Knock Sensor Calibrations - What voltages for built engines
Dec 19, 2011, 01:20 AM
#1
Newbie
Thread Starter
iTrader: (8)
Join Date: Dec 2007
Location: Western PA
Posts: 92
Likes: 0
Received 0 Likes
on
0 Posts
Knock Sensor Calibrations - What voltages for built engines
I'm mostly just trying to see if anyone has their calibration details handy...
I did a pull up to 8000rpms in vac and managed to get voltages as high as ~3.71v. I used that pull to see what the "highest" noises were per breakpoint, inputed them as the calibration and then set up the knock control.
I really am mostly interested in seeing what sort of voltages everyone is seeing on a built engine. The car has filled engine mounts, a dampener, it's .020 over w/Wiseco pistons, Manley rods, balance shaft deletes, etc... I have the timing uber conservative currently because I was a little surprised how much the sensor was reading. Then I did the 8000rpm pull in vac and decided maybe I needed to be a little less apprehensive - I've just have always read that on E85 there's usually no knock, then catastrophic failure.
Anyone happen to have any real world experience on what their sensor is reading on a built setup? Is something up w/the setup, sensor, or something else... or is seeing a voltage reading that high near a common value?
I also have it set up right now such that it's...
knock rtd max: 4.92*
knock rtd/v: .55*/v
knock restore rate: 15rev
knock spark advance: 1.05*
knock fuel add max: 5.08%
knock %rich/volt: .62%/v
knock restore rate: 15rev
knock decrease fuel: 1.95%
Any information would be appreciated! Thanks again!
I did a pull up to 8000rpms in vac and managed to get voltages as high as ~3.71v. I used that pull to see what the "highest" noises were per breakpoint, inputed them as the calibration and then set up the knock control.
I really am mostly interested in seeing what sort of voltages everyone is seeing on a built engine. The car has filled engine mounts, a dampener, it's .020 over w/Wiseco pistons, Manley rods, balance shaft deletes, etc... I have the timing uber conservative currently because I was a little surprised how much the sensor was reading. Then I did the 8000rpm pull in vac and decided maybe I needed to be a little less apprehensive - I've just have always read that on E85 there's usually no knock, then catastrophic failure.
Anyone happen to have any real world experience on what their sensor is reading on a built setup? Is something up w/the setup, sensor, or something else... or is seeing a voltage reading that high near a common value?
I also have it set up right now such that it's...
knock rtd max: 4.92*
knock rtd/v: .55*/v
knock restore rate: 15rev
knock spark advance: 1.05*
knock fuel add max: 5.08%
knock %rich/volt: .62%/v
knock restore rate: 15rev
knock decrease fuel: 1.95%
Any information would be appreciated! Thanks again!
Dec 19, 2011, 08:56 AM
#2
It does get hard with a built motor as they do make more noise. On my 680hp galant, I had a max of 3.6 volts in the knock cal.
So here is how you set your knock cal.
Put boost at wastegate, get afr's a little on the rich side, for E85 like 11.5 in boost. Take the timing down just a little bit, and do a WOT pull to redline. Make sure you have a good ECU log. From that log build your Knock Cal. As you turn up the boost you will need to up your knock cal, however dont be afraid, just keep upping it.
The thing to look for are big spikes in the knock, if you are not seeing big spikes most of the time you are ok and you might just be to agressive with your knock cal. The big thing with the knock cal, is make it as safe as you feel comfortable.
For me after a while of running, I would open up my knock cal and keep giving myself more room.
Hope this helps.
So here is how you set your knock cal.
Put boost at wastegate, get afr's a little on the rich side, for E85 like 11.5 in boost. Take the timing down just a little bit, and do a WOT pull to redline. Make sure you have a good ECU log. From that log build your Knock Cal. As you turn up the boost you will need to up your knock cal, however dont be afraid, just keep upping it.
The thing to look for are big spikes in the knock, if you are not seeing big spikes most of the time you are ok and you might just be to agressive with your knock cal. The big thing with the knock cal, is make it as safe as you feel comfortable.
For me after a while of running, I would open up my knock cal and keep giving myself more room.
Hope this helps.
