RE: New VFA Tuning Method rocks!
Very quick update. So had a print on after my post a couple hours ago. Not too bothered about quality so had it fast at 0.25. A bit of noise so needed to quiet it for a while so threw on stealth mode mid print. Can you spot the point I put on stealth mode? It went quiet but I also ended up getting the vfa’s back… trying to rack my brain as to why
RE: New VFA Tuning Method rocks!
Very quick update. So had a print on after my post a couple hours ago. Not too bothered about quality so had it fast at 0.25. A bit of noise so needed to quiet it for a while so threw on stealth mode mid print. Can you spot the point I put on stealth mode? It went quiet but I also ended up getting the vfa’s back… trying to rack my brain as to why
I noticed that as well on the slower perimeter filling. Have you run the phase stepping calibration? That seemed to help in my case.
RE: New VFA Tuning Method rocks!
It was a whim, and then I won't have a printer anymore... the C1 is a good machine, but the design is a bit below what was done before.
Yes, like why I'm scraping off powder coating off the frame to ground the X-motor housing so the printer doesn't crash.
RE: New VFA Tuning Method rocks!
I did do the phase stepping post tuning but will try again later to see if it makes any difference
Very quick update. So had a print on after my post a couple hours ago. Not too bothered about quality so had it fast at 0.25. A bit of noise so needed to quiet it for a while so threw on stealth mode mid print. Can you spot the point I put on stealth mode? It went quiet but I also ended up getting the vfa’s back… trying to rack my brain as to why
I noticed that as well on the slower perimeter filling. Have you run the phase stepping calibration? That seemed to help in my case.
RE: New VFA Tuning Method rocks!
did support suggest this? I'll have to note it when I build mine
It was a whim, and then I won't have a printer anymore... the C1 is a good machine, but the design is a bit below what was done before.
Yes, like why I'm scraping off powder coating off the frame to ground the X-motor housing so the printer doesn't crash.
RE:
did support suggest this? I'll have to note it when I build mine
Not directly. There is a quite a long thread on the subject and the conclusion is that in cases where the motor housing is not grounded, crashes can occur on long diagonal prints. See here: Core One crash/restart for details. Just make sure both X and Y motor houses are connected to ground. It's sufficient to either remove the powder coating underneath one screw head or use a star washer when mounting the motors to the frame
RE: New VFA Tuning Method rocks!
Hello, while trying to adjust the belts, I don't know what happened, but I couldn't get to the Y axis test, if the play between the left and right stops is zero, I can't tune the frequencies 98 and 92 hz, at best I get 98 / 87 ... The Prusa support contacted, told me to exert a twist on the rail by blocking a stop with a key to mechanically compensate for the squareness defect, finding this solution barbaric, I interposed washers on the right support of the X rail, and now I no longer have play between the left and right stops, I was able to set the frequencies to 98 and 92 hz, and my printer passed all the tests successfully. It worked for me, if that can help ...
MK3S + MMU2 migrée vers une MK4 +MMU3 , puis vers MK4S MMU3, en attente du kit upgrade Core One, Elegoo Mars, bricole sur Home assistant, arduino, 8266, Esp32, kicad, Freecad et Fusion 360
RE: New VFA Tuning Method rocks!
Hello, while trying to adjust the belts, I don't know what happened, but I couldn't get to the Y axis test, if the play between the
left and right stops is zero, I can't tune the frequencies 98 and 92 hz, at best I get 98 / 87 ... The Prusa support contacted, told me to exert a twist on the rail by blocking a stop with a key to mechanically compensate for the squareness defect, finding this solution barbaric, I interposed washers on the right support of the X rail, and now I no longer have play between the left and right stops, I was able to set the frequencies to 98 and 92 hz, and my printer passed all the tests successfully. It worked for me, if that can help ...
Your rail is now not square when the printer thinks it is.
RE: New VFA Tuning Method rocks!
Your rail is now not square when the printer thinks it is.
Possible! But it was the PRUSA support that showed me this method, (video in support) and everything works wonderfully now, the circles are perfect.
MK3S + MMU2 migrée vers une MK4 +MMU3 , puis vers MK4S MMU3, en attente du kit upgrade Core One, Elegoo Mars, bricole sur Home assistant, arduino, 8266, Esp32, kicad, Freecad et Fusion 360
RE: New VFA Tuning Method rocks!
Prusa support suggested washers? They have documented squaring the rail via brute force, which I presume is the method you rejected. Washers is not an equivalent approach.
RE:
I never wrote that Prusa suggested the washers, only twisting with a wrench... don't invent... The washers are my idea, mechanically it comes to the same thing and I can go back without breaking anything.
For me, this discussion is closed.
MK3S + MMU2 migrée vers une MK4 +MMU3 , puis vers MK4S MMU3, en attente du kit upgrade Core One, Elegoo Mars, bricole sur Home assistant, arduino, 8266, Esp32, kicad, Freecad et Fusion 360
RE: New VFA Tuning Method rocks!
Mate I think there's a miscommunication here, @cjd was just pointing out that even though you've fitted washers, your gantry still isn't square. You've just tricked the printer into thinking that the gantry is square.
