PDA

View Full Version : Barry's Evoscan - Removing the Knock



BraindG
28-05-2012, 10:13 PM
Now that i have evoscan installed, i have some questions, stupid as they may be...

On the Graphs and Maptracer menu, am i expected to input all the Power / Torque settings for the VR4? if so, what have others put in?

EDIT: now found this (http://www.clubvr4.com/forum/showthread.php?41712-Evoscan-Definitive-Settings) thread.

53943

regarding knock count, i want to set an alarm (if recommended?) - What are other using here?

53944

K, thats all for tonight :)

Thanks

EDIT: List of threads of relevance.

Evoscan Question (http://www.clubvr4.com/forum/showthread.php?37055-Evoscan-Question) - Discusses O2 Sensor reading 0.05856.
evoscan knock - anyone got some logs for comparison (http://www.clubvr4.com/forum/showthread.php?47482-evoscan-knock-anyone-got-some-log-files-for-comparison) - discusses possible causes for knock.

BraindG
28-05-2012, 10:29 PM
Graph question time...

Can someone explain to me why i jump to 15th gear at points? :rolleyes4 :smart:

53950

BraindG
28-05-2012, 10:33 PM
Sorry, your going to hate me soon :)

Can someone explain why the fuel consumption is so erratic just before i start slowing? - is it cos i take foot of accelrator?

53951

BraindG
28-05-2012, 10:43 PM
Sorry, your going to hate me soon :)

Can someone explain why the fuel consumption is so erratic just before i start slowing? - is it cos i take foot of accelrator?

53951

I've reviewed all 4 logs, what i notice is that when releasing the power (taking foot off pedal) there is a massive spike in fuel consumption - i'd have expected this for stamping on the pedal but not releasing - might have to start changing my driving habits..

BraindG
28-05-2012, 10:54 PM
Last one for tonight, off to play with electronics now :D

i note alot of knocksum when flooring car as high as about 18 :thinking: - is this the correct value to be measuring knock with? - and... should i be worried?

53952

Kenneth
28-05-2012, 11:16 PM
Gear is usually calculated by RPM / Speed. It is only useful while in gear and the clutch engaged. As soon as that is not the case, the RPM becomes independent of speed and the gear calculation will be wrong.



Graph question time...

Can someone explain to me why i jump to 15th gear at points? :rolleyes4 :smart:

53950

Kenneth
28-05-2012, 11:18 PM
How is fuel consumption being calculated? Do you have a VTA bov?


I've reviewed all 4 logs, what i notice is that when releasing the power (taking foot off pedal) there is a massive spike in fuel consumption - i'd have expected this for stamping on the pedal but not releasing - might have to start changing my driving habits..

BraindG
28-05-2012, 11:18 PM
Gear is usually calculated by RPM / Speed. It is only useful while in gear and the clutch engaged. As soon as that is not the case, the RPM becomes independent of speed and the gear calculation will be wrong.

cheers Kenneth, very clear now - makes sense, thank you.

Kenneth
28-05-2012, 11:19 PM
Yes, KnockSum is the correct value.

18 is bad. You should be aiming for 0. The OCCASIONAL count of 1 shouldn't be a problem though.


Last one for tonight, off to play with electronics now :D

i note alot of knocksum when flooring car as high as about 18 :thinking: - is this the correct value to be measuring knock with? - and... should i be worried?

53952

BraindG
28-05-2012, 11:30 PM
Yes, KnockSum is the correct value.

18 is bad. You should be aiming for 0. The OCCASIONAL count of 1 shouldn't be a problem though.

after some advise from Jason, i've been searching for "false knock" threads, found one here (http://www.clubvr4.com/forum/showthread.php?42175-Hooray!-I-have-got-Evoscan-working!-But-I-don-t-like-what-I-see). Few things that could lead to false knock, dodgy clutch, maybe engine mounts (im changing both this weekend) also HT Leads and plugs, even possibly fuel filter or pump.

i'll lower boost tomorrow and do same experiment, then stay on lower boost for a while if it goes away.

BraindG
28-05-2012, 11:38 PM
How is fuel consumption being calculated? Do you have a VTA bov?

[Speed]/(513*4*[InjDutyCycle]/100*0.015873)

Kenneth
28-05-2012, 11:40 PM
Why does he think it could be false knock? If you are getting knocksum when flooring it, chances are that it is real knock. You should be trying to correct it as such. If you can't get rid of it, then start looking for false knock.

18 is high and about what you would expect if you put your boost up to 15psi with only a standard intercooler.

BraindG
28-05-2012, 11:43 PM
Why does he think it could be false knock? If you are getting knocksum when flooring it, chances are that it is real knock. You should be trying to correct it as such. If you can't get rid of it, then start looking for false knock.

18 is high and about what you would expect if you put your boost up to 15psi with only a standard intercooler.

well in Barry's wee mind, dropping the boost may eliminate or rule this theory in? - if on lower boost im getting high knock, then i'd presume (under the same driving conditions, same stretch of road) its false knock.

p.s. im at 1.1 bar and on standard intercooler, 99 ron.

Kenneth
28-05-2012, 11:48 PM
[Speed]/(513*4*[InjDutyCycle]/100*0.015873)
So I assume this is default for an evo and should be

[Speed] = RPM
513 = injector size (so not correct for a VR-4)
4 = Cylinder count (so should be 6)

I have no idea what 0.015873 is meant to be for though

I guess you will want to check that IDC and Speed are correct and are giving useful information. Might pay to log them next to your fuel consumption so you can bench check to see if the values are sane.
Also, if you are logging 2 byte values there is possible read errors (values change quicker than the MUT requests are sent so values can change between 2 different values being received to the logger)

Kenneth
28-05-2012, 11:51 PM
1.1 bar with a standard intercooler and I can almost guarantee that it is real knock.

Yes, first step is to reduce boost. You can tune out the knock with ignition timing changes and/or fuel too.

false knock should NEVER be your first theory.

well in Barry's wee mind, dropping the boost may eliminate or rule this theory in? - if on lower boost im getting high knock, then i'd presume (under the same driving conditions, same stretch of road) its false knock.

p.s. im at 1.1 bar and on standard intercooler, 99 ron.

BraindG
29-05-2012, 12:20 AM
So I assume this is default for an evo and should be

[Speed] = RPM
513 = injector size (so not correct for a VR-4)
4 = Cylinder count (so should be 6)

I have no idea what 0.015873 is meant to be for though

I guess you will want to check that IDC and Speed are correct and are giving useful information. Might pay to log them next to your fuel consumption so you can bench check to see if the values are sane.
Also, if you are logging 2 byte values there is possible read errors (values change quicker than the MUT requests are sent so values can change between 2 different values being received to the logger)

Ok so based on this (http://www.clubvr4.com/forum/showthread.php?23550-stock-injector-size) thread, standard sizes are 390cc.

Im confused now actually, there are two calculations in evoscan.

53953

MPG (US) - [Speed]/(513*4*[InjDutyCycle]/100*0.015873)
Liters/100Kms - x*[InjPulseWidth]*20*6/[Speed]/1200

Firstly, i cant see how set one or the other as default and Secondly - neither much use to us here in the uk.

Anyone configured this for UK readings?

BraindG
29-05-2012, 12:32 AM
Why does he think it could be false knock? If you are getting knocksum when flooring it, chances are that it is real knock. You should be trying to correct it as such. If you can't get rid of it, then start looking for false knock.

18 is high and about what you would expect if you put your boost up to 15psi with only a standard intercooler.

Tbh, we exchanged 2 text messages, wasn't really a full discussion :)

Anyways, looking closer, the high knock occurs after lifting of the gas to change gear and then it dissapears when gear engaged and i've floored it again - Having read Ben's thread (http://www.clubvr4.com/forum/showthread.php?42114-Detonation-what-does-it-sound-like-for-the-uninitiated&p=458888&viewfull=1#post458888) (i'm no expert on detonation btw) i would have though knock occurs during load? - or am i looking at a build up of heat / sparks in bore between gear changes causing knock?

53954

Edit: Scratch that - i see now that there is a build up of knock leading up to the gear change, the slow decline in TPS is me easing off a little (chicken)

Kenneth
29-05-2012, 12:33 AM
Injector duty cycle (IDC) is how much the injectors are open as a percentage. At 50% then the injectors are open 50% of the time, regardless of what frequency.

Injector pulse width (IPW) is how long they are open for this injector cycle. You need the frequency for this as it IPW * frequency is how much fuel get delivered in total.

IDC should actually be sufficient for tracking fuel usage changes. (Useful if you don't care about actual MPG figures)

Kenneth
29-05-2012, 12:39 AM
If it is only on lift off, then it could be false knock. That graph resolution is much better than the first one

Do try reducing boost though as you want to eliminate that possibility first.

