Will we get real input shaping (with accelerometer)?
 
Notifications
Clear all

Will we get real input shaping (with accelerometer)?  

Page 2 / 2
  RSS
Rainer
(@rainer-2)
Reputable Member
RE: Will we get real input shaping (with accelerometer)?

Maybe it's my English.  As you can see, it's not my first language.  This may lead to misunderstandings. 

By no means did I want to blame the speed of the Babulab solely on the reduced masses.  But only that enables higher acceleration and thus speed. 

Which is the disadvantage of every Bettslinger and thus also the MK4.  That's why the Prusa can only handle about 4000mm/s acceleration, the Babulab or other corexy 20000mm/s acceleration or more. 

That's the only reason he's faster.  And an MK4 will never reach that speed.  Input Shaper just makes it look good. 

I understood your previous answer to mean that you used input shaper and were surprised that it didn't get faster.  I wanted to be nice and help, but I'm sure that won't happen again.  If I misunderstood, I apologize. 

But what really annoys me is the constant accusation that everything is meant to make Prusa better. 

If someone is reading along here and I'm technically wrong, please clarify it here.  I have genuine interest in learning.

Posted : 30/08/2023 8:57 pm
Thejiral
(@thejiral)
Noble Member
RE:

I am not necessarily disagreeing with your main point, I want to add however, that the disadvantage of bed slingers scales inversely with bed size. At the size of the Mk4 it is generally said as far as I know, that actually the disadvantage of the bed slinger in terms of moved mass is actually small. It is getting much more unfavourably for the bed slinger with larger sizes. CoreXY designs also have downsides, much longer belts for example. I needed 4.8 m of belt for the A and B drives together, of my 350 mm Voron 2.4.

Also, the Bambulab doesn't have a static print bed. It does move along the Z-axis (like most corexy printers do). Speed requirements are less stringent there but they still matter. So the print object mass does also impact the Bambulab. z-hop does matter for printing speed and artificats quite a bit. 

If you want a truly static bed, you have to look at something like a Voron 2.4. Not saying that is an overall superior design but it is a truly static bed design. 

This post was modified 1 year ago by Thejiral

Mk3s MMU2s, Voron 0.1, Voron 2.4

Posted : 31/08/2023 6:40 am
Rainer liked
Rainer
(@rainer-2)
Reputable Member
RE: Will we get real input shaping (with accelerometer)?

I am not necessarily disagreeing with your main point.....

Which point do you not agree with if I can ask?

I claim that input shaping does not contribute directly to the increase in speed.

But, among other things, a small mass, it allows higher acceleration and thus higher speed. However, because this increases the resonances sharply is rarely used.

Input Shaper eliminates these resonances by making minimal opposing compensation movements of the print head.This works even if you slowly print and want even more beautiful prints.

In summary, only with input shaper can the higher speed be used sensibly. Input Shaper does not contribute to the physical speed.

Could one say that this corresponds to the facts?

I fully agree to what you wrote. But if I had posted it that way I would be able to defend Prusa again.

One could also say that it can be more difficult to create a Corexy Input Shaper profile. Because both axles are hanging together. There it makes much more sense to use an accerolometer because errors in the structure are visible there.

Something is also explained in the linked video from Nero3D. After all, he is a member of the Voron team and should have enough experience with it. And I hardly believe that he is interested in doing Prusa better.

Posted : 31/08/2023 7:54 am
Thejiral
(@thejiral)
Noble Member
RE: Will we get real input shaping (with accelerometer)?

^^
I am not disagreeing with anything you wrote, just wanted to add that size is a big factor when comparing bed slinger vs CoreXY. 

Inputs shaper does wonders to eliminate ringing, indeed it is actually working best at moderate speeds, because there it has practically no downsides. 

Mk3s MMU2s, Voron 0.1, Voron 2.4

Posted : 31/08/2023 12:16 pm
Rainer liked
Rainer
(@rainer-2)
Reputable Member
RE: Will we get real input shaping (with accelerometer)?

Ok thanks, then I misunderstood again sorry.

Posted : 31/08/2023 4:39 pm
CupertinoGeek
(@cupertinogeek)
Active Member
RE: Will we get real input shaping (with accelerometer)?

Happily, I say Good bye…. 

Posted : 09/09/2023 7:43 pm
Page 2 / 2
Share: