answer please

[XXSTAR]

I know how you feel.

did you try new gas and SEAFOAM this has help my FI troubles, I have not vbeen on a long trip yet, but I am leaving in a couple of days, I will let you know when I get back.

Ron

[MrBadExxample]

This evening I spoke with Tim Morris, probably the best Honda tech in my area. He didn't know of any failure involving the XX knock sensor or any related problem that creates a knock sensor fault. I asked if something like an out of balance counter balance shaft or something else could create a vibration that created a knock-like condition. He was doubtful it would. He also thought the OEM manual contained a procedure for testing the knock sensor. Basically he reiterated what was stated in this thread. In his experience Honda's fault codes are very accurate.

The other mechanic I spoke with, Bee Johnson, knew someone he could call and get the resistance and voltage parameters of the knock sensor. I'll check back with him in a couple of days.

[XXSTAR]

This evening I spoke with Tim Morris, probably the best Honda tech in my area. He didn't know of any failure involving the XX knock sensor or any related problem that creates a knock sensor fault. I asked if something like an out of balance counter balance shaft or something else could create a vibration that created a knock-like condition. He was doubtful it would. He also thought the OEM manual contained a procedure for testing the knock sensor. Basically he reiterated what was stated in this thread. In his experience Honda's fault codes are very accurate.

The other mechanic I spoke with, Bee Johnson, knew someone he could call and get the resistance and voltage parameters of the knock sensor. I'll check back with him in a couple of days.

ask him if the CCT could cause enough knock or vibration to light the FI.

I don't see how, but ask him if what I did could fix this. I changes the gas and run SEAFOAM through it, the light has not come on yet. I should know more after this weekend. Ron

[bajjer9]

This evening I spoke with Tim Morris, probably the best Honda tech in my area. He didn't know of any failure involving the XX knock sensor or any related problem that creates a knock sensor fault. I asked if something like an out of balance counter balance shaft or something else could create a vibration that created a knock-like condition. He was doubtful it would. He also thought the OEM manual contained a procedure for testing the knock sensor. Basically he reiterated what was stated in this thread. In his experience Honda's fault codes are very accurate.

The other mechanic I spoke with, Bee Johnson, knew someone he could call and get the resistance and voltage parameters of the knock sensor. I'll check back with him in a couple of days.

ask him if the CCT could cause enough knock or vibration to light the FI.

I don't see how, but ask him if what I did could fix this. I changes the gas and run SEAFOAM through it, the light has not come on yet. I should know more after this weekend. Ron

Dunno if this has any relevance to your issue but when i put the factory pro ignition advance on my 97 it became very sensitive to different gas sources. I ran seafoam and chevron on a regular basis. if i used shell gas it would knock everytime like clockwork. I dont understand why exactly but I was aware of it

[R1000]

[.......

Dunno if this has any relevance to your issue but when i put the factory pro ignition advance on my 97 it became very sensitive to different gas sources. I ran seafoam and chevron on a regular basis. if i used shell gas it would knock everytime like clockwork. I dont understand why exactly but I was aware of it

It was because the margin for knocking is reduced with an advancer. If the preignition is just on the margin, the engine will ping with low quality fuel.

[bajjer9]

[.......

Dunno if this has any relevance to your issue but when i put the factory pro ignition advance on my 97 it became very sensitive to different gas sources. I ran seafoam and chevron on a regular basis. if i used shell gas it would knock everytime like clockwork. I dont understand why exactly but I was aware of it

It was because the margin for knocking is reduced with an advancer. If the preignition is just on the margin, the engine will ping with low quality fuel.

Thanks, I understand the concept of pinging but did not understand why shell gas would cause it to knock. Shell is not cheap gas by any means...anyway my point was more or less along the lines of what you said...different gas or the addition of seafoam might be enough to eradicate the issue all together

[gharknes]

Till you get the problem fixed could you disconnect the wiring harness from the knock sensor and plug a resistor into the wiring harness. This is just a guess, but if the knock sensor is suppose to have a certain resistance till it detects pinging, this could work. Of course this will only be a temporary solution, if indeed the knock sensor is bad.

