RE:
Would have been easier to simply say strum the top and bottom belts that go across the linear rail. Unless I'm missing something, that was never clear in the revised instructions.
RE: New VFA Tuning Method rocks!
This new 'strobe' tuning method can't come soon enough.
It really is a complete PITA right now using the current method.
Prusa's priority right now should be a simple, foolproof way to get the gantry perfectly square, followed by a simple foolproof way to get the tension set....using the current methods there are just too many variables that the user can't control that can throw the readings off.
I'm sorry but the current method works really well. If there is no way to obtain the 6Hz difference I'm pretty confident that something is wrong.
- Tuning instructions are not properly followed (Nextruder in the front right from the users perspective, strum the top and bottom belts that run along the linear rail)
- Tuner is used with a phone or device that has its own noise reduction running. Try using it with a proper microphone on a laptop. Check the frequency response while making different noises, you should see a clear response. If your voice makes a clear response but a finger snap does not, there is some kind of noise reduction going on.
- Make sure that both belts are tensioned at the same time since in a Core XY system, the belts have an effect on each other.
- There is something wrong with the belts. Are the motor pulleys correctly installed? Are the same number of teeth pulled through the Nextruder carriage on all four belt ends?
In my case, the gantry was square before tuning and perfectly square after tuning with the belts with 0,5Hz of the (new target. I used a proper microphone connected to my laptop and used the web tuning application. It worked really well and I had no issues. I really feel that if it's impossible to get to the target frequencies, something is wrong and it's not the tuning method...
RE: New VFA Tuning Method rocks!
This new 'strobe' tuning method can't come soon enough.
It really is a complete PITA right now using the current method.
Prusa's priority right now should be a simple, foolproof way to get the gantry perfectly square, followed by a simple foolproof way to get the tension set....using the current methods there are just too many variables that the user can't control that can throw the readings off.
I'm sorry but the current method works really well. If there is no way to obtain the 6Hz difference I'm pretty confident that something is wrong.
- Tuning instructions are not properly followed (Nextruder in the front right from the users perspective, strum the top and bottom belts that run along the linear rail)
- Tuner is used with a phone or device that has its own noise reduction running. Try using it with a proper microphone on a laptop. Check the frequency response while making different noises, you should see a clear response. If your voice makes a clear response but a finger snap does not, there is some kind of noise reduction going on.
- Make sure that both belts are tensioned at the same time since in a Core XY system, the belts have an effect on each other.
- There is something wrong with the belts. Are the motor pulleys correctly installed? Are the same number of teeth pulled through the Nextruder carriage on all four belt ends?
In my case, the gantry was square before tuning and perfectly square after tuning with the belts with 0,5Hz of the (new target. I used a proper microphone connected to my laptop and used the web tuning application. It worked really well and I had no issues. I really feel that if it's impossible to get to the target frequencies, something is wrong and it's not the tuning method...
Yeah I know all of that.
Having to go and get my laptop from another room, trying to ensure there is absolutely zero background noise...
It just all feels a bit amateurish to have to jump through all these hoops to get a £1k printer to even print. Even after completing the above I can strum the same belt and see different readings without changing anything.
I'm growing very fond of this printer...when it works it works really well. Apart from this setup song and dance I don't really have any problems with my Core One.
RE: New VFA Tuning Method rocks!
Would have been easier to simply say strum the top and bottom belts that go across the linear rail. Unless I'm missing something, that was never clear in the revised instructions.
So now that little misunderstanding is cleared up, what readings do you get for the upper and lower belts?
If you've got equal tension at the sides (old location), it should be a short step from there to get 98/92 at the new location.
(Images from Prusa Knowledgebase)
RE: New VFA Tuning Method rocks!
Yeah I know all of that.
Having to go and get my laptop from another room, trying to ensure there is absolutely zero background noise...
It just all feels a bit amateurish to have to jump through all these hoops to get a £1k printer to even print. Even after completing the above I can strum the same belt and see different readings without changing anything.
With my laptop/microphone set-up, I can strum the belts repeatedly and see readings within a range of 0,2Hz. If that is not the case, something is going wrong with the tuning method. Either the microphone is not close enough to the belts or there is some form of filtering present? Or maybe you strum so hard that the belt actually touches the linear rail or something else. I use the same allen key that I do the tensioner adjustments with to strum. Just one strum, pull the belt back with the key and let the key slide of carefully.
RE: New VFA Tuning Method rocks!
With my laptop/microphone set-up, I can strum the belts repeatedly and see readings within a range of 0,2Hz. If that is not the case, something is going wrong with the tuning method. Either the microphone is not close enough to the belts or there is some form of filtering present?
