Road King Fuel Pump Issue

[Hoople]

Not to sure about coil messages since I have a brand new coil, wires and plugs. And the new coil measured the exact same as the old.

If you boot the ECM with the Run switch in Run while having the system relay removed, you will immediately generate a 24 and 25 error code.

A 16 code can be caused by an intermittent ECM power relay. Inspect and clean/burnish relay pins. Also check max voltage reading across battery with engine running at 3500 RPM.

For 35 & 42 you may want to inspect the ECM connector. Use only micro-motion fretting lube on those pins.

 

[FLHPI]

Interesting findings this morning, thought I would bounce off of you all.
Inspecting the Voltage Regulator wiring I found a "Fix" from the previous owner. A chaffed positive wire from regulator to Main Circuit breaker which had been taped up. After removing the tape I can see that the insulation was rubbed off, but the wire is still intact, however I seen signs of corosion.

Service Bulletin:M-1099
Voltage Regulator Cables
(FL Models Only)
Chafed voltage regulator cables can cause low voltage
problems. To diagnose, visually inspect the entire length of
the voltage regulator cables for any signs of contact or
chafing. Inspect the cables for contact with the front and rear
edge of the lower front motor mount, with the machined
surface at the bottom of the crankcase (under the cam
cover), and with the bottom rear oil filler spout screw. In fact,
wherever there is possible contact with the powertrain or
road debris, further cable strap the cables as necessary to
prevent damage. For more information, see the 2000 FLT
Models Service Manual and/or Service Bulletin M-1077.

I Also performed Regulator Bleed test from the manual:
Unplugged stator connector, Connect one test light lead to ground and touch the other lead to regulator pins one at time, if light glows, replace regulator.

I do not have a test light, but used my meter instead. I am getting 12.5V on both leads of the Regulator. (Bike Off, Ign Sw off)

Stator Ground and Resistance checks were good.

My thoughts are to replace the regulator, 1.) instead of cutting, repairing, heat shrinking the damaged lead. 2.) It supposedly failed the bleed test

Could this be the cause of trouble codes? If ECM was getting bad voltage readings from regulator?

Thoughts?

 

[FLHPI]

Hoople, sorry I did not see you post until after I posted.

It is very possible that I booted ECM with relay taken out during my testing.
I have replaced both system and fuel pump relays with new ones. and inspected connectors. I have also inspected ECM connector, they appear to be very clean.
Male connectors are shinny at contact points and females are also clean and shinny.

 

[FLHPI]

I bothered me that I did not perform the Regulator test per the manual with a test light.
(I think I loaned mine out) I made one from a small automotive bulb and some clip leads.
Tested it across the battery.
Then I performed the Bleed test again. I do NOT get a glowing bulb from either lead on the connector going from the Regulator to the Stator. I do get +12.5VDC when I test those leads with the meter to ground. But I think this is a false reading as these should be the AC leads coming from the stator.
So maybe I am going down the wrong path with this one.

 

[Hoople]

A bleed test is used in conjunction with an output voltage test. Measure the output of the regulator at the battery during idle and also at 3000 rpm. If that test fails, the regulator is bad. If you pass that test, then perform the bleed test. If you fail a bleed test using a test light, it means one of the final diodes is shorted allowing AC to enter into the electrical system. (see attachment) If one of the final diodes is open, you won't have AC coming into the system, but you also will never have enough output to pass test #1.

Having AC come into the electrical system will cause all sorts of strange things to happen.
The HD bleed test "tries" to tell you if you have a missing output phase along with amount of AC getting through to the battery. The test is better than nothing at all but to perform a truly accurate bleed test you would need to view the waveform of the output using a scope..

Harley Davidson Community
View the picture I attached with my post. It shows what you are testing with a bleed test.

Here is the picture. Harley Davidson Community

 

[FLHPI]

Hoople, Do you have time for another question?

I went through the wiring with a meter, doing wiggle tests with the meter contacting the ECM connector (carefully) and each sensor per the schematics.
All continuity checks went well.

I was also checking for shorts to ground, by grounding one lead of meter to chassis ground and checking each connector on the ECM connector (unplugged from the ECM)

I have attached images with my shots of meter.

