Maxed Out My VE Table, Still Lean? | AEM
18 posts / 0 new
Last post
FreakshowFPV
Offline
Last seen: 3 months 2 days ago
Joined: 04/13/2020 - 13:08
Maxed Out My VE Table, Still Lean?

So long story short, I got wrangled into doing some base tuning on my Dad's project.  I have no real experience tuning with AEM or any other ECU and its been going well enough trying to learn it as I go.  I feel like I understand a good deal of what I'm doing.  As the tune has been progressing, I keep venturing further into rpm's, throttle position and in our case, boost pressure.  This has been going fine until yesterday when I hit the hard 200 VE limit.  I still have some cells I need to tune, but I'm maxed out and its still lean.  My VE numbers are around 100-105 in the idle areas and hit 200 around 6000 rpms and 90% throttle.  I need to tune to 7500 rpms minimum and it just keeps getting leaner if I go further on a data logging session. 

 

Is there anything I can do to change the scale for more resolution high in the rpm/load range?  

AEM_NS
Offline
Last seen: 2 hours 7 min ago
Joined: 06/06/2014 - 09:31
Something along these lines

Something along these lines was just recently discussed: https://www.aemelectronics.com/forum/understanding-ve

FreakshowFPV
Offline
Last seen: 3 months 2 days ago
Joined: 04/13/2020 - 13:08
I appreciate the attempt, but

I appreciate the attempt, but that doesn't really relate to my issue.  I understand the function of the VE table values.  The problem is I cannot raise my values past 200 to achieve a proper AFR in certain areas of my map. Can I adjust the scale down somehow?  My lowest values are 99-107, but I need more fuel than is provided by a value of 200.  If I could adjust the scale down, my idle VE numbers could be, hypothertically, like 50 and then the values that are 200 now could be, hypothetically, like 150.  Thus giving me more resolution at the top of the scale.  

EvoNate
Offline
Last seen: 2 months 2 weeks ago
Joined: 09/16/2016 - 18:02
The answer was in the

The answer was in the discussion posted. If your VE values are that far off, your injector data is probably incorrect. Verify the Injector Offset and Flow Rate data under the Injector tab is correctly configured for the injectors you're using.

FreakshowFPV
Offline
Last seen: 3 months 2 days ago
Joined: 04/13/2020 - 13:08
Oh jeez, I'm dense.  If I

Oh jeez, I'm dense.  If I alter the injector data, will the VE table automatically adjust its values to compensate?  Like say I change the injector information so that they flow 20% "more" will the VE table adjust its value by 20% as well or will I have to redo the portions of that map that are fueled correctly?

 

Thanks for the help.

AEM_NS
Offline
Last seen: 2 hours 7 min ago
Joined: 06/06/2014 - 09:31
No the VE table won't

No the VE table won't compensate for a change like that - why would it? The engine's actual VE doesn't change with injector data but since whatever injector data you've told the ECU isn't true (since your injectors appear to be flowing a lot less then your injector flow table says), you can offset both the injector flow table and the VE table by the same amount and fueling will stay the same. So since you're maxed out at 200 VE and realistically you should be around 100, if you decrease you injector size by 50%, you can also decrease you VE table by the same amount and everything will stay the same. That's how you "re-scale" the VE table when you're way off due to bad injector data.

So to further elaborate this point, when you decrease your injector flow by 50%, the ECU is going to compensate by increasing injector pulse by 50% because if you decrease your flow rate by half, you'll need to increase your injector pulse by half for the same VE in order to have the same AFR.  Since you don't actually want 50% more fuel, you then decrease your VE amount by the same 50% to get the injector pulse scaled back down to normal.  And notice that in order to get more fuel, you need to tell the ECU the injectors flow LESS fuel, not more.  Make sense?

