First off, I have an '01 ZX3 with the Zetec/ATX and have an XCAL 3 w/ Tom's tune.

I'm wondering what others who datalog with the first "KNOCK SENSOR RETARD LIMIT" DMR in the SBAF6 processor code list see as far as values or on the graph while doing a normal cruise datalog. My issue currently is when I log the KS (on either the stock tune or on Tom's tune, doesn't matter), it shows that I'm always knocking and pulling -5 degrees of timing usually right around 2000 rpm or above. The problem is though, I'm not hearing or feeling any detonation at all. I believe my KS is bad or something else making noise is setting it off but not positive. My graph would show a straight line at 0 when idling and almost always goes straight down to -5 when it hits the 2000 rpm mark and eventual goes back to 0. But there is never an in between, always at 0 or a straight vertical to -5 with the occasional advancement.

So I was wondering what should I be seeing on the graph or in values as far as the KS? Should it mostly read at 0 with very little timing being pulled at any rpm range with advancement here and there or what? I'd appreciate any advice or suggestions on this issue from others who know what I'm talking about. Its causing me to not be able to get tuned and I'd like to change that asap. Its impossible to get a proper tune when the KS decides to max out pulling timing.

Thanks in advance,
Steven

Bump for some help!

I'd really like to know what I should be seeing on the graph when I log the knock sensor.

You shouldnt see anything 0

Tom

We used to see false knock in boat engine EFI from loose motor mounts, etc. I think even a couple of loose knock sensors. Make sure you don't have anything like that.

Well today I found a few new things but the problem is still there.

First thing I found was the wire that goes from the battery to the alternator was rubbing on the stud where the knock sensor mounts. I moved it but the car still shows knock during a datalog.

I decided to log the car while free reving not in gear and in drive while on the jack stands still and the KS read 0 all through out the rpm's, couldnt get it to show knock. Then I go to put the car on the ground and go for a cruise. Well the same thing as before happens anywhere above about 2000rpm while accelerating in a gear, the KS goes off.

I guess I'll play with it more tomorrow and see if I get anything to work correctly. Is it unheard of to have a KS go bad? Starting to think its the problem.

EDIT: I tried transbraking while on jackstands to create load and the KS shows knock. Graph looks the same as if driving it on the street. So now, I think I've got this narrowed to a load issue. Gonna try a new set of Motorcraft spark plugs tomorrow and see what happens, cant hurt anything. I believe now its either ignition related or the KS itself is way too sensitive. I'll update tomorrow. Oh and I've run 87oct, 91oct, and 96oct all with the same knock results on the stock tune.

If the knock sensor was bad or going bad wouldnt you feel a major power loss?

I read the life of the knock sensors are like 10 years. I read you can trick the sensor temporarily to trouble shoot problems. emergency solution not a long term solution.

We talked about all the things that need to be checked and I dont see where you have tested any of them

Tom

We talked about all the things that need to be checked and I dont see where you have tested any of them

Tom

Ok,

The first thing you said was run 5 gallons of 91oct and 5 gallons of 100oct which averages out to 95.5 but since I had a little less 91oct in the tank already it came out to be about 96oct. Knock still showed in the datalog, just like the 87oct log, after wasting over $50 for a 3/4 tank of gas (remember jobless high school kid here)... That rules out bad gas though, right?

Next you said to get a piece of garden hose and listen around for metal noise while someone holds the rpms around 2000rpm. Well since I can only do that in place and already stated the KS doesnt pick up anything while revving in place, then what am I supposed to hear? I already said there's nothing out of the ordinary. Acording to datalogging, this only happens when driving around or transbraking on jackstands. Must be a load issue now, right?

And that's all you've said.

You will only see knock after a set load and RPM , you have to have load to have/see knock

I was asking you to check thing to see if you had something loose and vibrating and that would cause the knock sensor to go off

I can adjust the sensitivity of the knock sensor but I would like to know for sure it isnt something else

Tom

is this an NA focus?

is this an NA focus?

Yes it is.

How many miles on the motor? What bolt ons do you have.

I had this same issue when I went FI the first time. I ended up tuning it myself and pretty much eliminating all the knock by just pulling timing from the BKT.