If the solution was as simple as putting a washer in , then thats what Prusa support would advise.
Look if you're, happy with the printer, and the way it's printing, then good luck to you...but the reason Prusa support advise to square the gantry (with force if needed), is because its very important that the gantry is truly square.
RE: New VFA Tuning Method rocks!
Do people just eyeball it when they square it up or?
RE: New VFA Tuning Method rocks!
Having doubts about this following your comments, I took two angle measurements between the Y guides and the X rail, without washers: 89.10° with washers, 89.92° for sure. I only used off-the-shelf washers; I could have them machined to get 90°... just kidding. The initial squareness error was therefore well compensated for. I spent two hours disassembling, measuring, reassembling, measuring, and readjusting, for a comment with no basis other than a subjective impression. Thanks anyway. And if Prusa doesn't recommend this solution, it's because bending a frame beyond elasticity is invisible, while using spacers is visible and doesn't look serious for the business...
MK3S + MMU2 migrée vers une MK4 +MMU3 , puis vers MK4S MMU3, en attente du kit upgrade Core One, Elegoo Mars, bricole sur Home assistant, arduino, 8266, Esp32, kicad, Freecad et Fusion 360
RE:
And to come back to the main subject of this post, I received and mounted the accelerometer and calibrated the input shaper and the phase step, it is obvious that there has been a clear improvement with the noise (almost no more horrible squeaks) I found what I had on my MK4S, but the VFA which has decreased has not completely disappeared.
MK3S + MMU2 migrée vers une MK4 +MMU3 , puis vers MK4S MMU3, en attente du kit upgrade Core One, Elegoo Mars, bricole sur Home assistant, arduino, 8266, Esp32, kicad, Freecad et Fusion 360
RE: New VFA Tuning Method rocks!
The position of the washer compensates for the bracket not being square. This will result in the linear rail being square to the gantry if done right as you have noted. Which I think is a nice solution to getting the linear rail square versus the brute force approach. That said to compensate for that washer the X and Y carriages will now be slightly offset from each other. As already noted this will get you past the homing issues but your printer believes you’re print head is somewhere else. Likely just a couple millimetres off. For the most part this will just result in your prints being a little off the position on the plate. You may never see an issue until you print closer to the limits of the print volume on the xy.
This is one reason Prusa doesn’t recommend it, as it will result in the printer not running to the advertised spec rather than it being because of aesthetics. That offset in the long run could lead to uneven wear on the gantry. Also each customer that comes to them will likely have varying ranges of offset angles from 90degrees. So even if we disregard the other points, while this solution worked for you and you’re happy if another customer has a more skewed bracket they could need multiple washers to achieve the same result, which is hard to create a standardised solution for.
These are the reasons the recommendation is to attempt to return the L brackets to 90degrees by bending. This is really the only consistent way of suggesting a fix to customers without having to ship out new brackets… which could just bend again if belts are tightened inconsistently. The flaw is in how easily those brackets bend.
if I was to guess whenever they release the “Core One S”, the gantry will have a couple modifications… those linear rail L brackets getting thicker or reinforced being one.
RE: New VFA Tuning Method rocks!
These are the reasons the recommendation is to attempt to return the L brackets to 90degrees by bending. This is really the only consistent way of suggesting a fix to customers without having to ship out new brackets… which could just bend again if belts are tightened inconsistently. The flaw is in how easily those brackets bend.
You're absolutely right but I think we're wasting our time here. There's an old saying 'You can lead a horse to water, but you can't make it drink'
Side note, I think rather than these brackets actually bending, some aren't perfectly square to begin with. I don't think the forces the printer is putting on them is enough to bend steel.
RE:
@ WDLandry Good explanation! We agree that the rail supports are of questionable quality.
MK3S + MMU2 migrée vers une MK4 +MMU3 , puis vers MK4S MMU3, en attente du kit upgrade Core One, Elegoo Mars, bricole sur Home assistant, arduino, 8266, Esp32, kicad, Freecad et Fusion 360
RE: New VFA Tuning Method rocks!
Side note, I think rather than these brackets actually bending, some aren't perfectly square to begin with. I don't think the forces the printer is putting on them is enough to bend steel.
Agree. I measured mine before installing them, and they were certainly off by a degree or two.
The material is quite soft; the brackets are easily bent while holding them with two fingers. But I agree that plastic deformation during printing seems unlikely. Some elastic deformation is another matter, as witnessed by the fact that unequal belt tension can easily "adjust" the gantry.
RE: New VFA Tuning Method rocks!
Are you all just eyeballing this or are you squaring it some other way?
Side note, I think rather than these brackets actually bending, some aren't perfectly square to begin with. I don't think the forces the printer is putting on them is enough to bend steel.
Agree. I measured mine before installing them, and they were certainly off by a degree or two.
The material is quite soft; the brackets are easily bent while holding them with two fingers. But I agree that plastic deformation during printing seems unlikely. Some elastic deformation is another matter, as witnessed by the fact that unequal belt tension can easily "adjust" the gantry.