| Author |
Message |
    
Etennuly
| | Posted on Saturday, July 24, 2010 - 01:39 pm: | 
|
For you guys just getting on board with this problem, there are threads going back over two years about the topic. The only thing to date to make the bike go down the road over 70 mph on a hot day has been this fan switching. Many of us have changed ECMs and a book full of other things to attempt to fix it, many times with temporary success. It is the temporary success that has been so confusing.
Thanks for joining in. |
Jlnance
| | Posted on Monday, July 26, 2010 - 07:17 am: |
|
Thanks for having a thread waiting for me. |
Ksc12c
| | Posted on Monday, July 26, 2010 - 10:48 am: |
|
"I cut what is the ground wire from the ECM to the fan motor(black/orange wire). I use a ground through a switch, on the wire to the fan motor. The wire out of the ECM is not involved, it is taped back".
So the Fan circuit is isolated from the ECM? I wired my override switch a little differently. I just spliced one side of the switch into the Black/Orange wire so the ECM is still part of the equation. If my circuit works, wouldn't that eliminate the ECM as part of the problem and identify it as an overheating issue? I've only ridden the bike once since installing the switch so I can't really confirm all my Skip/Spark problems have be resolved. |
Jlnance
| | Posted on Monday, July 26, 2010 - 12:40 pm: |
|
ksc12c - No, you've taken the fan current out of the ECM as well. It is going to flow through your switch to ground, rather than flowing through the switching transistors in the ECM, and out through the ECM ground. |
Jlnance
| | Posted on Monday, July 26, 2010 - 10:28 pm: |
|
Went for a ride tonight. The bike was flawless. I haven't changed anything. I see why this drives people nuts. But I'm not going to complain about a great running bike. |
Etennuly
| | Posted on Monday, July 26, 2010 - 10:52 pm: |
|
 U A lucky man! How long have you been fighting this problem? Like a couple of weeks?  
I hope no one else has to have it happen without catching this thread. |
Jlnance
| | Posted on Monday, July 26, 2010 - 11:05 pm: |
|
A month, but it's only happened twice. The first time I thought it might just be a fluke. After it happened again, I knew it was something "real." |
Garrcano
| | Posted on Tuesday, August 10, 2010 - 02:08 pm: |
|
Signal/Voltage on fan during low speed.
Vertical scale: from -2 VDC to +14 VDC
Horizontal scale: 10ms per division
So the low speed is a pulsing between 8.2 VDC (during 15ms) and 12.4 VDC (during 10 ms).
On the high speed there is no pulsing, only 12.4 VDC (battery was 12.7 VDC, so I supose a voltage drop on the gates of the C-Mos(?) output). |
Garrcano
| | Posted on Tuesday, August 10, 2010 - 03:26 pm: |
|
Ok, assuming that the ECM-output only checks if it stays high or low in each moment (as description service manual page 4-80) an option could be this:
I dont't know if the impedance of the relay is important for the output in the ECM.
I haven't tested it yet, so be careful. You will use this on your own risk and I deny any responsibility !!!!! |
Etennuly
| | Posted on Wednesday, August 11, 2010 - 08:56 pm: |
|
 Wow! Ya done gone an went super electro-techno on me! Graphs an charts an all! How is a body man to keep up?
I gotta go paint something so I can think straight. |
Ksc12c
| | Posted on Thursday, August 12, 2010 - 10:38 am: |
|
So, the relay would trick the ECM into thinking the Fan Motor is operating correctly when in "Assisted" mode? Is the pulsing on Low speed the cause of the Skip Spark? |
Od_cleaver
| | Posted on Thursday, August 12, 2010 - 12:17 pm: |
|
Garrcano,
That is a lot of electrical noise on the signal you measured. I wonder if the ECM is getting screwed up because of all the voltage transients that it generates controlling the fan motor at the lower speed.
Did you have the other lead of your scope probe on a good ground nearby?
You must be measuring fan terminal 2 (Bk/O) to ground, correct?
Do you have a scope picture of this signal with the fan running at high speed?
Do you plan to install your relay circuit? If you do, PLEASE tell me if the CEL comes on when your switch is in the “Assisted” position.
