Tuning my 347 with rear mount turbo | Page 8 | Ford Explorer Forums - Serious Explorations

  • Register Today It's free!

Tuning my 347 with rear mount turbo

Check out my histogram for knock.
Anything in green is nothing to worry about.
Hows this for awesome?

upload_2017-6-10_16-40-23.png
 



Join the Elite Explorers for $20 each year.
Elite Explorer members see no advertisements, no banner ads, no double underlined links,.
Add an avatar, upload photo attachments, and more!
.





That is awesome!
I never even thought of making one for knock/timing
Great job, I may steal this idea. lol.
How did you decide on specs for knock activity?
 






Before I dump too much timing. at .75 load I am commanding a/f of 12.34. I saw an a/f of 13.2 in the log at this load level. Too lean.
I'll get that maf transfer function closer in line to where it should be. Fattening up the fuel will probably help the detonation issue.

If getting the a/f down to what I'm commanding doesn't drop the detonation, I might command an even richer a/f in the fuel table.
 






I didn't use any spec's, just the raw voltage. It might not be a bad idea to multiply the voltage by something as a formula in the livelink analog 1 spot, but this seems to work great.

I think the vampiure unit has a max of 1 volt output, so seeing the numbers in raw format is ok.
 






Ok, guys, interesting tuning thing found with the vampire unit and knocking. Keep in mind that the knock sensor brain does not even activate till it see's 5psi vacuum

Here's my borderline knock table. Decipha stated that MBT for aluminum heads at low load cruise was 45. I know there is lots ov other variables, but that is where I kind of adjusted my spark around.
upload_2017-6-15_22-49-3.png


Son, here is my histogram showing knock. I had to step on the gas at just the right moments, but have a look:
upload_2017-6-15_22-53-18.png

Red is bad. lol Red is knock, or false knock?

So, this tells me that you can actually touch on low load areas in transition to high load right from the base. How could I have a low load condition under low vacuum? Well here it is, you clearly can.
With the vacuum port off the Vampire brain, it it always sensing, and this knock does not happen under cruising conditions. Only quick transitions to wot while the vehicle is already moving.

.So, I reduced timing in the offensive area.
upload_2017-6-15_23-2-11.png


I didn't log after this, but I did end up going for a drive. I can tell you for certain that this helped quite a bit. The Vampire unit is still retarding timing under certain conditions, but far less. I may very well end it at the factory timing (Numbers are black) in this area when I am done.

I don't want to make any further changes to spark till I actually log what rpm/load the knock is coming from. No point in guessing at this point.

Very interesting, and worthy of pointing this out. Once I'm thru this process, I know there will be absolutely no reason to visit a dyno to find more power.
 






i may have to beg and throw some money at you to borrow that thing when your done tuning. then again, when is tuning ever done, right?
 






We just need to see if we can get a knock sensor with connector, knock sensor adapter, and the connector that plugs in to the vampire unit. Then, wiring, and swapping the brain over isn't difficult.
Moving knock sensor and wiring back and fourth is a big deal. Its not easy to get at that stuff.
 






@4pointslow
I have a bit of an issue. Its small, but interesting.
Tuning via ltft histogram, I would end up with a datapoint on my MAF thats lower than the one previous to it.
The LTFT is perfect in Neutral (datapoint 124), but in drive I'm looking at multiplying datapoint 151 by .96.
This will make datapoint 151 lower than datapoint 124 (Both are at 623 currently). Doesn't this create other issues? Am I ok to do it in this case? Leave it alone and let adaptive learning do its thing?

upload_2017-7-5_8-51-40.png


upload_2017-7-5_8-53-1.png
 






Keep in mind that there are glitches sometimes in the datalogs. They can come from voltage spikes during the recording.
I have looked at them in excel as I assume you have too.
If there seems to be erroneous data, just leave it alone.

If there are no glitches I would tune it off of the readings in drive. We spend more time in drive at idle than neutral in idle under regular usage.
With that said I will add more.
I like my mixture to be a tad on the rich side, so I would prefer it to say I need a .96 adjustment (and not do it), than a 1.04.
So I tune to try to make the PCM have to take a little fuel away, again I say a little.
Why? a slightly richer mixture at Idle will run slightly smoother.
A slightly richer mixture at WOT burns cooler in the combustion chamber, further away from detonation.
Again I say slightly richer because an extremely richer condition leaves fuel behind that can end up in the crankcase through normal blow-by gases from normal ring end gaps.