Dec 19, 2011, 09:30 AM
#3
here is how i set up knock tables...i am sure people do stuff differently.
here are some things you need to do:
***set up internal logger @ 250 samples / sec & log knock#1 volts as a fast parameter.
***turn your boost down to ~15 psi max, be certain that your ign timing for that level is extremely conservative = no knock possible
***be sure ur fueling is fine for that boost level
***verify that you can make a clean 3rd gear pull, fueling is ok etc.
*** conduct 2 or 3 3rd gear pulls from 2k-max revs
upload & export the internal log, overlay all 3 runs to make sure there is repeatability. i like to plot them & capture the noise around the knock cal break points.
take those peak values & multiply them by 1.23, then you have a knock cal table for your engine.
the "musts" are that your basing your profile from an internal log @ 250 samples/sec & YOU ARE NOT KNOCKING
i run 1.64*/V for knock retard. if your knocking, pull the timing quickly.
i dont use the fueling, as the knock already happened & the fuel is just getting there too late imo.
i have a 2.3L & the highest voltage i saw doing this testing was ~2.5v, my stock block 2.0 was very similar too, ~2.4 observed voltage as tested above.
i have seen motors that had the oil pick up tickling the bottom of the oil pan & i know that should agrivate the knock sensor.
GL, aby
here are some things you need to do:
***set up internal logger @ 250 samples / sec & log knock#1 volts as a fast parameter.
***turn your boost down to ~15 psi max, be certain that your ign timing for that level is extremely conservative = no knock possible
***be sure ur fueling is fine for that boost level
***verify that you can make a clean 3rd gear pull, fueling is ok etc.
*** conduct 2 or 3 3rd gear pulls from 2k-max revs
upload & export the internal log, overlay all 3 runs to make sure there is repeatability. i like to plot them & capture the noise around the knock cal break points.
take those peak values & multiply them by 1.23, then you have a knock cal table for your engine.
the "musts" are that your basing your profile from an internal log @ 250 samples/sec & YOU ARE NOT KNOCKING
i run 1.64*/V for knock retard. if your knocking, pull the timing quickly.
i dont use the fueling, as the knock already happened & the fuel is just getting there too late imo.
i have a 2.3L & the highest voltage i saw doing this testing was ~2.5v, my stock block 2.0 was very similar too, ~2.4 observed voltage as tested above.
i have seen motors that had the oil pick up tickling the bottom of the oil pan & i know that should agrivate the knock sensor.
GL, aby
Last edited by Aby@MIL.SPEC; Dec 19, 2011 at 09:32 AM.
Dec 19, 2011, 11:40 AM
#4
Newbie
Thread Starter
iTrader: (8)
Join Date: Dec 2007
Location: Western PA
Posts: 92
Likes: 0
Received 0 Likes
on
0 Posts
here is how i set up knock tables...i am sure people do stuff differently.
here are some things you need to do:
***set up internal logger @ 250 samples / sec & log knock#1 volts as a fast parameter.
***turn your boost down to ~15 psi max, be certain that your ign timing for that level is extremely conservative = no knock possible
***be sure ur fueling is fine for that boost level
***verify that you can make a clean 3rd gear pull, fueling is ok etc.
*** conduct 2 or 3 3rd gear pulls from 2k-max revs
upload & export the internal log, overlay all 3 runs to make sure there is repeatability. i like to plot them & capture the noise around the knock cal break points.
take those peak values & multiply them by 1.23, then you have a knock cal table for your engine.
the "musts" are that your basing your profile from an internal log @ 250 samples/sec & YOU ARE NOT KNOCKING
i run 1.64*/V for knock retard. if your knocking, pull the timing quickly.
i dont use the fueling, as the knock already happened & the fuel is just getting there too late imo.
i have a 2.3L & the highest voltage i saw doing this testing was ~2.5v, my stock block 2.0 was very similar too, ~2.4 observed voltage as tested above.
i have seen motors that had the oil pick up tickling the bottom of the oil pan & i know that should agrivate the knock sensor.
