RE: Bad Vibrations
It's simply because otherwise you would have had to develop a completely new loveboard (the board on the extruder). So you can use the same one from the Mk4 and the design is also the same. The Accelerometer Kit costs almost nothing and can be connected in just a few minutes. In addition, it is not needed in 99% of cases. We don't even know yet whether the vibration problem is caused by it or whether it has another cause. So let's wait and see.
It is simply meant that the picture shows the effect of phase stepping more clearly, because bad motors were used especially for the picture to make it clearer. With the motors that are installed in the printers, however, the difference will be less clear, because motors like those used for the comparison in the picture are not installed in the devices, as 100% of the motors are tested beforehand and such motors are sorted out. This is only to prevent you from wondering at home why the difference is smaller for you.
I think you misunderstood Jo's post.
Can you help me understand it correctly then? He shared a very striking example of the benefits of phase stepping, but with the disclaimer that it would be nowhere as impactful in the units they were now building:
Keep in mind that on the production XLs the difference won't be that visible if at all - we test 100% of the motors going out since the start of the production to be pretty much perfect. But the results from this test are just too cool not to share!
I understand that to say that the "100% tested" motors don't benefit from phase stepping as much, if at all, since motors with excessive tolerances will not be used. How would ou interpret it?
RE: Bad Vibrations
It's simply because otherwise you would have had to develop a completely new loveboard (the board on the extruder). So you can use the same one from the Mk4 and the design is also the same. The Accelerometer Kit costs almost nothing and can be connected in just a few minutes. In addition, it is not needed in 99% of cases. We don't even know yet whether the vibration problem is caused by it or whether it has another cause. So let's wait and see.
But the standard xBuddy board has an "Accelerometer" port, J29 (an SPI port). Isn't that where the accelerometer is connected?
I agree about the "wait and see"; hope we will see some updates soon. Does Prusa's tech support or R&D team monitor this forum, or even chime in occasionally? I was hoping to see them here, in particular for a newly released machine.
Within just a few days of shipping out the first units, we have learned about
- a phase stepping feature which was known to be necessary in the XL but was omitted in the Core One,
- overhangs which print worse than in the MK4S, apparently because despite all the fanfare the build chamber is not kept at the optimum temperature,
- loose belts in a newly shipped unit,
- units with nasty rattling and x-axis oscillation in the prints,
- units without firmware,
- misrouted cables in the toolhead,
- a unit with a non-working NFC/WiFi/filament sensor.
I am willing to cut Prusa some slack for a new product, but they need to fix this stuff quickly. And it would not get them extra credits for "look what a great company", but would only be catching up with what they should have gotten right in the first place.
RE: Bad Vibrations
- a phase stepping feature which was known to be necessary in the XL but was omitted in the Core One,
- overhangs which print worse than in the MK4S, apparently because despite all the fanfare the build chamber is not kept at the optimum temperature,
- loose belts in a newly shipped unit,
- units with nasty rattling and x-axis oscillation in the prints,
- units without firmware,
- misrouted cables in the toolhead,
- a unit with a non-working NFC/WiFi/filament sensor.
- We don't have information on if this is a phase stepping issue or if it can be fixed with firmware
- I've only seen a single instance of someone complaining about overhangs, many others haven't had the issue and show great overhangs
- Belts can come loose in shipping and can be tightened
- The nasty rattling can hopefully be fixed with the fix for the first point and is probably the same exact thing
- The firmware comes on the USB and this is explicitly stated on the instructions
- A misrouted cable happens. Prusa is sending the user a new toolhead and the issues is being fixed
- I haven't seen the non-working sensor so I can't comment on that one
Overall all of these issues seem small and very fixable. Prusa has proven themselves time and again to fix issues like these, and they are all things that have happened with other companies. Most of these issues are being pointed out in a very nit-picky way and are simple issues that happen to have been missed in QA. This happens all the time with products worldwide. The devices are made by humans, checked by humans and humans are not perfect.
RE: Bad Vibrations
Yes thats the Accl Port. You can connect the accl direct with this port or you can connect a bridge cable ,which cames with the accl set, and connect it to a plug ont the wifi board which you can reach every time from the outside.
It's simply because otherwise you would have had to develop a completely new loveboard (the board on the extruder). So you can use the same one from the Mk4 and the design is also the same. The Accelerometer Kit costs almost nothing and can be connected in just a few minutes. In addition, it is not needed in 99% of cases. We don't even know yet whether the vibration problem is caused by it or whether it has another cause. So let's wait and see.
