*** Knock, do you have any?

Took another log today, and I'm consistently seeing that early boost feedback knock correction.

 

Any suggestions on what my next best step is to correct it? I don't think it was there with the previous revision of Cobb's stg2 map so I was thinking about going backwards a map, or I could try stepping down even further to the 91ACN oct map (even though I'm running good 93oct). Advise?

datalog18.csv

Took another log today, and I'm consistently seeing that early boost feedback knock correction.

 

Any suggestions on what my next best step is to correct it? I don't think it was there with the previous revision of Cobb's stg2 map so I was thinking about going backwards a map, or I could try stepping down even further to the 91ACN oct map (even though I'm running good 93oct). Advise?

 

FBKC across > 1 log at nearly identical RPM/load would probably indicate you need to change something. If it were me, at a minimum I'd probably edit that map's Primary Ignition table, take out out 1-2 deg from that ~2.6 g/rev & 3200 rpm region, smooth neighboring cells, flash the revised map and do another couple of pulls in similar conditions. The more expensive proposition, but one which could yield you better output and a better safety margin, would be to get a good pro/e-tune.

 

What gear were you in? Your sampling rate is somewhat slow-ish so it's possible that you're getting a brief boost pike right at that 3200 rpm transition that isn't showing up in the logs. Other alternatives would be to revert to the earlier map you had (as you proposed above) or try switching to a different brand 93 octane gas.

Been chasing phantom knock for like a year.

 

Turned out to be a bad fuel pump or o-ring in the fuel pump assembly. Upgraded pump and got my first clean learning view ever.

 

Just wanted to throw that out there since I hadn't seen fuel pump as a cause for knock before until now.

Hey guys,

I've got a MY06 3.0RB (the H6) which according to subaru has a 2 busted catalytic converters on it.

Car seems to be running fine, no noticible drop in performance, but I have only had it for the last 3000km of it's life.

 

Are these the readings you would expect to get off a car with that sort of problem, or is there something else going on here?

 

Any help or advice you guys could give me would be much appreciated.

I'm very new to all this stuff, and it's pretty complicated if you're not sure what you're looking at.

 

Thanks

Just wanted to take this moment to thank Dave B. @ CryoTune Performance for his great help to dial-in an ideal 1st stage tune on my beloved LGT.

 

I would recommend Dave to anyone looking to play it safe and get a Professional road tune for their ride, regardless of what make of vehicle they drive. Dave is an expert in datalog analysis and provided ample safety cushion for the 5EAT Stg1 tune (esp. in Target Boost and Base timing) to reduce risks of blowing a motor due to the presence of a massive restriction in the downpipe. He truly went the extra mile and ensured that the tune was safe for the long haul by having me drive around for a couple of weeks without resetting the ECU, and then recording a couple of WOT logs in 3rd gear. He was always prompt in sending the next rev. and very patient and detailed in his explanation of what he was doing. The high quality of his service certainly justified his modest price of $150

 

For anyone interested, we determined that the phantom knock was caused by one or both of the following:

1. A small boost leak

2. It appears something is making the ecu pick up knock in low loads that we have not been able to adjust. At 2400 rpm at light load and transitioning into acceleration which is causing the trans to kick a tad and cause the feedback. There is no way we are running too much timing or lack of fueling in that range.

 

So now the scenarios where I'm used to seeing knock, and not worried about, are:

 

1. Transitioning into acceleration (with Auto trans AKA 5EAT)

2. Letting go off the pedal

3. Front driver-side suspension rattle/clunk (happens seldom)

4. General knock noise during cruise/low engine load

Just wanted to take this moment to thank Dave B. @ CryoTune Performance for his great help to dial-in an ideal 1st stage tune on my beloved LGT.

 

I would recommend Dave to anyone looking to play it safe and get a Professional road tune for their ride, regardless of what make of vehicle they drive. Dave is an expert in datalog analysis and provided ample safety cushion for the 5EAT Stg1 tune (esp. in Target Boost and Base timing) to reduce risks of blowing a motor due to the presence of a massive restriction in the downpipe. He truly went the extra mile and ensured that the tune was safe for the long haul by having me drive around for a couple of weeks without resetting the ECU, and then recording a couple of WOT logs in 3rd gear. He was always prompt in sending the next rev. and very patient and detailed in his explanation of what he was doing. The high quality of his service certainly justified his modest price of $150

 

For anyone interested, we determined that the phantom knock was caused by one or both of the following:

1. A small boost leak

2. It appears something is making the ecu pick up knock in low loads that we have not been able to adjust. At 2400 rpm at light load and transitioning into acceleration which is causing the trans to kick a tad and cause the feedback. There is no way we are running too much timing or lack of fueling in that range.

