David_17
|
| posted on 24/4/11 at 04:41 PM |
|
|
Air/fuel ratio with megasquirt
I'm starting to get fed up with megasquirt now 
I've got it running on 4 cylinders properly now, and fixed the rev limiting problem.
Now when the engine is first started, it idles at around 14.7 afr, which is good. But when looking at the laptop in megasquirt after a very short time
running, the afr drops to around 11. Eventhough the engine sound and speed don't seem to change.
Which means, if i then tune it to the 11, to make it back to 14, it runs too lean.
Has anyone experienced this before? Is there a simple fix, or am I on another wild goose chase that's been going on for months, trying to get
the stupid thing running right???  
I'll upload a video ASAP, to show what's happening, if my above paragraph doesn't make much sense.
Hope someone can help.
Dave.
PS. I've checked the wiring and earth connections, and they all seem fine.
[Edited on 24/4/11 by David_17]
|
|
|
|
|
MakeEverything
|
| posted on 24/4/11 at 04:46 PM |
|
|
Funnily enough, i recognised this before it happened to me. Ive been reading This which
clarified for me and explains why it happens.
You need to adjust the After Start Enrichment (ASE) Settings by the sounds of it.
[Edited on 24-4-11 by MakeEverything]
Kindest Regards,
Richard.
...You can make it foolProof, but youll never make it Idiot Proof!...
|
|
|
paulf
|
| posted on 24/4/11 at 05:36 PM |
|
|
The after start enrichments may need setting up but it is best to get it running correctly when warmed up first, then adjust the after start settings
from a cold start.Get the Afr correct for a steady idle when hot, it doesn't have to be 14.7 as you may find it idles better at 13 and it will
have little effect on consumption, at idle a change of 1 or 2 points on the map makes a big difference to the mixture and if too weak you will have
problems with flat spots coming off idle. Remember that you need to adjust the box in which it idles, plus you may then find that the adjacent bins
need to be adjusted slightly also.
Also make sure accel enrichment is turned off when first setting up as it can confuse things at first until the map is somewhere near at steady
speeds.
Paul
|
|
|
mark chandler
|
| posted on 24/4/11 at 06:17 PM |
|
|
Also you may finf the lambda sensor is cooling down outside of limits so will mis-read if left idling for long, if you hold a high idel for thirty
seconds then let it drop does it return to the original settings?
Is megasquirt set up to adapt to lambda values as it could be correcting?
|
|
|
David_17
|
| posted on 24/4/11 at 06:28 PM |
|
|
Thanks for all your advice guys, it helps alot to us beginners.
I'm sure i've got afterstart and ego correction off.
The video's finished uploading here... http://www.youtube.com/watch?v=j2uypm_wqI8
It's as if the sensor is reading correctly, then all of a sudden starts giving false rich readings. As leaning the mixture to 14.7 after
it's dropped makes is pop and bang as it's too lean.
Could the sensor be overheating or something? How can i test it?
|
|
|
coozer
|
| posted on 24/4/11 at 07:01 PM |
|
|
I fannyied on with mine for ages and it would never stay stable...
Installed the the Omex I bought a year ago last week and its smooth as silk and thats without any mapping, just the base map supplied by Northampton
Motorsport.
Going to Automark next week for a mapping session and after a long run out yesterday can hardly wait for the results!
1972 V8 Jago
1980 Z750
|
|
|
MakeEverything
|
| posted on 24/4/11 at 07:14 PM |
|
|
quote:
Originally posted by David_17
Thanks for all your advice guys, it helps alot to us beginners.
I'm sure i've got afterstart and ego correction off.
The video's finished uploading here... http://www.youtube.com/watch?v=j2uypm_wqI8
It's as if the sensor is reading correctly, then all of a sudden starts giving false rich readings. As leaning the mixture to 14.7 after
it's dropped makes is pop and bang as it's too lean.
Could the sensor be overheating or something? How can i test it?
Check the ASE Taper settings and the ASE Percentage Settings. It looks as though its changing the fuelling after a time period. Dont chase the red
herrings like i nearly did. I spent today (on and off) changing the ASE settings on mine, and it runs sweeter than it has ever before. Im still in
early stages like you as well mind.
Kindest Regards,
Richard.
...You can make it foolProof, but youll never make it Idiot Proof!...
|
|
|
BenTyreman
|
| posted on 24/4/11 at 07:24 PM |
|
|
Create a log of the start up through to the AFR change then post it. First thing to check is if the AFR change is due to a change in injector
pulsewidth/spark advance or if its a faulty lambda reading.
|
|
|
ashg
|
| posted on 24/4/11 at 07:57 PM |
|
|
only worry what your aft is above 80-85deg. anything below and its chcking in extra fuel to warm the engine up. as i keep telling people dont tune to
an afr number, tune to what the engine is happy with.
a really good book to buy is dave walkers haynes book on efi, it goes through nearly everything you need to know to setup from scratch
Anything With Tits or Wheels Will cost you MONEY!!
