jeepguy553

Okay...I was looking at my VE%2 table and I just realized how to use DD2000 to get a good start on this...and it may involve a completely new VE1 table. How's this?
I have the engine set up on DD2000 already and it gives me a really nice VE curve right outta the hole. Right? So, I can take that VE curve and plug it into the VE2 table and start working the VE1 table from there. The only pitfall to this that I can see is that the numbers that DD2000 gives are WAY higher than what is currently in the VE2 table. So, to preserve the programming, what should I do? Should I split these VE numbers (from DD2000) in half and put them into the VE2 table? After that, I could work the VE1 table as the variable and leave the VE2 alone.
Does this sound even close to right?

jeepguy553

Now I am home and I decided to give it a whirl with the idea I presented above. I took the DD2000 VE table and put it into the BIN. The numbers didn't match exactly in terms of RPM, so I just kinda extrapolated as best I could. I put the numbers in as they came from DD2000 and applied a divide term of 2.5 to the entire table. Here's the result. This is the actual resulting table. The graph is next and it looks a WHOLE lot better than the original...even better than what I started with. I'll post a shot of the original later.

jeepguy553

Here's the graph of the above table. I think it looks a lot better than the original. Let's see if it runs better.

jeepguy553

I'll run it and adjust as necessary using BLM/INT. Any suggestions and/or flames are welcomed and appreciated. If this works, I will be absolutely ecstatic.

Low C1500

DD2000 will give you good starting point, but you should be able to take a stock VE table and have it closer than DD can get you with only 2 data runs. Your running the 746 right? Well what I do with my 747 to get close VE is as follows:


Pretty much disable the intergrater,
Drecrease my prop gains by 75%,
increase the prop duration,
set the TCC to lock at really low speeds,
slack off the TV cable,

Its been a few months since I nailed down VE, but this method (I may have missed something) allows me to cruise up hills in drive, at low rpms and high loads, So I can high pretty much everything from 1000rpm to 3000rpm. And 3 or 4 runs of 15 min each got me really close.

jeepguy553

Hmmm....thinking....
What is the prop duration? I haven't seen that in my 8746 stuff. Am I missing it?
I am more or less starting from scratch because I am not running this with a GM vehicle. Long story...the TBI system is on an AMC 360 with a 3-speed TF727 tranny and a Dana 300 transfer case in a Jeep CJ7. I have it running pretty well, but as with all things automotive...more performance is the goal. I have already tried with a sotck VE1 and my idle VE1 numbers at 800 x 60 and 800 x 40 are at 9.xx. Way too low IMO. That's why I decided to look at the VE2 as another place to start moving things around. I had heard that DD2000 would give a good start point in the cal by approximating the VE curve (and TQ curve) of the engine. So I decided to give it a whirl. I didn't get a chance to run it tonight...kids, ya know. If I can manage to skate out of work/class early enough tomorrow, I'll probably try it. If it works...great. If not...well...I'll still look like a noob. I think there may already be some mods to it, but I want to experiment at idle and mid-RPM to see what it does like this. It will probably fall flat on its face.

Low C1500

The 746 has all that stuff, it just isn't as hacked as the 747. I assume your using winaldl for a scanner. Try paying more attention to the O2 logs, make changes using them.

jeepguy553

Okay...what do I use as the swing point on the O2 in the absence of a WB? 0.450V? My logs show a range of about 0.120V all the way up to 1.012V. There are a couple right around 0.450V. If O2 is a better way of doing this, then I am there!

Low C1500

O2 isn't better, but I would do say 5 or 10 runs using the O2 values, just manually change the table by a little in the off areas. This will give you feel for, how much change does what. Then use the blm method.

jeepguy553

Gotcha...thanks! I'll let you know how it goes this weekend.

kdrolt

The DD2000 gives you the total VE curve for the engine. VE typically has its maximum in the neighborhood for the rpm where peak torque is made. I doubt your engine has it's torque peak near 5000 rpm. There are two different VE tables stored within the ECM and you are only plotting one of them. (Aside: I don't know why there are two tables, but then again I'm a tuning newbie)

Can you plot what your combined VE curves look like?

jeepguy553

Peak TQ is 410 @ 4000rpm for my engine (AMC 360)
I'll plot it out with Excel this afternoon and post it later.

Mc Firebird

Jeepguy

I tried this in an earlier thread and was told that the ve tables can be used to counteract things like resonant standing waves in the rubber fuel feed pipes !!!

when i used excel to plot ve1+ve2 as a graph there were some weirdo humps and dips and this was given as an explanation.

as grumpy always says "give the car what it needs" there is no way DD2000 is going to know the real world stuff like that.

i think you gotta drive the car to tune it.

jeepguy553

I know that...the VE2 is IMO just giving a better foundation to start tuning. I know there's no way to tune it without driving it. If I implied otherwise, it was accidental. I have been driving it and tuning it. My idle VEs were WAY down...like <10 in the VE1 table and dropping as I kept going. I finally decided that there had to be something that could be done to fix it using the VE2 table. It seems to have worked. Here's the new table as of yesterday

jeepguy553

It still needs work in the mid and upper areas, I know. I was just concerned with getting the idle stuff worked out finally. The rich smell was beginning to **** me off.
I didn't get a shot of the current VE1 table. I'll do that when I get home tonight. It looks a lot different now.

Mc Firebird

excellent news then .

what method are you using to tune the VE, i have searched and there seems to be a few different methods. using int / disabling int / forcing open loop / modifying blm boundaries etc etc.
I dont know if i am comfortable with these as yet. at present i just drive about and log then use "narrow" values to correct. what about you ?

jeepguy553

Actually, I haven't touched the INT or BLM boundaries and I haven't disabled the INT either. I use them both. I am using wide values for my tuning right now because that is where the majority of my data is being output from WinALDL. I haven't seen much narrow data for some reason. Maybe someone here can shed some light on that. It is definitely getting better and better. Idle BLMs were just about perfect ranging from 125 to 128 yesterday with the new VE2 table and 11.xx burned into the VE1 at 800 x 60 and 800 x 40. I may just leave that part alone at this point. It's just about dead on perfect and I think if I screw with it it'll get worse. I also found the flat spots...it was from the abrupt changes in the VE1 table as it went from idle (with a value of 9.xx) to the next steps up in the table at 14.xx and above. IOW, there was a pit in my VE1 and the ECM was having to climb out of it to get to the right point in the table to run the engine at higher RPMs. Likewise when coming off a high RPM back to idle, the ECM was tripping back into the hole.

jeepguy553

Just noticed something else about all this. If I had been intelligent engough to look at my logs REALLY closely, I would have noticed that the thing was running BLM < 128 all the way through the 800 RPM range. Doesn't that point to VE2 as a good place to start tuning since it affects the entire RPM range?
LowC1500...I started looking at O2 ranges and found that the swing point looks like the standard .450V. I got a couple of cells that registered 0.451V during the last logging session. It looked really neat to actually see that in a real world scenario. Thanks for the input.