GL, aby
here are some things you need to do:
***set up internal logger @ 250 samples / sec & log knock#1 volts as a fast parameter.
***turn your boost down to ~15 psi max, be certain that your ign timing for that level is extremely conservative = no knock possible
***be sure ur fueling is fine for that boost level
***verify that you can make a clean 3rd gear pull, fueling is ok etc.
*** conduct 2 or 3 3rd gear pulls from 2k-max revs
upload & export the internal log, overlay all 3 runs to make sure there is repeatability. i like to plot them & capture the noise around the knock cal break points.
take those peak values & multiply them by 1.23, then you have a knock cal table for your engine.
the "musts" are that your basing your profile from an internal log @ 250 samples/sec & YOU ARE NOT KNOCKING
i run 1.64*/V for knock retard. if your knocking, pull the timing quickly.
i dont use the fueling, as the knock already happened & the fuel is just getting there too late imo.
i have a 2.3L & the highest voltage i saw doing this testing was ~2.5v, my stock block 2.0 was very similar too, ~2.4 observed voltage as tested above.
i have seen motors that had the oil pick up tickling the bottom of the oil pan & i know that should agrivate the knock sensor.
GL, aby
Hey Aby,
Thanks for always responding to my AEM threads
. I'm a noob. Ok... so the setup has a 44mm Tial w/20.305# spring. So without pulling it apart that's sort of my min pressure. With the weather being cold around here the car is constantly boost creeping though. At "wastegate" pressure (i.e. hooked directly to the mani) the car creeps up to 28psi. It's a Hafe cast manifold, so that may not be super surprising.
I'm looking at a log right now that is a day or two old - at WG press...
It's run out to ~8500rpms in 2nd and 3rd, AFRs right around 11.0, 5.15* of timing @ 6520 @ 23.8psi --- timing stays almost straight flat out to 8449 @ 28.59psi. Saw 3.16v up top. Oddly enough, as mentioned before, I did a pull completely in vac and saw 3.71v - which is what made me feel "ok" about seeing anything up to that value.
@TheBoz, I'm glad to hear that your voltage isn't much different than what I'm seeing. The 3.71v just seemed crazy high to me (conceptually). There was a point when I didn't look too much into the knock cal and the car was breaking up anytime I'd get near 7000rpms (I kept taking timing out and adding fueling w/no avail), then realized I wanted to log knock correction and saw it was correcting a ton because of how the cal was setup before.
Once I started using what I roughly thought the cal should be based on the vac pull I did the car stopped breaking up and could go to 8500 (9000 redline). Of course I tried to get up to 30psi and see if I could get it stop creeping... and as soon I tried that it broke up again (though I think now it's spark - because of no correction, stilll being safely rich and boring timing).
Thanks guys!!
Dec 19, 2011, 01:35 PM
#5
Evolved Member
iTrader: (7)
Join Date: Feb 2007
Location: MoTown,GA
Posts: 574
Likes: 0
Received 0 Likes
on
0 Posts
I also see a few 3.5 volt spikes at times.... I have my knock table set to 4.1 volts... I have had real knock, i could actually hear it. i saw spikes of 4.8 volts. Also as boz mentioned, it the voltage goes up but the spikes look square on the log then you are good but if you see more than 1 or 2 sharp spikes remove timing there and make sure it goes away.
Dec 19, 2011, 03:40 PM
#6
Newbie
Thread Starter
iTrader: (8)
Join Date: Dec 2007
Location: Western PA
Posts: 92
Likes: 0
Received 0 Likes
on
0 Posts
I'm not too great at paint... but I'm sure you will all get the picture from the log. As you can see in this log, the fueling and timing both are pretty conservative. I also had relatively "low" voltage readings in this log relative to some others that I've had...
@BrowningDerek, I see you're a stroker... but it's cool to hear your experiences because your car has sort of a similiar setup.
Evo9 - built as mentioned above, running a Magnus v5 twin rail (950s, 850s - may need more fuel
), 75mm Boomba, S362 .70a/r, Hafe T3 manifold, full custom 3" o2 housing, 3" RSR turbo-back, 1.75" dump tube, custom made 2.5" hot side piping - 3" cold side piping, Buschur Racing FMIC, Buschur double pumper, -8an feed lines (from factory feed), -6an return lines to FPR, -4an from FPR to stock return line.
Have a Dynatek Arc-2 and 300m COP that needs to go on... I think the Dynatek may be fried though. I need to find a testing strategy for one to check before I send it in. Seems like the car was blowing spark out with "better" dwell settings on OE wires + coils around 31-32psi.
