Knock sensor again...

[silverfox.xx]

I now realize why the ks wire is basically unshielded from the ks to the ECU.

i cannot comment on the possibility of the alarm, no experience on that.

On 1999/2000 models the KS wire though of a shielded type from the ECM to the connectors under the rubber boot, pre the KS connector.. this shield is not earthed.
On 2001 pre cat version and thereafter cat versions, the KS screen is earthed. Hence I did mine.

My KS is actually Ian Robbos, he has mine, both have done reasonable mileage (mine near 50k miles).. I have never heard on here anyone changing a KS doing a shits worth of good/difference, hence I don't think its that. I can assure you that my bike started to sound a bit rough/cam chain rattle/clutch basket rattle caused by slight uneven running on tickover, well before the KS light became a permanent feature.

Your are most welcome to borrow the KS I have spare, its one I got from a scrap yard, and its worked ok on Ians and my bike. If it works you can have it for the ?'s I paid (think it was ?35, but I'll check).

Please try running the bike with the KS disconnected, then reset the fault code after.

The ECM when going faulty seems to eliminate some of the rough running, and whatever goes phutt in it does so gradually. I believe that the only reason you eventually get a KS code then is that when the ECM goes through its check, the signal in the ECM is broken, so it throws up the fault code 25. I appreciate you have not got to that stage yet.

My email is cwroper@hotmail.co.uk, if you still want to forward your address, or my son in Appleshaw near Andover is down next week end and he could bring it back with him.

 

[smedly]

mine is a 99 bird and the KS has always been earthed, also confirmed on the circuit diagram, not sure why yours (silver) was not earthed, it was a screened cable.


PS the alarm comment was just kidding ????

 

[silverfox.xx]

mine is a 99 bird and the KS has always been earthed, also confirmed on the circuit diagram, not sure why yours (silver) was not earthed, it was a screened cable.


PS the alarm comment was just kidding ????

No mate the KS wire screen was not earthed, mine was as per Haynes manual for 2000 model. :dunno: I have to say though that the engine/frame number relates to a 1999 model, yet first reg'd in 2000. Could explain it !! On mine originally the screened ends were just taped up.

 

[smedly]

mine was 100% earthed without me touching it, in fact I honestly couldn't tell you where it is earthed too but it is earthed, one of the very first things I checked and constantly checked as I scratched my head for 2 years over fault 25. but there you go can't explain it, oh and I'm going by the Honda service manual not Haynes, here is the green wite attached to the screen from the 99 schematic from honda

 

[silverfox.xx]

mine was 100% earthed without me touching it, in fact I honestly couldn't tell you where it is earthed too but it is earthed, one of the very first things I checked and constantly checked as I scratched my head for 2 years over fault 25. but there you go can't explain it, oh and I'm going by the Honda service manual not Haynes, here is the green wite attached to the screen from the 99 schematic from honda

With or without my glasses I can't see my honda manual diagram s0s:

 

[SpanishJohn]

Smedley,

I can assure you, the 2000 FI bike ks is not earthed (the wire is blue).

The operation of the knock sensor is actually pretty much fundamental to the engine, The ECM will continue to advance ignitiion timing until the voltage from the ks hits a programmed 'pivot point' (a hard value programmed into the ECM) at which point it stops advancing and start retarding. If the voltage falls below this hard value, then the ECM starts advancing again etc. It is fairly obvious from this that should something (ks, ECM or wiring) go slightly out of spec then the results could be 'interesting'. I suspect that the ECM has some degree of safety margin buillt in such that the 'pivot point' is some way off actual engine knock. However, if the ks output is reduced by 10% (due to a shift in fundamental freq) the ECM will advance ingition timing by 10% more than it would otherwise have done which could lead to vibration (and not necessarily knock).

Spanish

 

[smedly]

I am not going to get into a continued debate about earthing of the screened (blue wire) from the KS, all I can tell you is that mine on a 99 model was 100% earthed if yours is not then sobeit. The official Honda service manual (as posted) clearly shows it earthed, maybe on later models they decided not too, I have worked with electronics all my working life, there is little I do not know about simple auto motor electrics, I have also built and raced my own cars in competition, you have a great resource here to help you through your problem with me and many others on this forum, don't come here looking for help then tell us we are all wrong otherwise it's pointless as you know best. and yes I am a straight talker no offence meant, you would like me if you met me in person

The knock sensor signal is a very distinctive spike when pinking (knock) occures, I would very much assume that if the ecu advances the ignition and does not detect Knock it will flag an error fault 25 and retard the ignition to a safe level and maintain that safe level until the faulty condition is rectified, you have recieved advice from people on here that appear to know about these things, you have neglected to answer a few simple questions that may assist you find out what is wrong with your bike, I wish you all the best and hope that you can get to the bottom of your issue, perhaps when you have replaced your KS we may be able to find out what is actually wrong with your engine, just my opinion as stated before - you are barking up the wrong tree.........good luck and look forward to helping you again

 

[smedly]

With or without my glasses I can't see my honda manual diagram s0s:

zoooooooooooooooom

lol

 

[smedly]

from the official Honda manual

 

[SpanishJohn]

The knock sensor signal is a very distinctive spike when pinking (knock) occures, I would very much assume that if the ecu advances the ignition and does not detect Knock it will flag an error fault 25 and retard the ignition to a safe level and maintain that safe level until the faulty condition is rectified, you have recieved advice from people on here that appear to know about these things, you have neglected to answer a few simple questions that may assist you find out what is wrong with your bike, I wish you all the best and hope that you can get to the bottom of your issue, perhaps when you have replaced your KS we may be able to find out what is actually wrong with your engine, just my opinion as stated before - you are barking up the wrong tree.........good luck and look forward to helping you again

Smedley, The ECM will advance the ignition until the signal from the ks tells it to stop, at which point it will start retarding the ignition. This implies that the signal from the ks is not a single spike at the knock frequency, but is instead a progressive signal. If this signal is attentated by 10%, then the ECU will advance by more than it would normally have done so. If you do not believe me then go to the very back of the Honda service manual and read for yourself. I too have a degree in electronics.