BraindG
29-05-2012, 12:51 AM
If it is only on lift off, then it could be false knock. That graph resolution is much better than the first one

Do try reducing boost though as you want to eliminate that possibility first.

Will do, ill drop the boost to 0.9 and do the same run - so under "normal" knock conditions, would you expect to see them underload?

looking at those graphs, they come in as im easing off, but not dropping gas to change gear. looks like i wasnt planting pedal to floor between each gear change, rather slowly decreasing before gear change.

Kenneth
29-05-2012, 12:56 AM
It could be going through a particular load / rpm cell that is causing knock then.

I would drop boost down as far as you can, then log. 0.9 is only a small drop and why do something twice?

I always log RPM, Load and Knocksum together so that I know if I should ignore the knock count.

BraindG
29-05-2012, 01:12 AM
I would drop boost down as far as you can, then log. 0.9 is only a small drop and why do something twice?.
True, ill switch it off for standard boost



I always log RPM, Load and Knocksum together so that I know if I should ignore the knock count.

This Load setting?

53955

Kenneth
29-05-2012, 01:23 AM
No.

Do you have a modified ROM? If so, use 2 byte Load (Requests 00 and 01), otherwise you want ECULoad (request 1C)

BraindG
29-05-2012, 01:25 AM
No.

Do you have a modified ROM? If so, use 2 byte Load (Requests 00 and 01), otherwise you want ECULoad (request 1C)

mope, standard rom - ok, ECUload it is.. ok thats me for tonight, thanks Kenneth!

BraindG
29-05-2012, 07:13 AM
I cannea seem to get the "Calc Power & Torque" option to work, I get "object reference not set to an instance of an object" when clicking the button.

53957

I've configured the Settings as follows (based on Jason's Reply (http://www.clubvr4.com/forum/showthread.php?41712-Evoscan-Definitive-Settings&p=682444&viewfull=1#post682444))

53956

I found the following (http://www.clubvr4.com/forum/showthread.php?37737-Evoscan-BHP-graph) thread mentioning a problem with it not working, but i think it differs to mine.

Any ideas?

BraindG
29-05-2012, 09:14 AM
Here's another example of the knock coming off load, this was taken this morning with the boost controller off, i presume this means it's on default boost or if i turn it off at 1.1 it stays at 1.1? - its a HKS EVC 5.

53958

Colin Wiltshire
29-05-2012, 04:38 PM
Following this with interest as I've just downloaded evoscan, saving up for the cable now.

foxdie
29-05-2012, 07:45 PM
Firstly, in my own defence, I did tell Barry by SMS that 18 knock is "moderately high" and something should be done about it, however when he told me it happens whilst changing gear, I immediately thought "it's potentially false knock", I did give him a warning it could "possibly" be false knock but not to assume it is ;)

That said, after actually seeing those graphs I'd say it'd be beneficial at this point to check AFRs in those knocking areas, it's hard to gauge what the cause is, it could be genuine knock or it could be a worn mount or clutch issue. Regardless, Kens advice is sound, start by reducing the risk and lower the boost to about 0.7-0.8 bar and see if it still happens.

(By default boost, that means turn the boost off so the car runs on the wastegate spring)

BraindG
29-05-2012, 08:01 PM
I already ran the test with boost off, 2 posts up - still getting 18 knock :(

I can't remember if I did maf mod, so going to check that and run again with no boost (boost controller switched off)

However the knock only appears to be there in later gears, not 1st or 2nd - if you review the graphs again you may notice - is thst normal?

however, today drivin home, built up area with speed bumps and a few T junctions I found the knock spiking really high, with no excess RPM (max about 1500-2000) - the knock alarm would sound turning left or right out of a T-Junction, aswell and completely randomly when going over speed bumps.

Further thinking today, the BOV could be affecting the fuel consumption, see one of the earlier posts.

foxdie
29-05-2012, 08:29 PM
It does sound like false knock however it'd be foolhardy to assume it is.

I'm hoping someone else can add more to this thread, maybe someone with more insight / experience with false knock. I haven't seen it yet but then every car is different.

BraindG
29-05-2012, 08:32 PM
Like anything in life one should never assume but, I still don't understand - under what conditions would 18 knock appear whilst pulling away from a t junction, max 2000 revs and keepin under 30mpg.

If that could be discussed it may give some clues, so what (not false knock) could cause 18 knock at those low conditions?

I'll post the CSV later when I boot up pc.

BraindG
29-05-2012, 10:46 PM
Yet another full tilt graph, i have attached the xls aswell - note that you will have to "unhide" the other criteria.

53961

Not sure what happened to my low rpm/speed log - its missing, will have to do another tomorrow.

Kenneth
29-05-2012, 10:48 PM
Looking at your latest graph, knocksum does still follow load.

Knock can happen at just about any load (though there is bound to be a cut-off) if the fuel and ignition are not correct.

Personally I would go out and get a 600x300x76 intercooler installed. Relatively cheap and gives you some scope to improve performance.

I have always found excessive knock when logging on vehicles with a standard intercooler.

BraindG
29-05-2012, 11:04 PM
Oh, found one...

53963

53964

BraindG
29-05-2012, 11:06 PM
Looking at your latest graph, knocksum does still follow load.

Knock can happen at just about any load (though there is bound to be a cut-off) if the fuel and ignition are not correct.

Personally I would go out and get a 600x300x76 intercooler installed. Relatively cheap and gives you some scope to improve performance.

I have always found excessive knock when logging on vehicles with a standard intercooler.

Just so happens, i have one coming soon.

I will be installing clutch and poly mounts before that though - so what ill do is run same test post installation and see what happens - then again once the FMIC is installed.

I just find it a bit weird that i get so much knock at low rpm/speed - i was actualll getting it on idle and just reving up to 4k - would you also expect that?

BraindG
29-05-2012, 11:14 PM
Actually, regarding the idle revving thing i just mentioned - here is the graph..

53966

Seems just the first rev spiked with knock, the rest of the spikes you can see cap at 2 knock - situation was, went shopping, came back, started car, got evoscan ready then the graph begins.. could just be build up of fuel or something that cause the first spike.

Kenneth
29-05-2012, 11:15 PM
I wouldn't expect that much knock doing just that.

BraindG
29-05-2012, 11:16 PM
I wouldn't expect that much knock doing just that.

Which you referring to Kenneth, the low revs or idle graph?

Davezj
29-05-2012, 11:24 PM
Barry I can't do pm with iPad, sorry for the hijack.

I am going ton have to call off the intercooler temp testing tomorrow, my father in law has fallen and broken his leg, he has gone to hospital and getting operation tomorrow, so we will be visiting him, tomorrow evening.

We will have to rearrange it.
Dave

BraindG
29-05-2012, 11:26 PM
No probs Dave, can maybe leave it till sat, no hurry :) - all the best to father in law!

Kenneth
29-05-2012, 11:26 PM
The one which shows ~20 knock at ~70 load.

Knock can happen just about any RPM and it doesn't require huge load. Without some more diagnostic information it is difficult to make any judgements on why you get knock or if it IS real knock or not... could even be a combination of the two.

BraindG
29-05-2012, 11:28 PM
The one which shows ~20 knock at ~70 load.

Knock can happen just about any RPM and it doesn't require huge load. Without some more diagnostic information it is difficult to make any judgements on why you get knock or if it IS real knock or not... could even be a combination of the two.

Well, i uploaded all the info in xls form, just need to unhide the data - if thats of any use?

Kenneth
29-05-2012, 11:41 PM
Don't see anything obvious sorry. My only thought is that if you are revving freely, there IS a chance you have false knock. That is still just one scenario though.

Do you have a flashable ECU?

BraindG
29-05-2012, 11:44 PM
Don't see anything obvious sorry. My only thought is that if you are revving freely, there IS a chance you have false knock. That is still just one scenario though.

Do you have a flashable ECU?

unfortunatly im stuck with a 7201 :( - so can't fiddle with anything.

lets see what happens post clutch/poly engine mount change, then again once i put on the new intercooler.

Thank you for checking Kenneth, value your time.

Kenneth
29-05-2012, 11:58 PM
Let me know if you decide you want to buy a 7202/3 flashable ECU, I have one which I don't really need (had it in case I blew up a couple doing mods, but have only managed to stuff up one :) )

foxdie
30-05-2012, 10:04 AM
unfortunatly im stuck with a 7201 :( - so can't fiddle with anything.

lets see what happens post clutch/poly engine mount change, then again once i put on the new intercooler.

Thank you for checking Kenneth, value your time.

If you don't mind a trip down to Brum I can reflash the 7201 ECU for you, although long term you'd probably be better off with a 7202/7203 :)

BraindG
30-05-2012, 10:10 AM
Cheers Jason, let's see how things are after my plan of attack, which is..

