help rolls transition from a tuna to a tuner

[hiddeous]

Rolls, where are you located?

[rolls]

Adelaide

Its ok Grant is going to flash my PCM for me.

[JETURBO]

Didn't you get a sweet as Keep Paying Money tune .......

[rolls]

Didn't you get a sweet as Keep Paying Money tune .......

Overall I was pretty happy with it, eg boost curve and fueling was much better than before, but definitely needed some love for the low load/idle/coldstart which I find unless they keep your car for a week most places don't do or do poorly.

The more I learn the more I realise how little others (and I) know. I now trust a grand total of zero people to tune any of my cars now. The problem is I'm in list as well! lol. Also not having access to a dyno or proper knock detection equipment means I still have to rely on others. At least I know what I've touched though.

[misk]

it sucks when you learn a little, you only realise how much more there is to learn!

[rolls]

it sucks when you learn a little, you only realise how much more there is to learn!

The problem is there is little information and the tuning platform sucks. Tuning a motec for example which can be as complicated as you like is much easier as you have proper logging (map tracing) and you know exactly how all the maps relate to each other.

People not willing to share what they have learnt is also a very foreign thing to me.

[rolls]

So I wrote a small calculator that helps you calculate your slopes. Basically my high slope had fueling with 0-1% max trims when cruising above 60kph or any load at all but idle was lean and very light cruise was rich.

The lean idle and rich cruise meant I needed to tweak my offset, breakpoint and low slope, the problem is these affect the high slope quite dramatically (breakpoint especially). The calculator will calculate new low/high slope if you adjust your offset or breakpoint to maintain the same fueling at a given injector ms (eg full load). Makes tweaking offset and breakpoint to get cruise/idle fueling perfect much much easier.

Basically all you need to enter is a ms for your "full load" which will make the calculator ensure the new and old high slopes intersect at this point. You can then adjust the breakpoint and it will auto recalculate your low and high slopes. Another thing it can do is keep the breakpoint the same but bend the high/low slopes left or right to add or subtract fueling mainly around the breakpoint. Or simply adjust the offset only if you want to only add fueling at the idle area.

Instructions:
Figure out your full load injector ms, eg if you get 60% duty cycle at 6500rpm this is about 11ms from memory. This setpoint is to ensure your new curves will give you identical fueling at this point.
Idle needs richening? Increase offset only.
Cruise is too rich but idle and full load is fine? Put 0.95% into the "Shift break point intercept" or adjust the breakpoint.

May not be useful for you guys but was useful for me as my high slope was already dialed in really well, the low slope/offset was pretty average giving me poor trims. This made it very easy to tweak around the low load points knowing I wasn't messing with my high load fueling.

https://dl.dropboxusercontent.com/u/...%20slopes.xlsx

[rolls]

https://www.dropbox.com/s/9ozdywlqeu...trims.hpt?dl=0

So I've played around with the tune a fair bit the past week or so and I've made lots of improvements but I'm still stuck on a few points.

Here is where I have the tune currently, fixed the cruise trims and return to idle bobbing around.

Issues that remain are cold start fuel trims are woeful causing it to go really lean for the first minute of driving. I believe this is due to the cold start map being wrong (they had richened the stock map from 1.05 to 0.98) and this was causing closed loop to over correct (I think), eg it would idle at 13:1 when in open loop (first 40 seconds) then long term trim would go to 0.80 to bring it back, it would overshoot and go lean at ~17-18:1 at cruise for about 30 seconds. I've just restored the cold start maps from a stock file to try tomorrow.

If I do a hot start all of the above issues do not exist.

Idle long term trims are about 1.08 -1.11 even if I massively fudge the speed density map in that section. As soon as I get on the accelerator it goes to 0.99 - 1.0 and trims are close to perfect the rest of the time. I'm guessing the injector is simply not linear there and the 2 slopes struggle to approximate it.

When hot (>200F) the idle/holding ~1.5krpm in neutral is a tad rough/poppy as well, it seems to love 14.2 and not 15.0 at a hot idle, unsure how to get it to actually idle there. Increasing offset or minimum pw to fudge it just ruins trims everywhere else so currently I'm stuck on how to get rid of the roughness when hot.

Other issue is some bucking around 2k on deaccel, not sure if it is under fueling or overfueling as trims are close to perfect at 0.99 If I fudge the offset to get it lean it is smoother (pops more) but then this breaks other areas.

I assume once I get my trims close to perfect everywhere all the other issues will effectively solve themselves so that is what I've been doing currently. Is this a good approach?

[turbotrana]

Will try your calculator sometime.

