Kukunin
Newbie
Karma: +1/-2
 Offline
Posts: 20
|
 |
« on: March 24, 2018, 11:37:30 AM »
|
|
|
I have a New Beetle with 1.8T AWV engine. According to a dump of my ECU, it's 06A 906 032 KQ, I believe it's Bosch ME7.5.
But the answer should be probably the same for all line of ECUs.
Is there a way to measure constant and average fuel consumption? I guess not since there is no such group in VCDS.
But still, is there a way to perhaps calculate it, for example, based on air consumption and lambda correction?
I want to get some indicator for my tuning changes, to know did I increase or decrease fuel consumption and how much.
PS, A little background. Currently, my car has huge fuel consumption. It's about 10l/100km or 23.5 MPG on a highway. I checked everything on my hardware, so I guess there is something with ECU firmware since I had a chip tuning from a random guy.
|
|
|
|
|
Logged
|
|
|
|
|
vwaudiguy
|
|
« Reply #1 on: March 24, 2018, 01:02:37 PM »
|
|
|
Why not just calculate miles driven by fuel consumed?
|
|
|
|
|
Logged
|
"If you have a chinese turbo, that you are worried is going to blow up when you floor it, then LOL."
|
|
|
Kukunin
Newbie
Karma: +1/-2
Offline
Posts: 20
|
|
« Reply #2 on: March 25, 2018, 02:18:56 AM »
|
|
|
That's the case, but it has drawbacks:
- it's pretty inaccurate on < 100 miles
- it takes time
- you waste gas if you don't need to drive somewhere
Of course, if there is no alternative, this is the only option. But If I'm going to tune my firmware, I need to do a lot of tests, a couple per day, that would be much simpler if there is fuel indicator from ECU
|
|
|
|
|
Logged
|
|
|
|
|
prj
|
|
« Reply #3 on: March 25, 2018, 07:34:18 AM »
|
|
|
Flash stock file and check consumption then.
What you are suggesting is not feasible at all, because there is a million things that will affect the consumption apart from tuning.
The only way to find the correct timing is by adjusting timing in realtime on the dyno - but if you have stock CR the OEM already did that for you, so don't touch the timing map anywhere but WOT/knock limited area.
Other than that, make sure your fuel strategy is reasonable. Don't request lambda 0.8 at 1000 rpm even at WOT - lamfa should request rich at WOT only from the point where your turbo is actually capable of making boost. Always request 1.0 at part throttle and let BTS kick in for component protection only when needed. Use lambda based timing offset to prevent knock in these conditions and so on.
|
|
|
|
|
Logged
|
|
|
|
Kukunin
Newbie
Karma: +1/-2
Offline
Posts: 20
|
|
« Reply #4 on: March 25, 2018, 07:49:34 AM »
|
|
|
I'm thankful for this tips. They sound advanced for me, but I'm sure I'll figure out.
A little background, the ECU was chipped by not me, I don't have a contact with that guy, and also I suspect he did a poor job (he just flashed, not measurements, not tests, no before/after etc). Also, unfortunately, I don't have a stock firmware, I found it only for similar ECUs.
So, to summarize, there is no way to report actual fuel consumption from an ECU, but it's not needed.
|
|
|
|
|
Logged
|
|
|
|
|
prj
|
|
« Reply #5 on: March 25, 2018, 09:02:58 AM »
|
|
|
You can log instantaneous fuel consumption from ECU, that is not the problem, but it's useless for doing actual tuning.
If I remember, I'll check on Monday if I can find an original file for your ECU.
|
|
|
|
|
Logged
|
|
|
|
Kukunin
Newbie
Karma: +1/-2
Offline
Posts: 20
|
|
« Reply #6 on: March 25, 2018, 09:12:46 AM »
|
|
|
Thanks a lot. Can I know how can I read the fuel consumption from ECU? That's still might be useful, if not for tuning but for personal usage
|
|
|
|
|
Logged
|
|
|
|
|
prj
|
|
« Reply #7 on: March 25, 2018, 10:22:23 AM »
|
|
|
Thanks a lot. Can I know how can I read the fuel consumption from ECU? That's still might be useful, if not for tuning but for personal usage
Look at the FR around KVB, there is a ram variable you can log. Sorry, don't remember off the top of my head. |
|
|
|
|
Logged
|
|
|
|
|
nyet
|
|
« Reply #8 on: March 25, 2018, 10:34:45 AM »
|
|
|
Instantaneous fuel consumption (over time) is always te*rpm
mileage (over distance) is gearing dependent and inversely proportional to te
ymmv (pun intended)
|
|
|
|
|
Logged
|
|
|
|
Kukunin
Newbie
Karma: +1/-2
Offline
Posts: 20
|
|
« Reply #9 on: March 26, 2018, 01:38:37 AM »
|
|
|
Thanks
I'm very sorry for a silly question, but what does te stand for? Can't google it.
|
|
|
|
|
Logged
|
|
|
|
|
KasperH
|
|
« Reply #10 on: March 26, 2018, 01:40:02 AM »
|
|
|
Injector pulse width.
|
|
|
|
|
Logged
|
|
|
|
|
nyet
|
|
« Reply #11 on: March 26, 2018, 04:04:08 PM »
|
|
|
In any case, you are approaching this incorrectly. The right question is: are you lambda 1 less often than stock? If so, why.
The odds of you getting any useful information out of fuel consumption data is zero, since it is almost indistinguishable from noise (that is to say, follows torque request almost 1:1). You're looking for variations that are NOT following torque request EXACTLY (deviation on top of that noise).
What you really want to know is how much time your ecu spends requesting less than lambda 1 compared to stock.
|
|
|
|
|
Logged
|
|
|
|
|
prj
|
|
« Reply #12 on: March 26, 2018, 10:43:34 PM »
|
|
|
Also, I don't have stock file for you.
Try backing up your ECU and cross flashing to something else to at least have a baseline.
|
|
|
|
|
Logged
|
|
|
|
Kukunin
Newbie
Karma: +1/-2
Offline
Posts: 20
|
|
« Reply #13 on: March 28, 2018, 02:01:43 AM »
|
|
|
Anyway, thanks a lot for the attempt. Will try to flash a similar ROM
|
|
|
|
|
Logged
|
|
|
|
|