Haynes Roadster (Finished)
Exocet (Finished & Sold)
New Project (Started)
|
|
|
omega 24 v6
|
| posted on 24/4/11 at 08:21 PM |
|
|
Good adice from Ashg.
mines WILL NOT idle at anything over 13.8afr. I reckon it's because of heat soak and fuel vaporisation IMHO. As already said do not get hung up
on the numbers. If you ever tuned cars with carbs/dizzy then you'll know when it sounds right.
Also it's 12 seconds in and Instant so ASE looks like the culprit. Remeber that extra unburnt fuel shows up as extra O2 in the afr so any ASE is
flooding the engine and showing a high FALSE afr reading IMHO
[Edited on 24/4/11 by omega 24 v6]
If it looks wrong it probably is wrong.
|
|
|
David_17
|
| posted on 24/4/11 at 08:38 PM |
|
|
Thanks everyone.
I'll check the ASE and do a log file tomorow.
I've been doing a bit of reading, it could also be my sensor is too close to the engine. The recommended length is 1m after the head, well
mine's about 30cm :|
|
|
|
paulf
|
| posted on 24/4/11 at 09:11 PM |
|
|
It wont be the sensor overheating as many tin tops have the sensor close to the engine and it happens just a few seconds after starting, could it be
that initially it is reading incorrectly and then when it drops to 11 is when the sensor is at operating temperature and giving a correct reading?.If
you switch of and restart straight away does it doe the same thing?
Paul
quote:
Originally posted by David_17
Thanks everyone.
I'll check the ASE and do a log file tomorow.
I've been doing a bit of reading, it could also be my sensor is too close to the engine. The recommended length is 1m after the head, well
mine's about 30cm :|
|
|
|
David_17
|
| posted on 25/4/11 at 08:36 PM |
|
|
quote:
Originally posted by paulf
It wont be the sensor overheating as many tin tops have the sensor close to the engine and it happens just a few seconds after starting, could it be
that initially it is reading incorrectly and then when it drops to 11 is when the sensor is at operating temperature and giving a correct reading?.If
you switch of and restart straight away does it doe the same thing?
Paul
quote:
Originally posted by David_17
Thanks everyone.
I'll check the ASE and do a log file tomorow.
I've been doing a bit of reading, it could also be my sensor is too close to the engine. The recommended length is 1m after the head, well
mine's about 30cm :|
Hit the nail on the head there. It shows 14.7 while it's warming up, then reads 11 once the sensor is up to temp. I'm not convinced that
11 is the correct reading, but i have no way to prove it. I've tried calibrating the sensor, and checked it's the right settings in the
megasquirt.
Whatever settings i put into my VE table, it's always way too rich.
Wish I knew what i was doing.
Here's a log file of a steady drive... https://www.yousendit.com/download/VnBwWWV0NEhreEEwTVE9PQ
|
|
|
paulf
|
| posted on 25/4/11 at 09:01 PM |
|
|
What wide band are you using? the megasquirt base settings need to be set up to suit the particular model of wide band.
Paul
|
|
|
David_17
|
| posted on 25/4/11 at 09:16 PM |
|
|
Its a techedge one, which ive clicked in the list in the project properties.
|
|
|
flak monkey
|
| posted on 25/4/11 at 09:33 PM |
|
|
FWIW the AFR reading at idle is almost irrelevant. I got hung up on trying to get my duratec to idle at 14 for the MOT....
Never ran properly like that, so I ran it at an indicated 12 afr at idle with the intention of fiddling with it at the MOT station to get it right. In
the end it quite clearly wasnt really 12AFR as it was less than 0.75% CO and 200HC at the MOT test without a cat....
Basically, get it hot (until the fan comes on if possible) and let it idle. Lean it out a bit at a time until it just starts to misfire, then increase
the fuelling again by around 1 or 2 points and that will be about the leanest idle you can get. Your ignition timing needs to be right too, start with
about 8deg at idle and thats a good starting point for most engines.
Sera
http://www.motosera.com
|
|
|
matt_gsxr
|
| posted on 25/4/11 at 10:26 PM |
|
|
It looks to me like your sensor goes to 15.6 AFR when it is unhappy. It does it when you start up (mine behaves strangely at start up, until it
settles down), but also at 1759 which is more disturbing and wierd.
High AFR on over run looks pretty likely.
Not sure I believe the values at 1755 (i.e. AFR of 19 but reving up to 5kRPM), actually you seem to be lean when at higher RPM. But maybe car engines
can handle these low AFR and work fine. If not then put some more fuel in the table at these higher RPM (to my mind your fueling looks a bit flat,
i.e. I would have expected things to increase a bit more from 3kRPM) but I am used to bike engines.
Not sure whether this could be a leak in the exhaust (sucking in air perhaps?)
You probably want to turn off your Gego until things are working right, it hops in once or twice.
You need to connect up a temperature sensor or fix your cooling (presently temps are 101.7deg throughout)
I am no expert, but I have looked at a few of these with my car.
Matt
|
|
|
.jpg)