I was NOT getting OL (open loop) on pin 26 (white/blk wire) of the ECM, which is also Pin 86 of the fuel pump relay. I was not getting a BEEP for continuity, but was also not getting OL. See attachment.

I got this same reading checking IGN FUSE Gray wire to ground.

Trying to chase this down, I found that when I unplug the Bank Angle Sensor I get OL on both these test points. (White/Blk wire and Gray Ign wire)

I have two sensors and tried both, also tried the sensor in different positions.

Does this mean that the BAS is some sort of Resistor to Ground, which gives some reading, but not Continuity to ground?

Secondly, is this what you would expect to see?

I have gotten my hands on a second ECM. I am going to try swapping them out, but I wanted to go thru the harness one more time, looking for shorts.

 

[Hoople]

So at this time, does the bike perform ok or are you still having a problem with it dying. Entry point #29 said the bike was running. You had a few DTC but the bike was running. You did not know the reason why it was running. Seemed to fix itself but it was running.
So is the bike running and fixed and your just trying to understand the reason why?

You closed by saying: "I am calling today a success though, because it started/restarted and I rode it. Now just to continue trouble shooting until I get all the kinks worked out."

So what is the status?


As far as your question regarding Pin 26 of the ECM and ohm meter.
Getting a meter reading instead of OL may be normal. First remove the fuel pump relay and place the Run/Kill switch in the KILL position. You should now read OL (infinity) on ECM pin 26. If you do read OL, re-install the fuel pump relay. Do you still read OL? IF not, consider this normal. If you do read OL, now flip the Run/Kill to RUN. You should now get a meter reading which is normal.


As long as you read OL with fuel pump relay removed and the Kill switch in Kill position, your meter test for Pin 26 of ECM (black/white) tests normal.
If you don't read OL with kill switch in kill & fuel pump relay removed,, you have a problem.

Be aware it is risky to use the meter range your using for a test like this. Your using diode test/beeper position which is incorrect. This meter range (beeper/diode test range) uses a fixed 2K ohm range. You will read OL as long as the resistance is anything greater than 2K. That's not good. You can have a ton of leakage (lets say 3.5K of resistance) and you will think it is infinity when it is not. Use the ohm position instead. You will get a true reading in this position. The range (200,2k,20,200k 20meg) will be auto displayed in the window. Infinity on this meter can be considered to be anything greater than 19.99 meg ohms. Best way is to force a 200K range. This way OL is greater than 200K which is plenty, and you won't read skin resistance if you touch the probes or bike while testing.
Don't move on unless you understand this. Otherwise all readings you take can be incorrect or at least misleading.

Also, what is status of regulator output test. What is voltage across battery during idle, 1500 rpm,,2500 rpm, 3000 RPM. With a single phase set-up, it Would be nice to see something like 13.2 at idle and a peak of 14.4 (low to mid 14's) @ 3000 rpm.

 

[FLHPI]

Current Status is:
I can start the bike on first try, it runs, throttle response is good, idle is acceptable, little low, but doesn't die. I have not rode it out of the garage. That is where the good news ends. I do still have codes and an engine light (codes 24,25,35,16,41)
(front coil, rear coil, tach, bat volt, crank sensor) I have installed new: coil, wires, plugs and crank sensor)

Here's were it gets bad. When I shut it off after the first start the next start is very hard. Turns over fine, but takes so long to get it to fire. During which I get backfire through the exhaust and wheezing on the intake. Similar to the original problem if I let it sit for a period of time, hour or so. I can go back to a great first start, but as soon as I shut it off, subsequent starts are near impossible. Plugs are both fouled black.

The second part of the testing I was trying to accomplish was just trying to see if any wires had a short to ground, that I could not see after visual inpection. So my thought was to connect the meter to ground and start touching connections that should not be ground to see if I would get a beep. I just happened to notice a reading on the meter at one point. And yes when the switch is off or the relay is out, I do not get a reading. I was not using the readings for any diagnostics, I was just listening for continuity beeps when I saw something on the display.