1. return MAF screw to original position
2. test run.
3. clutch/engine mount change
4. test run.
5. bigger intercooler on
6. test run.
7. new fuel filter
8. test run..

if after that i am still getting high knock on no and 1.1 boost then i have to consider more options, 1 being reflash ecufocusing on ignition timing changes and/or fuel.

wintertidenz
30-05-2012, 10:25 AM
Are you running a heat range colder plugs with the increased boost?

BraindG
30-05-2012, 10:26 AM
Are you running a heat range colder plugs with the increased boost?

i honestly cant remember what plugs are in there tbh - good point, i should check.

EDIT: Need a set of NGK BKR7EIX (http://www.sparkplugs.co.uk/pages/database/plug-detail-ccp.asp?partno=BKR7EIX)

Davezj
30-05-2012, 01:27 PM
ebay the NGK part number, there were loads on there last time i looked and relatively cheap as well.

they seem quite expensive now about £6 a plug, when i got mine, i am sure i only paid anout £3 a plug.

http://www.ebay.co.uk/itm/6x-NGK-Iridium-IX-Spark-Plug-BKR7EIX-2667-/230793990142?pt=UK_CarsParts_Vehicles_CarParts_SM&hash=item35bc64cffe

maybe £38 for 6 delivered is not so bad. but work shopping around.

BraindG
30-05-2012, 06:51 PM
i honestly cant remember what plugs are in there tbh - good point, i should check.

EDIT: Need a set of NGK BKR7EIX (http://www.sparkplugs.co.uk/pages/database/plug-detail-ccp.asp?partno=BKR7EIX)

Just pulled a plug out to check, I already have them - although I've no idea when they were put in, I'll search forums for clues..


54004

BraindG
30-05-2012, 07:43 PM
Ok, so i returned the MAF Screw to its original position (based on a secondary unmodified MAF i have) - I went for a couple runs and got max 5 knock - much better.

54005

I then pulled over and reset ECU, then got a max of 10 Knock - i only did one run and it was about 3 minutes after reset, i should go for a longer drive and get ECU relearnt - then try again.

54006

Let's see what happens tomorrow, nothing to get excited about yet.

adaxo
30-05-2012, 08:16 PM
Barry, you do that runs just now?? its a bit colder and damp outside now so that could help to reduce knock me think??

BTW great thread, watching with big interest as I got evoscan for ages now but never really have a clue what means what

BraindG
30-05-2012, 08:26 PM
Yeah I already considered that, the previous runs we're done just after sun set, 10ish I think - that run was done when outside temp was 26, didn't log what the temp was on previous runs. But you are right, it has to be taken into consideration.

I'll check the spreadsheets and get te inlet temp and reply.

BraindG
30-05-2012, 08:31 PM
BTW great thread, watching with big interest as I got evoscan for ages now but never really have a clue whadt means what
Tbh, I really don't know what's what either - but as part of my Job I spend hours searching through performance logs on servers. Looking for issues - so that bit I can do, just don't fully understand what everything means and or what effects they have on each other. But, we have some very knowledgable people on here that can provide some guidance.

BraindG
30-05-2012, 10:45 PM
Ok, some temp figures..

#32 (http://www.clubvr4.com/forum/showthread.php?63922-Barry-s-Evoscan-questions&p=704840&viewfull=1#post704840) - Max 15 Knock - Airtemp, before flooring it 84 - Airtemp at highest knock 78.8
#52 (http://www.clubvr4.com/forum/showthread.php?63922-Barry-s-Evoscan-questions&p=704974&viewfull=1#post704974) (After MAF Fix) - Max 5 Knock - Airtemp, before flooring it 89.6 - Airtemp at highest knock 86
#52 (http://www.clubvr4.com/forum/showthread.php?63922-Barry-s-Evoscan-questions&p=704974&viewfull=1#post704974) (After ECU Reset) - Max 10 Knock - Airtemp, before flooring it 96 - Airtemp at highest knock 91.4

need to run more logs tbh, as the post ECU Reset i think is a bit bogus, the car was sat still for a while heating up in sun whilst waiting for reset, then i was on idle for a while and did that test only a matter of minutes later - and not fogetting that the ECU is still in baby mode and needs to learn.

but looking at the first two, i got lower knock at higher temp post MAF Fix - again, i'd need to log some more consecutive runs to get a better overall view.

edit: but lowest knock occured where temp difference between "before flooring it" and "highest knock" was at its smallest, unrelated?

BraindG
30-05-2012, 10:49 PM
As a side note, i was doing some reading on the WWW regarding knock causes - most of which we have covered in this thread however, there was a couple additional points that were not mentioned yet.

1. Carbon build up - How would i clean this without stripping engine? (redex?)
2. Injectors need cleaning.

Thoughts on those two topics?

Kenneth
30-05-2012, 11:51 PM
If it were me, I wouldn't bother until getting that big intercooler in. No point in chasing remote possibilities until you address the most obvious/likely cause.

BraindG
30-05-2012, 11:54 PM
If it were me, I wouldn't bother until getting that big intercooler in. No point in chasing remote possibilities until you address the most obvious/likely cause.

Agreed - I'm just trying to build up a strategy in my head of next logical steps, in the chance that doesnt resolve - although, the googling i did before all suggest getting a bigger intercooler as a first step (just like you are suggesting). The reason im talking about the above was because two of the articles i read discussed the carbon build up due to using lower octane fuel, which i have done on a number of occasions.

Kenneth
31-05-2012, 12:55 AM
Came across this which was interesting: http://www.fiatforum.com/bravo-brava/127009-redex-engine-cleaner.html#post1328722

Might be worth a shot at least! By the sounds of things you can just buy white spirit and use that.

As for injectors, I think taking them out and having the properly cleaned (and flow tested) is probably the most worth while way to go. If you are going to do it, do it right.


Agreed - I'm just trying to build up a strategy in my head of next logical steps, in the chance that doesnt resolve - although, the googling i did before all suggest getting a bigger intercooler as a first step (just like you are suggesting). The reason im talking about the above was because two of the articles i read discussed the carbon build up due to using lower octane fuel, which i have done on a number of occasions.

BraindG
31-05-2012, 07:43 AM
To expand on the two points made, incase others like me dont understand..

1. Carbon build up.

If the carbon build up is allowed to continue you will potentially have deposits of carbon glowing red hot igniting the fuel as it enters the engine.
2. Injectors need cleaning.

If injectors are blocked this will affect the amount of fuel getting to engine and therefore affecting the air/fuel mix ratio.

Thanks for the link Kenneth, for the money its worth giving a go and i'll do this asap - this info affects my plan of attack,

01. Return MAF screw to original position - Done see post #52 (http://www.clubvr4.com/forum/showthread.php?63922-Barry-s-Evoscan-questions&p=704974&viewfull=1#post704974)
02. Add a cold airfeed to get more O2 to filter.
03. Clutch/engine mount change
04. Add Redex engine additive to clean carbon deposits
05. Bigger intercooler on
06. New fuel filter
07. Clean fuel injectors

BraindG
31-05-2012, 09:14 AM
Rather than using redex, has anyone ever used Amsoil Quickshot SE (http://www.performanceoilsltd.co.uk/amsoil_quickshot.html)?


Degraded fuel presents a major maintenance issue in small engines and powersports equipment. It can form varnish, gum and insoluble debris that clog carburetors, fuel injectors and fuel filters. Carbon buildup can form on the tops of pistons that leads to pre-ignition, rough idling and poor throttle response. Fuel-related problems are only expected to intensify in the coming years as the ethanol content in pump gasoline continues to increase.

AMSOIL Quickshot SE (AQS) is a premium fuel additive formulated to thoroughly clean and restore peak performance in small engine and powersports equipment fuel systems. It also stabilizes fuel between uses and during short-term storage periods. Its revolutionary technology focuses on three major fuel-related issues plaguing small engines and powersports equipment: ethanol, water and dirty pump gas.

BraindG
31-05-2012, 04:41 PM
Ok, so ive done about 40 miles post ECU reset and i thought i would give it another run, again no boost and same stretch of road, but different weather conditions - this time it was wet and 15 degrees outside temp (as shown on heater control unit)

54041

Max 1 Knock - Airtemp, before flooring it 93 - Airtemp at highest knock 80.6

I did the run twice, both times pretty much the same result.

