PRO R Torque Managment Tuning

[Brycter]

Hello Guys,

We know this is taboo in the powersports industry to show weakness and a hole in our capabilities of tuning a vehicle.

Nobody likes to say we need help. Nobody likes to be called a noob or feeling like one. Especially when we have been supplying this world with products and services for many many years.

We consider ourselves lucky to be surrounded by so many strong capable manufacturers.

Now that is out of the way. We have spoken to and received alot of phone calls of customers over the last month.

Each having a different brand to turbo system or super charger. Each and everyone has some what the same problem. Some worse than others. Some are lucky running flawless. So they say.

The inconsistency is all about the closed loop torque settings and torque management with these turbo systems and super chargers.

Closed loop Torque management is something new to all of us in the powersports world.

Up until now we have been able to control tuning through the Ve table and mass airflow throttle body calculations. Then add in some pid senerio boost control and timing. Then boom it works. Now there is a new world out there and we don?t think we have all the information needed to do this right.

We want to get a discussion going on your solutions, fixes, over rides and thoughts. That way we can figure out the best solution. Hear you success and failures. Also we are hoping HP tuners will read this and give us more support and unlocked capabilities inside the firmware. Sooner over later.

Items like special features added, Live tuning added, better table explanations, would be so helpful.

Since at the end of the day we are all working with the same product that is given and need more information.

Hp tuners is a platform that grows with us a giving feed back and working to let them know what we need. Their success is hinged on our success. Exspecally being new to the powersports market.

Here is a problem that we are seeing: We are hitting a torque limit that is shutting off the injectors. Just like we turn the key off and back on again. Or if the conditions change the tune goes slightly out of calibration (ie wet sand vs dry sand or big truck tires vs stock tires) these machines are stupid fast and are hitting so hard they will be the top dogs for a long time.

We have changed every number combo we can think of for 21 straight days. We have found no solution, or any methodology/flow math for the torque management tables. Who has this knowledge that will share it and teach us this new information?

I am ready to learn and have taken my adhd meds. I have my note book ready, my thinking cap on, and just took off the band aid.

This is not a bash session. Or mine is better than yours. We can work together and we can grow as an industry. This is our livelihoods and we want it around for a long time. Many hands make light work if we can let it happen.

Thanks.

[jaz|TUNING]

I haven't messed with a Pro R yet, but have had plenty of tuning in the automotive side with torque based tuning. If there is a map available to look at, I'd love to take a look at it.

[VFORCEJOHN]

it'll get easier. this is so hard to explain in a way to make sense to you yet.

I started doing this 7 years ago. reading and writing by sending simple commands and replys by code to a connected ecu. then getting the binary file and deconstructing it to what was understandable. and it all goes back to the days of the basic Tandy computer programming from the 80s.

Tq management is nothing more than saying (your engine at this RPM, MAP, TPS, MAF requires this much fuel) this much fuel x this much air = projected torque (calculated from what i call the airflow table)

if your VE table (how efficient your engine is at making the HP)

so by math, we know it takes X amount of fuel and X amount of air to make 1 hp. if your making 100% efficiency. then the tq table translation should equal your actual output.

Polaris uses this strategy to control CVT SHIFT RPM by capping the LOAD into your throttle axis values which appear directly before each map. exceed that requested load (turbo models are 197 max load at 100 throttle which equates to around 8500 rpm), and itll hold you there or pull back some to keep RPM down.

the MG1 is about the EASIEST ecu to figure out. FAR easier than a Mitsubishi ECU!!!!! Thats what I learned on. now I know why HP don't support them......They throw 2 of these scenarios at each cylinder pending TPS or rpm SD vs Alpha crossover. this is TOUGH to manipulate as they ratio off all variables vs target them


so. set your afr TABLES. Datalog the STFT and LTFT trims. apply them to the Air correction tables to achieve a correct requested AFR and then smooth the TQ tables to control power output.

works every time.

