Intermittent ignition problem while cranking engine during starting.

[mikesim]

Your explanation is partially correct. However, the winding does not go open or infinite ohms. The heat creates a condition called "Temperature Rise" and the resistance of winding raises. The coil is supposed to have a given impedance range, for discussion for purposes lets says it's supossed to be 300 to 440 ohms. If the coil resistance is at for this example 480 ohms after soaking in the heat the ECU will be not seeing the correct voltage threshold levels from the sensor and simply ignores them. Thus creating the "No Start" connection.

Here's some light reading for those interested in knowing more about coils and temperature rise. Estimating Temperature Rise of Transformers

The Crankshaft Positioning Sensor or Pulse Generator is basically a transformer generating a signal for the the ECU.

Your theory doesn't explain why a CKP will function properly for many, many miles and then suddenly fail. It has had numerous heat soaks throughout the course of its life, why doesn't it fail every time it gets hot? If you examine the output of the CKP with a scope when it has failed, you will not just see a weak signal, you will see no signal thus signifying an open circuit in the device.

Mike

 

[Brillot2000]

Your theory doesn't explain why a CKP will function properly for many, many miles and then suddenly fail. It has had numerous heat soaks throughout the course of its life, why doesn't it fail every time it gets hot? If you examine the output of the CKP with a scope when it has failed, you will not just see a weak signal, you will see no signal thus signifying an open circuit in the device.

Mike

I am sure that there isn't enough collected field data on this particular sensor to make any sound determinations regarding the failure modes. No mechanic is going to use an oscilloscope to analyze the signal levels. The fact is that the mechanic is going to simply replace the part and chuck the old one in the trash. It's not going to be shipped to Japan for the Honda's engineers to do a failure mode analysis on the returned part.

The Pulse Generator is a magnetic device with a small magnet used in its construction. SWAG (Scientific Wild Azz Guess) would say that the magnet is weak and fails when heated for thousands of cycles. Magnets do stop being magnetic at a certain temperature. That would be my best guess on why there would be "No Signal" at the time the sensor is soaked with heat.

 

[Macka]

I have been discussing this thread with my friend who is a retired EFI Specialist.
Without having any data to go on, it is his educated guess that the Sensor MAY be going open circuit when it gets hot.
I have been wondering whether it may be the opposite condition and the winding MAY be shorting internally when hot.
The scenario that I imagine is engine oil eventually leaking into the Sensor as the sealing of the unit breaks down over time.
This MAY cause the insulation on the windings to swell and break down eventually resulting in an internal short circuit of the windings which could kill the signal output.
Both of the above conditions would cause the signal to stop.
The only way to confirm it would be to test the Sensor output when it is actually hot and causing a no start condition.
My friend prefers to test this type of sensor using Frequency or Duty Cycle as it is a more reliable indication of operation than voltage.
Also, you do not need a scope or "Honda" Adapter.
I do not know what a normal output would be but I will test it next time I have the bike out.

Macka

 

[ARKNT]

I doubt there has been much heavy testing on NT sensors that show this issue. It is, (at least starts out) intermittent and that by itself makes it difficult to get to the bottom of.
There could be manufacturing issues causing high resistance or high resistance to develop.
It could go totally open when hot.
As Macka stated you could also get a short in the windings or a partial short that only takes out some of the windings, thus reducing the signal strength.
It even entered my head of a slight possibility of the heat causing the crankcase to expand and increase the distance from the sensor to the high points of the crank rotor (reluctor) once again decreasing the output.
There is maybe even a heat related connector problem.
We may never know for sure.

Does anyone know if these same sensors are used on other bikes?

Brad

 

[Turbo DV8]

Since my ckp sensor failed under 16 thousand miles, Which is less than the maintenance interval mileage for some maintenance items, maybe the part number for the pulse generator should be added to the consumables list sticky!

 

[Macka]

I am not aware of anybody having to replace their CKP sensor a second time.
I replaced mine at 34,oookm (21,ooo miles) and the problem has never returned (now 93,oookm).
It seems that the UK owner have much less trouble with their CKP Sensors, possibly due to the much cooler climate.
It is uncommon for their maximums to exceed 30C.
As far as I know the part number has never changed but it is curious that the problem never seems to return.
I have not bought a spare but I still keep the old one just in case I need it in a hurry.
It was still working 99% of the time.
Generally, if Honda has a revision to a part (improved version or different supplier) the last digit on the part number will change ie: 000 to 001 or 002 to 003.
000 indicates the original factory part from concept to current supply.
I know for sure that this applies to car parts from Honda but am not sure whether this applies to bike parts as well.
I did check the magnet on my old unit yesterday and it is still quite strong.