From another forum

--------------------------------------------------------------------------------

Originally posted by OffroadX:

I bypassed the knock sensor with a 560K-ohm resistor (normal spec is 500K-620K) earlier in the week so that when the sensor is replaced during the mod work, the before/after numbers wouldn't show any improvement due to the proper knock sensor reading. After a couple of dyno pulls with the resistor in place (and no trouble code in the system), took the resistor out and reconnected the knock sensor. Subsequent pulls showed no change in output, but the knock sensor code was back in the system.

Pretty conclusive proof that you're not missing a thing with a bad knock sensor code in the system eh?

FYI,

Brent

--------------------------------------------------------------------------------

Just an FYI, the knock sensor produces a voltage and sends that back to the ECU, which interprets a higher voltage as knock. The knock sensor test method, as you know, is to measure the resistance across the sensor, and hope for about 560 Ohms. If you simply bypass the sensor with a resistor (or simply unplug it), you *should* get this code:

P0327 Knock Sensor 1 Circuit Low Input (Bank I or Single Sensor)

If the sensor is bad, usually it will give this code:

P0328 Knock Sensor 1 Circuit High Input (Bank I or Single Sensor)

If you add a resistor into the circuit in line across the sensor, approx. 1 meg ohm by quick math, that *should* keep the voltage under approx. 2.5 volts, which is the ideal voltage per the service manual. A quick way to test the knock sensor is measure resistance from ECU pin 64, to ground, which is the same as measuring directly across the knock sensor, but insures all wires are properly connected and not shorted as well. To test it's actual operability, you need to measure voltage from pin 1 of the knock sensor (ECU pin 64) to ground. The Service Manual states approx. 2.5 volts is ideal.

As far as why you did not get a trouble code, I cannot explain.

Source (Good Read)

it is my understanding there are 2 different types Od KS, the one on the Bird does not have a ballast resistor, it reads OC on a meter,

could be wrong but I know mine reads OC

also as I mentioned before a guy with the same problem as me replaced the KS and it didn't fix it

[MrBadExxample]

I talked with Bee Johnson this afternoon. Here's what he said. A faulty knock sensor should not put the ECU into safe mode, which seems to be what your ECU does. Apparently the bike should run fine even with a faulty knock sensor. Check for continuity on the blue wire that runs from the knock sensor to the ECU. If that's not the problem, check all the grounds. Otherwise he said you might have a bad ECU.

[XXSTAR]

well just got back from a 2,400 mile ride. the FI light never came on, don't ask me what I did.

could have had bad gas. I don't think so.

I think it was the cold weather. I will know next fall.

looks like if it was the gas it would have run rough.

when the light would come on, it was in the morning, after it got up to say 65 it never came on.

the light was so sproratic I really don't know.

I do know it did not come on this trip., Ron

[airborneXX]

well just got back from a 2,400 mile ride. the FI light never came on, don't ask me what I did.

could have had bad gas. I don't think so.

I think it was the cold weather. I will know next fall.

looks like if it was the gas it would have run rough.

when the light would come on, it was in the morning, after it got up to say 65 it never came on.

the light was so sproratic I really don't know.

I do know it did not come on this trip., Ron

That's because we unplugged the light when you were busy playing around in room 123.

[gharknes]

I talked with Bee Johnson this afternoon. Here's what he said. A faulty knock sensor should not put the ECU into safe mode, which seems to be what your ECU does. Apparently the bike should run fine even with a faulty knock sensor. Check for continuity on the blue wire that runs from the knock sensor to the ECU. If that's not the problem, check all the grounds. Otherwise he said you might have a bad ECU.

Afaulty KS must put the engine in safe mode as it cannot function safely, the KS allows the ECU to maintain ignition advance right on the edge of performance at all rev levels and throttle positions, if the ECU is not getting a correct KS signal it has know way of knowing what the setting should be so it retards the ignition to a safe level, this results in poor performance, the KS is a fundamental component of all FI systems, the checks you mention I have already performed

All I need to know is the test routine the ECU performs to determine if the KS is good or bad, I do know that the test occurs at 3000revs but that's it, TBH I suspect that my ECU is faulty.