Thanks for the info on the meter readings. What I was trying to accomplish with in the diode setting was just continuity. I put the meter on both ends of a connection and worked the wires back and for listening for the "beeping" to stop, to see if I had a break in the wires inside the insulation that I could not see. These tests all went well, I did not find any wires that had apparent breaks in them.

So this is kind of where I am at right now.
1.) The bikes starts on first try, runs decent in the garage, have not road tested.
2.) If I shut it off, restarts are almost impossible unless I let it sit for at least an hour
3.) Will not Start at all with my PCIII installed (it's ether blown or does not like what's going on currently)
4.) Checked for shorts (checked for broken Injector wire)
5.) New Coil, wires, plugs (although plugs are fouled black now, I clean them during testing)
6.) New Crank Pos sensor
7.) New in tank Fuel supply (was showing signs of tank rub, but had not rubbed through)
8.) New Voltage Regulator (had a previous wire repair that I did not like)
9.) Voltage is 13.6-8 (as high as 14) at high RPM, 13.0-13.2 at idle
9.) Engine Light is lit with codes (24,25,16,35,41)
10.) New Start\Stop switch and Bank Angle Sensor (because I had it from prev owner)(triple checked for pinches)

I did get my hands on a used ECM with exact same P/N, I am going to try that next, I just wanted to do one more check for shorts before installing it.

Other things on my mind: Cam Sensor, Eng Temp Sensor
Things that have been corrected since the original problem:
Fuel Pump works all the time, every time.

I hate to give up, but maybe it's time for Scanalyzer or Road Tech,
just heard so many horror stories about taking these issues to the dealer.

 

[Hoople]

I wish there was an easy way to clear codes so you could start with a clean slate You will have codes even if the problem has been corrected. Historical codes won't clear until 50 clean cycles have taken place. The check engine light means a code is current (happening now).
The cam sensor is a hall effect logic level switch You certainly can have a heat related issue with that switch. What it will do is allow spark firing on TDC of the exhaust stroke. Injectors will also fire 180* out of phase with the valve train. It would be near impossible to start the bike without having back firing through the exhaust & maybe carb.
I would heat up the bike to the point that you know a restart is going to be hard. Then hit the cam sensor with an inverted can of compressed air so it shoots the liquid out. Don't go nuts with it because the freon will take it to minus 20 degrees in a hurry. Hit the mounting plate and let the chill travel into the sensor if you can. If it's a thermal issue w/ the cam sensor, this may tip you off. There is also a meter check you can make to test the cam sensor.

Do not re-install any aftermarket box (PCIII) right now. Get the bike to perform correctly while in a factory bare bones state.

It would be nice to know what the fuel pressure was. Not that I think it may be the problem, but I would just like to remove that question from the mix. Those black plugs bother me.

If the spare ECM has no change on the way the bike runs, your not far away from fixing the root cause of this problem. I would conduct the cam sensor test.

 

[FLHPI]

Put everything back together tonight.
Installed the new (used) ECM
Started on first try, engine light went off.
Let it run for about 3 minutes.

Shut the bike down. Tried a second start right away, same thing as before.
Hard start, backfire, wheezing in the intake.
Now I get a Code 56. (Crank and Cam Sensor out of sink)

Since I have replaced Crank Sensor. the logical step is to try a Cam Sensor.

Trouble shooting Code 56 in the Manual has me check Batt Voltage for 12.6
Tested after the first subsequent failed start. Which was a few minutes cranking with no start.
(I get 12.6 with key off) (12.2 with ignition and run switch on, and headlight, etc.)
Maybe this battery is going bad? Charging system has been verified.
But it sure starts well on the first try after sitting.

Manual Also say to check for physical damage.
Not thinking this is it since the bike will start and run, just not subsequent starts.

Funny though, trouble shooting Code 56 in the manual never mentions Cam Pos Sensor.
But that was my first thought.

I will also inspect the cam pos sensor tomorrow. I can replace that.
The mechanics wire trick makes it very easy, I did the Crank sensor in less than 1/2 hour.

Maybe I will try a new battery, I have had the bike for 3 years and do not know when it was replaced.


As you said, I think I am getting there. Thanks again for the help. It really is appreciated, just to know someone is out there to bounce ideas and offer suggestions.