Sorry for all of the questions, but my goal is to remove the fan and not have the CEL come on. Thanks
All (for comment to correct me),
I have hated this fan from day 1 and would like to eliminate it entirely. Yes, I agree with others that it helps to cool the rear cylinder but with the right hand scoop is it needed? I base this on the following:
1. When the fan isn’t running, it will impede airflow. Remove the fan and the airflow at speed will increase.
2. Per members of this board, with the comfort kit ECM flash on ’08 and above the fan does not run on shut down,. >> Oil coking in the rear head at shut down must be marketing BS.
3. Jinance inadvertently ran his bike without the fan and had excellent results >>> “I think I did 900 miles. All with no fan in 90+ degree temps, pushing the bike hard. NOT ONCE DID THE BIKE GO INTO SKIP SPARK MODE. “ |
Garrcano
| | Posted on Thursday, August 12, 2010 - 12:32 pm: |
|
Ksc12c: The pulsing makes run the fan at a slower speed. IMO nothing to do with the run/Skip which drives the rear spark plug.
I hope the CEL doesn't come in assisted mode.
(Message edited by garrcano on August 12, 2010) |
Garrcano
| | Posted on Thursday, August 12, 2010 - 12:46 pm: |
|
That is a lot of electrical noise on the signal you measured. I wonder if the ECM is getting screwed up because of all the voltage transients that it generates controlling the fan motor at the lower speed.
Did you have the other lead of your scope probe on a good ground nearby?
No. The noise could come from ignition, cam position signal, power supply of the scope, or a mixture of all
You must be measuring fan terminal 2 (Bk/O) to ground, correct?
No, I'm measuring the fan voltage, between battery positive and fan terminal 2 (Bk/O)
Do you have a scope picture of this signal with the fan running at high speed?
No, I've made the measuring, but I didn't took any picture because it was a 'noisy' straight horizontal line at 12.4 VDC
Do you plan to install your relay circuit? If you do, PLEASE tell me if the CEL comes on when your switch is in the “Assisted” position.
Not at this moment. I don't have any spare ECM and I need the bike |
Garrcano
| | Posted on Thursday, August 12, 2010 - 01:40 pm: |
|
Sorry for all of the questions, but my goal is to remove the fan and not have the CEL come on. Thanks
IIRC you can mask with ECMSPY the errorcode 36. So the CEL won't come on without fan.This is also an option for Etennuly and his switch.
(Message edited by garrcano on August 12, 2010)
(Message edited by garrcano on August 12, 2010)
(Message edited by garrcano on August 12, 2010) |
Garrcano
| | Posted on Thursday, August 12, 2010 - 02:11 pm: |
|
Ok, second try, I've changed the three position switch (no need for OFF) and added the Etennuly-Switch (S2, more helpful IMO).
I accept any type of suggestions for improving. |
Etennuly
| | Posted on Thursday, August 12, 2010 - 05:02 pm: |
|
What do you think of my plan to install a 'psudo' fan to let the ECM think it is functioning normally?
I am thinking about putting a fan at the oil cooler, wired as if it were the factory fan. The ECM should "see" and be able to control a load on it's circuit as if it were running the original one .....right? Not that it would matter if it ran at all, but the ECM should be tricked into a "normal" fan condition and not light the CEL. Besides, this would be cheap and easy to do. |
Garrcano
| | Posted on Friday, August 13, 2010 - 10:01 am: |
|
It's an other option, but it only works if the ECM checks high and low state on the output.
If it detects amount of load, you will need a matching fan (and my version in this case will not work). |
Etennuly
| | Posted on Friday, August 13, 2010 - 02:29 pm: |
|
I was wondering if matching it would be the case. No problem. Buy a spare and store it in a position right by the oil cooler and hook it up to the ECM.
