Auto tune

[Roverdose]

ive only use lambda control once, worked ok! not the best on the market but perfectly good.

omex has excellent lambda control for simmilar money to an s40. but features and ease of use are much better on the dta.

Drew

[kutle]

Yes, if you put zeros in vipec it will make pretty good basic map, if you put in it basic map it would make much closer to finised map. On the other side if you put zeros in DTA you will have zeros. If you put basic map it wouldn't autotune it not even close like vipec.
I know that dta is computer, but software is working with parameters that you write in, and dta software is total rubbish. If you hade carbs before than ok,dta is better,but oem 20years old ecu's have better algorithams.

[Antral]

Begining to sound like that Jaguar guy

[VR6Turbo]

DTA build race ecus, race engines are almost imposable to get to run closed loop they rev up and change load point to fast.

But I can't get it to work properly on a simple road car engine, well, 3 different engines actually.

it also seams to me that some people can get the DTA to work fine and some cant, this says everything about the people using it and nothing about the ECU.

I would hardly class 2 days solid on the dyno, writing a spot-on base map, as an inability to use the ECU.
The fact remains that in my experience, both on a P8 Pro and an S80 and with 3 different lambda controllers, DTA's closed loop just cannot be relied upon for truthful and / or consistent corrections. And I am not alone.
Even Allan himself admitted to my mapper that it has it's weaknesses, especially when using the older LC-1 WB, where the P8 Pro, WB display and workshop Motec PLM all read a different AFR. How the hell can you even begin to configure closed loop when you don't know which WB to trust?

for example if you go to Bosch and buy one of there ecus you have to start from nothing! there isent even the maths to work out RPM in them, fantastic if you want it to be flexible, rubbish if you want to plug in and drive!

Yes but once the Bosch has been configured, it's consistent forever and a day thereafter, and is precisely why they are chosen by the OEs.

the first bike engine i dynoed with DTA was on the dyno for 2 days! to get the injector angle and high low injector blend map correct i think took about 4 hours running time!

Yep, same here. We built a lovely open loop map that was perfect. Perfect on the dyno, perfect on the road. But perfect only on that day and conditions, which is why I needed CL in the first place, to make the fuelling more consistent over summer and winter. So on the same day, closed loop switched on, it all goes to shit almost immediately and no amount of parameter adjustments seemed to make much of a difference.

See, what's interesting about the Motec is built in WB and if you use a standard LSU4.2 probe, then you're onto a winner with consistency immediately. Oddly enough, OEs also have onboard WB. Having to rely on a 3rd party WB is where some inaccuracies can start to creep in, but isn't the sole cause of the poor CL performance.

Don't get me wrong, my car is far from undrivable and doesn't kangaroo down the road, belching out black smoke or anything. It's just the finer control I would like, particularly in idle and closed loop lambda. Hopefully some improvements in these areas will follow in due course. I do like the DTA overall, it's far from a terrible product and the software is so easy to use.

[kutle]

I also like some things in DTA but would suggest to Allan that because obviously they are not enough competent to resolve all problems with firmware it would be best to open source code and allow others to improve it.
As their funcions are obsolete nobody actually can stole somethig from them, it can only be improved and make that ECU more usable because it is not fair to sell anybody unfinished ECU.

[Jon K]

LC-1 WB, where the P8 Pro, WB display and workshop Motec PLM all read a different AFR. How the hell can you even begin to configure closed loop when you don't know which WB to trust?

That's not a DTA issue at all. You have ground offsets on your devices. My NGK AFX and my DTA AFR reflect exactly. If you mix two widebands up they are sure to read off from one another. You have to use a single wideband that you trust. Outside of a lab there is no "absolute" AFR Meter. Once you start referencing several you are sure to get ground offsets and circuit discrepancies.

[Jon K]

Begining to sound like that Jaguar guy

It's not

Jaguar guy came up as from Croatia and this guy is from UK.

[VR6Turbo]

LC-1 WB, where the P8 Pro, WB display and workshop Motec PLM all read a different AFR. How the hell can you even begin to configure closed loop when you don't know which WB to trust?

That's not a DTA issue at all. You have ground offsets on your devices. My NGK AFX and my DTA AFR reflect exactly. If you mix two widebands up they are sure to read off from one another. You have to use a single wideband that you trust. Outside of a lab there is no "absolute" AFR Meter. Once you start referencing several you are sure to get ground offsets and circuit discrepancies.

Nope, it is not. After ripping out the LC-1 and fitting a PLX Devices M300 (using exactly the same ground points), the DTA, PLX display and the Workshop Motec all agreed. Funny that. We even had Allan on the phone whilst mapping it with the LC-1 and he couldn't answer it. Not long after that, an LC-1 preset mysteriously appeared in the Lambda parameters page.

[Jon K]

LC-1 WB, where the P8 Pro, WB display and workshop Motec PLM all read a different AFR. How the hell can you even begin to configure closed loop when you don't know which WB to trust?

That's not a DTA issue at all. You have ground offsets on your devices. My NGK AFX and my DTA AFR reflect exactly. If you mix two widebands up they are sure to read off from one another. You have to use a single wideband that you trust. Outside of a lab there is no "absolute" AFR Meter. Once you start referencing several you are sure to get ground offsets and circuit discrepancies.

Nope, it is not. After ripping out the LC-1 and fitting a PLX Devices M300 (using exactly the same ground points), the DTA, PLX display and the Workshop Motec all agreed. Funny that. We even had Allan on the phone whilst mapping it with the LC-1 and he couldn't answer it. Not long after that, an LC-1 preset mysteriously appeared in the Lambda parameters page.

Why do you need a preset? I always set LC-1s to 0v = 10 AFR and 5v = 20 AFR, then set it in DTASWin. If it's not a ground offset then it's a bad voltage to lambda table.

[Antral]

Begining to sound like that Jaguar guy

It's not

Jaguar guy came up as from Croatia and this guy is from UK.

Jon, I wasn,t implying it was him I know full well where the clown came from.