pcmhacking.net

mag wrote:understand that blm disabed (all set at 100) but thought blm history would still show which cells have been visited during ve learn & therefore which might have been changed.

Have attached bin.

With the changes to the VE table does the engine run better? What were your BLMs like before VE Learn? Its taken a fair bit of fuel out but will need some manual smoothing of the VE table if you are wanting to keep using it. The 20kPa range has not been touched as your target AFR is 14.8:1 and not stoich, the low RPM areas that are stoich would be unlikely to get used.

Also changing the VE learn setting called 'VE Learn Params - MAP Radius (4.3kPa MAX)' back to the default setting of 2kPa will make VE learn slower but more accurate.

VL400 been away for a week & this first chance to reply. Have a couple of real long runs now & also fiddled with DFCO. have not down loaded the cal yet so don't know what the ve table looks like. got DFCO working by increasing MAP max to 30 (was 20) & lowered RPM min & max (1300/1400). DFCO now cuts in on closed throttle on down hill, interesting that if a/c is on DFCO remains off. must be extra load of a/c just pulls MAP above 30.

other question I have now, does anyone know the effect of E10. thought 10% would have little effect but read the other day that 10% ethanol shifts afr to 13.8:1. if true then E10 would need a complete new tune.

in the past i have used e10 on more than one ocasion and you really can't notice a change in perfromance other than using more fuel ..... if it was 13.8 ( as apposed to 12.8 i am guessing) you'd definately know about it!

Holden202, have used E10 quite a lot myself without noticable change in performance but when I read that E10 stoichiometric ratio is 13.8:1 makes me think that a tune based on 14.7:1 would be seriously off.

yeah but the oxygen sensor doesnt adjust the mixture to 14.7:1. it reads stoich of whatever fuel is runnin thru it, so the blms will fix it a bit anyway.

agree blms would try to compensate but thought open loop fuelling would a fair bit off & might explain why ve learn seems to be pulling my ve table quite dramatically.

if anything, it would increase VE with VE learn on... it would be running too lean and try to richen it up. I dont think thats whats casuing VE to be pulled.

the difference between 13.8 and 14.7 should give you roughly a 7% change in VE ...... that is assuming the ve table is right before this!

Have had a chance to analyse recent long run logs & ve learn is certainly trying to pull a lot of fuel out at particular rpm/map points. At some rpm/map points it is also adding fuel. Large changes in the ve table values appears odd beacause I used the ve table from the factory bin for the engine & doubt the values would not have been as far off as they appear now. Why would they now change by so much under ve learn.

Should I expect to notice a change in cruise performance from a ve value change of magnitude 30.

For the dummy in the class, what is the range limits of ve learn. VL400 you have made the comment that ve learn can only do so much & that the base ve table has to be close. A ve value change from 65 to 39 is pretty big, if this is within ve learn range can I leave ve learn to finish the task.

have a look at the ve learn settings, theres settings for this ....

low rpm limit - will not use ve learn under this rpm .... similarly there is a high limit

rpm radius - how far from the cell numbers it can go for changes .... eg. 50 rpm so it has to be within 50 rpm of the cell otherwise it wont adjust.

map radius - same as above but on the map axis.

minimum ve value - will not adjust any ve table number below this value - there is also a max ve value too.

the other thing to look for is trends, are all values in the area around that being pulled down ?

a change of 30 is huge in any area if your running a tune that should be pretty close.

having said that i've had issues with ve learn screwing with idle ve's by big values because at high vacuum at idle it was sucking vapour out of the charcoal canister and making it too rich so was leaning out ve table to compensate .... was able to set the minimum rpm limit to fix this though.