VEMS Community Forum
Technical => Ignition => Topic started by: AVP on July 31, 2009, 02:54:45 pm
-
Hello,
i wanted to ask this query i have just to rest my mind really,or to understand things a bit better.
I have been getting some funny sounds from the engine,only on 6th gear pulls,something that i had with motronic as well, which sound as loud 'tsk-tsk' sounds,happening rapidly during hard acceleration. The car with both ECUs was running fine and pulling hard without any weird EGT indications either during these acceleration incidents and there are no other signs of anything being wrong other than those sounds im describing.
Now these are quite distinct and loud,it is not like you would need headphones to hear them or they are mumbled through interference or anything,so im wondering if these are really knocks or something else,since to my understanding,if knock was to be that loud,it would be more or less devastating for the engine anyway, isnt that right?
Also a knock sensor should be able to pick up something that is so loud in pitch the driver can actually hear it clearly as well!!
I will try and make a sound file too
My spark settings on those rev range that i have heard this is about 16-17deg, on boost of 300kpa and as i said,this is only happening on 6th gear pulls.Also im always using 70:30 water/meth injection along with BP97octane fuel, which makes me think that this should not be knock related.
can someone please tell me if knock can actually be that loud and have all those secondary corresponding symptoms?
I have made a few changes on the spark table,and lamda table just adding a bit more fuel on that region.The problem is that when i do that,and have get lamda richer, i sometimes get too rich on 3d gear and the car stumbles,while it is ok for 4th gear and above pulls.
lamda on the pulls was: 0.88 to 0.84 on the 3000-4000rpm region i did the test
thanks
vasilis
-
You only have 0.88-0.84 at 300kpa?
I´d not run it leaner then 0.8
Also, that is very likely knock/det/ping. Check your Manifold temps as I´m sure during the 6th gear pull they are higher then the 3rd or 4th gear pulls.
Your running MAT dependend retard I´m sure? It may not be enough, the latest firmware has tables for this where you can retard the ignition more then has been the norm as the intake temps go higher. Thus allowing you to create a better safety net.
-
on the 2500-3000rpm area i have that lamda,because when i was on 3d or 4th gear, it was getting too rich and was couphing.After that it goes to 0.80 and finally to 0.75lamda as revs go higher. EGTs are about 800-850 on 3000rpm and do not go more than 930 towards the higher revs (i havent tried 6th gear pulls anyway more than 5000rpm though but on 5th gear pulls it stays low).EGT is on the turbo hotside.
My inlet manifold temps are no more than 20 to 30C and that is because after 1bar of boost the water/meth injection comes in and keeps everything nice and cold( i havent checked them during pulls and i will try to). The 1.1.27 version has the mat retard implemented which at those inlet temperatures is supposed to keep the degrees of advance at about -1degree of timing than the setted one.
Im waiting for an official 1.1.5x version from VEMS that would allow that.
Here is the thing. ON 3d gear and 4th gear, i get the turbo spooling at around 3500 and 3200rpm but on 6th i get it from 2700rpm.
when i build the maps, if i run 0.80lamda from 2500(so as to cover the 6th gear area), the car goes rich on a 3d gear WOT from low rpm and stumbles,but obviously runs ok on 5th and 6th gear.
If i do the other way around and make that area leaner,then i risk to have detonation on 5th and 6th gear. As it is now, it seems that i only get that with 6th gear only,so i will try and richen things there as well.
Could i also be getting knock if the advance is too low? does 14 to 16deg sound low to you?
-
14-16 does sound high for 300kpa. but it´s pretty hard to say. As engines can be so different.
Can you log the 3rd gear pull against 6th gear pull. Maybe there is a chance of adjusting the rpm range there so to tune out
the problem. I´d be interested in looking over the log.
-
i will log a few pulls now.
i will also lower the deg.of advance on the time that turbo sets in as well.I have lower timing there and gradually but slowly bring it higher as revs go up. I have noticed that as i rise the advance slowly i get better EGTs as well,which makes me thing im safer on higher revs.Also it is richer there anyway.
I have altered the fueling on that initial area of revs and see what will happen.
-
here are my current spark and lamda tables,with the alterations that i have done
(http://i117.photobucket.com/albums/o59/vasp3690/lamda1.jpg)
(http://i117.photobucket.com/albums/o59/vasp3690/spark1.jpg)
-
What about the fuel map.
-
fueling map is close to the lamda and on the rich side on the top bins.I use EGO correction closed loop with 15% +/-
-
That doesn´t matter,
I need to see it to make a educated guess on why your engine is over fuelling or under fuelling at various conditions.
As you said it can´t be close to the lambda map if it reacts in two different ways in two sets of gears.