Considering the conditions, the air temp on inlet was higher than on the hotter days :rolleyes4



#32 (http://www.clubvr4.com/forum/showthread.php?63922-Barry-s-Evoscan-questions&p=704840&viewfull=1#post704840) - Max 15 Knock - Airtemp, before flooring it 84 - Airtemp at highest knock 78.8
#52 (http://www.clubvr4.com/forum/showthread.php?63922-Barry-s-Evoscan-questions&p=704974&viewfull=1#post704974) (After MAF Fix) - Max 5 Knock - Airtemp, before flooring it 89.6 - Airtemp at highest knock 86
#52 (http://www.clubvr4.com/forum/showthread.php?63922-Barry-s-Evoscan-questions&p=704974&viewfull=1#post704974) (After ECU Reset) - Max 10 Knock - Airtemp, before flooring it 96 - Airtemp at highest knock 91.4


I will do this again when i go home, once round the route with no boost, once with 1.1 bar - THis should give me two examples of each condition.

Question is, has returning the MAF screw to original position and ECU reset caused this, or the outside temp? It's the first time i have had 1 knock during this whole monitoring process i've undertaken.

aboo
31-05-2012, 08:08 PM
Noticed you have a pod filter bolted directly to the maf. Could that be causing issues as well?

adaxo
31-05-2012, 08:26 PM
Barry just to clear things in my mind, that temp figures is in Fahrenheit??

BraindG
31-05-2012, 08:33 PM
Yep, it is in Fahrenheit

I have some interesting results from tonight, did 7 runs with some conclusive results.

Will post results later on

adaxo
31-05-2012, 08:36 PM
Yep, it is in Fahrenheit

I have some interesting results from tonight, did 7 runs with some conclusive results.

Will post results later on

/woot

I just wait for my miss to go back from work and will do few as well

Next thing to ask knowledgeable ppl, what that 'ecu load' means??

BraindG
31-05-2012, 10:47 PM
Ok, did some tests today - i felt they were quite interesting...

So, incase you've not kept up, currently the car is on 99ron with standard boost, the ECU was reset lastnight -( knock levels maxed at 1 in the previous tests. )

Today, i thought i would run a variety of tests, firstly i did 2 runs on normal boost to get a baseline, then i was going to switch to 1.1 boost, here's the graph - it shows 4 runs, 2 on normal boost 1 on 1.1 boost which instantly gave Knock at which point i decided to go back to normal boost ( instantly knock levels were low again)

54110

This proves that running 1.1 boost on standard kit causes knock.

what about the MAF Mod? - i decided unscrew the MAF screw and i ran two runs - look at the knock!

54111

Ok, so on standard boost with MAF Mod, knock increased... not good, so i returned the screw to normal and did another two runs - un-oh, still getting knock!

54130

I didn't expect this, i thought it would go back to normal - decided to reset ECU and hey presto... no knock again.

54131

Interesting.

So in summary..

On standard boost, after ECU reset i got no Knock
On 1.1 boost, after ECU reset i got Knock
With MAF mod, i got Knock.
After MAF fix, i still got knock
ECU reset fixed the Knock.

So, i would always recommend resetting ECU if you have knock and retesting - its like the ECU remembers and therefore continues to ignite/perform as before.

BraindG
31-05-2012, 10:54 PM
All my tests are being performed on the same stretch of road...

54145

BraindG
31-05-2012, 10:57 PM
Next thing to ask knowledgeable ppl, what that 'ecu load' means??

I think its something to do with the fuel pump and how its delivering?

Kenneth
31-05-2012, 11:07 PM
ECU Load is the calculated amount of effort that the engine is putting in.

The effort is directly related to how much air is going through the system, so the main input for load is the air flow sensor. The load is then adjusted based on a few things such as air temperature and barometric pressure.

ECU Load and RPM are the axis for the fuel and ignition timing lookups, so anything which alters that can potentially have a harmful effect. (such as the MAF mod, which is now not necessary due to the ability to flash the ECUs)

adaxo
31-05-2012, 11:36 PM
ECU Load is the calculated amount of effort that the engine is putting in.

The effort is directly related to how much air is going through the system, so the main input for load is the air flow sensor. The load is then adjusted based on a few things such as air temperature and barometric pressure.

ECU Load and RPM are the axis for the fuel and ignition timing lookups, so anything which alters that can potentially have a harmful effect. (such as the MAF mod, which is now not necessary due to the ability to flash the ECUs)

Thanks for that, so its basicly means what is say in its name:), just wonder why is oscillating up to 160?? is that percentage??

Anyway not see a point to start new thread with my logs so Barry here for you my two logs just for comparison maybe:d??

LPG

54177

Petrol
54178


looks like I got a tiny wee knock but is in weird 'places' not under load??

BraindG
31-05-2012, 11:42 PM
nice, we can have a chat about this on sat, show you how to use it with excel too. - btw, you hardly have any knock :D - and you are brutal on the loud pedal

adaxo
31-05-2012, 11:46 PM
and you are brutal on the loud pedal

Its just for science :laugh:/rally

Davezj
01-06-2012, 12:02 AM
have you both read merlin's guide to ecu tuning the vr4.

it is very useful and explains a lot of the info you need. it is not complete, and if you use kenneths ecu mods a lot of the detail will not apply but the concept and the explanation of how the tables interact with eachother is very useful. not to mention how various peramiters are worked out and what they are based on, Knocksum and ecu load for instance.

it is either on this forum or it is on OZVR4 but there are links to it.
i read it about 7 or 8 times to try and get my head round what is going on and made notes for myself to explain in terms that mean something to me explaining what everything does.

It really is not something you want to be jumping straight in on, you need to read, understand, and reread and realise you got the wrong end of the stick, digest some more info, leave it for a couple of hours, reread it, start making notes for yourself, realise you still haven't got it, reread it and then it will start making some sense when the notes you are making start painting a picture that you can understand.
well that is what i did, and it took me all day to get the basics of the merlins manual. you will need to understand merlins manual before you start playing with kenneths ecu mods.
once you start getting it, it will all become clear.

if there was a flow chart showing how all the tables interacted with eachother it would make things much simpler but i don't think there is anything like that out there.

Kenneth
01-06-2012, 12:05 AM
160 (just less actually, 159.375) is the maximum load you can display without doing the 2byte load mod.

1byte = numbers between 0-255. The load formula 0.625 x VALUE where value is the number between 0 and 255

My understanding is that load is roughly proportional to the absolute volumetric efficiency (VE) of the engine. So at 1bar absolute (0 boost, 0 vacuum) you should get around 100 load.

lateshow
06-06-2012, 08:36 PM
Use 390x6 or little less and you will get the right fuel consumption.....

BraindG
12-06-2012, 02:22 PM
Ok, back to this topic again...

The Clutch/engine mount change was aborted as I had the incorrect friction (http://www.clubvr4.com/forum/showthread.php?63227-Which-Clutch-!&p=705573&viewfull=1#post705573) plate, therefore this has now been moved down the list.

01. Return MAF screw to original position - Done see post #52 (http://www.clubvr4.com/forum/showthread.php?63922-Barry-s-Evoscan-questions&p=704974&viewfull=1#post704974)
02. Add a cold airfeed to get more O2 to filter - Done, no difference logged (not uploaded any data on this sorry, but logs showed significant knock on 1.1bar)
03. Add Redex engine additive to clean carbon deposits - Done (http://www.clubvr4.com/forum/showthread.php?64090-Mpg), will do some runs this evening and provide some logs for you.
04. Bigger intercooler on.
05. New fuel filter
06. Clutch/engine mount change
07. Clean fuel injectors

BraindG
12-06-2012, 02:24 PM
Use 390x6 or little less and you will get the right fuel consumption.....
Oh, is this for the formula within evoscan? - what does it mean, could you break it down for me please, i'm a little unclear.

lateshow
12-06-2012, 06:39 PM
Yes for the formula... It says 4 x 513 in the formula, you just replace that with about 6x 390 or actually a little less would be in order, then you're set to go to see the momentarily consumption about right. :)

BraindG
12-06-2012, 07:02 PM
Yes for the formula... It says 4 x 513 in the formula, you just replace that with about 6x 390 or actually a little less would be in order, then you're set to go to see the momentarily consumption about right. :)

Oh ye, remember why now - its the injector size and quantity.

Spirit
12-06-2012, 10:26 PM
Barry, sorry if anyone else has mentioned it.....but have you tried different fuel ?

At a guess are you running with Tesco 99 ?

I cannot recall if it was the Jack Daniels VR4 or the Evo 8, but I was getting knock readings of 6/7 with Tesco 99. I put in Shell VPower and my readings dropped to a max of 1 right away !

Worth a try.

Kenneth
12-06-2012, 11:57 PM
so what are your knock readings now with the bigger intercooler?

BraindG
13-06-2012, 10:36 AM
ok, the results are in, all runs performed on Shell Optimax - Thread discussing intercooler install can be found here (http://www.clubvr4.com/forum/showthread.php?64172-Barry-s-FMIC-Install).

