[djustice]

I have been trying to tune a part of the fuelmap that gives me issues during acceleration.
Going wot from idle or just after about 1000-1100 rpm, what i think is happening is acceleration enrichment is disableing learn when it is active hence this area will never have self learning. i have been manually adding fuel to a portion that is from 80 kpa and up to 105 kpa in the 900 rpm to 2000 rpm area with little change. i have attached the my cfg file and a datalog in the link below.
the datalog has 2x heavy acceleration events one rigth after the log starts and one at very end of the data log, both times experienced stumbeling and hesitation til i eased off the throttle.
if i roll into the throttle and gain more rpm on my way to full throttle there is no issue.
After this log i did a +5% extra fuel to base map in the affected area, but that did produce much worse stumble. so i reverted back.

If anyone is intrested i would love to have feedback on the datalog and base map. there has only been cruising and a few full throttle pulls to about 3500 - 4000 rpm so far. and a lot of transfer learing to base + manual smoothing (no smooth all button used) https://www.dropbox.com/s/92bre9z7ip...20log.rar?dl=0

[Scott65]

If possible, try setting AE to zero. It sounds like a too rich condition. If you get the ve table right, AE gets much easier to figure out.

[djustice]

Think i am headed in the rigth direction, after watching some of Joe Simpsons tuning video's on the holley system, basically if a sharp acceleration evenet dont clear up within a second, it is the base table. with that in mind i carefully looked at my log and found both the base table and accel enrinchment to be wrong in this area. Accel enrinchment appers to to be too lean spiking to 20 AFR , then after the accel event ends i am on the base table that is delivering as much as 9.5 AFR.
Just pulled out over 5% fuel off the base table (in the affected area) and smoothed down the surrounding area, car is 95% better now.
Posting a picture of the heavy acceleration event on the datalog to what i belive was the issue.

[JLMounce]

What does the Sniper use for acceleration enrichment? If it's like the FiTech there should be 4 different variables that effect enrichment. What you have going on is the classic lean tip in followed by rich condition oscillation that seems to plague these systems.

In FiTech land ther four variabls are as follows:

Accel Pump - Slow reacting accelerator enrichment, based on MAP. This is for typical driving throttle imputs
Fast Accel - This adds fuel based on Alpha-N which is a combination of TPS and RPM and is for quick transient throttle applications.
dTPS Acc Gain - This is a modifier to Fast Accel based on how quickly the TPS values increase. Increasing the gain, increases the amount of fuel applied in transient operation
dTPS Gain Max - This sets the upper limit of fuel added during quick transient operation

As I stated, I don't know exactly how the Sniper software handles this. The Holley software has a much larger VE table and may not use Alpha N as a result. So this all my by wild speculation for you.

There is also an accel decay which will come in to play here. Based on the log you've posted, this is how I would tackle the issue.

1. Your fast accel is not providing enough fuel, quick enough. I would increase the gain function first and see what it does
2. If gain alone will not take care of the lean spike, you'll need to also add fuel via the accel pump feature
3. Continue to add gain or fuel until tip in becomes acceptable. If you're close to commanded but either a bit lean or a bit rich, you may wish to stop there
4. If you've swung to a rich tip-in situation, this is where the Gain Max function comes in. The transient is working correctly, you can work with the max function to limit it's effect.

Finally the decay function comes in to play to smooth out that rich oscillation. If the tip-in is still a little lean, but you don't feel it, then again works it's way rich, you may need to decay the accel pump quicker to get rid of the rich situation after tip-in.

Look for similar functionality if available.

[djustice]

Thank you JLMounce , seems there are som similar features between fitech and holley sniper.
The Sniper system is purely VE based and LB/HR ofc. when it comes to acceleration enrichment there is almost too many tables, there is 6 tables and a RoC blanking value: AE = Acceleration enrichment, the tables:
TPS based:
-AE vs Rate of change (this table also has a Roc (rate of change) blanking value, this acts as a filter to filter out unwanted enrichment
-AE TPS vs coolant temp
-AE correction vs TPS

Map based:
-AE vs map rate of change
-MAP AE time vs coolant
- MAP AE vs coolant

As i am led to belive the most useful and most used tables are the TPS based ones. That beeing said i think i have some Base map tuning to do still, basically i want that to be as close to the commanded AFR as possible before i change the AE values. This would be acomplished quickly on a hub dyno possibly in first gear as the th400 most likley would kickdown or up/downshift during low rpm high load.
The car seems to work better in this regards with som heat in the intakemanifold, this beeing a singleplane with no heat crossover, and a seperate watercrossover i think i migth have some atomisation issues before i get som temp in the intake too.

[JLMounce]

This is one area where I do personally believe that the FiTech's Alpha N map shines over the Holley's software, especially in a typical consumer category vs a "pro-sumer" or super user.

MAP is too damn slow for transient events if the VE table isn't perfect to begin with, which to Scott's point, I think is the issue here. The VE table needs to be solid in this scenario so that AE can modify and filter in very small increments. This is what causes the oscillation as the system tries to recover, but goes too far in the opposite direction.

It's a bandaid on these things, but really the MAP sensor should actually be...IN the manifold. As opposed to being in the throttle body. Yes in the FiTech and Sniper it's still below the butterflies, but I do worry about accuracy and response times of the MAP sensor based on this compromised placement.