I as well never felt or heard any knock until I tapped into the knock sensor from the ecu harness and listened through my stereo.

Wish i could help you. It's a crappy thing to deal with.

I as well never felt or heard any knock until I tapped into the knock sensor from the ecu harness and listened through my stereo.


What the heck!? [80?]

How many miles on the motor? What bolt ons do you have.

I had this same issue when I went FI the first time. I ended up tuning it myself and pretty much eliminating all the knock by just pulling timing from the BKT.

I as well never felt or heard any knock until I tapped into the knock sensor from the ecu harness and listened through my stereo.

Wish i could help you. It's a crappy thing to deal with.

Interesting. Well my car has 123,5xx miles and for bolt ons, it has stock SVT header, cat, flex, and flex back exhaust, and stock SVT airbox with K&N drop in filter along with Taylor 8.2mm plug wires. That's all I'm going to do for NA, currently saving for 5 speed swap and then turbo after that. [burnout]

What the heck!? [80?]

He's reffering to a setup like this where you can listen to your knock sensor through your in car stereo. Link to Illinipo's thread: http://www.focusfanatics.com/forum/showthread.php?t=207523

Today, I simply just unbolted the KS from the mount and left it plugged in. I put a towel over to insult it and had my dad rev it with the A/C on/ headlights on to create load. Well the KS still showed knock even though it wasnt bolted to the block. Tom or anyone, do you think this could point to the KS being the issue? Not really sure if thats the problem.

If the KS is not the culprit, then I'm just going to break down and re-clearance my buckets back into spec to get rid of the tappit. That's the only thing I can really do because that's the only real engine noise I hear. Alternator, timing belt/ and related pullies, serpentine belt/ all related pullies, spark plugs, plug wires, fuel filter have all been changed. Old datalogs from before all that stuff was done showed knock then so it cant be any of those things I dont think.

Oh and I found the typical collasped PCV hose running to the intake manifold end. Will replace that asap. That's probably what was causing my rough idle.

The knock is sensor is pretty much a microphone, so yeah it probably would still show knock.

The KS should not of shown knock if it was not bolted on , You removed the sensor just below intake runner 2 in the block on a stud that kinda looked like a doughnut with wires ?

Tom

The reason I said the KS would or could show knock if not attached to the block would be it was rattling on something, but I guess if he wrapped it up in paper towels that could keep it secure enough.

But that's odd it would still show knock, maybe an electrical connection?

The KS should not of shown knock if it was not bolted on , You removed the sensor just below intake runner 2 in the block on a stud that kinda looked like a doughnut with wires ?

Tom

Yes sir, that's the one and it showed knock in the datalog even though it was not bolted to anything.

The reason I said the KS would or could show knock if not attached to the block would be it was rattling on something, but I guess if he wrapped it up in paper towels that could keep it secure enough.

But that's odd it would still show knock, maybe an electrical connection?

Yea that's what my dad and I were thinking next. I'll do a continuity test next time I get a chance and confirm the wiring is good. It does run in a harness that goes in between the 1 and 2 intake runners where it could of chafed on the rough edges of '01 and up intake manifolds. As far as I can see though, it looks fine.

Thanks for the help so far guys. [thumb]

When you had the sensor off did you look at it close with a magnifying glass.

Does it rattle if you shake it?

Try changing it

Tom

^^right, do the easiest thing first. It's much harder to track down a short or something that might not even exist.

^^right, do the easiest thing first. It's much harder to track down a short or something that might not even exist.

They are one of the more expensive sensors though

They're not too bad actually. Just ordered one from RockAuto for about $28 plus shipping. I'll guess we'll see what happens when it gets here.

They're not too bad actually. Just ordered one from RockAuto for about $28 plus shipping. I'll guess we'll see what happens when it gets here.

REALLY! [80?] I thought they were around $100. Was it a PITA to get to it?

REALLY! [80?] I thought they were around $100. Was it a PITA to get to it?

Its basically right next to your oil filter.

Its basically right next to your oil filter.

oh you have the older motor. I have the duratec i think mine is more work.