Before IC, I did 4 runs.


Normal Boost
Normal Boost
1.1bar
Normal Boost

Outside temp - 16C
Max Knock - 14@ 6:59:13 PM

54517 54515

Post IC, I did 6 runs.


Normal Boost - but very carefull on throttle (incase of loose pipes)
Normal Boost - a bit quicker, still nervous of loose pipes
Normal Boost - quicker still
Normal Boost - Aborted, traffic!
Normal Boost - full throttle
1.1bar

Outside temp - 15C
Max Knock - 13@ 9:34:06 PM

54518 54516

Summary: Althought I am still getting high knock, the quantity of the knock is lower (please see the 1.1bar screenshots)

What are peoples thoughts on this now?

Davezj
13-06-2012, 01:33 PM
i don't think an intercooler upgrade in this coolish weather is going to make any difference to knock with regard inlet temps.

which intercooler did you buy and ebay add link would be good so i can see if it is the same one i bought. then when i fit mine i can let you know if it is any more efficient than the stock factory one. you would imagine it is, but from the test i have done with the factory intercooler the outlet temp is approximatlely half the inlet temp with in a certain range of temps.

inlet temp (deg C) outlet temp (deg C) All aprox values from watching UTCOMP screen
30 18
40 20
50 25
60 30
70 35
80 43

80 is max UTCOMP can measure. and is not full boost pressure por any prelonged runs, a few small squirts of WOT at motorway speeds.
but as the inlet temp climbs towards 80 the outlet temp does start to struggle to stay at half temp. so maybe the max efficiecy of the intercooler has been reached at this point.

maybe this is where the bigger intercooler is going to help and maintain the half inlet temp for longer up the temp range or even reduce the outlet temp by more then half.
I will do a thread on it once i have done my upgrade.

does anyone know what temp the inlet needs to be before it starts causing issues.

Davezj
13-06-2012, 01:41 PM
i have just noticed the time on the graphs and had a thought on the subject,
barry you have done your runs very close together.
when you do runs that close together there is a continuing heat build up and it may affect you results, i think you need to leave a few minutes between runs to let thing settle down before the next WOT to give 100% meaningfull data.
the engine and engine bay will be getting hottter and hotter on every run you do if the time between is extended out more.
but i don't know how much effect thiswill have your results.
just a though for the future data gathering.

BraindG
13-06-2012, 03:06 PM
Cheers Dave, yeah - was thinking about your reply on the way to work this afternoon. I think because i have done all my tests like this so far, i should continue to do so - Also, maybe taking a break between run's isn't a real world test, i mean if you out and about giving it some welly then I would like to be in position to have removed the knock and not worry about it, if that makes sense?

Anyways...

I added a pipe from fog light to Mushroom this afternoon, went for a drive - This means i now have a 4" pipe (tumble dyer) and the snorkle providing fresh air to mushroom.

Also, before i continue, i've made a bit of a mis calculation on my high boost runs, i presumed it was still 1.1 bar, turns out its more like 0.8 - So i fiddled with my knobs and set A to 1bar and B to 0.8bar - So each of the times i have reffered to 1.1bar, it was actually 0.8 :o oops.

Ok, so first run with pipe..

Outside temp 15C
max knock 8 @ 1:45:05 PM

54531

I also had Vanessa video the EVC during this run ( i was checking the boost level, which confirms i was running 0.8 all the time and not 1.1 /Facepalm)


http://www.youtube.com/watch?v=715_8JY077c

ok, second run, now running 1.0.

Outside temp 15C
max knock 10 @ 1:48:40 PM

54532

I've attached the full log if you are interested.

BraindG
13-06-2012, 03:14 PM
The plan thus far.

01. Return MAF screw to original position - Done see post #52 (http://www.clubvr4.com/forum/showthread.php?63922-Barry-s-Evoscan-questions&p=704974&viewfull=1#post704974)
02. Add a cold airfeed to get more O2 to filter - Done, no difference logged (not uploaded any data on this sorry, but logs showed significant knock on 1.1bar)
03. Add Redex engine additive to clean carbon deposits - Done (http://www.clubvr4.com/forum/showthread.php?64090-Mpg), See first part of thread for figures #84 (http://www.clubvr4.com/forum/showthread.php?63922-Barry-s-Evoscan-Removing-the-Knock&p=707380&viewfull=1#post707380) under heading "Before IC, I did 4 runs."
04. Bigger intercooler on. - Done, Althought power through the gears is better, knock still there - results #84 (http://www.clubvr4.com/forum/showthread.php?63922-Barry-s-Evoscan-Removing-the-Knock&p=707380&viewfull=1#post707380) - FMIC Install Thread (http://www.clubvr4.com/forum/showthread.php?64172-Barry-s-FMIC-Install).
05. New fuel filter
06. Clutch/engine mount change
07. Clean fuel injectors

K, so as you can see, i'm running out of options now :)

Few questions.

What are the chances of the false knock theory now?
What influence does the dump valve have on all of this? (i was speaking to a scooby friend at work, his tuner has to remap cars with dumpvalves to compensate for the loss of pressure)
What else can I do?

Colin Wiltshire
13-06-2012, 04:09 PM
What influence does the dump valve have on all of this? (i was speaking to a scooby friend at work, his tuner has to remap cars with dumpvalves to compensate for the loss of pressure)
What else can I do?
[/LIST]


That is a good question, As I also have high knock with large fmic, fuel filter changed, waiting for cleaned injectors. I was wondering how much difference a vta has on knock compared with the original recirc?

BraindG
13-06-2012, 04:10 PM
K, in the name of science :vulcan:

I wanted to see what the difference in airtemp was before and after the tumble dryer pipe i added was (in addition to the snorkle) , it's positioned beside foglight and is fed up to the mushroom filter.

Pre pipe installation - as you can see the temp never drops below 80.

54536

Post pipe installation - At higher speeds the temp drops to 66.2

54535

Thought i'd share that with you :)

BraindG
13-06-2012, 04:11 PM
That is a good question, As I also have high knock with large fmic, fuel filter changed, waiting for cleaned injectors. I was wondering how much difference a vta has on knock compared with the original recirc?

Have you got a atmospheric DV?

Colin Wiltshire
13-06-2012, 04:12 PM
Have you got a atmospheric DV?

Yes, it vent to air. Blocked off the original pipes

Davezj
13-06-2012, 05:54 PM
barry i have a spare stanard recirc vavle you can try if you want to.

BraindG
13-06-2012, 06:30 PM
Dave, I think I may borrow that for a few test runs. I'd have to compare to your car, I can't remember how to plum it in presuming you don't have a aftermarket BOV?

BraindG
13-06-2012, 07:14 PM
Actually, I have an adjustable BOV, recon I could tighten it right up so it doesn't release?

VR457
13-06-2012, 08:44 PM
Can i ask if you have tried the MAF mod, reset the ecu, then go for a run? At the moment it seems that you put the MAF to normal and then saw a 0 knocksum but that would not be the same as resetting?

BraindG
13-06-2012, 08:48 PM
I did do a reset after putting maf back to normal, there was an improvement but not enough to be happy.

I've not reset the ECU since though, my thinking being that the car would relearn ad-hoc - there has been an improvement with the bigger intercooler though. If you compare the recent logs to the ones from earlier on.

I might go for a run in a moment and tighten up the BOV.

Nick Mann
13-06-2012, 10:05 PM
Barry, looking quickly at a couple of graphs above, it looks like you are getting knock at gear changes. What happens if you short shift then drive through the rev range? What happens if you feed in the throttle more on gear changes? What happens if you do a run from say 2000rpm to the redline and then suddenly lift? What happens if you have the car holding a constant 3500 rpm and then slam the accelerator?

I suspect but can't say for sure, that you have false knock. The irritation with that is finding what is causing it. Soggy engine mounts, a loose bolt, or an aggressively sprung clutch? Or plenty of other things that haven't sprung to mind yet. The problem is that because the ecu senses knock, it will pull timing.

VR457
13-06-2012, 10:14 PM
I did do a reset after putting maf back to normal, there was an improvement but not enough to be happy.

What i meant was the opposite, resetting after modding the maf...

BraindG
13-06-2012, 10:22 PM
Thanks nick, I'll do a few runs as you suggested - I want to do as many tests as possible before I change the clutch and engine mounts.

Not sure if you been following the entire thread, but early on I spotted that the knock was appearing on gear changes, which at the time suggested false knock.

I really hope it is false knock but, before this I want to make sure I have covered all possibilities, just incase.

Next thing to try is adjusting the BOV, do it doesn't dump, run some test you suggested, which I presume is to see whether the knock appears when engine is rocked?

After this it's the clutch and mount change.