 

So now the scenarios where I'm used to seeing knock, and not worried about, are:

 

1. Transitioning into acceleration (with Auto trans AKA 5EAT)

2. Letting go off the pedal

3. Front driver-side suspension rattle/clunk (happens seldom)

4. General knock noise during cruise/low engine load

 

Interesting... I'm seeing the same thing at first start. Knock sum goes up by one or two during on or off throttle events.

Yep, welcome to the world of 5EAT phantom knock.

you guys aren't special, I've got it on a 5mt.

We ARE special, in an autotragic kind of way.

Hey guys this is my first post and I am some what of noob to the world of Subaru. I have had my 05 LGT Limited 5EAT for a little over a year. It has about 130k on it.

 

Mods: Cobb AP 91 octane, SPT CBE, Cobb SF intake, Cobb DP and waiting to put on an Grimspeed upipe and intercooler.

 

I have been experimenting with the logging on the AP and have been getting some minor readings for Feedback knock and Fine Knock. I have been reading this forum for a few weeks now and trying to figure out how to read and diagnose the logs by myself but I am having trouble. I have posted some logs of just cruising and some pulls. If anyone has a chance to take a look at the logs and maybe tell me what I am seeing or even if I am logging correctly that would be greatly appreciated.

 

I've noticed that the Feedback knock is coming in when I'm not boosting high, mainly when not boosting at all but its not always there. The Fine Knock hasn't shown anything different then -0.7. I previously took the car back to a stage 1 tune from the OTS stage 2 map and the Feedback knock readings went down in range from about -6.0 to nothing higher than -2.0.

 

The First Log is with the stage 2 OTS map sorry its kinda long but there seems to be a bit of fine knock in it and feedback knock. Also my DA goes negative not sure if that is good...... this pull was at about 60mph

 

The second and third logs are stage 1 OTS Cobb map. Not as much of either knock but still evident. these pulls were from about 40-50 mph.

 

The fourth log is also the stage 1 OTS map just at cruising speeds not too high of boost load.

 

Thank you for your help!:)

datalog20.csv

datalog29.csv

datalog36.csv

datalog28.csv

Hello sticky thread bump. Welcome to the forums.

 

You're at 130K miles -- congrats. If you're on the original turbo, I hope you have made sure it's healthy! Read up on common failure modes, but don't scare yourself too much with other people's sob stories; just make sure you have enough oil.

 

That stage 2 map looks pretty clean to me if I'm interpreting that first log correctly. (disclaimer: I read a lot of datalogs and learned basic tuning a few years ago, and haven't paid as much attention lately.) Here's my rationale:

 

- knock events are near 2000 RPM or below

- knock events are not under heavy throttle

 

I use open source logging myself and tend to add a couple of other parameters. I usually like to pay attention to knock under higher loads as a priority, and I'm not sure from these logs what that is.

 

A couple other questions:

- Can you log engine load? It'll help you judge whether the knock event is worth caring about.

- Are these logs from 3rd gear pulls at under 2k rpm to near redline? Those are usually the most useful.

- Do you have a Learning View snapshot to go with these logs?

 

From the RomRaider world, here's a guide that might help:

http://www.romraider.com/forum/viewtopic.php?f=33&t=6156

 

From that post:

"Knock sensor activity below 3000 RPM and 1.0 load is normal, even on stock tunes. "

 

Hope that helps.

I did some more logs from third gear 2-2.5k to redline. I am not sure how to measure engine load or display learning view but I was able to do some 3rd gear pulls earlier today. Not so sure if the results are what I want to see but here we go. Stage 1 Cobb OTS Map, data log 40 appeared to be showing the most feedback knock readings. Thank you for the links to Romraider I found a nice data logging video

datalog37.csv

datalog38.csv

datalog39.csv

datalog40.csv

I did some more logs from third gear 2-2.5k to redline. I am not sure how to measure engine load or display learning view but I was able to do some 3rd gear pulls earlier today. Not so sure if the results are what I want to see but here we go. Stage 1 Cobb OTS Map, data log 40 appeared to be showing the most feedback knock readings. Thank you for the links to Romraider I found a nice data logging video

 

A couple things.

 

1) You indicate you're running a Cobb OTS map. I presume that means you are using an Accessport? If so, you should be able to select Engine Load as a logging parameter.