Macka

 

[Turbo DV8]

Irony (definition):

Walking into my local bike shop this afternoon to order the pulse generator for my NT700V, and seeing for sale on the floor an NT700V! It must have been the OTHER Deauville Honda sold in the US.

 

[DirtFlier]

[...if Honda has a revision to a part, the last digit on the part number will change...not sure whether this applies to bike parts as well...]

Yes, it also applies to bike parts.

 

[Turbo DV8]

DirtFlier, you mentioned that the connector for the sensor is located close to the radiator, so it's easy to access. Just follow the wires up. I can see the connector with the flashlight in the mirror, but how in the Sam Hill do you get to that connector without removing the entire right side main Fairing, and all that entails? I for the life of me cannot see it, because all the plastic seems to be connected together and only removable if the entire right side fairing is removed. The connector is in front of the black inner shroud which is attached to the main fairing. And I cannot get to it from the front by the radiator. I can barely see the connector up there with a mirror and flashlight, let alone being able to reach it with my hands. Any suggestions from folks would be appreciated.

 

[DirtFlier]

I looked at the attached photo of when I did it and the right side fairing half is still on the bike but perhaps I'd removed some of the lower bolts so it could be gently pried outwards enough to get my hand on the connector. This is just a guess at this point since it's been so long since I changed that part.

 

[Turbo DV8]

That's the picture I had saved, and what I was going on when I estimated the severity of the job. But I just cannot see anyway to partially remove anything, and get to that connector. Any screws that would need to be removed to wiggle stuff around, would have to be accessed from areas that you can only access if the whole fairing were removed. I'll just bite the bullet and take the fairing off.

But Jesus... Methinks once this is done, next spring this bike is as good as sold. I bought a newer bike to forestall having to do too many mechanical repairs compared to my CX 650 Turbo. I find it totally unacceptable to have to disassemble half the bike with only 15,000 miles on it just to remove a connector to replace a part that should not have failed at such a low mileage. And all for a bike that I don't particularly enjoy riding in first place. Just venting...

 

[DirtFlier]

After removing two fairing bolts (one is lower and the other is central), gently pull the fairing outward to provide enough wiggle room to get your hands on the connector.

 

[Turbo DV8]

I removed the two 10 mm hex head bolts which hold the fairing to the frame, but that gained me nothing. The connector is behind the radiator, but in front of the expansive black plastic heat dam which goes behind the thermostat, and hugs the cylinder head all the way down. That is held to the main fairing from the inside with two Phillips head screws. They were somewhat of a son of a gun to get out from the inside, but once done, did give enough wiggle room, in combination with the two 10 mm bolts being removed, to get the connector undone. Thanks.

 

[Turbo DV8]

The right crankcase cover slid out easily without having to loosen the rear exhaust header.

The oil pump chain FreePlay is half an inch, which causes the heart to skip a beat when one considers that the specification for free play in the Honda CX engine oil pumps is only one tenth of an inch. But there is no way to adjust the free play, nor is there any specification for free play, so I guess it is good to go.

On the sensor itself, prior experience in the School of Hard Knocks has taught me to use a sealant on the wire harness inner rubber grommet, even though it isn't stated to do so. I have ThreeBond liquid gasket 1194, semi drying liquid gasket, which I will use on that, along with the crankcase cover. It appears that Honda used more of a black silicone sealant, so I'm hoping the 1194 sealant should do the trick.

 

[Turbo DV8]

I'm failing to see how the crankshaft position sensor setup can actually tell the computer the angle of the crankshaft. The sensor is just picking up pulses from that wheel with the 11 knobs on it. How does this convey the actual crankshaft angle? Or is it just an engine speed sensor?

 

[ARKNT]

It looks to be an engine speed sensor. All the knobs (reluctors) would allow the computer to receive a more accurate signal of speed.
The position I believe is received from camshaft sensor. It has one reluctor for front cyl and one for rear cyl.

Brad

 

[seagrass]

Turbo DV8, I am feeling your pain with this job but it is still much easier than replacing the oil feed line to the CX Turbo ;-)

Seagrass

 

[mikesim]

I'm failing to see how the crankshaft position sensor setup can actually tell the computer the angle of the crankshaft. The sensor is just picking up pulses from that wheel with the 11 knobs on it. How does this convey the actual crankshaft angle? Or is it just an engine speed sensor?

Crank "angle" sensor is a bit of a misnomer as you have surmised. It's reference pulses coupled with the output of the cam sensors does however inform the ECU of the relative positions of the pistons with regard to TDC.

Mike