Shooting for near 700awhp.
PS - how's your drivability w/the Magnus + Boomba on AEM? I think I've spent more time on that than anything else thus far.
@BrowningDerek, I see you're a stroker... but it's cool to hear your experiences because your car has sort of a similiar setup.
Evo9 - built as mentioned above, running a Magnus v5 twin rail (950s, 850s - may need more fuel
), 75mm Boomba, S362 .70a/r, Hafe T3 manifold, full custom 3" o2 housing, 3" RSR turbo-back, 1.75" dump tube, custom made 2.5" hot side piping - 3" cold side piping, Buschur Racing FMIC, Buschur double pumper, -8an feed lines (from factory feed), -6an return lines to FPR, -4an from FPR to stock return line.Have a Dynatek Arc-2 and 300m COP that needs to go on... I think the Dynatek may be fried though. I need to find a testing strategy for one to check before I send it in. Seems like the car was blowing spark out with "better" dwell settings on OE wires + coils around 31-32psi.
Shooting for near 700awhp.
PS - how's your drivability w/the Magnus + Boomba on AEM? I think I've spent more time on that than anything else thus far.
Last edited by SloEvo12; Dec 19, 2011 at 03:43 PM.
Dec 19, 2011, 03:54 PM
#7
is that log image you posted from a internal log?
if so, what is thhe logging rate set too?
also, after you let off the throttle, expect to get some voltage spikes, as everything is asked to do something different, whether its piston being load differently, gearbox, drivetrain, etc...
care you email your log to look at? pm sent
if so, what is thhe logging rate set too?
also, after you let off the throttle, expect to get some voltage spikes, as everything is asked to do something different, whether its piston being load differently, gearbox, drivetrain, etc...
care you email your log to look at? pm sent
Trending Topics
Dec 23, 2011, 07:38 AM
#8
Evolved Member
iTrader: (7)
Join Date: Feb 2007
Location: MoTown,GA
Posts: 574
Likes: 0
Received 0 Likes
on
0 Posts
I'm not too great at paint... but I'm sure you will all get the picture from the log. As you can see in this log, the fueling and timing both are pretty conservative. I also had relatively "low" voltage readings in this log relative to some others that I've had...
@BrowningDerek, I see you're a stroker... but it's cool to hear your experiences because your car has sort of a similiar setup.
Evo9 - built as mentioned above, running a Magnus v5 twin rail (950s, 850s - may need more fuel ), 75mm Boomba, S362 .70a/r, Hafe T3 manifold, full custom 3" o2 housing, 3" RSR turbo-back, 1.75" dump tube, custom made 2.5" hot side piping - 3" cold side piping, Buschur Racing FMIC, Buschur double pumper, -8an feed lines (from factory feed), -6an return lines to FPR, -4an from FPR to stock return line.
Have a Dynatek Arc-2 and 300m COP that needs to go on... I think the Dynatek may be fried though. I need to find a testing strategy for one to check before I send it in. Seems like the car was blowing spark out with "better" dwell settings on OE wires + coils around 31-32psi.
Shooting for near 700awhp.
PS - how's your drivability w/the Magnus + Boomba on AEM? I think I've spent more time on that than anything else thus far.
@BrowningDerek, I see you're a stroker... but it's cool to hear your experiences because your car has sort of a similiar setup.
Evo9 - built as mentioned above, running a Magnus v5 twin rail (950s, 850s - may need more fuel
), 75mm Boomba, S362 .70a/r, Hafe T3 manifold, full custom 3" o2 housing, 3" RSR turbo-back, 1.75" dump tube, custom made 2.5" hot side piping - 3" cold side piping, Buschur Racing FMIC, Buschur double pumper, -8an feed lines (from factory feed), -6an return lines to FPR, -4an from FPR to stock return line.Have a Dynatek Arc-2 and 300m COP that needs to go on... I think the Dynatek may be fried though. I need to find a testing strategy for one to check before I send it in. Seems like the car was blowing spark out with "better" dwell settings on OE wires + coils around 31-32psi.
Shooting for near 700awhp.
PS - how's your drivability w/the Magnus + Boomba on AEM? I think I've spent more time on that than anything else thus far.