[Scott65]

Using an entirety tpsdot based AE strategy is simplest, if that's an option. Map can change (both legitimately and not so much) in many different scenarios. Using tpsdot is keeping things honest. It should only actuate with TPS change, and be adjustable by RATE of change, which is important to the task of controlling an adjustable amount of fuel for different scenarios. (i.e. how hard/fast you're stomping the loud pedal)

[djustice]

Originally Posted by Scott65

Using an entirety tpsdot based AE strategy is simplest, if that's an option. Map can change (both legitimately and not so much) in many different scenarios. Using tpsdot is keeping things honest. It should only actuate with TPS change, and be adjustable by RATE of change, which is important to the task of controlling an adjustable amount of fuel for different scenarios. (i.e. how hard/fast you're stomping the loud pedal)

Yepp it is clear looking on the log that TPS based input enrichment would be the fastest, going by map would induce lag.

[Formulajones]

I went through this exact thing with the Sniper Stealth on my Chevelle.

From idle, a WOT stomp of the pedal would create a massive bog. It would recover if I lifted off the throttle slightly.

Datalogs showed something I didn't even think possible, but on the main fuel table the curser at idle would shoot straight up to the top left corner of the table. I thought it would follow a more linear curve towards the right as RPM increased but that's not the case when you whack the throttle hard enough. It goes straight up in the idle column.

I had a couple issues. First I had to make sure I was commanding a reasonable AFR up in the top left corner of the table that would support WOT. So that is in the mid 12's. The second issue is that my car idles best when I command a 14.2 AFR (13.7 on my other AFR gauge) and in gear it idles up around 74-75 KPA at 900 rpm (aggressive camshaft)
So to make this work my fuel table has to be about 14.2 AFR at 900 RPM and from about 78 KPA and up it has to quickly ramp up to 12.6 AFR in the 900 RPM column up to 100 KPA. So when looking at the graph that area looks like a pretty steep ramp. It's pretty common to have a fuel table that is very low in the idle areas where VE is low, and ramps up around it.

This part, as Scott and Jlmounce has already eluded to, has to be dialed in first before thinking about messing with AE vs TPS tables.

Speaking of the "pump shot" tables I had mixed and opposite results vs what most say should work. After my base table was dialed in I did have some light tip in throttle lean spots while driving.

When I increased AE vs TPS Rate of Change it did nothing for me. I made stupid crazy increases on that table from right off idle and up and it never got rid of the lean tip in stumble.

However, when I just slightly increased AE vs Map rate of change I instantly noticed better light throttle tip in. That's the table that made the most difference for me.

Your ROC blanking is just something you have to play with. Stock tunes in Snipers used to come with 15 but Holley has now changed the base tunes to 7 ROC blanking.

Luckily mine being 3 years old had the older base tune setup with 15 ROC blanking because 7 would have never worked. You just have to keep lowering that value while idling until you see the Sniper going in and out of closed loop. When you see that happening, it's too low, bump it up a number or 2 and leave it.

I ended up with 9 on my ROC Blanking value. 7, the recommended number I see thrown around, and now in the standard tune, was too low on both cars.

[djustice]

Originally Posted by Formulajones

Datalogs showed something I didn't even think possible, but on the main fuel table the curser at idle would shoot straight up to the top left corner of the table. I thought it would follow a more linear curve towards the right as RPM increased but that's not the case when you whack the throttle hard enough. It goes straight up in the idle column.

Yepp i was amazed too, i tougth it would atleast drift 200-300 rpm before it reached 100 KPA, but mine is straigth vertical.

Atlest i have been a belivere of this tuning guide map i have seen on the Holley EFI forums, and taken it as gospel, well it is not wrong but i just never imagined i would be tuning the rarely used portions.



And thanks for the AE map accel tip. i'll dig into it down the road if things still are bogging

[Formulajones]

Originally Posted by djustice

Yepp i was amazed too, i tougth it would atleast drift 200-300 rpm before it reached 100 KPA, but mine is straigth vertical.

Atlest i have been a belivere of this tuning guide map i have seen on the Holley EFI forums, and taken it as gospel, well it is not wrong but i just never imagined i would be tuning the rarely used portions.



And thanks for the AE map accel tip. i'll dig into it down the road if things still are bogging

Yep, even with the issue fixed in the fuel table, with copious amounts of fuel up in the top left corner of the table, you can stab the throttle, and rpm obviously spikes up, but the curser in the table still goes straight up to the top left corner of the fuel table in the 900 rpm column and then finally moves to the right, as if it can't keep up with the rpm trace. Doesn't make sense to me but that's what it does, making that part of the fuel table crucial to tune properly if you want crisp WOT performance right off idle. Even with a really tight converter the RPM still climbs faster than what that curser shows, and with a 3500 converter that flashes immediately, well I shouldn't have to explain what that does, lol.

So it ends up making your fuel graph jump up increasingly fast in that area and looks nothing like the pretty fuel graphs everyone posts up. I see all those beautiful looking fuel graphs and think to myself, "how do they whack the throttle from idle when the cells at the very top left corner still only have 10-12 lbs per hour of fuel, that's just going to fall flat on it's face.

It's become apparent to me that those pretty fuel graphs are from people that putt around in the car and obviously don't put the engine in all scenarios of driving. So you have to take that tuning guide map with a grain of salt. I mean, there is some basis for it, but parts of it just isn't going to work in the real world.

[Formulajones]

Originally Posted by djustice

And thanks for the AE map accel tip. i'll dig into it down the road if things still are bogging

You're welcome. As I understand it, it's not typically how one should go about tuning a lean tip in stumble. That is supposed to be done with the AE vs TPS, but that table just would not do anything for me at all. I got silly with it and just moved it really high across the board and it changed nothing. So I put that back near stock settings and started playing with the AE vs Map and the very first change I stayed conservative, at just a couple percent increase, I instantly noticed better tip in throttle off idle and while cruising.

I don't know why that is but that's what worked for mine. Oddly on dad's car I did not have to mess with these tables, and his engine is a much larger and quite a bit more HP than mine is

Best part is if it doesn't work just revert back with a few key strokes