-
ok here is the fueling map
(http://i117.photobucket.com/albums/o59/vasp3690/ve1.jpg)
-
definitely get a log of both situations,
3rd gear pull from 1500rpm
6th gear pull from 1500rpm
-
ok
going out now
im going to see if i can do that with ego on and off
-
my thought is: ouch way to advanced timing on way too lean. lambda 0.77 and maybe 5 spark advance at 300 kpa even at lower revs, specially less advance 3000-4000
-
Offtopic: can someone tell me in few words the difference between VE table and Lambda target table?
-
i did some runs, and the log files are quite large.After studying them i realised that there are some instants where for some reason for a given VE value (this time it was 99) although most of the times it gives me a value of lamda of 0.78 for example, on a couple of instances it went 0.70 and maybe more, thus making the car run rich.
Now i dont know why this happens occasionally,although both lamda and VE are supposed to be giving out something different. Im not sure if this lamda value is due to water injection,as the amount im spraying is not that large anyway.
Out of all tests,i managed to get rid of that sound.I have reduced spark to 13+ deg.and i have given more fuel on those areas of highest torgue as well.
I do get however at some point if i go WOT on 6th gear,an either missfire or some kind of rich mixture stumble as the boost needle passes through the 1.5bar area,and as it surpasses it , it stops and car runs fine after that. I have checked and noticed that before,as i made the mixture leaner,this feeling went away and corrected itself.It seems now that it needs finetuning on the 250kpa area to make it work properly
-
Where the lambda is different between gears is the RPM, MAP, PW, TPSacc, Vbatt, Gammae, Gammaair all the same,
i.e is the ECU doing exactly the same thing?
You can upload the logs to a website for people to download. Also put your msq file with it.
There may be some compenastion function not proberly tuned between the different conditions in each gear.
Without seeing the logs and msq it´s harder to assume the problems occuring causing a missfire.
-
here is a download link for the .rar file of the msq and the excel file of the datalog.
please let me know if you can see something that i dont!
also my email is: [email protected]. I would be happy if you could tell me if i need to make a change
thanks
vasilis
-
I can tell you what I see and that is no link :D
-
oops!! forgot to to do the copy/paste...
i will try and send you the rar on your email.
-
Offtopic: can someone tell me in few words the difference between VE table and Lambda target table?
The VE table is just that, it's a map of the engines Volumetric Efficiency, or how well the engine is breathing The Lambda map is what you alter to get the desired AF ratio at that particular site. The VE map should be tuned so EGO correction is doing very little, or nothing at all. (Read a few posts by dnb, he's spent a lot of time analysing and tuning his VE table).
Back to the original post, that published VE table doesnt look right to me, seems as if EGO is having to work overtime which is never a good thing. EGO should only be there as a small tweak, it should never be used to cover a bad mapping job, a 15% swing is rather large. Spend some time looking at the logs and tweak the VE accordingly. Also, if you are only using up to the 10x range, consider altering the req fuel to get greater resolution.
-Gavin
-
Thanks Gavin,
how can you tell if the VE map is wrong?
with EGO off i tuned the engine up to 1.5bar at least only with the VE map and datalogs. The only part of the map that has been done with less accuracy is the high boost high rpm,and that is because there is no available road to do that tuning around.Even so, on datalogs of 4th and 5th gears up to 6500-7500rpm the VE map is correct.
I could alter the reg_fuel for better resolution as you say.That is always an option.
-
The VE table isn't wrong, it just looks rather coarse/unrefined, and I think a lot of other enrichments have been masking the true VE table which has tripped me a fair few times before now :) This is why gunni is so interested in the log to see the gamma variables (total corrections) as obviously these are altering between a slow sped accel run (6th gear) to a high speed accel run. The accel enrichment being the biggest culprit of all! This is why having dyno time is a real good thing, can hold a steady load for a while to ensure everything has settled then give the engine what it wants (MBT, right fuel). Once thats done, i tend to handle the transient with road tuning. Once you get to the high load portions tho, I still extrapolate the known values, as even I won't hold an engine at 2bar boost at 6000rpm to map lol.
Try and publish the rar file tho, I think the vems.hu file uploader is still running. I've been amazed with how you've handled and documented this project so far, be nice to get you closer with the calibration :)
-Gavin
-
many thanks,
i will try and post the rar file.im not at home comp. at the moment so i cannot post the file.
I always think the accel. enrichments have much to do with it,but i could never think how that would effect the map that much far after the pedal has been pressed.
what i mean by that is,that if i understand correctly,the enrichment and the settings of it,have to do with how fast you press the gas pedal and how hard.But they only last for the amount of sec you have specified. I get those engine roughness quite far away in time and revs from the moment that i go WOT.So i would think that accel. enrichment shouldnt have been a problem.(in theory)
When i go WOT on 3d from 2500rpm i get a roughness on 6000rpm for example, because of lamda being possibly less than 0.70.Now the accel.enrichment should have ended by then.That happens at about 260pka cause the other problem i have is that i cannot get VEMS to control adequetly the boost valve on the audi.I have tried many things,but on lower gears it just cannot hold boost on high rpm.It only holds it better on higher gears.
