I would like to hear your suggestions about this.I have read a few books and spoken to a few people already regarding tuning for cruising conditions and light load (up to 140kpa) and i can find 2 options.
1) Use the best timing for optimum torque and keep the lamda around 1.00-1.04 (same thing that motronic does on our old audis which didnt come with a wideband)
2) As best lamda for economy is said to be around 1.16, tune for that area and possibly reduce the timing slightly to avoid engine 'struggle' or misfire.
So far i have tried:
i) keeping lamda from 1.13 to 1.10 from 30 to 102kpa, then bringing it up to 1.03 to 120kpa and timing from 2000-3000rpm going in those areas from 33 to 38deg roughly. The car did not stuble, knock or misfire. Consuption was worse than stock motronic but decent.
ii) Brought lamda to 1.08-1.03 in same kpa areas and increased timing by a couple of degrees, reaching about 39.50 on 3000rpm(sort of same as my motronic map used to have). I immediatelly noticed that the car had more torque than before when off boost and was more agile. I havent managed to see any significant difference in consuption either way.
iii) same setup in timing but lamda being kept at 1.11 to 1.09 on the above kpa areas. The car although not heard knocking or anything, it felt to be less torquey and on a motorway run, it consumed quite a lot more fuel. I suspect the leaner option wasnt working for the engine.
I havent tried a long journey with the (ii) option yet, but so far, in the city it seems to be ok as far as consuption, around the motronic levels.
In a week or so i will have the 1.1.81 version and i suspect that with the extra mat vs tps table i will be able to avoid increasing my VE map on the cruise areas to avoid playing around with the warm up enrichments, as my EGO is set to keep substracting fuel rather than enrichening, and that overall will have some negative effect on the consumption. In addition, cold weather and cold starts are not helping either i suppose.
It would be interesting to see other people's views regarding tuning those areas in VEMS. What do you opt for client's cars?
I know that aftermarket ECUs are not meant to provide a cruising optimum as they are meant for motorsport, but since vems has become the people's aftermarket ECU and it is being used on lots of everyday cars, it would be nice to be able to take advantage of all it's feauters to be able to beat old motronic technology in maintaining a good consuption level!
thanks
You can only do what your engine wants.
Trying to run a engine very lean that is not designed to do so will not net the gains it does on designed for that engines,
The major problem is your fuel atomization and distribution in the cylinder.
I tune all road engines to lambda 1.0 for cruise, 1.02 for light cruise
0.9 for around 100kpa, and 0.85 for up to 150kpa-ish, and then 0.8 above that.
would a different type of injector help in that department? of atomization that is
It´s more to do with the squish band and the amount of tumble into your cylinder.
anywone else care to give us an input?
Lately I've done some observations with the Audi S4/S2 engines on gasoline, I've seen that lambda 1.02-1.03 is a good mixture, and about 27-29 degrees advance while cruising. This seems to give better results than the stock Motronic at speeds between 70-130 km/h. Any leaner than give the same results = no gain, but I've not been super-scientific about this. Just made overall changes and let the owner of the car observe fuel consumption (fill up the tank) during long trips.
It's quite important on some turbo engines to let them run lambda target 1.0 at pressure up to 120-130 kPa because that's where they are very quick if they press the throttle just a bit at speeds in excess of 100 km/h, the turbo is always ready above 3000 rpm. Not doing it will make the fuel economy worse compared to Motronic.
Remember that the cylinders are quite small on this engine, just over 80 mm, with often high c/r. They don't seem to want much ignition advance on low loads. It would be interesting to know what the stock ignition timing map looks like.
i have that but its based on 'load' values which do not correspond to kpa(map) equivalents.
Ideally we need someone with an engine like that, doing a datalog on vagcom and at the same time observe a boost gauge to try and best estimate the equivalent values.
my original motronic had higher timing due to slightly smaller compression ratio. The original RS2 timing along with the ADU cams have less timing than mine, but the original pistons come with 9.3:1 CR while mine was close to 9:1
i find out that lambda 1.07 is best for consumption.
that is what im trying this week. ill report back!
i think im at the edge of the timing that i can be, as well as lamda going to 1.08 on light acceleration doesnt go well with that.
so i reduced the timing a bit from 102kpa to 104 kpa and also made those areas a bit more rich(ie about 1.04-1.01 lamda)
testing continues...
good results so far. I think it drives much better than before. I lean it out till 78kpa and start enriching it again from 90 and above. I have added a row of lines for lamda in the 1.1.27 for values below 100kpa which helps. Obviously i need to see how effective that is in the tiny lamda table that we have on 1.1.81.... but hey, what can i do?
Never really understood whats the big deal using at least a 12x12 lamda table instead. How difficult would that be for software and tuners...?
It´s just not needed when you look at the value range you are using as above say 160kpa you don´t need to change lambda above that so that could be your highest axis value
yes,thats another solution.,although i dont know why, but i feel that i need to use one line for the top lamda value for some reason... its psychological i guess! :D
The big difference from earlier firmwares like 1.1.27 is that now with more recent 1.1.x you have separate axis for each table. That makes a big difference in how many rows/columns you actually need now that it's possible to put the rows/columns where you actually need them.
can you clarify what you mean excactly?
the size of the table is fixed, right? or you mean because we have the ability to use different values for rpm and kpa on the different tables that gives us flexibility? yes, that is true
The latter. When all tables are the same large size, that can feel good but when you're forced to use the same rows/columns you're screwed.
i understand.
i made a lamda table that goes only up to 140kpa. its more workable. ill see how it is once the ecu arrives.
ECU will be here on tuesday, only i wont, so i will have it next saturday.
I must say that the latest configuration regarding the cruise seems to be giving me the best results so far. So i have leaned out the 78kpa line from 1500-3500, and then gradually get the lamda back to 1.02-1.03 till 100-120kpa. That, so far has given me a decent little over 11l/100km in the city, which is ideal, since on the motorway it should be less. Also on 56kpa i take it to 1.05 and lower than that 1.07.
On the 1.1.81 firmware i will make sure the VE map is very accurate, so that the EGO works as minimally as possible, and i will make sure that the MAT vs TPS map compensates for heat soak and hotter inlet temps. That should also allow for less lost fuel during EGO adjustments.
you should use the lambda table to set the target, then you can keep the EGO working normally throughout the entire table.
You'll find that the VE table will be fairly flat throughout.
yes, the only reason that i have on 1.1.27 the VE slightly on the rich side is to avoid having lean spots due to heat soak of the IAT sensor. Generally if my VE map is ideal and spot on, it works well until about 15-18C IAT and then EGO starts to add fuel instead of keeping zero or removing fuel. I found that on hot days, the car was struggling, because the EGO was not fast enough to add the necessary fuel, and therefore i was seeing quite lean spots until EGO kicked in. So after those few instances, i made the VE in a way that EGO always takes away fuel rather than add.
now with the MAT vs TPS table, i can keep VE ideal for the most occasions, and add more fuel on hot IATs etc.