goes away when the fuel temp warms up, while you seem to be describing the problem when the coolant is warm (coolant temp above 150). Is that right? | ||
| Zac4mac |
Bob- your last line is describing the two failure modes. Yours is firmware related while Njdevil's is defective hardware. Yours was fixed with a new IC with Rev 6.2 firmware. Mike needs his IC to see 800-1400 Ohms from the fuel sensor, not the Out of Range values it IS seeing, for his lights to go out. Either by replacing the pump/sensor, by adding a smaller resistance in parallel with the sensor or ignoring it as it doesn't do anything but turn on a light. Zack Also - Mike/Njdevils - I think the pressure is supposed to hold at 500-517 on the static check. As my pump faded out, it would drop pressure faster and faster, until actual failure when it wouldn't hold while running. Keep an eye on your pump. (Message edited by zac4mac on December 11, 2011) | ||
| Timebandit |
Zack, thanks for being so patient in explaining the difference between the two failure modes. Somehow I managed to misinterpret what you wrote the first time. I've got it now. | ||
| Njdevils1990 |
2nd code came up yesterday P0628 Fuel Pump circuit low and still have B1005 im taking her to harley on the 8th so im hoping its just the pump but preparing to have to get the IC replaced also. | ||
| Zac4mac |
Mike - New pump assembly should fix you up. At the end of all the stuff in Diag Mode is the firmware rev. If you have 6.1, yes they should replace it. | ||
| Timebandit |
I put off having the IC replaced until end of season because I wanted to keep riding the bike without interruption. I had the work done in December 2011 after I decided to put the bike up for the winter. I brought it straight home without any extensive testing and put it in storage. We had a couple of unseasonably warm days in January, so I went out for some 3-hour backroad joy rides in the 50-F weather. Much to my amazement, the bike now seems much smoother in the low RPM range. Now it's much easier for me to pass through a town and keep a steady 35 mph on the speedo without engine surging. I've also noticed that the bike just seems to run more smoothly. I can't explain those observations. I had the latest fuel map, M3HUS142, for the entire riding season. It was installed with the Program 507 harness upgrade. For reasons that I can't explain, the bike seems a lot smoother now that I have the new instrument cluster, especially at low RPM. Maybe there's a new map that I wasn't aware of. Maybe it's just the cold air. Maybe it's something else that I just can't identify. Too early to tell. (Message edited by timebandit on January 14, 2012) | ||
| Kenm123t |
Programming Ports exist for this reason owners change the set up of thier bikes! Motorcycles are not med equipment Epa mandated the Emmissions profile of the engine. Most riders will change that based on modifications. |
