Does AMS support Direct Drive Wheels?

[Immutef]

Will it support Direct Drive Wheels like the OSW out of the box?

 

[Spinelli]

I noticed some people are still using Damping, Filter, and Inertia from MMos when they should be using it from Granity instead.


All downloads to setup/update IONI OSW (not sure about Argon). Updated to newest versions as of Dec. 7, 2016:

OSW (except MMos)

• IONI Firmware v1.5.4
• Granity v1.10.1
• STM32 ST_LINK Utility v4.0.0
• USB to RS485 Drivers v2.12.24

- Click the downwards pointing arrow at the top-right of the screen to download


MMos

• MMos FFB Firmware v0.99.2
• MMos FFB Tool v0.91

- You shouldn't need to download anything else listed in that link - just the 2 files (MMos Firmware and MMos Tool). If there's a newer version listed compared to what I wrote then download the newest version instead.


NOTE: You may not need to install #3 or #4, it depends if Windows automatically installs and if you use the RS485 thingy. The RS485 Driver is to make that little PCB thing with the phone-jack on one end and USB on the other end work. It plugs into your PC (USB) to allow Granity to communicate with the wheel. I had to install this myself on my system (Ollie IONI OSW). I also installed STM32 just in case the Windows version is older.

NOTE: Apart from using ST Link to update firmware, check in the ST Link auto-update section to see if it installs any updates.

NOTE: MAKE SURE TO setup your particular wheel's settings in Granity or load your motor's DRC file! I have not included this!

 

[Spinelli]

In-case anyone's interested, these are my current settings for the F309 and my Large Mige.

Granity:
Goals
MUL: 50
DIV: 50
Machine
MCC: 15.375 A
MCC: 20.500 A
Tuning
TBW: 1500 Hz (experiment in combination with MMos Filter and RealFeelPlugin.ini SmoothingLevel)
NFF: 7.0 Hz
NFA: -3.3 dB
NFQ: 1.0
TED: 0.60 %
TEF: 0.70 %
TEI: 2.50 %

MMos
Filter: 5 (experiment in combination with MMos Filter and RealFeelPlugin.ini SmoothingLevel)
Min Force 0 %
Max Force 80%
All Effects: off

RealFeelPlugin.ini
MaxForceAtSteeringRack=-4500.000000
FrontGripEffect=0.500000 (1.0 is way too much but 0.5 possibly not enough, I'm going to try 0.65)
SmoothingLevel=4 (experiment in combination with MMos Filter and RealFeelPlugin.ini SmoothingLevel)

In-game:
Minimum FFB: 0%
FFB Strength 80%


For Small Mige:
- You may have to use different TED, TEF, and TEI settings but it doesn't hurt to try the above.
- NFF: 2.5 Hz
- NFA: -2.3 dB
- NFQ: 0.5
- If you want to run the same torque as me then set your small Mige to run 22.9 Nm so you would set your MMC to 14.723 A (MCC to anything lower). You won't really be running 22.9 Nm because of the following three reasons:
A. the MaxForceAtSteeringRack has been raised by 50%
B. in-game FFB lowered to 80%
C. MMos Max Force lowered to 80%.
Although it's probably not exact, I think about it the following way for simplicity's sake: 22.9 Nm * 0.80 * 0.80 = 14.66 Nm. You really won't even be hitting 14.66 Nm since the F309 doesn't really reach 4500 (the MaxForceAtSteeringRack #).

If you don't feel comfortable setting your OSW to 14.723 A (due to using a Small Mige or some other part which may not be made to handle 14.723 A) then set both MMos and In-game FFB higher and then lower your MMC to compensate in-order to end-up with the same max-torque of 14.66. I made some examples below:

• MMC @ 14.723 A, MMos @ 80%, In-Game @ 80% = 14.66 Nm (not recommended)
• MMC @ 13.041 A, MMos @ 85%, In-Game @ 85% = 14.66 Nm (probably OK too)
• MMC @ 12.740 A, MMos @ 86%, In-Game @ 86% = 14.66 Nm (recommended)
• MMC @ 11.632 A, MMos @ 90%, In-Game @ 90% = 14.66 Nm

I can't remember the official max limits of the Small Mige and IONI Pro non-HC but I'm pretty sure option #2 (13.041 A) is safe. The Small Mige is rated for 20 Nm which would be MMC @ 12.855 A so 12.740 should definitely be safe. Many people run 13-14 A with the Small Mige just fine but 12.740 A should definitely be safe from the research I did.