I'm wondering what others who datalog with the first "KNOCK SENSOR RETARD LIMIT" DMR in the SBAF6 processor code list see as far as values or on the graph while doing a normal cruise datalog. My issue currently is when I log the KS (on either the stock tune or on Tom's tune, doesn't matter), it shows that I'm always knocking and pulling -5 degrees of timing usually right around 2000 rpm or above.

SBAF6 has two DMRs that are described as KNOCK SENSOR RETARD LIMIT. I've talked to SCT about this at least 10 times over the last 3 years and all they have done was change the name of one of them, not the description. You need to take a closer look at the names, or download the latest livelink if you cannot see the difference I outline below.

KNK_RET_LIM is the first one. There is a table in the ECU with axes of RPM against load that simply limits the amount of knock retard you can get. It is -5 in almost all cells, with a few at low and high rpm at -4. The reason you are seeing -5 on this DMR only when you are moving or transbraking is the sensor is not allowed to be active at all until 0.359 load, and you would be hard pressed to see that much load without the car in gear (Tom already metioned this earlier in the thread). So basically whenever you are driving, as you describe, this should rightly show -5.

KNK_SEN_RETARD is the second one, the one you actually want to log, this is the actual knock retard. This one you ought to be able to at least see it ramp down (instead of jump down) to -5, and ramp slowly back up (instead of jump up) when you let off the throttle, if it is indeed pulling full knock retard.

Well im not sure what your looking for and i dont even know what all this says. I probobly dont even have the same setup logs as you.

Here is the last one i ran a few nights ago.

All of my KS settings are just flat all the way across
http://i5.photobucket.com/albums/y191/BowerR64/LiquidFocus/Secondlog.jpg

Bower you have a Duratec and your KS readings will be diff

He has the correct KS logged in his list

Tom

He has the correct KS logged in his list

Tom


the first "KNOCK SENSOR RETARD LIMIT" DMR in the SBAF6 processor code list

Not if he is using the first one.

But what the hell do I know (nothing apparently)

Bower you have a Duratec and your KS readings will be diff

He has the correct KS logged in his list

Tom

ok i thought he just wanted to see some other logs.

SBAF6 has two DMRs that are described as KNOCK SENSOR RETARD LIMIT. I've talked to SCT about this at least 10 times over the last 3 years and all they have done was change the name of one of them, not the description. You need to take a closer look at the names, or download the latest livelink if you cannot see the difference I outline below.

KNK_RET_LIM is the first one. There is a table in the ECU with axes of RPM against load that simply limits the amount of knock retard you can get. It is -5 in almost all cells, with a few at low and high rpm at -4. The reason you are seeing -5 on this DMR only when you are moving or transbraking is the sensor is not allowed to be active at all until 0.359 load, and you would be hard pressed to see that much load without the car in gear (Tom already metioned this earlier in the thread). So basically whenever you are driving, as you describe, this should rightly show -5.

KNK_SEN_RETARD is the second one, the one you actually want to log, this is the actual knock retard. This one you ought to be able to at least see it ramp down (instead of jump down) to -5, and ramp slowly back up (instead of jump up) when you let off the throttle, if it is indeed pulling full knock retard.

You and Tom are both right. Thanks for the awesome detailed response though!

It turns out the right KNK SEN RETARD was being logged. I had the view on the "long name PIDs" setting instead of the "16 Char Name PIDs". In the long name veiw, both DMR's show up exactly the same as "KNOCK SENSOR RETARD LIMIT" but for some reason the correct one to be logged comes up first in the PID list. In the 16 view, the correct one comes up second on the list? [dunno]

But yes, the graph shows a short little spike (about 0.6-1.0 advancement) then drops straight down to -5. Just waiting for my new one to see if anything changes.

You and Tom are both right. Thanks for the awesome detailed response though!

It turns out the right KNK SEN RETARD was being logged. I had the view on the "long name PIDs" setting instead of the "16 Char Name PIDs". In the long name veiw, both DMR's show up exactly the same as "KNOCK SENSOR RETARD LIMIT" but for some reason the correct one to be logged comes up first in the PID list. In the 16 view, the correct one comes up second on the list? [dunno]

But yes, the graph shows a short little spike (about 0.6-1.0 advancement) then drops straight down to -5. Just waiting for my new one to see if anything changes.