But the standard xBuddy board has an "Accelerometer" port, J29 (an SPI port). Isn't that where the accelerometer is connected?
I agree about the "wait and see"; hope we will see some updates soon. Does Prusa's tech support or R&D team monitor this forum, or even chime in occasionally? I was hoping to see them here, in particular for a newly released machine.
Within just a few days of shipping out the first units, we have learned about
- a phase stepping feature which was known to be necessary in the XL but was omitted in the Core One,
- overhangs which print worse than in the MK4S, apparently because despite all the fanfare the build chamber is not kept at the optimum temperature,
- loose belts in a newly shipped unit,
- units with nasty rattling and x-axis oscillation in the prints,
- units without firmware,
- misrouted cables in the toolhead,
- a unit with a non-working NFC/WiFi/filament sensor.
I am willing to cut Prusa some slack for a new product, but they need to fix this stuff quickly. And it would not get them extra credits for "look what a great company", but would only be catching up with what they should have gotten right in the first place.
RE: Bad Vibrations
- We don't have information on if this is a phase stepping issue or if it can be fixed with firmware
- I've only seen a single instance of someone complaining about overhangs, many others haven't had the issue and show great overhangs
- Belts can come loose in shipping and can be tightened
- The nasty rattling can hopefully be fixed with the fix for the first point and is probably the same exact thing
- The firmware comes on the USB and this is explicitly stated on the instructions
- A misrouted cable happens. Prusa is sending the user a new toolhead and the issues is being fixed
- I haven't seen the non-working sensor so I can't comment on that one
Overall all of these issues seem small and very fixable. Prusa has proven themselves time and again to fix issues like these, and they are all things that have happened with other companies. Most of these issues are being pointed out in a very nit-picky way and are simple issues that happen to have been missed in QA. This happens all the time with products worldwide. The devices are made by humans, checked by humans and humans are not perfect.
I don't understand this strong impulse to defend Prusa. Yes, they are a likable company. They have advanced the state of 3D printing a lot over the past decade, have generously shared their developments, and have an honest business model. I cancelled a Bambu order in favor of a Core One just a few weeks ago, after Bambu came out with their firmware lockdown plans.
But Prusa also has a history of messing up their initial product launches. This time they got the first ship date right, but they did not manage to get to a reasonably mature design and stable manufacturing process. I think Prusa should be concerned about that, because as they fall behind regarding innovation, really solid engineering and production quality is one of the advantages they should try to maintain. And I think it is entirely justified to call them out on the deficiencies.
I don't want to go through your bullet point list once more, trying to argue each point. Many of them are judgement calls or need a final technical assessment. (Although I would venture a bet that omitting the acceleration sensor will turn out to be a mistake, maybe even in two respects -- forfeiting a way to detect and compensate for loosening belts, and forfeiting the possibility to implement phase stepping.)
But please don't claim that the missing firmware is by design and in line with the documentation -- take another look at the user manual:
6.3. Updating the Firmware
The Prusa CORE One is ready to print once assembled and powered on. To make sure you have the most up-to-date version of the firmware with the latest features and settings, we recommend checking prusa3d.com/drivers. You can perform the firmware update after you complete the initial Selftest.
RE: Bad Vibrations
I don't understand this strong impulse to defend Prusa. Yes, they are a likable company. They have advanced the state of 3D printing a lot over the past decade, have generously shared their developments, and have an honest business model. I cancelled a Bambu order in favor of a Core One just a few weeks ago, after Bambu came out with their firmware lockdown plans.
But Prusa also has a history of messing up their initial product launches. This time they got the first ship date right, but they did not manage to get to a reasonably mature design and stable manufacturing process. I think Prusa should be concerned about that, because as they fall behind regarding innovation, really solid engineering and production quality is one of the advantages they should try to maintain. And I think it is entirely justified to call them out on the deficiencies.
I don't want to go through your bullet point list once more, trying to argue each point. Many of them are judgement calls or need a final technical assessment. (Although I would venture a bet that omitting the acceleration sensor will turn out to be a mistake, maybe even in two respects -- forfeiting a way to detect and compensate for loosening belts, and forfeiting the possibility to implement phase stepping.)
But please don't claim that the missing firmware is by design and in line with the documentation -- take another look at the user manual:
6.3. Updating the Firmware
The Prusa CORE One is ready to print once assembled and powered on. To make sure you have the most up-to-date version of the firmware with the latest features and settings, we recommend checking prusa3d.com/drivers. You can perform the firmware update after you complete the initial Selftest.