I bet that would be a noisy deal though! Every you stop you would probably be subjected to: "What is THAT noise?!" |
Garrcano
| | Posted on Saturday, August 14, 2010 - 06:31 am: |
|
Mounted yesterday the circuit and took an 10 miles testing tour:
a. All works fine.
b. Something similar to the low speed is still there. First I thought "nice", BUT what happens is an oscillation in the contact of the relay due to the the low speed modulation of the ECM. This oscillation IMO will reduce the life of the contacts in the relay. So if anyone has a suggestion to eliminate this pulsing in the relay or has an idea to use a solid state relay, it would be appreciated. |
Od_cleaver
| | Posted on Saturday, August 14, 2010 - 12:28 pm: |
|
Garrcano,
You seem to be schooled in electronics. So, I have some thoughts for you to consider:
1. You show a diode around your relay coil to handle the inductive voltage. Buell would have put a diode in the ECM or the inside of the fan motor to handle the inductive voltage spikes (or BEMF) when the fan motor is switched off. You should add a reverse polarized diode across the motor or confirm Buell’s design. If the diode is in the ECM, you will have a lot of arcing across your relay contacts. They will not last long without this diode.
2. If you want to stay with your relay design, you could control the relay coil with a PNP transistor (emitter to 12V, the collector to the coil and the other side of the coil grounded – keep the diode across the coil). The ECM would connect to the base of the transistor. Part of the base circuit would be a resistor to 12V and a capacitor to ground. The RC time constant would be used to delay the turn off of the relay. Your scope picture shows an off time of 10msec. This would make the R and C values reasonable.
3. You could toss the relay and use a PNP power transistor to control the fan. The benefit would be similar to the relay. The high fan current would be kept out of the ECM and potential ground shifts would be removed.
Hopefully that makes sense. If it doesn’t, feel free to ignore my ideas. PM me if you want me send you a drawing of the circuit. Words sometimes escape me.
Another question for you:
Since you tested your idea (I applaud you), did the CEL turn on with the switch in the assisted position?
If it did not, then the ECM must not be looking at fan motor current. It must be looking at the voltage level at pin 6 to decide if it should post an error. |
Garrcano
| | Posted on Saturday, August 14, 2010 - 01:11 pm: |
|
You seem to be schooled in electronics.
Yes, but it´s many years ago and without practise.
1. You show a diode around your relay coil to handle the inductive voltage. Buell would have put a diode in the ECM or the inside of the fan motor to handle the inductive voltage spikes (or BEMF) when the fan motor is switched off. You should add a reverse polarized diode across the motor or confirm Buell’s design. If the diode is in the ECM, you will have a lot of arcing across your relay contacts. They will not last long without this diode.
I supose you mean the red diode. Location is because it's easier to implement at the pins of the relay (betatesting phase).
+++ jpeg +++ 587179 +++ Fan control fifth try +++
Since you tested your idea (I applaud you), did the CEL turn on with the switch in the assisted position?
No CEL on assisted mode.
If it did not, then the ECM must not be looking at fan motor current. It must be looking at the voltage level at pin 6 to decide if it should post an error.
That's what I was hoping out of the interpretation of the service manual, because I didn't had an idea how to fool the ECM for load controled error.
(Message edited by garrcano on August 14, 2010)
(Message edited by garrcano on August 14, 2010) |
Garrcano
| | Posted on Saturday, August 14, 2010 - 01:52 pm: |
|
Ok here the scheme
|
Jlnance
| | Posted on Sunday, August 29, 2010 - 05:22 pm: |
|
Very cool oscillographs  |
Darth_villar
| | Posted on Wednesday, October 13, 2010 - 02:54 am: |
|
Any more input on if these fixes worked?
My bike has the axtell 88" kit, ceramic coated headers, heat shield inside the frame, and comfort kit.
My fan turns off almost immediately after short spirited rides, but when I get up to 76mph sustained for a few minutes, I get the skip spark.
I thought it was just the AZ heat, but tonight it was 80 degrees and I got skip spark. Which leads me to believe that it is not heat related, but more of an electrical problem somehow... I need to datalog a ride and get the actual observed engine heat from the sensor, maybe sensor data is being corrupted?
Phil
p.s. Oil cooler is clean and clear. |
Darth_villar
| | Posted on Thursday, October 14, 2010 - 08:14 pm: |
|
Interesting development, hooked the bike up to Direct Link, and the engine temp was jumping all over the place. That can't be normal...
Going to replace the sensor, see if that helps at all.
Phil |
|