[gharknes]

well something strange happened

Went for a ride yesterday and instead of the usual FI light on after 3000 revs - it stayed off and it stayed off for 30 miles, I gave it loads and it still stayed off, thinking this can't be the bulb must be faulty I switched off and on and the light was working fine, stopped the bike at the hospital for about 4 hours came back out rode for 2 minutes and on it comes again and has been the same since

anyone manage to get the info I requested about the KS test ?

[gharknes]

Fixed

so far so good and if it comes back at least I know the area of the problem - what did I do ?

turned the ECU upside down, so wires and connections are laying a different way

as I suspected this is not a knock sensor problem as I checked and double checked the wiring, it is another undetermined signal from another sensor that is required for the KS to operate correctly.

[Digito]

I know its an old thread, but may be it helps others.

I have the same problem, dont waste your money in anything but the ECU.

Error 25 is not knock, its a fail in the system that detect knock, not knock per se, KS, wiring, or ecm, and, sadly, its always the ecm.

Honda Aquatrax from 2002 to 2007 share the same knock sensor and a very similar ecm, same problem, ecm.

By the way, when you have fi light on and error 25, ecm its in fail safe mode, may be you did not notice, but ecm retards timing at 9000 rpm and drop the power more than 20hp.

Check this out.

http://www.cbrxx.com/general-cbr-xx-discussion/28039-efi-error-25-another.html

http://www.cbrxx.com/general-cbr-xx-discussion/28063-fi-light-error-25-limits-power-proof.html

Hope it helps.

[redxxrdr]

Wow,

Old post . But information is always appreciated.

I was just thinking about riding with X1rider (Mike). And XxStar (Ron). I never met Brett, but still miss him.

And this post shows up.

Here's to old friends who never entirely leave us.

I hope that we ride together some day.

And to new friends. Thank you and welcome.

Edited December 23, 2015 by redxxrdr

[superhawk996]

sadly, its always the ecm.

I don't buy it. Knock sensors fail, sometimes just because they're wet and come back once dry. The wiring can have issues as well. If there's been no failure of a sensor or wiring on a XX ever then maybe it's something special.

As far as testing, I've tested them on cars. Put on a timing light and see where it's at, tap near the sensor with a light hammer or something and look for a retard. If it retards you know it's working. I don't know of an electrical test that can be done to know the sensor is working but I assume the computer looks for an acceptable resistance to determine that it's dead. With some cars the computer will force a knock during diagnostics and set a code if it doesn't detect the knock.

[mikesail]

sadly, its always the ecm.

I don't buy it. Knock sensors fail, sometimes just because they're wet and come back once dry. The wiring can have issues as well. If there's been no failure of a sensor or wiring on a XX ever then maybe it's something special.

As far as testing, I've tested them on cars. Put on a timing light and see where it's at, tap near the sensor with a light hammer or something and look for a retard. If it retards you know it's working. I don't know of an electrical test that can be done to know the sensor is working but I assume the computer looks for an acceptable resistance to determine that it's dead. With some cars the computer will force a knock during diagnostics and set a code if it doesn't detect the knock.

Nope, not resistance. The piezo unit has basically infinite resistance, it only generates a voltage with deflection of the crystal.

[superhawk996]

So does the computer listen to normal operation to know the sensor is working?

My only experience with on-board diagnostics and knowing how it worked is with old Fords. During diagnostics the computer prompts you to floor it (dynamic response test I think it's called) while the timing is advanced and it looks for a reaction from the sensor....at least that's what I understand of it. Maybe the old sensors used different technology or the new computers are programmed to monitor non-knock signals.

[mikesail]

So does the computer listen to normal operation to know the sensor is working?

My only experience with on-board diagnostics and knowing how it worked is with old Fords. During diagnostics the computer prompts you to floor it (dynamic response test I think it's called) while the timing is advanced and it looks for a reaction from the sensor....at least that's what I understand of it. Maybe the old sensors used different technology or the new computers are programmed to monitor non-knock signals.

I do not know how this ECU works, I suspect it looks to see some minimal noise level which would occur during normal operation to validate the sensor.