In regards to cold start this is how I sort of do it. I use stock cold start parameters on 98 and saw a "pro tune" the other day on E85 use stock cold start parameters (did a compare and stock from I could see).
Remember that the injectors are on Hi slope on cold start. So what I do is dial the hi slope first (sort of, you are always jugling) so the car imediately starts and you can floor it (moderately) from cold and all runs cleanly with is no hesitation (maybe a teency of hesitation). I start from lean (hesitation on cold start,hi number on hi slope and reduce hi slope number till the cold start is good). Fuel base lamda numbers should be close.

This needs to be done over several/many mornings. At the same time you are dialing low slope and inj offset when warm by seeing what ratios you are getting at idle and what ratio you are getting at very light load cruise (below the breakpoint, roughly under 2ms depending on the injectors, you need to calculate this).

Once drivability is dialed in then tweeking SD tables above 4000rpm (this is what the pro did) or what I also do is play with the Fuel base numbers to get the ratios that I want on hi load/rpm bearing in mind that the numbers in the table wont necessarily match what is actually there.

This is a rough description of how I scale injectors ignoring any data that there may be. For simplicity I just use one number for injector offset for all voltages.

Thinking that you just need to dial the hislope to match the air/fuel ratios you want at hi/rpm/load is not the way to do it IMO as it will mess with the cold start.

[rolls]

At the same time you are dialing low slope and inj offset when warm by seeing what ratios you are getting at idle and what ratio you are getting at very light load cruise (below the breakpoint, roughly under 2ms depending on the injectors, you need to calculate this). .

The problem is your high slope is shifted when you change offset/low slope/break point. So if you dial in your high slope then adjust anything at all you put it out, hence the calculator. For example if you change your breakpoint at all, you massively change your fueling on the high slope.

The only way I can see to do it is use something like my calculator or do it in the following order.

Offset
Low slope
Greakpoint
Highslope

If you tune it in any other order you'll be chasing your tail forever.

[turbotrana]

The breakpoint is a set and forget thing. You either accept the injector data from the supplier or calculate it yourself. The calculation may not be absolutely accurate but you know roughly where is should be, i.e fuel flow around 2 to 2.5ms just where the fuel flow starts to become linear.

So that leaves you with three figures you need to juggle. So yes, change one, it changes the others. But you sort of know what direction to go into when reading air/fuel ratios -trims at various points in the map and you just keep on juggling untill you are happy with it.

More of an art than a science when scaling injectors from scratch ignoring any data out there.

If you were juggling 4 numbers then you would be chasing your tail. With three, ie low/hi slope and injector offset you will get there.

[rolls]

http://www.hptuners.com/forum/showth...483#post432483

Posted a thread about O2 bias voltage, can someone please explain it to me?

[rolls]

Someone able to explain the common causes for lean tip in/rich tip out when cold even with O2 feedback disabled? Only seems to happen when cold, I've left O2 feedback disabled and driven the car hot and I get some rich tip out but no lean tip in. I can mask the problem but I want to understand what the main causes are. Are there other tables surrounding transient fueling that would be affected by a larger intercooler?

Most of the things I can think of will make it always rich/always lean.

[rolls]

Ok well I improved it massively. Seems my map per airmass map had been quite heavily adjusted down low, perhaps they had down a recalc with their poor injector data.

I put the map back to stock, logged cam angle vs rpm and the LTFT with at least 2000 counts per cell and adjusted it. Got most cells between 0.99 to 1.01 except idle at 10 degrees cam which seems to need a massive adjustment, at 0 or -10 degrees it is fine.

Doing this fixed the lean in/rich out during warm conditions however I still get a lean tip in during cold start most of the time, it is irregular and only if I change gears really fast at 2k, even the smallest delay during changing gears and the lean tip in is gone.

Any help for what else affects transient fuel? What values would be used in the calcs?

[rolls]

Wrote some custom graphing software to help with tweaking the SD maps

http://www.hptuners.com/forum/showth...d-density-maps

[rolls]

Not sure if you guys are aware but Eric can add in DMRs for the aussie strategies in HPTuners VCM scanner. He added the following in for me, if yours doesn't already have them if you ask nicely I'm sure he will add them for you.

Trans Current Gear
Injector Pulsewidth
Manifold Charge Temp
MAP
MAP at Zero Airmass ( I think I have this mapped wrong)
MAP Per Airmass (also mapped wrong)
Air Load
Average Air Mass
Fuel Source
Borderline Knock Spark
MBT Advance
Spark SOurce
Idle Speed Source
Idle Adapt (STIT)
Knock Retard
Torque Source
Cam Angle
Boost Error
Desired Boost
Wastegate Duty Cycle
TCC Slip
Att Applied
ETC Vacuum
ETC Throttle Angle Error
ETC Effective Throttle Area
Throttle Angle Source