 

*** 2) You indicate in your above-quoted post, and in your prior post, that you were/are running a Stage 1 Cobb OTS map. Your prior post also indicates you're running a Cobb DP. The DP puts you at Stage 2. If the DP is still installed, you should -not- be running any Stage 1 map, esp not under boost.

 

3) You have single count FLKC stored in the 4800-5200 rpm / 5200-5800 rpm & presumably high engine load cells. Small FLKC isn't uncommon for OTS maps but if that doesn't learn itself out to 0 over time, I'd recommend looking into it (protune, etc).

 

4) The multi-count FBKC appears to happen as you come off the throttle. Might be something rattling as you transition from loaded to unloaded condition, or could be 5EAT 'phantom knock', not sure though.

Edited April 10, 2015 by serx7

This was helpful reading for me trying to determine harmful knock events.

 

 

It comes up as a PDF file and I cant figure out how to link or attach the file, so Google - Subaru Knock Monitoring FAQ v2 - COBB Tuning

Ok guys i need some help. I recently picked up a Btssm and one of the first things i noticed was knock, which has had me concerned ever since. I have searched to no avail for anything loose causing it. Last night i did some logging after doing a compression check on the drivers side cylinders and plugs. I looked afterword and noticed i was an idiot as my IAM had dropped considerably and really hope i didn't hurt it more than it might already be.

 

08 OBXT 5eat

catless dp, stock mid, sti muff

tgv deletes

dw65c

vf52

perrin inlet

perrin drop in

open source

fuel tank was probably around 1/4 (could be an issue) running 93, tuned on 91

 

Attached is one long log split into 2. Something appears very wrong to be me but i am also very new to this. I am talking with my tuner but haven't had time to meet yet. I guess i am asking for ideas on problems and if i should stop driving it at this point.

 

Also is this log useless, i have been trying to figure this out for about a week now and the logs i pulled last night are the only ones i have seen with the IAM dropped, maybe i missed hooking something up?

Book1.csv

Book2.csv

3 logs I did last week after noticing the knock initially.

BtSsm_20150407_223456_knock.csv

BtSsm_20150407_223759.csv

BtSsm_20150407_224004_knock.csv

imprezkid: It seems that you're getting some substantial FLKC corrections above 5500 rpm despite IAM being down at 0.562 (see rows 926-935 in Book1.xls). Assuming 'knocks' == knock sum, it looks like that log caught actual knock there, as 'knocks' increments from 6-7, then from 7 to 13.

 

The large (presumably) Feedback Knock Correction seen in BtSsm_20150407_224004.csv at rows 155-176 might be indicative of shift knock. The timestamping in this file is odd though. At row 154 (right before the big timing corrections are applied), the timestamp is 15708 ms. The very next row 155 (where large timing correction is applied and 'knocks' increments) has timestamp 20100 which looks to be ~5 sec later. Not sure what's up w/ that.

 

I'd def run these logs by your tuner, and would recommend staying out of boost esp above 5000 rpm.

 

Can you grab a Learning View, or a brief log showing your fuel trims? (AF Learning #1A, B, C, D).

I will see if i can grab one, i talked to my tuner last night and he asked me to look for any maf related leaks, so i cleaned that and in the process noticed my inlet had slipped quite a ways so i am not driving it till i get that fixed, id love for that to be the problem but i have a feeling im not that lucky. It is however T-bolt time on that thing. I really don't understand why there is no lip where the inlet hooks on.

Been fighting with some temperamental knock around 1,5000-2,000rpm at .50 g/rev to 1.20 g/rev, FBKC just goes nuts from -5 to -12* of timing pulled. This mainly happens in 4th gear at 30mph going up a hill (lots of them here) or 5th gear at 40mph. What's interesting it doesn't happen for the first 7-10minutes (even if car is fully warmed up). But starts knocking easily towards the last 2 minute of my drive.

 

I tried pulling 5-10* of timing from those areas in the tune and it just made the car very choppy and hesitant. Went back to the stock timing map for that portion, drives fine but same knock.

 

I thought it was the heatshields, which were loose, removed all of them (but the turbo one which I tightned back up), and it's still happening. It really makes me wonder if the cylinder develops a hot spot, but at that low load/rpm just doesn't seem likely.

 

Also I noticed that if I self induce knock (via free revving while clutching in) it will pull FBKC timing and then it's much more likely to keep on pulling timing as I start driving. Makes me wonder if my DMF is falling apart .