I REPEAT! you do not have ignition problems. get the 2150's, put them where they should be and thank me later......
I plan on trading the boomba TB for a mil spec after christmas. Its just too big.....
Dec 23, 2011, 10:10 AM
#9
I would also ditch the dual fuel rails, and just go for the FIC 2150's. They are freeking awesome. The do idle rich as hell on pump gas, but if you are running E85 they are so win it is not even funny.
Here is my cal and log files from when I made 680hp on E85 running FIC 2150cc.
http://forum.aempower.com/forum/inde...c,26972.0.html
Here is my cal and log files from when I made 680hp on E85 running FIC 2150cc.
http://forum.aempower.com/forum/inde...c,26972.0.html
Dec 23, 2011, 02:06 PM
#10
Evolved Member
iTrader: (7)
Join Date: Feb 2007
Location: MoTown,GA
Posts: 574
Likes: 0
Received 0 Likes
on
0 Posts
Well Boz, What i did was put my stock injectors on top and the 2150's on the bottom. Run the 2150's full time on E85. But on pump gas just use the stock injectors for idle only so i can get 14.7AFR then kick in the 2150's asap. I havent tested this out yet but it should work.
Also, IMO the secondary injection on the AEM S1 is not very accurate. I used the "staged ratio calc" to get my multiplier and even tweeked in in some using the logs. The ecu was not responding to fuel map changes like it should, my fuel maps were weird looking but it held 12.0 AFR under boost. After i put the big injectors where they should be and tuned it normally it was all good.
Maps looked smooth and car ran smoother.
When i took my car to AWD to use their dyno ,Mike said to me " your gonna loose a motor using that duel injection on the AEM" So im just guessing im not the only one having problems with it. Maybe the OP can try what i have on my car and see how that works. It will sure run better under boost AND the stock ignition will go further.
Magnus V5 duel injection = poor fuel atomization and poor running engine. I saw a manifold somewhere that had the duel injectors side by side close to the head, similar to a Ford GT. I think it was a BMC race manifold. Thats what i want for christmas.
Also, IMO the secondary injection on the AEM S1 is not very accurate. I used the "staged ratio calc" to get my multiplier and even tweeked in in some using the logs. The ecu was not responding to fuel map changes like it should, my fuel maps were weird looking but it held 12.0 AFR under boost. After i put the big injectors where they should be and tuned it normally it was all good.
Maps looked smooth and car ran smoother.
When i took my car to AWD to use their dyno ,Mike said to me " your gonna loose a motor using that duel injection on the AEM" So im just guessing im not the only one having problems with it. Maybe the OP can try what i have on my car and see how that works. It will sure run better under boost AND the stock ignition will go further.
Magnus V5 duel injection = poor fuel atomization and poor running engine. I saw a manifold somewhere that had the duel injectors side by side close to the head, similar to a Ford GT. I think it was a BMC race manifold. Thats what i want for christmas.
Dec 23, 2011, 02:33 PM
#11
Newbie
Thread Starter
iTrader: (8)
Join Date: Dec 2007
Location: Western PA
Posts: 92
Likes: 0
Received 0 Likes
on
0 Posts
So you guys are suggesting that the issues are caused by the twin rail setup? I haven't had any real issues hitting any AFRs I could request thus far... and my fueling maps look pretty standard in my opinion? I have the 950s as the primaries and 850s in the secondaries. It definitely makes a lot more sense to have the larger injectors in the primaries to me... since they are the only ones that are going to get proper atomization and are the only ones aimed into the intake port at a reasonable angle. Do the secondaries just tend to puddle in/on that long injection tube?
For the staged injection I literally had just about no issues at all... just hooked it up, turned it on, set the max duty cycle, ran the wizard... never looked back. I'm not sure what to think now.
On another note, the Boomba is really offering some challenge as far as tuning. I suspect that's why you're thinking of switching back to a Mil-Spec? I contacted Boomba and they asked me some questions, but then never got back to me. I've spent hours just messing with the accel pump and throttle fuel offset. It's still not really "good" but it's reasonable at the moment.
For the staged injection I literally had just about no issues at all... just hooked it up, turned it on, set the max duty cycle, ran the wizard... never looked back. I'm not sure what to think now.