I don't understand this strong impulse to constantly bash Prusa in every post that's made. Secondly I'm not defending them, I'm replying to the comments that are made with inflated perceptions of issues.
I don't want to go through your bullet point list once more, trying to argue each point. Many of them are judgement calls or need a final technical assessment.
That's quite a deflection. If you aren't happy with the product, don't buy it it's that simple.
But please don't claim that the missing firmware is by design and in line with the documentation -- take another look at the user manual:
I also recommend you follow your own advice and read the manual as it explicitly states to plug in the usb stick that has the firmware:
https://help.prusa3d.com/guide/4-calibration-first-run_825413#825475
You don't like the core one, that's fine you've made it extremely obvious with your many posts about how horrible it (and Prusa) are. I'm just pointing out the issues you are stating are quite inflated and in some cases completely incorrect.
RE: Bad Vibrations
The specs on the CoreOne pages does indicate that it has phase stepping. I'm curious how it will accomplish that without any included accelerometer as well. I'm still keeping my orders as I figure Prusa will work out these kinks pretty quickly.
RE: Bad Vibrations
According to the developers on Github, phase stepping is active by default. However, the calibration for this is currently hidden as it is still being worked on. Although the XL also has phase stepping, it also has 1.8 degree stepper motors, while the Corer One, like the Mk4, has 0.9 degree stepper motors. That is the reason.Here is the link to the Github repository:
https://github.com/prusa3d/Prusa-Firmware-Buddy/issues/4461
The specs on the CoreOne pages does indicate that it has phase stepping. I'm curious how it will accomplish that without any included accelerometer as well. I'm still keeping my orders as I figure Prusa will work out these kinks pretty quickly.
RE: Bad Vibrations
[...] If you aren't happy with the product, don't buy it it's that simple.
[...] You don't like the core one, that's fine you've made it extremely obvious with your many posts about how horrible it (and Prusa) are. I'm just pointing out the issues you are stating are quite inflated and in some cases completely incorrect.
Look, the thing is, I want to like the Core One. I ordered it (as a kit) and already paid for it in full. I don't see a good alternative for me, since the other players are either Chinese companies with business models that give me headaches, or companies which are too far on the "fiddling with the printer is the hobby" side for my taste.
After long deliberation, I had made my peace with the fact that the Core One is behind the competition regarding features and innovation. I had convinced myself that this was outweighed by the fact that Prusa (both the company and the printers) are solid in every respect. And I am now disappointed to see that this isn't quite the case. And I am disappointed to see that -- except for the unlucky people who have the new printer and can't get it to produce clean prints, or to print at all -- everybody just shrugs it off and considers it "par for the course" for Prusa.
Maybe I am lacking the years of experience and (eventual) satisfaction with Prusa, from back when they were also technology leaders in the home 3D-printing market. The Core One would (hopefully will!) be my first Prusa printer. But it can't be Prusa's strategy to sell only to loyal long-standing customers. I very much hope they are still aiming to, and able to, gain new customers as well. And to that end I think they need to make sure that the competitors don't get ahead of them regarding the launch of mature, stable products as well.
RE: Bad Vibrations
Oh interesting. So they are saying that phase stepping calibration will be coming. Will we need to order the accelerometer to take advantage of that feature though? I assume we probably would.
RE: Bad Vibrations
Oh interesting. So they are saying that phase stepping calibration will be coming. Will we need to order the accelerometer to take advantage of that feature though? I assume we probably would.
I would say you'd need the accelerometer, I'm hoping I can wire my existing third party one to work as shipping is prohibitively expensive to NZ for the official one.
RE: Bad Vibrations
Yes, you have to buy it separately. But it doesn't cost much and can be set up relatively quickly.
https://www.prusa3d.com/de/produkt/accelerometer-set/
Oh interesting. So they are saying that phase stepping calibration will be coming. Will we need to order the accelerometer to take advantage of that feature though? I assume we probably would.
RE: Bad Vibrations
Makes sense. I just added the accelerometer to my order as well. I noticed that on the product page it does say it is compatible with the CoreOne.
RE:
I would say you'd need the accelerometer, I'm hoping I can wire my existing third party one to work as shipping is prohibitively expensive to NZ for the official one.
