RatOS refusing to accept my Z-Offset overwrites

[wet palm]

It's been happening since a while now, but RatOS just will not accept my manual Z-Offset.
I'm also unsure if it it doe to me being in a print, but the "SAVE CONFIG"
It is stupidly annoying, since it will hammer down layers at up to 0.2mm too low, and having to manually increase it each time I print is just something that should not be needed for a machine like this..

[dusk oracle]

Are you on RatOS RC4? If not all of the beacon / z-offset has been redone since RC3. With the new setup I have found it so good I have not needed to do any manual z-offset.

[wet palm]

I am still on RC3, but it seems to have somewhat fixed itself? I'll try later to update to RC4 and see if this issue persists

[dusk oracle]

RC4 is such a major upgrade. Just follow the upgrade document as a lot of the beacon configs are not compatible.

[wet palm]

Will definitely try

[wet palm]

Now it decided that it's incredibly funny to just set my currents to incomprehensibly stupid values whenever I restart my printer [the rest is like 0.3, 0.1, 20, you name it]

[dusk oracle]

Something is messed up with the install. Find the time to upgrade to RC4

[wet palm]

I did, and it does not show up

When Ii hit "Check for updates", I do not get anything for RatOS Core

[dusk oracle]

RC4 is major upgrade. It should be called 2.2 or even 3.0 so it requires a reimaging. https://os.ratrig.com/docs/upgrading_rc3

[wet palm]

Lovely

[dusk oracle]

Backup your old configurations with the debug under configurator and if you have a spare sd card flash the new OS on that so you have the old one in case you missed anything.

Yah it takes some time but it made my vc4.0 feel like a new printer. It just works between the new beacon code and the adaptive heat soak. No baby stepping z offset and no need to babysit it either. Just send a job and it does its thing

[wet palm]

Sounds great lol

Hm.. Does it need to be an SD Card? I got some rather snappy USB's, if that's an option

[since I essentially had 3 MicroSD in my whole life, I just never use them]

[dusk oracle]

I think it works but I have not tested. Do you have a spare usb port on the pi?

[wet palm]

Yeah I do

Though I think with 2.1 there where some issues about USB boot.. Eh, worst case it explodes

-# [again]

[wet palm]

Seems fine so far, but for some mysterious reason Klipper refuses to start currently, will see what that is..

[wet palm]

Error evaluating 'gcode_macro START_PRINT:gcode': gcode.CommandError: Missing START_PRINT parameter. EXTRUDER_OTHER_LAYER_TEMP parameter not found.

This is straight from Orca into RC4

[dusk oracle]

Check that all of the gcode settings in orca are right. https://os.ratrig.com/docs/slicers#orca-slicer

the gcode lines are long in the doc above. Use the copy button or make sure you get every bit of it. Sounds like something is missing in your Orca setup

[wet palm]

Yeah I found that

That seems to have fixed it, I will provide an update once my Beacon calibration is done

Tho one thing I noticed is the z-tilt probing like this (blue) instead of how my Z motors are placed (red)

[dusk oracle]

its a change with RC4, the reason being is the largest thermal defection in the gantry si in the middle so they are not using the middle anymore to take measurements for z-tilt. The idea is they can still figure out the adjustments needed but avoid the area of the gantry that could be problematic.

[wet palm]

Ah, I was figuring something similar (since I found the idea of it randomly rotating the probing by 90° pretty odd), but good to know that this is intentional and actually beneficial

[wet palm]

Also, how long does the Beacon Calibrate usually take? It's been ~60 minutes and it's still probing.

-# not complaining, just wondering

[dusk oracle]

hour or so for a 500. It does like 3-4 probes for every spot it probes. It only does this when you make the mesh

[wet palm]

300mm HY, and it's been doing quite a lot of probing lol

It's also not exceeding tolerances, it's just very extensive it seems

[dusk oracle]

it is. part of that is the true zero correction they added. https://os.ratrig.com/docs/configuration/beacon_true_zero_correction

[wet palm]

I see

Love they're adding this many features, RC3 felt very buggy and unpolished, but RC4 feels very smooth this far

-# [now if only they're support was actually good and not rivaling BL's live chat......]

Oh, also; I figure this is calibrated automatically by what I've read? Though, do I need to worry / change anything?

[dusk oracle]

I do not know. @compact atlas any thoughts on this showing "quite low". I do not think I have seen that

[compact atlas]

What hotend are you using? What modifications does your printer have compared to stock?

[wet palm]

UHF+ and only frame braces

[compact atlas]

0.035 is indeed suprisingly low. For Rapido 2+ UHF we normally see values between 0.06 and 0.08. Are you using a V6 or a volcano nozzle, and what kind of nozzle is it?

[compact atlas]

A small correction: when creating a compensation mesh, the first point is probed 3+ times, but all other points are probed only once unless the single probe produces a suspicious value, in which case the point is then muilti-probed as a robust fallback. And this is not part of the ture zero correction feature, it's the cotemporal compesation mesh generation feature. True zero correction is the 6 extra probes when doing the true zero probe just before printing starts.

[wet palm]

I am using a Brass V6

[wet palm]

What ever it is, it is working great 😅

I think I had one small issue with the first layer so far, a couple mm wide spot on a thin area so nothing drastic (and I assume by plate was just greasy), so this beats RC3 by miles

[wet palm]

Ok so something really interesting just happened with the dynamic heat soak; it all went well up to ~95%, then it started to go up reaaaaally slow, then it just stopped at 96% and about 10 seconds later it finished

[compact atlas]

The progress indication is not linear beacuse it's impossible to completely predict ahead of time when the soak will complete. Above 95% it enters the "waiting for last bit" phase of progress, and that can complete rapidly. The behaviour you descirbed is normal and expected.

[wet palm]

Ah, didn't know that

I thought it was a pre-determened time based on the first layer length / complexity

[compact atlas]

Indeed, it dynamically monitors what's actually going on by continuously sampling beacon proximity readings.

[wet palm]

Ohhhh, that's really clever! Happy to learn all about this, it really gives you a lot more understanding & patience if things seem weird knowing whats actually happening in silence

[compact atlas]

It's actually quite complex. There's a lot of maths and signal processing involved.

[wet palm]

I can imagine

It does a fabulous job tho!

[compact atlas]

Thanks 🙂 It took a lot of R&D to develop and tune the algorithm

[wet palm]

although I did admittedly get some significant warping issues even with plain old PLA on RC4.. But this might be user error

[compact atlas]

Warping is a different thing. Even with great soaking, it will depend on bed adhesion and indeed part geometry.

[wet palm]

It shows, especially that real human effort went into it

[wet palm]

Yeah, but I never had this on RC3 and prior, it has only started on RC4, tho I suppose it is due to the first layer being actually set correctly for once where now my plates 12-month cleaning cycle is starting to come a little short ^w^

I will try to properly clean it (dish soap + iso bath) and will let you know if I get anymore warping, if so I will just utilize more of my good old 3DLac lol

Mission accomplished, now onto testing