On another note, the Boomba is really offering some challenge as far as tuning. I suspect that's why you're thinking of switching back to a Mil-Spec? I contacted Boomba and they asked me some questions, but then never got back to me. I've spent hours just messing with the accel pump and throttle fuel offset. It's still not really "good" but it's reasonable at the moment.
Dec 27, 2011, 05:22 PM
#12
Evolved Member
iTrader: (7)
Join Date: Feb 2007
Location: MoTown,GA
Posts: 574
Likes: 0
Received 0 Likes
on
0 Posts
Yes and no. It is normal to have up to 3.8 volt spikes at times on the knock sensor. The easy way is count each spike that sticks out as 1. If you see more than 2-3 close to each other then it is knocking.
Yes, the fuel sticks to the wall of the tube. Heres the deal, My best friend is a fluid engineer. I was telling him about my misfire issues and my injector setup. Now what tiped me off was he said "When you have poor fuel atomization the O2 will read rich but the engine will show lean symptoms"... The light bulb came on for me and i fixed my problem after a year of searching. NICE!
Tuning with the boomba was not a problem. The reason i want a smaller one is because big throttle bodys in general make the car hard to drive slowly. I went a little to close to race car with some of my build so im just trying to get my drivability better.
Yes, the fuel sticks to the wall of the tube. Heres the deal, My best friend is a fluid engineer. I was telling him about my misfire issues and my injector setup. Now what tiped me off was he said "When you have poor fuel atomization the O2 will read rich but the engine will show lean symptoms"... The light bulb came on for me and i fixed my problem after a year of searching. NICE!
Tuning with the boomba was not a problem. The reason i want a smaller one is because big throttle bodys in general make the car hard to drive slowly. I went a little to close to race car with some of my build so im just trying to get my drivability better.
Dec 27, 2011, 11:36 PM
#13
Newbie
Join Date: Dec 2011
Location: usa
Posts: 7
Likes: 0
Received 0 Likes
on
0 Posts
is that log image you posted from a internal log?
if so, what is thhe logging rate set too?
also, after you let off the throttle, expect to get some voltage spikes, as everything is asked to do something different, whether its piston being load differently, gearbox, drivetrain, etc...
care you email your log to look at? pm sent
if so, what is thhe logging rate set too?
also, after you let off the throttle, expect to get some voltage spikes, as everything is asked to do something different, whether its piston being load differently, gearbox, drivetrain, etc...
care you email your log to look at? pm sent
Dec 28, 2011, 07:26 PM
#14
Evolved Member
iTrader: (7)
Join Date: Feb 2007
Location: MoTown,GA
Posts: 574
Likes: 0
Received 0 Likes
on
0 Posts
One other thing to note, for some reason the knock retard will kick in before the voltage goes above the threshold table. So keep an eye on the knock retard and tweek the table if needed to keep it from pulling timing early.
Dec 30, 2011, 06:25 PM
#15
Newbie
Thread Starter
iTrader: (8)
Join Date: Dec 2007
Location: Western PA
Posts: 92
Likes: 0
Received 0 Likes
on
0 Posts
Now I have conservative timing a more aggressive knock control setup. It definitely starts pulling timing before it reaches the threshold though. I was beating my head off the wall for a while because of that.
I think the car is going to get 1200cc primaries and 950cc secondaries now. Right now it's 950s and 850s. It's going to get a double batmowheel 362. Right now around 30psi it's hitting like 170% duty cycle. It'd definitely hit 180 before I could even max out the 362 non-BMW. So, I'm going to step up the primaries even more and continue to expect those to carry most of the weight/atomization. I'm not sure that I'm on board with giving up the Magnus at this point. The second I saw it I knew that atomization would be an issue given the spray pattern of a typical injector and the second rail tube. Of course, whenever air is flowing past the tubes there should be a pretty strong venturi effect that should assist in the process (which I'm sure was factored in given the angles on the tubes). I will however count on the primaries being 100% of whatever their actual flow rate is... the secondaries, I'm almost more so thinking of being "50%" of what they actually are. I think that's more how they will perform with the location. Just my .02 on the matter I suppose. And if I do 1200s as primaries and 950s as secondaries it should be more than enough to support 700whp without any hiccups in my mind?