Expose slope_control field on TMC2240.

[liampwll]

The chip has a default value of 0, corresponding to 100V/µs. The default is overridden with 2 here, corresponding to 400V/µs to approximately match the TMC2209. This should also solve some overheating issues as it lowers the power dissipation at a 50kHz chopper frequency by around 1W.

This is currently untested.

[Sineos]

Would it make sense to expose DRV_CONF register bits 1..0 as well. This seems to be a nice way to tune the driver closer to the actual motor current (see data sheet, page 47).
I'm currently using 01 instead of the default 11 on my z-axis and at least have the feeling that it somewhat reduced z-banding (could be wishful thinking, but at least no negative effect observed).

[liampwll]

It probably makes more sense for Klipper to do that internally based off the current specified by the user, but it wouldn't hurt to have an override that the user can specify.

[nefelim4ag]

Slope control defines how fast the MOSFETs gates will be charged/discharged.
*Didn't know that, so it is for someone who will also struggle with that.

For now, I can confirm the information about power dissipation:

The green line is fan PWM. 0->2 looks like a lot of difference (83C->69C).
*With fan forced to 100%, there will be 83C->62C

[nefelim4ag]

Would it make sense to expose DRV_CONF register bits 1..0 as well. This seems to be a nice way to tune the driver closer to the actual motor current (see data sheet, page 47). I'm currently using 01 instead of the default 11 on my z-axis and at least have the feeling that it somewhat reduced z-banding (could be wishful thinking, but at least no negative effect observed).

2A. next one 1A

It is calculated and set automatically, based on your desired current.

[wight554]

Have to check regressions, but did speed and resonance tests so far on mellow drivers and results are amazing.
I've got temp drop from from 100-110C to 70-75C on 1.414A.
Was able to raise current to 2.5A motors RMS (~1.77A) with temp at 85-90C level vs overtemp error before the change.

Have to mention that I use LH Stinger Ebox with 2 8025 fans blowing on all electronics inside and TMC Autotune enabled (it has different change to drop temps on 2240, by dropping freqs iirc, so idk if these play well together)

[liampwll]

The lower frequency in the TMC Autotune module shouldn't conflict with this, but I suspect it should also no longer be required with this change for most users.

…

On Mon, 30 Dec 2024, 23:38 Volodymyr Zhdanov, ***@***.***> wrote: Have to check regressions, but did speed and resonance tests so far on mellow drivers and results are amazing. I've got temp drop from from 100-110C to 70-75C on 1.414A. Was able to raise current to 2.5A motors RMS (~1.77A) with temp at 85-90C level vs overtemp error before the change. Have to mention that I use LH Stinger Ebox with 2 8025 fans blowing on all electronics inside and TMC Autotune, it has different change to drop temps on 2240, by dropping freqs iirc, so idk if these play well together — Reply to this email directly, view it on GitHub <#6771 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AANZQUNCNAOCHDG6K5IQGID2IFSIHAVCNFSM6AAAAABUJU56HKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDKNRVGYZTMMBXGE> . You are receiving this because you authored the thread.Message ID: ***@***.***>

[KevinOConnor]

Thanks for working on this. We can certainly add a new driver_SLOPE_CONTROL config option.

However, I'd be leery about setting a default that differs from the Trinamic default, unless we can find some strongly worded statement from Trinamic that indicates the setting should be changed. That is, I'm unsure we'll be able to select a better default that Trinamic can.

If we can find a strong indication from Trinamic that the default should be different, then we'll need to document the change in docs/Config_Changes.md .

-Kevin

[liampwll]

The only things that are impacted by this setting outside of the switching losses should be EMI and power supply ripple. As noted earlier, setting this to 400V/µs matches the typical value of the TMC2209 that is widely used on stepsticks on DIY boards (see attached image), so I doubt that will cause any issues for users of DIY boards. It might be an issue for any non-DIY vendors that have gone through proper EMI testing or are pushing what the caps on their boards can handle, but if they're just blindly applying updates then that's their own problem.

I think the fact that no one has made a PR to expose this setting before now makes a good case for making this the default as it shows that most users will not read the datasheet to investigate why their drivers are overheating and will instead reduce the current or reduce the chopper frequency, which can both decrease performance.