[4SeasonsPower]

I can also confirm that there are issues with tuning forced induction ProR's using HPTuners. I'm having issues and I can confirm first hand that several of the popular companies with kits on the market are struggling with the same issue. My specific Pro R appears to be having the same issue with load/torque mangement and causes the throttle to act erratically, drop injectors and pull timing. I have attached several screenshots of a few tables if any of you guy want to offer some input. This is the stock file generated by HPTuners.

DesredTorque.jpg
ExtendedMonitor.jpg
FunctionMonitor.jpg
OptimumTorque.jpg
TorqueToLoad.jpg
TorqueLimit.jpg
VE.jpg
TB.jpg

[Brycter]

Hp tuners is working on the fix. We have sent over multiple data logs. They have added at least 60 different item to the beta software. We are going to get this done. Thanks to all that are helping.

[VFORCEJOHN]

POLARIS 1 CORRECTED.jpg

Im showing 67 maps that correlate to this in a 16x16 map.

What I'm also seeing is HP tuners correction value is used (which may not be correct??) original ECU values are about 25% higher than the corrected HP tuning factors.

Hp is using the math as follows to calculate the tables

0.073749*16 bit decimal ( Big endian values) =output value shown.

the conversion isnt making much sense to me? see above map using HP correction

Below I will load a torque map at what I believe the tq maps may be corrected to.

[VFORCEJOHN]

polaris 2 VFJ correction.jpg


THis is the values displayed using 0.1xdecimal=map output values

THis may not be correct either as when would the values shown require this engine being N/A to operate at double the load values? doesnt logically fit unless its a turbo model. which this file isnt.

[VFORCEJOHN]

vfj polaris correction 3.jpg

Using the values 0.05x DEC VALUE =output shown

I'l getting a much more logical map value which would correlate to a usable translation (pending HP PID conversion value shown for the scanner theyre using)

it may not be that the maps arent there.

its that the defined vaues are incorrect calculations




I wish HP tuners would allow a true "native value" to show the uncorrected ecu value or the HEXIDECIMAL values of the mapping shown when using ADVANCED parameters. THen use the SCANNER in NATIVE MODE or HEXIDECIMAL LOGGING mode to see actual values.

this would eliminate any of the improper correction factors and make easier progress for definitions when the values shown dont correlate to whats actually shown.

this would MAJORLY help issues like this. However this would be extremely complicated for the standard user who doesnt understand Hex/DEC factoring.

GOOD LUCK!

[VFORCEJOHN]

polaris all uncorrected.jpg


Here is the Fully uncorrected map with raw decimal values displayed.

this shows how much advanced math HP must apply to each map to make it user froendly for the end user.

However, they may not calculate them properly (which is the biggest challenge they will always face)

Not easy stuff by any means!!!!

[4SeasonsPower]

Thanks a ton for the input John.

Since my last post HP Tuners has released some updated parameters in the latest BETA release. I'm certainly not the authority on the topic, but I feel like the root cause of the problem is the ECU's torque model. Certainly driver demand tables play a part in the entire picture, but I wonder if we are all struggling to find the correct balance of throttle body airflow, MBT timing, optimal torque tables, etc. It seems to me that because we are adding forced induction to an engine that has an ECU with a torque model originally intended for N/A, we are now running into an a problem where it is trying to control this torque with inaccurate data. This what is causing us to see erratic throttle angles, injectors being shut off, and ignition timing being pulled, because the ECU is trying to control an engine which is making drastically different levels of torque than it thinks are present.

I'm currently working on my own tune and have noticed a dramatic improvement solely from a refined MBT map. As I continue to refine these timing and airflow tables I would expect the torque model to become more and more accurate and therefore the ECU can do its job properly.