With that all said the target is to be within 3% of desired A/F.
So I would try to adjust it by 1% or .99 multiplier, but 4% is really only 1% away from target of within 3%. lol.
You can not have 151 data point lower than 124 data point, it will cause problems or even cut fuel.

Your options,
Leave it alone
try to lower both by the same amount
try to lower the one below the one that says it needs to be adjusted and see if that alone fixes the one above.
There may be more I am not thinking about.lol.

Hope I didn't make it more confusing.
 






lol.
Clear as mud.

I think I'll leave it alone for a while. I want the ltft's down low to be at 1.

This truck has a starting issue. It just does not like to start without touching the gas.
Of course I play with different crank fuel pulse widths and air charge during crank. I finally think I have it right, and after a few days it wont start without touching the gas. Very odd. I'm wondering if adaptive learning is retaining its memory thru ltft's and throwing off the start mixture?

It just seems to be a characteristic of this motor. It's always behaved like that. I'd sure like to figure it out though.
 






did you want to look at my logs and tunes for your start up? i have noticed now that i found that massive crack in the mount it starts way better with very minimal surging
 






No, I'm absolutely sure its just my motor.
Iv'e read and read. Iv'e tried all kinds of different techniques.
I did email and speak with Don Lasota about the issue, and he said that some motor combinations are like that.
Its just so finicky.
Its a small thing, but a bit irritating just the same.
 






Forgive my lack of attention, but are you still using the returnless fuel system, and if so the stock fuel damper? ( regulator looking thingy on fuel rails)
 






Yup. stock fuel rails and fuel damper still attached (Hooked directly to manifold vacuum now). A 340 pump in returnless config and 6an fuel line from pump to fuel rails.
No fuel starvation issues. One day I'll move to bigger rails just because..
 






Im still struggling with exhaust smell and problems starting when the motors warm.

I still think the issue is that I have the wrong 60lb injector file in my tune.

I add the correct injector file, and it's like my narrow band o2's quit working.

I asked Don Lasota for help. His best guess at this point is that my maf needs to me closer to my throttle body. So, I'm hoping to complete this project next weekend.

I'm at a loss as to why a change of injector file causes this issue, so I'm more than willing to try anything.

I'm thinking the wrong injector file is behind my issues. My motor is still doing great though. The rest is small stuff.
 












No, not this. I tried applying the ford racing 60#v2 injector file, and it would seem like the narrow band 02s wouldn't read and I get results completely different than what my wide band would be telling me.

Don Lasota sent me a different file with the correct injectors for my truck (the ones I couldn't get to work) and the results were the same as mine.

Don sent me another file with all learning shut off so he could see the maf and make sure the computer wasn't defaulting to load with failed maf table. His best guess is that the maf is too far away causing issues. The maf is under my truck, as I felt having it there in a nice long straight run just after the intercooler straightening the air would be an ideal spot. We will find out if its a problem, or not, after I move it.

Any way I look at it, its probably better for the maf sensor to be under the hood.
 






If the injector data is way different from the one you are using you could end up with a way rich or lean mixture when trying to switch back to the correct one.
A misfire caused by a rich or lean mixture will cause lots of O2 in the exhaust (from either type of misfire).
So the misfire could cause what looks like a lean condition when it really is rich. That would sure make it hard to tune the MAF transfer function.

It will be cool if moving the maf closer fixes the problem, but if it doesn't let me know.
The other thing I am thinking is do you know for sure what injectors you really have, have you been able to get the part number off them?
Just wondering because it is critical to get the right injector data entered into the tune or nothing seems to work right sometimes.
 









Join the Elite Explorers for $20 each year.
Elite Explorer members see no advertisements, no banner ads, no double underlined links,.
Add an avatar, upload photo attachments, and more!
.





Tim, with the correct injector data, the narrow band o2's just don't seem to work. Its really weird. I haven't calibrated my wideband for a while, but this is a different issue.

4point, I'm 99.99% sure I have the Siemens Deka 63lbs/hr EV6 Injector Part #108191. I'm sure these are the ford racing 60#v2 value file. I'm out of town till tomorrow evening, so ill try to remove that other .01% doubt I have on the injector part number then.
 






Back
Top