Kenneth
13-06-2012, 10:36 PM
Do you have a pod filter Barry? I got the impression from a couple of your posts, but can't rightly tell by them.

If so, are you able to get hold of a standard intake to try that?

BraindG
13-06-2012, 10:44 PM
I'm not sure Kenneth, anyone in Manchester got one?

Davezj
13-06-2012, 11:00 PM
i have got one but it is on the car. if you need to borrow it, it could be arranged.

i am having knock issues as well though.

BraindG
13-06-2012, 11:08 PM
It would be worth trying Dave, any chance we could meet up, I could grab it out of your car and do a couple runs. At least it would tick another thing off the list?

Nick Mann
13-06-2012, 11:22 PM
Yep - knock showing on sudden throttle changes can be (but isn't necessarily) false knock. However, big throttle changes can give real knock too. I'm assuming after reading the thread properly that you don't have a wideband? AFR's would be interesting to know.

Davezj
13-06-2012, 11:25 PM
i am sure we can work something out for this weekend.

BraindG
13-06-2012, 11:28 PM
Cool, Sunday is out of question for me, it's fathers day and I'm being treated.

If sat is ok with you I'll arrange to meet for an hour or two?

Davezj
13-06-2012, 11:34 PM
yes that should be fine, we can just swap inlets for a few days and you can go off and do wat you what to do. then when your finished we can swap back.

BraindG
13-06-2012, 11:36 PM
That could work, I do have a spare maf with a cone filter on it, maybe you could take that and at least then I can continue tests with the same parts?

Davezj
13-06-2012, 11:48 PM
yeh that sound like a plan.

aboo
14-06-2012, 06:43 AM
B
Do you have a pod filter Barry? I got the impression from a couple of your posts, but can't rightly tell by them.

If so, are you able to get hold of a standard intake to try that?I asked the same question in post 64/yes. Nice to be ignored :p

BraindG
14-06-2012, 08:18 AM
updated the plan of attack.

01. Return MAF screw to original position - Done see post #52 (http://www.clubvr4.com/forum/showthread.php?63922-Barry-s-Evoscan-questions&p=704974&viewfull=1#post704974)
02. Add a cold airfeed to get more O2 to filter - Done, no difference logged (not uploaded any data on this sorry, but logs showed significant knock on 1.1bar)
03. Add Redex engine additive to clean carbon deposits - Done (http://www.clubvr4.com/forum/showthread.php?64090-Mpg), See first part of thread for figures #84 (http://www.clubvr4.com/forum/showthread.php?63922-Barry-s-Evoscan-Removing-the-Knock&p=707380&viewfull=1#post707380) under heading "Before IC, I did 4 runs."
04. Bigger intercooler on. - Done, Althought power through the gears is better, knock still there - results #84 (http://www.clubvr4.com/forum/showthread.php?63922-Barry-s-Evoscan-Removing-the-Knock&p=707380&viewfull=1#post707380) - FMIC Install Thread (http://www.clubvr4.com/forum/showthread.php?64172-Barry-s-FMIC-Install).
05. Put on standard Air intake
06. tighten up BOV, preventing release of pressure
07. perform some tests to recreate engine rock, without gear changes, shortshift.
08. New fuel filter
09. Clutch/engine mount change
10. Clean fuel injectors

adaxo
14-06-2012, 08:34 AM
As mentioned above,

Perhaps tired gbox bearings can made strange noises, fouling knock sensor. Should we worrying if 'knock' appears at weird conditions, like not under load??

Dave/Barry can you pm me details of your meeting on sat?? got an issue with my car which I need 'a fresh eye' to look at it.

Davezj
14-06-2012, 01:43 PM
Cheers Dave, yeah - was thinking about your reply on the way to work this afternoon. I think because i have done all my tests like this so far, i should continue to do so - Also, maybe taking a break between run's isn't a real world test, i mean if you out and about giving it some welly then I would like to be in position to have removed the knock and not worry about it, if that makes sense?

Anyways...

I added a pipe from fog light to Mushroom this afternoon, went for a drive - This means i now have a 4" pipe (tumble dyer) and the snorkle providing fresh air to mushroom.

Also, before i continue, i've made a bit of a mis calculation on my high boost runs, i presumed it was still 1.1 bar, turns out its more like 0.8 - So i fiddled with my knobs and set A to 1bar and B to 0.8bar - So each of the times i have reffered to 1.1bar, it was actually 0.8 :o oops.

Ok, so first run with pipe..

Outside temp 15C
max knock 8 @ 1:45:05 PM

54531

I also had Vanessa video the EVC during this run ( i was checking the boost level, which confirms i was running 0.8 all the time and not 1.1 /Facepalm)


http://www.youtube.com/watch?v=715_8JY077c

ok, second run, now running 1.0.

Outside temp 15C
max knock 10 @ 1:48:40 PM

54532

I've attached the full log if you are interested.

Barry i have just had a look at this post again. by fitting the bigger intercooler you have removed some restriction from the boost pipework from turbo to throttle body, which has reduced the pressure seen by the boost gauge.
So restiction has been removed, but air flow rate has not been changed. meaning you are getting the same amount of air into the engine at 0.8bar as you were doing at 1.1 bar.
By turning the boost to the so called "original 1.1bar" you are actually increasing the air flow into the engine over and above what you had flowing before the bigger intercooler was fitted. by doing this you are not comparing like for like.

i believe you should be doing your knock test at 0.8bar and see if the kock is reduced, if so then the intercooler has helped to reduce knock. you can then turn the boost up further if you can, monitoring for knock as you do.

I am presuming you did not get a drop in performance after changing the intercooler and running at 0.8Bar?

if you had not changed the intercooler and just dropped the boost from 1.1bar to 0.8bar you would have felt a big difference in performance.

to put it another way what you have done is like keeping the original intercooler and turning the boost up to 1.4bar and expecting the knock level to reduce.

hope my logic is correct here.
i just looked at the situation from a different angle.

Davezj
14-06-2012, 01:45 PM
As mentioned above,

Perhaps tired gbox bearings can made strange noises, fouling knock sensor. Should we worrying if 'knock' appears at weird conditions, like not under load??

Dave/Barry can you pm me details of your meeting on sat?? got an issue with my car which I need 'a fresh eye' to look at it.

will give you a shout about saturday, not sure what the weather is going to be like, it does not look good, so i thought maybe a mini meet at karl's if he can fit us in.
i need to do a few thing to my car and a dry environment would be really helpful.

BraindG
14-06-2012, 01:56 PM
will give you a shout about saturday, not sure what the weather is going to be like, it does not look good, so i thought maybe a mini meet at karl's if he can fit us in.
i need to do a few thing to my car and a dry environment would be really helpful.
VR4Kaos - this possible karl?

Davezj
14-06-2012, 04:31 PM
Barry i have just had a look at this post again. by fitting the bigger intercooler you have removed some restriction from the boost pipework from turbo to throttle body, which has reduced the pressure seen by the boost gauge.
So restiction has been removed, but air flow rate has not been changed. meaning you are getting the same amount of air into the engine at 0.8bar as you were doing at 1.1 bar.
By turning the boost to the so called "original 1.1bar" you are actually increasing the air flow into the engine over and above what you had flowing before the bigger intercooler was fitted. by doing this you are not comparing like for like.

i believe you should be doing your knock test at 0.8bar and see if the kock is reduced, if so then the intercooler has helped to reduce knock. you can then turn the boost up further if you can, monitoring for knock as you do.

I am presuming you did not get a drop in performance after changing the intercooler and running at 0.8Bar?

if you had not changed the intercooler and just dropped the boost from 1.1bar to 0.8bar you would have felt a big difference in performance.

to put it another way what you have done is like keeping the original intercooler and turning the boost up to 1.4bar and expecting the knock level to reduce.

hope my logic is correct here.
i just looked at the situation from a different angle.

I have done so lateral thinking and rechecked my theory above, and it does not stack up.
for the pressure to change due to a rescriction being removed the pressure would ahve to measured at the restriction e.g. in the intercooler, and i am assuming the pressure is measured in the plenum chamber and the volume of that area has not changed.
so the other theory is the bigger intercooler is doing what it says on the ti,n and cooling the air more than the original intercooler and it is this reduction in temperature that is causing the pressure drop. air flow has still not changed but as the temp decreases so does the volume of the air and thus reducing the pressure.

So my initial concept of not increasing the boost back up to 1.1.bar was correct but not for the right reasons. knock should be measured at the reduced 0.8bar.
again i hope this is sound theorising.

adaxo
14-06-2012, 05:01 PM
VR4Kaos - this possible karl?