EvoNate
Offline
Last seen: 2 months 2 weeks ago
Joined: 09/16/2016 - 18:02
FreakshowFPV wrote:

FreakshowFPV wrote:

Oh jeez, I'm dense.  If I alter the injector data, will the VE table automatically adjust its values to compensate?  Like say I change the injector information so that they flow 20% "more" will the VE table adjust its value by 20% as well or will I have to redo the portions of that map that are fueled correctly?

 

Thanks for the help.

Your VE table will need to be adjusted when you change the fuel flow data. Reducing the VE values by the same percentage that you increased the fuel flow data may get you in the ball park.

AEM_NS
Offline
Last seen: 2 hours 7 min ago
Joined: 06/06/2014 - 09:31
EvoNate's on it. 

EvoNate's on it. yes

FreakshowFPV
Offline
Last seen: 3 months 2 days ago
Joined: 04/13/2020 - 13:08
Thank you and everyone else

Thank you two so much.  Sorry I didnt understand your original response correctly.  As I mentioned in the OP, I'm a total noob with EMS's.  The original tuner/fabricator my dad had on this project passed away and everyone else we've found since then has been, lets say, far less than adequate.  I really appreciate the help.  I believe you guys have given me the tools I need to correct the issue and its much appreciated.  I'll report back with results, but I'm expecting I understand enough to be able to proceed.

Slightly unrelated, its for a damn 3 rotor rx7 of all things, so i'm basically pissing my pants throughout this whole process. I really don't want to have to pull this motor and learn how to rebuild a rotary because I can't understand infinity tuner haha.    

AEM_SB
Offline
Last seen: 2 months 3 weeks ago
Joined: 06/07/2014 - 14:29
Another thing to consider is

Another thing to consider is whether the fuel system is able to supply enough pressure/flow that the injectors actually spray the expected amount fuel into the engine. This could be caused by the engine needing more fuel than the pump can supply. Most cars use electric fuel pumps, like the stock one or our AEM fuel pumps or other common brands like Bosch 044 / Walbro 450 / etc. The fuel pump needs enough electrical power (voltage and current) to deliver fuel; it's a pretty common problem to use insufficient wires or connectors supplying electrical power to the fuel pump. When the wiring and/or connectors isn't correct, the fuel pump won't receive enough power and won't deliver enough fuel. You can usually see this in datalogs if a fuel pressure sensor is installed and connected to the ECU for logging. 

 

For a rotary engine, there is also the variable of staged injection. In a stock setup those engines use two injectors per rotor (one primary and one secondary). The primary injectors (near the engine) are usually sized small enough to deliver the right amount of fuel to idle. The secondary injectors (further upstream in the intake manifold) are usually larger and only operate at higher RPM / load. You need to configure the ECU so it uses the correct injector flow & offset & fuel type for the primary and secondary injectors. You also need to configure the StagedSplit [%] map to split the fuel delivery between the primary and secondary injectors. Getting any of those settings wrong (flow, offset, fuel type, and StagedSplit) in the InfinityTuner software can result in the engine receiving less fuel than the pump might be able to deliver. The correct numbers to use in the StagedSplit map depend on the flow rate of the car's injectors. Post that info here, and a copy of your calibration, and I can help talk you through it. It's also described in the 'Staged Injection Tuning' section of the instructions PDF, somewhere around page 258. 

 

I agree it's a good idea to be cautious with a 3-rotor engine. Set the boost as low as possible, and dial things in at part throttle before doing any full throttle pulls. I can recommend at least one tuner who is experienced using Infinity ECUs on rotary engines, but can't guarantee they will be interested in remote tuning.

 

Hope that helps,

SB

AEM_SB
Offline
Last seen: 2 months 3 weeks ago
Joined: 06/07/2014 - 14:29
Also, this video on our

Also, this video on our Youtube isn't a complete how-to-tune guide but it might help understand why some numbers for VE are obviously 'right' or 'wrong' once you understand the related math/physics.

https://www.youtube.com/watch?v=Xou5u9DaH2Q&list=PLsqzudH4kSTvvYQfacvWsuni8nNUyL533

VE describes how well the engine flows air, and it's flat-out wrong to have numbers near 100% VE at idle with the throttle mostly closed. If you want to learn about tuning, EFI University, Horsepower Academy, and Evans Tuning are well-respected companies we have worked with. There is a lot more info available online now than 5-10 years ago.

