Input shaper concerns  

RE: Input shaper concerns

All valid concerns, and all reasons why I think an on board accelerometer is important. It should have been included as default as it's not very expensive, although I do hope we at least get the option to purchase one and fit as an upgrade considering the port is available......

RE:

Yes I think it should have come with accelerometers too.

But then there is another problem. There is only one port on the board and a bedslinger needs 2 accelerometers...

You would have to manually switch cables before every calibration. 

I think that's the main reason it comes without it.  Not the price. 

RE: Input shaper concerns

I wonder if an accelerometer couldn't be plugged into one of the USB ports as well (obviously with some firmware updates).  At least in the situation where you'd have it plugged in only for the calibration.  Would not be a pretty solution though that's for sure.

RE: Input shaper concerns

I think it is really unfortunate that one did not integrate the ability to handle to accelerometers, one way or another. As long as they do implement an input shaper where you can set the critical parameter by hand there is always the option to do manual calibration by calibration print. This is certainly more bothersome but yields comparable results to accelerometer based calibration, from what I have heard. 

Even if the implementation doesn't allow for that (which would be really strange). Such a calibration print could be used to determine the quality of the input shaper as it is. 

RE: Input shaper concerns

Manual input shaping calibration by trial and error ?.  If you ask me, bothersome is a huge understatement. You probably could spent days printing test after test trying to achieve what an accelerometer does in less than ten minutes with better accuracy, 

RE: Input shaper concerns

No, not trial and error. There are specific calibration prints for this very purpose, where input shaper settings are gradually changed during the print. With a calipher a single calibration print is enough to take the necessary measures and calculate the optimal settings. Granted this is not quite as elegant as with accelerometers but apparently is pretty comparable in the quality of the outcome.

Here is the manual for Klipper: https://github.com/Klipper3d/klipper/blob/master/docs/Resonance_Compensation.md
Also the calibration model is linked there. The whole manual is however designed for Klipper. If Prusa enables a proper input shaping system I don't see a reason why an equivalent calibration model and procedure should not be possible on an Mk4.

Of course, an inbuilt accelerometer is way superior and more efficient but if you have to open the electronics and manually install accelerometers for the calibration, than in fact, the calibration model could be just as easy to do or even easier. Given that you really need to do that only once or very rarely, as long as you don't modify the printer or change its location, it isn't such a big deal as long as the result is a good one.

RE: Input shaper concerns

I hope that Prusa will grant access to the values. Looks very promising.