Additional Info (not really important)
Regarding MaxForceAtSteeringRack
Niels said that each car's MaxForceAtSteeringRack has been reduced by around 20-30% from it's ideal point (this is because most people don't have high powered wheels). So, in order to raise something which has been lowered by 30% back to it's original value, you to raise it by around 43%, so I just use 50% to make it easier and give the car a bit more leeway in-case it experiences higher forces (different tracks, setups, etc.) So, the F309's default MaxForceAtSteeringRack is 3000 so, 50% more of 3000 = 4500.

Why I set higher MMC and compensate with lower MMos and In-Game FFB
Setting MMos and In-Game FFB lower than 100% seem to offer a more progressive build-up in Automobilista (and anything based on the ISI engine, eg.RF2). It may have something to do with the physics engine and/or tyre model but FFB forces often seem to increase too quickly/much relative to how little actually tyre-lock has been input.

 

[Deleted member 197115]

Hi Spinelli,
Is there any practical difference between 50/50 MUL/DIV and 100/100 (default), both result in 100%, wondering if it matters at all.

 

[Spinelli]

Hi Andrew, long time no talk!

I haven't tested 50/50 VS 100/100 in-game but a few months back 100/100, 50/50, and 1/1 all gave me different results on the desktop. You could hear the very faint motor vibrations more or less while the wheel was idling at centre depending which setting you put it to. I don't know about in-game, but something definitely changes between different numbers even if they still equal the same number.

I wish I could give you a better answer.

MCC also makes a difference. With my Amperage (MMC) @ 20.500, I tried MCC @ 15.375, 10.250, and 5.125 and they definitely make a difference. Very slight but still noticeable.

 

[Deleted member 197115]

Thanks.
One more, you seem to try matching MMOs max force with in game gain.
Any practical reason for that?
I am on Small Mige MCC 12.6A and just use 100% MMos and around 35-40% in game (AC), to leave some healthy headroom for in game signal clipping.
So essentially maxing out amperage and MMos gain and just adjusting force to my physical abilities/taste in game.
I remember doing all crazy stuff in the past, mixing and matching amperage, MMOs and in game gain, just to come to conclusion that as far as feedback goes there is no noticeable difference as long as final NM is the same.
The latter method relieves from hassle of tuning individual car gains to eliminate signal clipping, just fire and forget. Ended up removing my essential before FFBClip app.

 

[Spinelli]

There's definitely a difference between lowering FFB from the amperage and lowering from MMos and in-game. I find lowering from MMos and in-game make the FFB build-up or (ie. linearity) seem less aggressive, more gradual, while putting MMos and in-game to 100% and lowering through amperage has the wheel feeling like it builds up force too quickly and generally seems to behave slightly more aggressive/violent (in unrealistic ways).

I tested keeping game @ 100% and lowering MMos, and keeping MMos @ 100% and lowering in-game. I couldn't figure out which one I liked better so I decided to do a mix of both.

 

[Deleted member 197115]

You don't want 100% in game, leave some headroom for overall signal clipping, sometimes you have no indication of one but when damping get added (gyro in AC), it's there.
Tried 50/50 vs 100/100 MUV/DIV. Testing in MMos with desktop effects there is slightly smoother response with 100/100, but the difference is so small that it can be just in my head.

 

[Spinelli]

I don't get clipping in AM, RF2, and IR because I can use the RF2 and IR torque setting, and the AM MaxForceAtSteeringRack setting, to define clipping points. But, I agree, lower in-game. However, lowering in-game with AM I don't believe reduces clipping because clipping goes by whatever your MaxForceAtSteeringRack setting is set to and then lowering in-game just lowers those forces across the board but the clipping point is still the same (since MaxForceAtSteeringRack hasn't changed). I'm guessing though.

In Granity, test 1/1 then 5/5 then 50/50 and you should see a big difference.

 

[Deleted member 197115]

In Granity, test 1/1 then 5/5 then 50/50 and you should see a big difference.

I know you will feel the difference, my point was that 50/50 is not necessarily better than 100.
You might have seen this already but this is the best explanation of MUL/DIV I've found
http://granitedevices.com/wiki/Signal_path_of_motor_drive

Take a note on this.

Input multiplier. Purpose of this is to increase resolution of input setpoint to allow more fine grained velocity & acceleration control in trajectory planner.

Which explains why moving up in numbers make it feel smoother even that final division result is the same.

 

[Spinelli]

Thanks for that link. Ya, I'm going to experiment with different values. It also says that the divider can be used to slow down acceleration or velocity (which is often unrealistic and too much with DD servo wheels). So maybe instead of only lowering FFB strength feom in-game and MMos, I'll alsolower from MUL/DIV.

More experimenting needed