On 6th gear,although i go wot on 2500 or 2700rpm,it quickly rises boost but just as it passes through the 250kpa region,it stumbles a bit and as it crosses it,its alright again. I hear no knock on that region of any sort. When i reduced the lamda and VE bins,it went better.
-
http://rapidshare.com/files/262984293/a80avant.rar.html
here is the download link
-
Offtopic: can someone tell me in few words the difference between VE table and Lambda target table?
Also, if you are only using up to the 10x range, consider altering the req fuel to get greater resolution.
-Gavin
I like using Map multiplication as that does a great deal of background interpolation that you don´t have to worry about.
But that´s just me. Having a bigger range of numbers doesn´t make the tuning better because the vems is interpolation on so many levels that when the cells are right then no matter where between cells or floating around the rpm and map are they will be spot on.
This happens when the VE map is smooth and has a obvious 3d look and feel to it.
I have no problem tuning NA cars with MAP multiplication and only a 5x15 portion of the map.
-
have you seen the map now?
-
after further testing and listening, i get now no weird sounds from the engine,but after having reduced the spark i get higher EGTs now much easier!!
i get into a dangerous area around 5000rpm. when running on 0.74 to 0.73lamda on that region.
Do you think it is wise to get the spark a bit higher,or just increase the fueling a bit more?
-
You have got to stop thinking about EGT´s as #1.
High EGT´s don´t mean anything bad to the engine. AT ALL....
If your getting higher EGT´s because of retarded ignition that is 100% totally normal, Also 300kpa pressure is not like it´s normal or anything, it´s quite high actually, so to assume that you´ll not have problems with 300kpa pressure in the low rpm band is childish,
Just retard the ignition to be safe, richen up the fuel and leave it at that.
If you must fine tune for MBT and max power you´ll have to take it to a dyno, EGT readings will not help you one bit.
-
ok!
i will take your word for it.
i only need someone with more knowledge than me to fine tune the engine on a dyno now.
But since im going to be having new parts next year,ill leave it for then!
-
Cliff wrote that tuning ignition (finding MBT) can be also done by finding detonation - once you hear a knock, retard 3 deg and you are good. Isnt this good enough?
-
Cliff wrote that tuning ignition (finding MBT) can be also done by finding detonation - once you hear a knock, retard 3 deg and you are good. Isnt this good enough?
No.
As there are more tables in the ecu than just the ignition table. They change the final timing and they must be tuned accordingly.
-
it is just weird because on motronic i had 20deg on high rpm and more than that on lower rpm + i was geting +3deg when IATs were below 25C or something.
and now im down to 13-14deg on middle range and slowly climbing up to 16-18deg to the red line.
Why is that? Same car, same specs, different way of determening the spark vs boost. Motronic uses load from the MAF
-
Do you have a log from the motronic showing it using this higher advance?
-
yes i do. i have done a couple of vagcom logs,although the whole thing got stuck a few times,but i managed to see that even up to 5500rpm on 4th gear pulls it went up to 21deg advance(which was pretty much the same as the map i have)
-
Next saturday im going to a garage near where i live to recheck the timing with a strobe light.I want to make sure the trigger setting is accurate,and therefore have a better idea of the timing difference between motronic and vems.
-
we tried to do a timing check today,but it was a no go,due to the coil pipes configuration on the rs2 rather than coil cables.
does anyone know an easy way to get to the cyl 1 spark plug,while still running the engine?
-
Why don't you use old spark plug wire connecting it from the coil (pipe) to spark plug? That's the usual way to use timing light with COPs.
-
My way to check timing on 20VT is put clamp on primary coil wire. My strobe light is very sensitive in this case. Often it strobes not spark event but start of coil charging. Opening clamp by ~1mm helps to trigger on spark event.
20deg@2bar boost?! May be you have very good fuel. I had never more than [email protected]. And going higher usualy require overfueling to prevent knock. It is on RON98.
Are you sure that motrnic is maped at so high air masses per cycle?
Gints
-
motronic was mapped for up to 1.9bar overboost and 1.7bar to the red line,so 20-22deg advance is with probably 1.7bar boost.However i also got aggressive ignition advance with motronic adding 2 degs when IATs are lower than 25C as well. Which is why i thought that now that i have it no more than 16deg on 2bar and 17 at 1.8bar maybe my basic timing setting is wrong.
Also i recently got some pinging but that wasnt at high rpm,it was only during spool up and low rpm where i was getting the initial high boost.Then it cleared up as revs came up. Anyway,now that is sorted and i have reduced timing on that area plus increased fueling on higher revs to keep everything under control,but i will have to do this check.
does anyone know any cheap ignition cables that would fit the audi 20V so that i could perform that check?