[VFORCEJOHN]

can you link your PRO R oem file here so I can see what parameters they have accessed to? non modded file for a pro r 4 cyl I dont have any oem stock HPT pro r files to see and compare to.
\


on the turbo models (different ecu) they have 3x 1 axis maps for throttle and how tq is applied. one must be larger than the other 2 or it can cause a throttle error. this is gear dependent as well. There are also switchable maps being used as well.

scaling those properly is crucial. this unit is using 3d table vs 1 axis 2d table

also I believe the AXIS VALUES DISPLAYED have a crucial gain on how the setup is defined as well.

HPT seems to select the highest value and Percentage calculate. so as on my uncorrected map we see the raw values of 4096 or something like that at max throttle. this MAY very well be voltage or map LOAD Some companied use throttle values usually end at 819 (meaning 81.9% relative to idle opening which is usually equated to be 100% throttle on the axis) however without seeing the TRUE native value from the ecu decimal its too hard to tell what's what and what's corrected. I believe that may be MAP LOAD as calculated by Decimal/.333 INHG showing values from a single bar senspr from around 7 to 30 inHg. so its possib;e those axis values may need to be searched for and all changd to work with boosted apps.

I have to do this for the KRX 1000 and it is extensive to do. thats why boosted apps are not functional on that even with a map sensor and rescale as defined by Dynojet. they cover 1 map sensor adjustment of literally DOZENS of map sensor scaling axis values. which then 1 is changed and the rest remain stock goofing up the whole show with fuel timing throttle cut backs.

to overcome this. youd have to search for ALL parameters and rescale EACH. then the corresponding map immediately following it.


heres a vague idea of what I mean. tough to explain. (viewed as hex to decimal raw values) parenthesis aren't there but I just separate for explanation.

16 16 (2000 4000 5000 6000 7000 ......16 parameters for the ROWS) ( 0 21 103 205 615 1024 ... (16 parameters for columns) (next is the map data strung out in rows 256 map cell values)

[4SeasonsPower]

John,

Thanks for all your input, it makes alot of sense. I have attached a link to download my stock tune file below.

https://www.dropbox.com/scl/fi/ahucm...ydmffk5yq&dl=0

[jpatbx89]

did you ever get anywehre with this? so far the FIX seems to just not use HP tuners, and go with a standalone motec or EMU

[TJB94]

Can anyone help tune my Pro R? I keep throwing code P2177 system too lean off idle bank 1. It has a Pandemyk performance tune on it now, stage 2+ cams and Packard exhaust. I have the handheld device. Willing to pay if we can figure it out, I miss riding, I also live in the VI so no dealers out here.

[jacep]

Can anyone help tune my Pro R? I keep throwing code P2177 system too lean off idle bank 1. It has a Pandemyk performance tune on it now, stage 2+ cams and Packard exhaust. I have the handheld device. Willing to pay if we can figure it out, I miss riding, I also live in the VI so no dealers out here.

your handheld device are your referring to, does it work with hp tuners?

[Eric@HPTuners]

polaris all uncorrected.jpg


Here is the Fully uncorrected map with raw decimal values displayed.

this shows how much advanced math HP must apply to each map to make it user froendly for the end user.

However, they may not calculate them properly (which is the biggest challenge they will always face)

Not easy stuff by any means!!!!

100% of our Pro R maps are scaled correct.

[Eric@HPTuners]

I spent the good majority of time yesterday working with a tuner on this issue and we resolved it 100% with the parameters that are in the software.

The gist of the issue is that being a fully torque based control system, the torque and torque inverse model needs to be pretty accurate, and that in itself is pretty challenging. However, there are some tables you can fudge a bit to fix this issue up.

Being torque based means there are many checks on the system. If you make more torque than expected the ECU will go at lengths to get things back under control, including shutting off injectors entirely.

At any rate... You need to dial in the Torque to Load and Load to Torque tables, and then bump up the tolerance under Function Monitoring and Extended Monitoring. Specifically table IDs 12524 and 14236. Table 13164 also needs to be adjusted to reflects something closer to reality.

If you have done all of the above, and still have issues. Shoot me a PM and I will get to the bottom of it.