The ECM would never advance the ignition so far as to actaully cause a knock, that would be crazy, it would instead have some degree of safety margin. I am 90% sure my problem is the KS. I appologise if I have offended you fl4g71

Spanish

If you must know, my engine runs on Silkolene comp4, and I live in basingstoke, which is not too far off sea level (I guess).

 

[smedly]

knock is not progressive but it does have levels of severity, it is either there or not and is very distinctive, you can even hear it with your ear if allowed to continue long enough, lucky the ecu is there to stop it, and yes the ecu detects knock that is why it is called a ------ ------you seem to know a lot more about this stuff than me so I'll leave you to it, and I am 100% sure your problem is not KS, but there yee go


no idea what comp4 is, I go by viscosity to suit running conditions

 

[SpanishJohn]

Smedley,

If the signal was purely black and white, then the engine would destroy itself as the ECM was continuously advancing the signal until actual knock occurred, and then retarding it again. This would be madness. I suspect the signal to be more of a gaussian type curve where the limit is set some way of peak (knock).

PS. I always run 10W40 (part synthetic), even during the summer months, though I did consider running 15W50, but didn't in the end. I change my oil every 3K miles.

Spanish

 

[silverfox.xx]

w;;v Spanish John,

Whilst I cannot agree nor disagree with what you are saying... As I really don't know...

Once the ECM produces a fault code 25 'Knock sensor fault', or any fault that still allows the engine to run, the ECM reverts to a fail safe mode, which I assume prevents the continuous advance/retard you mention... hence the ECM cannot destroy the engine. From my experience of the safe mode power is down about 15% (I believe book says 25%) but I used mine for 700 touring miles while trying to sort the KS fault out.

According to the manual, the KS is there to detect knock that the ECM uses to adjust timing.

 

[andyBeaker]

It always worries me when an owner is 100% focussed on one thing that they think is causing a problem. Take a step back, have a good think about it and try not to focus on one aspect - it could easily turn out to be something completely different. remember the guy who everyone kept telling oil was getting into the bores as he had clouds of blue smoke and oil coming out off the pipes - seem to recall it turned out to be a missig washer on a US emission control system................

I know this probably doesn't seem very helpful but it is meant to be!!@tu*

Keep posting.....

 

[silverfox.xx]

It always worries me when an owner is 100% focussed on one thing that they think is causing a problem.

A bit like banking and investment... 8rfl@8rfl@8rfl@8rfl@

 

[smedly]

well you see I just find myself repeating what I am saying over and over, knock is either there or it isn't to varying degrees of severity but it is either happening or it isn't, the knock sensor is a microphone that sends noises to the ecu which is able to distinguish between normal engine noise and the distinctive sound of knock, it uses a set of audio frequency filters to achieve this, I suggest you do a simple google search on knock sensor and you will find plenty of information explaining how this works, if the knock sensor was indeed faulty in your engine the ecu would have detected this as it does several tests to make sure it is there and doing what it is designed to do, I and others cannot continue explaining this to you over and over, several suggestions where made to you about possible causes of your problem which you have chosen to ignore, several questions where asked of you to give us more information in an effort to try and understand and help with your problem, you have managed to answer 1 or 2 of them, based on what you have told us so far all I can suggest is you get some all weather gear and ride your bike when it's raining foggy and cold.

 

[SpanishJohn]

It always worries me when an owner is 100% focussed on one thing that they think is causing a problem. Take a step back, have a good think about it and try not to focus on one aspect - it could easily turn out to be something completely different. remember the guy who everyone kept telling oil was getting into the bores as he had clouds of blue smoke and oil coming out off the pipes - seem to recall it turned out to be a missig washer on a US emission control system................

Hi Andy,

I have had this problem with my bike now for over a year (I have owned the bike for just over a year - I suspect the prev. owner had the same issue). In that time I have taken many, many steps back and changed my opinion on what is causing the problem. I have had the bike apart so many times (balance shafts, valve clearances, CCT's, air filters, many oil changes, fuel injectors, throttle bodies, cooling system - my bike is practically brand new) and each time I put it back together the problem is still there :cry:. If I ride the bike in glorious weather I end up wanting to trade it. My hands, feet and privates are all numb with the vibes. If I ride the bike in fog, its as smooth as silk. I was considering taking the bike to Jaws and letting him have a go with it (I still may). You guys may well be right, it may have feck all to do with the KS, but It is one of the few things I have yet to change. I appologise if I have upset anyone, but please bear with me on this one....

Spanish

 

[Wolfie]

have you changed the tyres??

 

[smedly]

spanish that is the best post yet and you have actually shared some useful info, perhaps that should have been included in your first post, you must change the KS as you won't be happy until it is eliminated and you never know contary to what me and a few others have said - it may fix it.

Many of the sensors on the bike are passive or go through a very basic test, it could be one of them out of range e.g. the baro sensor, maybe it is not responding correctly to atmopheric pressure as I mention in an earlier post, barometric pressure changes with weather so there is a possible connection - think of how a barometer works tp predict weather, air intake temp is another possibility -IAT sensor, I am trying to relate to the conditions you have described, both these sensors are not tested by the ecu for how they are functioning just that they are there, temp and baro pressure are variables so the ecu just accepts that they are calibrated correctly and sending accurate information, if they are out of calibration they could possibly account for your symptoms - who knows

 

[andyBeaker]

Toys all put away then??

Let's get this sorted@tu*