Been fighting with some temperamental knock around 1,5000-2,000rpm at .50 g/rev to 1.20 g/rev, FBKC just goes nuts from -5 to -12* of timing pulled. This mainly happens in 4th gear at 30mph going up a hill (lots of them here) or 5th gear at 40mph. What's interesting it doesn't happen for the first 7-10minutes (even if car is fully warmed up). But starts knocking easily towards the last 2 minute of my drive.

 

I tried pulling 5-10* of timing from those areas in the tune and it just made the car very choppy and hesitant. Went back to the stock timing map for that portion, drives fine but same knock.

 

I thought it was the heatshields, which were loose, removed all of them (but the turbo one which I tightned back up), and it's still happening. It really makes me wonder if the cylinder develops a hot spot, but at that low load/rpm just doesn't seem likely.

 

Also I noticed that if I self induce knock (via free revving while clutching in) it will pull FBKC timing and then it's much more likely to keep on pulling timing as I start driving. Makes me wonder if my DMF is falling apart .

 

I'm getting something simular. There are two false knock points, one around 1800rpm and one at 2500rpm. It happens the most around 1g load, but will show up at almost any point. Timing is sane in this area of the map, and the ECU will max out the FBKC without solving the issue. You can also see then at highway speeds when coasting (injectors off) so it's NOT actual knock.

 

Building an "electronic detcan" so try and find the noise source.

I'm getting something simular. There are two false knock points, one around 1800rpm and one at 2500rpm. It happens the most around 1g load, but will show up at almost any point. Timing is sane in this area of the map, and the ECU will max out the FBKC without solving the issue. You can also see then at highway speeds when coasting (injectors off) so it's NOT actual knock.

 

Been fighting with some temperamental knock around 1,5000-2,000rpm at .50 g/rev to 1.20 g/rev, FBKC just goes nuts from -5 to -12* of timing pulled. This mainly happens in 4th gear at 30mph going up a hill (lots of them here) or 5th gear at 40mph.

 

I thought it was the heatshields, which were loose, removed all of them (but the turbo one which I tightned back up), and it's still happening.

Also I noticed that if I self induce knock ...it will pull FBKC timing and then it's much more likely to keep on pulling timing as I start driving.

 

^ Exactly the same as above. I can hear a heat-shield-like rattle around I think the front end when I'm alongside a wall at about .7 - 1.2g at the same RPM as reported. I lifted up the car earlier today and put a wrench to everything in sight, starter included and it's still doing it, though it maxed at only -7.55 FBKC on APV3 after all that. I almost want to somehow fasten the rear diff heat-shield since it was the only thing I could find that just might make the same noise. Good thread. Sub'd

Edited November 4, 2016 by Outback2.5XT-05

When the heatshields were on, if I were to hit the downpipe with my hand I heard it rattling. After I removed the header heat shields hitting the downpipe resulted in no more rattles. Still driving around today after a while I started getting major knock.

 

Sadly my previous years of logs were mostly on cold started engine so it's hard to know if it's being going on for a while. I did find some logs that had similar trends last year at 1700rpm and under .70 g/rev.

 

I'm gonna check my spark plugs, maybe it's a sign of them going out or being too hot of heat range. Seems unlikely at this low of load though.

 

Another observation, 2007 tune has higher rpm & load start times for feedback and fine correction monitoring. Though Feedback still is monitored below 1700rpm, but at .80 g/rev. Maybe this is common engine noise and Subaru had to just disable knock control in these spots?

Update on my false knock: After replacing the catback things are MUCH better. I'm not seeing nearly as much pulled just driving down the highway. There is still something there though, need to hook up the electronic detcan (super simple to make) and see if I can correlate the sounds.

Update to this, I had some acorns (damn mice) and other debris contacting it. I vacuumed up as much as I could, tried to remove the knock sensor but it there is not enough clearance without removing the throttle body.

 

This seemed to have reduce the false knocks by half, but I would still get some feedback correction around 1500rpm now. I'm now more convinced it's false knock so I raised the FBKC min load to be at .80 g/rev.

Can I get a second opinion on this? 187,000 miles, all stock.

 

https://docs.google.com/spreadsheets/d/1_NaUIOWVpWYp4zv6VdApxYP3Zc0Akf7q3fXO7GTsF_o/edit?usp=sharing

 

https://docs.google.com/spreadsheets/d/1KM1PbWG3V7uVPFgRIC5EVXMmyDv_SKCsTaL0d9nyrhg/edit?usp=sharing