If not possible, you guys are welcome to come to my garage, at least we can work on one car at the time and not getting wet. Half a mile from me is a good 'test road' as well:P

VR4Kaos
15-06-2012, 11:31 AM
No problem Barry i'm willing to let all cvr4 members have the use of my unit f.o.c just bring your cvr4 membership card to confirm you are a fully paid member of clubvr4 hehehe

BraindG
15-06-2012, 11:35 AM
Nice one, thank you Karl! - /grouphug

Davezj
15-06-2012, 01:37 PM
barry have you thought about my theory, that you should be doing you est now at 0.8 bar.

are the results directly after fitting intercooler that say they are 1.1bar are actually 0.8bar, if so then my theory does not hold up.
as you are still getting knock of about 12-14 counts, you are monitoring knocksum and not the raw knock value.

just went back to look at sum of you runs and it does say knocksum.

BraindG
15-06-2012, 02:04 PM
I have considered it, let's chat tomorrow.. the problem is, im not actually sure what it was set to before, the onyl way to find out would be to put original intercooler back on and test it.

Davezj
15-06-2012, 09:44 PM
ok,
so i need to bring leak testing stuff for adam

standard inlet pipework, and standard recirc vale and pipe for barry.


plus stuff for me.

can you think of anything else i need to bring.

BraindG
15-06-2012, 09:55 PM
Can't think of anything, what time we planning to meet?

Davezj
15-06-2012, 09:58 PM
about 9.30am
i have doughnuts to bring as well.

BraindG
16-06-2012, 12:29 AM
Kenneth

question...

Please compare the first two images in these threads #72 (http://www.clubvr4.com/forum/showthread.php?63922-Barry-s-Evoscan-Removing-the-Knock&p=705210&viewfull=1#post705210) and #87 (http://www.clubvr4.com/forum/showthread.php?63922-Barry-s-Evoscan-Removing-the-Knock&p=707448&viewfull=1#post707448)

i know for a fact that my TPS was 100% between each gear change, so why does my graph show a decline before each gear change?

we both have a manual, adam has uprated clutch and engine mounts, i don't...

Kenneth
16-06-2012, 03:10 AM
Have you tried sitting in the car and seeing what sort of TPS response you get? (I assume you can do a live graph?)
I have found that foot to the floor is one thing, but variation of pressure can be enough to change the TPS. It could be that the throttle cable could do with some adjustment.

Other than that, not entirely sure why it would taper off like that. Next time it might pay to be a bit more aware of it and seeing if you can link it to any particular activity (pedal pressure etc)

I assume your car is factory manual and doesn't have a ASC/TCL throttle body on it?

BraindG
17-06-2012, 12:04 PM
Yep, it's factory manual.

I sat on the drive before and with ignition on I pressed the pedal right down, evoscan showed the TPS as 100%, no drop off.

What I will do today is tighten up the cable and give it a wee run and see if I still get drop off on the TPS.

I've also decided to stop any more tests after this till I put in my new clutch and engine mounts.

If its still showin loads of knock after this ill resume with some follow up tests.

BraindG
20-06-2012, 09:03 PM
Taking on board what Kenneth mentioned about making sure I was pressing peddle correctly I went out for a run with this at the front for my mind.

Guess what? - the reason I was getting drop off was due to the way the peddle is being pressed, if I really really pushed hard the tps would flat line.

So I'm going to try bending the metal rod/peddle/thingy so I don't have to push really really hard.

Thanks Kenneth, at least that's one thing answered!

Barry

Davezj
20-06-2012, 10:06 PM
barry as standard the max tps you should see with evoscan is about 96% and the minimum should be about 12%

this has been discussed in other threads and these values seem to be the normal.

a few people have changed the formulas used by evoscan so they get 0% and 100% but the standard evoscan values are 12% and 96%.

but i do beieve these were autos so you would have to check up on this.

chris_dono
21-06-2012, 09:22 PM
barry as standard the max tps you should see with evoscan is about 96% and the minimum should be about 12%

this has been discussed in other threads and these values seem to be the normal.

a few people have changed the formulas used by evoscan so they get 0% and 100% but the standard evoscan values are 12% and 96%.

but i do beieve these were autos so you would have to check up on this.

Mine does this in evoscan (12% at idle, 96 ish at WOT) and it's an auto

lateshow
23-06-2012, 06:25 PM
Mine does this in evoscan (12% at idle, 96 ish at WOT) and it's an auto

And same here, i think thats quite standard. One gut adjusted his auto so that it would do 100% at full throttle and he said that it's shifting better now.

BraindG
11-07-2012, 02:23 PM
Ok, it's been a wee while but i've been waiting for clutch to bed in. /popcorn

To recap, the latest on the "attack-plan"

01. Return MAF screw to original position - Done see post #52 (http://www.clubvr4.com/forum/showthread.php?63922-Barry-s-Evoscan-questions&p=704974&viewfull=1#post704974)
02. Add a cold airfeed to get more O2 to filter - Done, no difference logged (not uploaded any data on this sorry, but logs showed significant knock on 1.1bar)
03. Add Redex engine additive to clean carbon deposits - Done (http://www.clubvr4.com/forum/showthread.php?64090-Mpg), See first part of thread for figures #84 (http://www.clubvr4.com/forum/showthread.php?63922-Barry-s-Evoscan-Removing-the-Knock&p=707380&viewfull=1#post707380) under heading "Before IC, I did 4 runs."
04. Bigger intercooler on. - Done, Althought power through the gears is better, knock still there - results #84 (http://www.clubvr4.com/forum/showthread.php?63922-Barry-s-Evoscan-Removing-the-Knock&p=707380&viewfull=1#post707380) - FMIC Install Thread (http://www.clubvr4.com/forum/showthread.php?64172-Barry-s-FMIC-Install).
05. Put on standard Air intake - Not yet
06. tighten up BOV, preventing release of pressure - to dangerous, not doing
07. perform some tests to recreate engine rock, without gear changes, shortshift. - Unable to achieve on multiple tests.
08. New fuel filter - DONE
09. Clutch/engine mount change - DONE
10. Clean fuel injectors


Lastnight, i went out for a wee run, with differing results - all graphs you about to see are on standard boost (EVC was switched off) and flooring it from 2nd gear where possible at around 30 mph.

I did 6 runs, the last two i ran into traffic so didnt focus graphs on those runs.

56181

1) Easy half throttle as it was first time ive actually floored it.
(Some knock at lowish revs?)
56182

2) Easy on 3rd gear, nailed the rest.
(Some knock at lowish revs?)
56183

3) Nailed it from 2nd to 5th

56184

4) Nailed it from 2nd to 5th.

56185

Summary: So, under extreme load, no knock - however under slower conditions i was getting some high knock.

The only thing i can associate between the first 2 runs and the last 2 runs are....

a) Less WOT
b) Both of the first runs require me to turn left after leaving the roundabout, the last 2 have an easier exit off the round about - but, my previous tests in step 07. saw me unable to recreate any knock -

What i will do over the next couple days is do a few runs from the same place - Couple runs on 50% wot, then 100% and take it from there.

Any other thoughts?

BraindG
11-07-2012, 02:34 PM
Just for giggles, cos i ave ze teknologie :)


Run 2 -------------------------------------

56188


http://www.youtube.com/watch?v=2AGrfPKEqdc

Run 3 -------------------------------------

56187


http://www.youtube.com/watch?v=Qp5fwwJJzZ4

Run 4 -------------------------------------

56186


http://www.youtube.com/watch?v=-7t_RHfmYuc

Nick Mann
11-07-2012, 10:16 PM
Barry, have you considered taking the car to a rolling road and asking them to check it over? They would tell you if it is real knock and how much to worry. They would also be able to repeat the test with different boost settings etc.

On graphs three and four above I can't see the knock trace? Normally you can see it even if it stays on zero for the whole graph. It would be nice to be able to look at revs and octane level too, although I'm not sure that would suddenly make me come up with an idea.

BraindG
11-07-2012, 10:45 PM
Barry, have you considered taking the car to a rolling road and asking them to check it over? They would tell you if it is real knock and how much to worry. They would also be able to repeat the test with different boost settings etc.

How much would something like this cost?


On graphs three and four above I can't see the knock trace? Normally you can see it even if it stays on zero for the whole graph. It would be nice to be able to look at revs and octane level too, although I'm not sure that would suddenly make me come up with an idea.

I've exported the data into excell, it seems the graphs feature in there won't show a line if there is no data - i checked the raw data and it all shows 0, bit weird, i'd have though it would atleasdt show the line along the bottom.

BraindG
11-07-2012, 11:26 PM
OK, done some more runs - i'm tbh getting a little fed up with this now /Hissy

I did 6 runs in total 4 with EVC switched off, 2 on 0.8 bar.

All 4 runs on normal boost gave no knock whatsoever, earlier in this thread I would pretty much always be getting knock, even with the EVC switched off.