After going through the same process on a NFC1 car of buying a new sensor and lashing the valves to min spec, with no change in the sensor activity, I would recommend you guys just lower the sensitivity from the get-go. The combination of listening to it and lowering sensitivity would be very effective.

But supposedly I don't know what I'm doing so you probably should just ignore my posts.

But supposedly I don't know what I'm doing so you probably should just ignore my posts.

Why do you keep saying that?

Maybe you should make that your sig so you wont have to keep typing it. [screwy]

My knock sensor has been turned off for over a year now. There not accurate. I have logs of it picking up noise from the wastegate and it would pull 2degrees. Also when I would shift fast it would think it was knock and pull 5 degrees then slowly put it back in. As long as your tuner knows what he is doing there should be no problems. Also I fill up at any gas station. Hess, Bp, sans club. Etc.

My knock sensor has been turned off for over a year now. There not accurate. I have logs of it picking up noise from the wastegate and it would pull 2degrees. Also when I would shift fast it would think it was knock and pull 5 degrees then slowly put it back in. As long as your tuner knows what he is doing there should be no problems. Also I fill up at any gas station. Hess, Bp, sans club. Etc.

How do you have it turned off?

I wondered about that myself, when i started reading about what they are and how they work. Then listening to sound clips i must say it sounded a little iffy to me. I then read where some one said they use some kind of audio filtering or somthing to single out the knock. It makes a little more sence till the OP said they are only $30. for the sensor now im starting to wonder.

If i raced my car i dont think ide want one pulling timming out but for how im using my car ill leave it alone.

You actually do want it to pull timing if you have knock and on the Zetec and Duratec they work perfectly on about 99% of the Foci

Its my opinion to leave the knock sensor on as it is a HUGH safety for your engine , yes you may see some knock on gear change but thats perfectly normal because of tip in timing adding timing on tipin in among other things

Please leave your KS on at all times unless you have a SVT

Tom

I just got my new KS and rear o2 sensor thanks to RockAuto.com, only took them 2 days! Once my car cools down, I'll be installing both things and report back if anything changes.

You actually do want it to pull timing if you have knock and on the Zetec and Duratec they work perfectly on about 99% of the Foci

Its my opinion to leave the knock sensor on as it is a HUGH safety for your engine , yes you may see some knock on gear change but thats perfectly normal because of tip in timing adding timing on tipin in among other things

Please leave your KS on at all times unless you have a SVT

Tom

I agree a knock sensor is a nice backup if something goes wrong (gas, boost,etc) but the focus ones are not calibrated right. they pick up the noise from the drums on the dyno, road noise, wastegate, and shifting.

Pulling 5.5 degrees (not 5 like i thought) is not normal. the car fell on it face everytime.

Also you know as well as i do that chris focus blew the ringlands with a J&S on it when you tuned it. there was deto marks on the piston head. I bring chris car up cause it had the J&S on it and still went. I would love if we had a company that would make a nice External knock system for our cars that get away with the factory one.

How do you have it turned off?

I wondered about that myself, when i started reading about what they are and how they work. Then listening to sound clips i must say it sounded a little iffy to me. I then read where some one said they use some kind of audio filtering or somthing to single out the knock. It makes a little more sence till the OP said they are only $30. for the sensor now im starting to wonder.

If i raced my car i dont think ide want one pulling timming out but for how im using my car ill leave it alone.

You have to have your tuner or if you have a prp you can turn it off in the tune

I agree a knock sensor is a nice backup if something goes wrong (gas, boost,etc) but the focus ones are not calibrated right. they pick up the noise from the drums on the dyno, road noise, wastegate, and shifting.

there was deto marks on the piston head.

I feel the KS is calibrated VERY WELL , I dont feel there is any way to get road noise to set off the KS , drum noise , dyno noise , waste gate because the kS works off a frequency , I have never seen this with the 100`s of Foci I have tuned and it is very rare to have to change the sensivity of the stock sensore , I am not saying yours does not have this issu as I havent seen how things are set

What marks did you see on the pistons from deto ?