That should be quite feasible as long as the accelerometer "speaks" SPI, using the same command set as Prusa's original. The accelerometer connector on the xBuddy board is a standard SPI port, and the accelerometer chip seems to be an LIS2DH12TR (ST Microelectronics). You probably want to use the same chip for full compatibility.
RE:
Yes, it is compatible. To ensure that the cable is long enough, there is a small rubber plug on the back wall of the Core One, which you have to remove in order to pack the cable through it. I didn't notice this at first and wondered how the cable could be long enough. At 600 mm.
Makes sense. I just added the accelerometer to my order as well. I noticed that on the product page it does say it is compatible with the CoreOne.
RE:
I'm willing to bet they'll have some sort of factory fix. I don't think they'll require you to buy an accelerometer to fix it. Maybe the fix won't be as good as it would be with an accelerometer, but it'll be improved.
One interesting thing I'll note is that I just watched the made with layers review and according to him the print quality is as good as the bamboo. So I guess watch it and make your own conclusion.
So maybe the problem is only with some machines and not all.
Edit: I posted a comment asking about the noise as it wasn't mentioned
RE: Bad Vibrations
I'm willing to bet they'll have some sort of factory fix. I don't think they'll require you to buy an accelerometer to fix it. Maybe the fix won't be as good as it would be with an accelerometer, but it'll be improved.
One interesting thing I'll note is that I just watched the made with layers review and according to him the print quality is as good as the bamboo. So I guess watch it and make your own conclusion.
So maybe the problem is only with some machines and not all.
If the noise and VFA ripple are indeed due to motor imperfections and can be compensated via properly calibrated phase stepping, Prusa could add a unit-specific factory calibration to the manufacturing process. I understand that these imperfections (in motor windings or belt sprockets) are long-term stable and should not change in shipping.
But the printer would need some long-term parameter storage which survives firmware updates and hard resets. I don't think that exists at the moment, but it could possibly be added in a future firmware. -- On the other hand, by the time they add in the labor costs for that factory calibration and a potential development effort, Prusa might find that just shipping the accelerometer with every Core One is the cheaper option? (And would give the added benefit of letting the user re-tune the input shaping when needed, e.g. in case of belt tension drifts over time.)
You are right, several Youtube reviews specifically mention how quiet the printer is, and/or that it produces very good prints without major VFAs. I think that's consistent with the hypothesis that unit-to-unit variances -- either in some components or in the assembly/alignment process -- are behind the issues in some units which were reported here. Maybe the review units did undergo an extra round of inspections or selection?
RE: Bad Vibrations
Hi, I am not very familiar with phase stepping. But do I understand correctly that phase stepping is related to input shaping and accelerometer calibration will solve issue? I understood previously maybe incorrectly that input shaping is just acceleration control correction and phase stepping is more of way of controlling the motor itself? Or is prusa accelerometer calibration do both ?
RE: Bad Vibrations
Nothing new here. Given how different the Core one is, I'm actually surprised not to see more issues. I got one of the first XL's and it wasn't pretty at the beginning. But now it's the machine I love most. I'm sure the Core One will be fine. Hopefully my first one is coming this week, and I can't wait to replace my Mk4S's, which is one generation of Prusa machines I have not been impressed with.
Formerly known on this forum as @fuchsr -- until all hell broke loose with the forum software...
RE: Bad Vibrations
Hi, I am not very familiar with phase stepping. But do I understand correctly that phase stepping is related to input shaping and accelerometer calibration will solve issue? I understood previously maybe incorrectly that input shaping is just acceleration control correction and phase stepping is more of way of controlling the motor itself? Or is prusa accelerometer calibration do both ?
Input shaping and phase stepping are two different things, but they are both calibrated using the accelerometer.
Input shaping will indeed modify the acceleration profiles. It reduces frequency components where the printer has mechanical resonances (and will hence amplify these components anyway). The resonances stem from the printer setup as a whole, where elastic parts (belts, frame parts) and suspended masses form oscillators.
Prusa has not really discussed what "phase stepping" does under the hood, to my knowledge, and phase stepping seems to be their proprietary name for it. From a superficial look at the source code, it seems to modify the timing of the individual motor (micro-)steps, to compensate for manufacturing tolerances in the motor itself.
The goal of input shaping is to avoid overshoots caused by accelerations while allowing the printer to move and accelerate fast. The goal of phase stepping is to avoid resonances (which would cause acoustic noise and/or ripple in the print) when the print head moves at specific constant speeds.