 

Hope that helps,

SB

FreakshowFPV
Offline
Last seen: 3 months 2 days ago
Joined: 04/13/2020 - 13:08
Thankfully the staged

Thankfully the staged injectors were configured correctly on the base map I was able to find.  The fuel pump is an aeromotive a1000 with -10an supply lines, with some beefy wires running from the solenoid straight to the battery.  Voltage drop is like .3v from the battery to the pump itself.  I've observed pressure drop from like 45psi down to 43psi during logs from a harder pull, but I'm pretty confident the fuel system is up to the task. 

I know that my VE numbers are completely fanciful compared to reality.  At first I was mostly just trying to get the car running well enough to do some shakedown runs.  The guy that installed the AEM left a lot of mistakes for me to clean up, the TPS being wired backwards being the first that comes to mind.  The tune was/is progressing just fine until the injector/ve issue prevented further progress.  Basically just keeping everything safely rich and very conservative on the timing for now.  I appreciate the offer, still trying to find someone semi-local to do the actual dyno tune.  The guy we had lined up moved back to Jamaica.   

 

Since I've got some knowledgeable people here, I'm currently using tps/load as the Y axis of my VE table and rpms as the x axis.  Shooting for 10-12psi of boost ultimately.  Does anyone have some compelling reason to switch to MAP for the y axis or should tps/load work well enough?  So far its seemed to work fine, but as previously stated, a bit of a n00b here. 

 

Edit - I'll post the injector configuration here shortly.  I do the tuning on a laptop and the posting here from my desktop.  I'll pull the files over and put them up shortly.  Afaik the actual injectors themselvs are 850cc all around(a decision made before I had anything to do with the project), as for how they are configured, i'll have to consult the files.  

Def
Offline
Last seen: 1 week 1 day ago
Joined: 09/14/2014 - 10:21
TPS as the y-axis of your VE

TPS as the y-axis of your VE table works well, so does MAP in my experience. Only downsides I've noted of each just has to do with lack of applicable sensor data on the VE table. If TPS is your Y-axis, there's nothing besides a linear pressure interpolation on engine VE, this means that if you start pushing a turbo harder at higher boost, your VE at 100% throttle will change if you don't have exhaust backpressure info feeding to the Infinity. For my Toyota 3S-GTE, at ~12 psi it'll run about 94-95% peak VE, then at 18 psi it'll run closer to 90-91% peak VE due to the higher EBP. I could input this directly on a MAP vs. RPM VE table... but on a MAP vs. RPM VE table, you've got many different throttle positions that can give you say 1 psi, or 3 psi. The flow in the intake manifold can be very different at some of these on-boost part throttle conditions, which can definitely change the engine VE a noticeable amount. 

 

So I've done both, and both can be successful, and have their own drawbacks/limitations based on what other engine operating conditions are assumed. You can work around this some with tuning methodologies, like going for a slightly richer 100% TPS row at high boost, but comes close to optimal at high boost. And a lot of times we're talking about changes of 1-2% fueling, which easily dealt with on feedback loops like the UEGO sensor. 

 

Getting reasonable injector data is the first part. My Injector Dynamic 1000 data in Infinity Tuner lists them as something like 780cc injectors, but the data works out well and was characterized by AEM themselves for how the Infinity treats the injectors.

FreakshowFPV
Offline
Last seen: 3 months 2 days ago
Joined: 04/13/2020 - 13:08
Sorry for the radio silence

Sorry for the radio silence the last couple weeks.  It's been final exam time and my hair was constantly on fire.  

 

After some digging through receipts I found out out the car is running 83 lb/hr (870 cc/min) injectors both primary and secondary.  As far as configuration in infinity tuner, there was somehow no injector selected for primaries and the secondaries were selected as 1000 cc/min.