Tom

I feel the KS is calibrated VERY WELL , I dont feel there is any way to get road noise to set off the KS , drum noise , dyno noise , waste gate because the kS works off a frequency , I have never seen this with the 100`s of Foci I have tuned and it is very rare to have to change the sensivity of the stock sensore , I am not saying yours does not have this issu as I havent seen how things are set

What marks did you see on the pistons from deto ?

Tom

This is just my experience, ik not everyones focus is the same but my car runs well with it off.

And there were marks and pits from deto in the top, but not trying o throw this off topic.

Got the new KS in and my datalogs still show damn knock. Nothings really changed except in some spots where it "ramps" up or down on the graph instead of a straght drop or rise.

Think I'll try doing illinipo's version of listening to knock through the stereo and then re-clearance my buckets to get rid of the extra valvetrain noise.

How is thins coming alon egg?

How is thins coming alon egg?

I'm still working on it slowly but surely. Tomorrow, I'm going to try taking the serpentine belt off and run the car long enough to see if I get knock or not. If nothing shows then it would eliminate possible noise from the serpentine pully, tensioner, water pump, alternator, A/C, power steering, etc. Hopefully that method can eliminate a lot of possible causes.

Now out of curosity, if I did have real unaudiable knock, then what would be some things that cause it? What if it was real knock while running 91 oct on the stock 87 oct tune?

I tried my test I posted above just now. Knock still shows, but at least I eliminated a bunch of possible causes. It cant be anything in the accessories.

Re-clearancing the cams will come this weekend. I'll continue updating this untill I find the cause.

When was the last time you had the clutch looked at or replaced?

Does it shift funny or clunk into gear when shifting? flex plate cracked maybe?

I would hate to think you have bad rods

When was the last time you had the clutch looked at or replaced?

Does it shift funny or clunk into gear when shifting? flex plate cracked maybe?

I would hate to think you have bad rods

This is an ATX car.... for now. But everything shifts just fine and feels fine. I heel a slight whistle when I decelerate in gear. Dont think its an issue though. Also I hear a sound that sounds like a rattle that knock would make coming from the back of the car, could the KS actually here it if the noise was coming from somewhere in the back? I started heaing it after doing my SVT brake swap way back when before getting my tune.

Got all the buckets off my friend so I'll be re-clearancing the cams when I get a chance.

I'm gonna go ahead and bump this back up since I just graduated high school and have all the time in the world to play with this right now. Have a couple questions...

1. With the KS unplugged, should I still be getting a reading from my KS retard on Live Link? It advances and retards still even though its unplugged.

2. As I said before, I got alot of different size buckets off a friend. Does it matter if they're used? I know it used to be if you were using a used cam, you'd need new buckets because of the wear patterns. Does this still apply to todays newer hardened materials, etc?

Thanks in advance!

Bump for my last two questions if anyone knows. Looking to work on this throughout this next week.

I wouldnt think you should still be seeing the KS making changes with the KS unplugged but the KS is proving to be interesting , I am doing a ZX2 that we have not been able to log the KS at all and he plugged in the cam sensor and now the KS is logging changes

You can use used buckets with new cams or used cams you will be fine

Tom

Interesting topic for me (it was my motor that had the KS issue mentioned above)

To be specific it's a cougar (zx3 style zetec) with a focus zetec swapped in; which is why the Cam sensor wasn't plugged in to start with (it's on the opposite cam).... no idea why it acted this way.

I hope you figure out your issues It mostly just takes time (I'm not done yet either)

eggyolk did you find out the problem?

eggyolk did you find out the problem?

No I never did find out the problem. I've re-clearanced my buckets back into spec on the tighter side, nothing. Replaced KS, nothing. Undid the serpentine belt and knock still showed so that elminated all the accessories. Since then, I've done a MTX swap and nothing changed eliminating the ATX. Ran 91oct with stock tune and also a mix of 4 gallons 91 oct with 5 gallons 100 oct, once again nothing changed. Checked the timing belt idler and tensioner when I had to pull the cams to clearance the buckets, it all felt/sounded fine. Really have no idea why it was doing that. I'm hoping to finish my built motor by summer so we'll see if it shows knock on a whole new motor.

After trying all that, Tom just turned down the KS sensitivity in the tune, now I only show knock very rarely but 99.5% of the time, it shows flat line.