Discuss all things tuning in this section. News, products, problems and results.
As the title says - im after opinions about the "optimum" (yeah - this might turn into a can of worms but hey) way to configure the fuel and ignition (and other) maps to tune a honda k20a engine.
The base map that haltech provide gives a single fuel and ignition maps for the entire rev and load range and have various other things i didnt like about it.
After studying the layout of a hondata ecu (which i didnt want because they dont have a good base map with data ripped from the jdm ecu's) - it has multiple maps for fuel and ignition that cover low cam and high cam (vtec off and vtec on) - and then they are broken into individual maps for cam angle (ivtec cam angle control) with maps covering 0, 10, 20, 30, 40 and 50deg
Yup - hondata has a #@^$ ton of maps that need tuning for an engine 2(vtec) x 6 (cam angle) x 2(fuel+ignition) maps in total to get the "perfect" tune
Haltech will let you setup fuel/ignition x cam angle which is acceptable
The benefits i can see for this is that - well - the cam angle is not going to perfectly track the target angle - it lags - it overshoots - it undershoots etc etc - and in the times it is doing that - its going to different interpolated fuel and timing maps on the hondata giving in theory the exact fuel and ignition needed for any given state of vtec / cam angle
Unfortunately i have ocd about some things - so im not overly happy with the haltech base map of 1 fuel and 1 ignition table for all - because theory and reasons more than experience or advice.
So i got a "base" dyno tune done at a workshop with a plex to make solid and safe ignition tables - setup the haltech for multiple cam angles for fuel and ignition maps and got back a car that im quite happy with how it drives etc. But ive been thinking bout it some more and i wanted the advice from people who have tuned haltech / hondata ecus and have experience with the and can offer an opinion
Should i keep trying to "copy" how honda/hondata have multiple maps - i feel that in the end of the day - with *a lot* of fiddling and tweaking to things - they could end up with better drivability - at the expense of massive amounts of time and effort
compress them all back to one map again for fuel and ignition - maybe splitting it into vtec on and off maps only and just use the dyno resulting cam angle map i got from my new base - but that means that any cam angle changes will require tweaking fuel and ignition cells at that point where the multiple maps once you have them done - you can do whatever you want to the cam angle table and it will "just work" because the data is there already in the multiple maps
So - opinions please - is it worth the extra effort on doing alllll the maps to gain a bit of drivability n stuff - or is it a waste of time and the average driver (me! lol) wont notice it on a road car (this is a daily drive jdm ep3 type r - not a track or weekend toy)
I also want to work on linking the cam pid tables against rpm as well not just a 1d coolant temp axis so i can get nice smooth response to the desired angle request that wont exceed the oil pressure thats required to get the cam to move there
Also want to link the injector firing angle against cam angle as well not just rpm because the "best" injector firing angle will change with cam angle as well as rpm - again letting you make any changes to cam angle and have things "just work" because the data is there already
The car is a play toy really - im experimenting with things but have no practical experience with a honda k20a - the closest ive fiddled with something like this in the past was a sr20ve nissan with nissans vvl (vtec)
My biggest concern is wasting too much time when the same results could be had different ways - so opinions please if you can back them up with theory or facts - im very interested in learning more from other people.
I may be a bit brief but in what scenario are you planning on running?
Using a single base map irrespective of VTC or high lift cam is perfectly acceptable as you will always be as that specific set point unless you map the VTC or vtec to be switchable which seems like an over complication.
To break it down a little bit the vtec switching point is going to be set independent of load and this wont change so at 4000rpm it will always be on the low cam and at 6000rpm it will always be on the high cam so having 1 map for low cam at 6000rpm is irrelevant unless you plan on having an economy type map.
The VTC should be mapped at linear set points and the results blended together to define the optimum VTC map - the fuelling and ignition table can then be fine tuned to suite this optimum VTC map a wideband controller should mop up and account for any VTC target errors.
the Hondata is probably more representative of an OEM calibration as the OEM will have to account for a multitude of variables, driving conditions, engine conditions and still be emissions compliant.
Tuning iVtec is a little more complicated than traditional variable timing and is an iterative process but worth putting the effort into especially in the part load and transient areas. I'm not 100% on Haltech but some standalone ECU's have user configurable scalars where you'll be able to adjust the fuelling or ignition based on whatever input you like so in theory you could add fuel to a base map dependant on VTC for finer control for example but again I'm not sure if this is a feature in 2500
Hi Scott, Thank you for your input here :)
The car is a daily driver i use to experiment with different settings - mainly datalogging for if i take it out to the track anytime and to see how its behaving.
99% daily driver tho.
From what your saying it sounds like i can get away with collapsing maps back down to 1 table for fuel and 1 for ignition - and the reason behind all the extra maps on hondata type setups is for better control over emissions during all the transition type events.
Well ive got the saved dyno tune - so i'll experiment with collapsing the fuel back down to one map for a start and go from there.
Im still super happy to hear anyone elses input as well - as its something to keep me occupied during these wonderful times we're in so please let rip with theories or facts or experience - it all adds up :)
Your methodology is on point and hats off to you for thinking about it.
There is a pretty well established method for tuning VVT and you might find these webinars useful.
You're quite right that injection angle should vary to suit the VVT map and assuming the VVT map is fixed is then, based on load and speed then the injection angle should use the same control parameters and share the same breakpoints. A good way to correctly calibrate injection timing is to monitor exhaust gas UHC content with more complete combustion ie. optimum injection timing giving the lowest values whilst still maintaining MBT - monitoring fuel stand off is also a good indicator of the correct injection timing angle but this is only really possible with an open inlet.
Many thanks - ive collapsed the maps down (took a while using the cam map to work out what fuel / timing maps i needed to pull data from etc) and to be honest - it seems to drive a bit nicer now with a single fuel and ignition map - so on with my next experiments - disable auto iat ve compensation and fiddling with the transitionary enrichments etc
Thanks heaps - i'll be investigating the injection angle after i sort out the fuel map a bit better - thanks for the links to the webinars - i'll be watching them over the weekend :)