Changes I have made, to recap are.

New clutch
Poly Bushes
New Fuel filter

Take 1 - 0.8bar -----------------------

56190

Take 2 - 0.8bar -----------------------

56191

Take 3 - 0.5bar -----------------------

56192

Take 4 - 0.5bar -----------------------

56193

Take 5 - 0.5bar -----------------------

56194

Take 6 - 0.5bar -----------------------

56195

No knock whatsoever on 0.5bar.

Could someone please summarise again what could cause knock at higher boost? - i've lost the plot a little now that this has been going on so long.

adaxo
11-07-2012, 11:38 PM
I will probably not help you much, but did you notice differences on ECU load between 0.8 and 0.5 Bar?

BraindG
11-07-2012, 11:39 PM
I will probably not help you much, but did you notice differences on ECU load between 0.8 and 0.5 Bar?

yeah, flatlines more - i'm not sure of relevance?

adaxo
11-07-2012, 11:47 PM
me too, just noticed that and maybe more knowledgeable members can see any relevance ?

Kenneth
12-07-2012, 01:51 AM
ECU Load has a maximum of about 160 to to the way that the data is stored. In reality, the ECU uses a different load value for calculations and that load value can go much higher.

To get this, you need to mod the ECU to get either 2byte load or use the 1byte load mod (divides 2byte load value by 4)

Knock (or detonation) is usually produced when the combination of fuel octane + ignition timing + air fuel ratio is such that the fuel burns in a manner which causes a large pressure spike in the combustion chamber.

BraindG
12-07-2012, 11:27 AM
Knock (or detonation) is usually produced when the combination of fuel octane + ignition timing + air fuel ratio is such that the fuel burns in a manner which causes a large pressure spike in the combustion chamber.

Question: ignition timing = Timingadv in evoscan

BraindG
12-07-2012, 11:45 AM
Ok, been discussing with a colleague at work and we've added some more values into the picture.

Please take a look at this 0.8 run, (take 2)

Note the point at which I get knock, also note the timing adv - Is anyone able to explain why this occurs before i change gear? - i'd have though i'm in steady state now, foot to floor, boost at same level etc, yet for some reason the timing is pulled and there's knock... fuel related?


56197

56198



http://www.youtube.com/watch?v=PQfr_cLCUrg

Davezj
12-07-2012, 01:15 PM
is the graph starting 2nd or 3rd?
the timing advance is reduced due to the knock not the other way around, if you displayed the octane level as well then i would expect to see it drop as the timing is being reduced.

have you read merlins guide?
we will have to go through the basic stuff that i was trying to get my head round, i have pulled it out of the guide and it seems to be the most relavent stuff, written in a simple way for me to understand. I would like to see if it makes sense to you before i post it up, i will write it up properly at some point.

i can flash the 2 byte load mod to 7202 ecu but that does help you much barry.

BraindG
12-07-2012, 01:27 PM
For consistancy im allways start my logging from 2nd, at about 3500 revs then I floor it.

I think i've scanned that Merlin guide, but nothings stuck in my head tbh.

I'm trying to figure out whats going on now, at top end of 4th, just before i change gear to 5th i get some knock, then the timing is pulled to protect. So what is causing that knock to occur, Air/Fuel? - I'm going to do 4 runs tonight at 0.8bar, maybe 4 at 1.0bar to get a better overall picture.

What else can i be adding (other than octane) to the graphs to help pin point the route cause?

p.s. next steps for me are injector cleaning and HT Leads.

adaxo
12-07-2012, 01:32 PM
Sorry if I lost a plot a bit but did you try to go with standard air box yet??

btw where you doing your runs?? I will happily come to hear your exhaust again :smitten:

BraindG
12-07-2012, 01:36 PM
Sorry if I lost a plot a bit but did you try to go with standard air box yet??

btw where you doing your runs?? I will happily come to hear your exhaust again :smitten:

Nope, but i did add the snorkle and a huge pipe from fog to mushroom, which brought the inlet temp down by 14 Degrees (http://www.clubvr4.com/forum/showthread.php?63922-Barry-s-Evoscan-Removing-the-Knock&p=707460&viewfull=1#post707460) - Which should be sufficient imho.

Nick Mann
12-07-2012, 01:39 PM
Being slightly lean in that area is the most usual cause of knock, assuming that the octane level of the fuel is good enough for the amount of ignition advance the car is running.

A rolling road session can cost anything between £30 and £150, depending on what you need out of it and the relative expense of the venue. You'd potentially only need a power run, so the price should be around £50-£80 if you turn up somewhere by yourself. If you can get a heap of people to a rolling road day for multiple power runs then the cost should come down a bit.

BraindG
12-07-2012, 01:46 PM
btw where you doing your runs?? I will happily come to hear your exhaust again :smitten:
:P

In my little private play ground (http://www.clubvr4.com/forum/showthread.php?63922-Barry-s-Evoscan-Removing-the-Knock&p=705192&viewfull=1#post705192) :o

Kenneth
12-07-2012, 10:24 PM
No, it is NOT sufficient. The problem with the mushroom filter (or more likely the adaptor) is that it alters the air flow to the MAF which is VERY sensitive to the air flow. The MAF reading is the primary component of Load and so if that is incorrect, your ECU is calculating fuel and timing incorrectly. If fuel octane, timing and air fuel ratio determine knock, what do you think the results are when 2 are calculated incorrectly?




Nope, but i did add the snorkle and a huge pipe from fog to mushroom, which brought the inlet temp down by 14 Degrees (http://www.clubvr4.com/forum/showthread.php?63922-Barry-s-Evoscan-Removing-the-Knock&p=707460&viewfull=1#post707460) - Which should be sufficient imho.

BraindG
12-07-2012, 10:35 PM
Ok, sounds clear.

Will source standard intake and retest before I continue.

Cheers Kenneth.

Davezj
12-07-2012, 10:38 PM
if you want to try a few runs with a factory air box i can bob round at some point and you can use mine, and see what happens.

BraindG
12-07-2012, 10:40 PM
if you want to try a few runs with a factory air box i can bob round at some point and you can use mine, and see what happens.

Yeah, I'd really like that Dave - you free tomorrow night? I'm off on holiday on sat, otherwise we can do week on Sunday, or after.

Davezj
12-07-2012, 11:24 PM
Yes, straight after work, I get out at 3.30. Pm me you address and postcode and I can pop round.
I am not sure I can remeber exactly where you live.

BraindG
12-07-2012, 11:26 PM
Yes, straight after work, I get out at 3.30. Pm me you address and postcode and I can pop round.
I am not sure I can remeber exactly where you live.

Cool, I'll message you work postcode tomorrow and we can meet there then go to my wee play ground.

Thanks Dave!

Davezj
12-07-2012, 11:29 PM
Ok see you tomorrow.

Davezj
13-07-2012, 06:46 AM
Mabe I will repeat the runs you do and compare the evoscan traces. Then it will give you a good idea if a fuel pump and pressure regulator will help your situation or not.

I am not sure mine will be any better than yours knock wise.
I will have to fill it with proper petrol before I try it though.

BraindG
13-07-2012, 05:58 PM
Big thank you to Dave for taking the effort and popping round, we swapped over his maf and standard intake and went for 6 runs, 2 x 0.5 and 4 x 0.8 runs. On all 0.8 runs there was severe knock, more than on my mushroom - thinking this was a little weird we discussed options.

When we got back to daves car we noticed te previous owner did the MAF mod, which might explain the high knock.

We'll regroup after my holiday and try again.

Thanks again Dave! Really appreciate your help.

Davezj
13-07-2012, 10:07 PM
yeh that was a suprise to me, the MAF screw was screwed out so the head was flush with the outside of the plastic treaded bit about 5mm wound out from where it should have been.

so i wound it back in to match barry's MAF i will have to go out in me car now and see it that has affected my knock measurement value.

lateshow
15-07-2012, 10:19 AM
There's no point in flooring it with 2nd, the best thing would be to floor it on third to see what it actually does. 2nd gear is much "shorter" so you wont see knocking. With longer full boosting (like 4th gear) you should see lot more. And knock sum 1-3 = like nothing :D

Davezj
15-07-2012, 11:02 AM
I do tend to do knock checking for about 40-80 in 3rd then repeat in 4th then repeat in 5th.
But has done it as one power run 2, 3, 4, 5 so he is continuing with that for comparison reasons.
Next time I see him I will suggest he does some 3rd gear only then 4th gear only runs to see if the knock is still there.
I think due to his recent clutch change the gear shifts are quite violent, but his knock seem to appear at the end of run when the boost starts to drop off and the clutch should not have anything to do with that.

It is a bit odd but I am sure he will get to the bottom of it at some point.