 

There are both 885cc/min and 850cc/min injectors on the drop down list for the injector Flow wizard.  If I were to select 850cc/min for both the primary and secondary injector flow, I assume I would just drop down my whole VE table by 15% to accomodate the flow rate change?  I'm not sure what effect having no primary injector flow selection makes in correlation to how the AEM has been computing pulse widths I guess? 

I know I'll still have very high VE numbers compared to actual real world figures, but dropping by 15% would give me 30 more points of resolution at the top of the scale. 

 

Thanks again.

 

edit - grammar

FreakshowFPV
Offline
Last seen: 3 months 2 days ago
Joined: 04/13/2020 - 13:08
I went ahead and changed

I went ahead and changed primary and secondary injector flow data to 850cc/m and adjusted my VE table down by 15%.  Went out for a quick logging session and everything definitely changed in a positive direction.  It ended up being a good bit richer than before at every point, even after the 15% downward adjustment.  I'll make some more coarse changes to the map to get it back in the ballpark, hitting mid 9 afr's under boost, which is crazy rich, even for a rotary, but I feel like I should be able to adjust damn near the whole table down by another 10% and then proceed back to adding boost and rpm's incrementally as I was before I maxed my table out.

 

Thanks again for all the help everybody.

FreakshowFPV
Offline
Last seen: 3 months 2 days ago
Joined: 04/13/2020 - 13:08
Here's a quick rolling 1-2. 

Here's a quick rolling 1-2.  Got some wheelspin in 1st as soon as there was boost around 4000rpms.  Hits 6psi at 3500 in third, I'm sure it'll do better once I put a real boost controller on there instead of the mechanical, bleed off, cheapie I'm using now. 

I cant for the life of me figure out how to embed youtube here.

https://youtu.be/4j3obToWmaU

 

 

AEM_NS
Offline
Last seen: 2 hours 7 min ago
Joined: 06/06/2014 - 09:31
All the wizard does with

All the wizard does with injector selection is set the injector flow and offset values (and peak and hold info if using peak and hold injectors). Once you change any of these values, the wizard will show that it's no longer matched to a particular selection but that doesn't mean there isn't injector data. There is - it just doesn't match the wizard. Look at the injector flow and offset tables from your original cal - that'll tell you what "injector size" was being used. And as you've just experienced, going down in injector flow rate for the same VE means that the injector pulse width will increase thereby giving you more fuel. Your previous injector data was quite literally the perfect example of a case where injectors flow less than what you've told the ECU (have 850's but cal is set for 1000's) and the end result is skewed (high) VE values. Once you get accurate injector data, things start to look more normal.

Not that you'd really want to do it this way but you could technically "tune" the injector data to hit a certain VE value. So for instance, say you decide that you want your largest VE value to be 120. If your current max VE value is 150, you'd need to decrease the VE table by 20%. You would then also decease your injector flow table by the same amount. Now this assumes you want your AFRs to stay the same. If you're still too rich, when you decrease the VE table by 20%, you would decease the injector flow table by a smaller amount in order to reduce the final injector pulse width.

It's kind of just a numbers game. You can tweak it however you want but one thing to be very aware of especially on a rotary is injector duty. Make damn sure you're not running out of injector in boost and at high rpm.

FreakshowFPV
Offline
Last seen: 3 months 2 days ago
Joined: 04/13/2020 - 13:08
I am definitely starting to

I am definitely starting to understand the relationship you're describing.  Thankfully 850cc/m x 6 injectors should safely support upwards of 650rwhp in this application and addtionally I am keeping a very watchful eye on injector duty cycles, so all is well there.  At 6psi, given the very conservative ignition timing, and my butt dyno, I really doubt the car is currently making more than 400rwhp and the ultimate goal is well below 650rwhp.  If I'm being honest, I just want the car to be trackable and safe, at which point my father can decide to sell it or pay someone who actually knows what they're doing to properly tune it.