But this is kinda my point mate. I have enough to worry about on a daily basis without trying to work out if my mic is close enough or if some form of filtering is present....or if ultimately something is actually wrong with the printer.
That's why I mentioned the strobe approach, hopefully it completely takes out all the environmental variables that can throw things off.
RE: New VFA Tuning Method rocks!
This suggests either that your gantry wasn't square before you started to tension the belts, or that the belt tension wasn't increased evenly. I've seen it myself - if I let, say, the right tensioner get too far ahead of the left one, it will open a gap on the left side of the gantry, such that adding a bit more tension to the left side results in both belts having an increased frequency.
For me, the solution has always been to slacken them both off again, check that the ganrty is square, then go more slowly and evenly with the tensioning.
Hi! Can you explain how you check and then correct the squareness of the support? Because I haven't seen any way to do this? Thanks
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:
Hi! Can you explain how you check and then correct the squareness of the support? Because I haven't seen any way to do this? Thanks
Put the extruder in the centre of the linear rail, and pull the linear rail to the front of the printer. Look at where the linear rail mounting brackets meet the tensioner mounting blocks in the front corners. If the gantry isn't square there will be a gap between the rail mounting bracket and the tensioner mounting bracket on one side or the other. This will be due to one or both of the metal right-angle brackets that secure the linear rail not being at a perfect right angle. The task is to manually bend those brackets to eliminate the gap.
If adjustment is needed, it is ESSENTIAL to fully release the tension in both belts so that there is plenty of slack in the belts - if you don't do this you may break a tensioner pulley mount.
Find an object that is slightly thicker than the gap that you're trying to eliminate, and place it as a spacer between the rail mount and the tensioner mount on the OTHER side to the one with the gap. Then pull on the side with the gap to permanently bend the brackets and close that gap. You may find you need to use more force than feels comfortable, but provided there is no belt tension you (probably!) won't break anything.
Remove the spacer object, check for gaps, repeat as necessary. Run the gantry to the back and check that both ends of the rail meet the motor mounts at the same time.
Once everything is square, you can tension the belts, but FIRST...
While you have the belts fully slack, remove the two tension screws and put a small amount of lubricant on the threads (keep the heads clean and dry). This will help to prevent the tension screws seizing on the nuts through 'galling'.
When tensioning the belts, go slowly and evenly (see k1mu's notes above), and aim to start building up to the 6Hz difference between the top and bottom belts from early on.
RE:
Would have been easier to simply say strum the top and bottom belts that go across the linear rail. Unless I'm missing something, that was never clear in the revised instructions.
So now that little misunderstanding is cleared up, what readings do you get for the upper and lower belts?
If you've got equal tension at the sides (old location), it should be a short step from there to get 98/92 at the new location.
(Images from Prusa Knowledgebase)
Had I known about the existence of this knowledgebase article, the issue would have been resolved many posts ago rather than getting a pedantic lesson on prepositions. I simply didn't know which position to check the belts. That is all. Anyway, issue resolved.
RE: New VFA Tuning Method rocks!
@k1mu's 'pedantic lesson on prepositions', as you call it, appears to have been an honest attempt to clear up an obvious misunderstanding, by going back to first principles. It was clear to all that something was being missed, but it wasn't clear what. One person's 'easier explanation' is another person's 'ambiguous or incomplete description'. And @k1mu's explanation evidently worked, even if you thought there was an easier explanation.
So what are your upper and lower belt tensions now?
RE:
@k1mu's 'pedantic lesson on prepositions', as you call it, appears to have been an honest attempt to clear up an obvious misunderstanding, by going back to first principles. It was clear to all that something was being missed, but it wasn't clear what. One person's 'easier explanation' is another person's 'ambiguous or incomplete description'. And @k1mu's explanation evidently worked, even if you thought there was an easier explanation.
So what are your upper and lower belt tensions now?
His explanation cleared up nothing and the knowledge base article, which I wasn't aware existed, did. All that was required in the beginning was a link to the article. Now problem solved because of the article; rather than a lesson on what English prepositions mean.
RE: New VFA Tuning Method rocks!
His explanation cleared up nothing and the knowledge base article, which I wasn't aware existed, did. All that was required in the beginning was a link to the article. Now problem solved because of the article; rather than a lesson on what English prepositions mean.
Ahem. Are you sure it didn't help?
Would have been easier to simply say strum the top and bottom belts that go across the linear rail. Unless I'm missing something, that was never clear in the revised instructions.
I'm not going to argue, but your particular brand of gratitude for the efforts of those trying (and succeeding) to help you is a bit off-putting.
RE:
His explanation cleared up nothing and the knowledge base article, which I wasn't aware existed, did. All that was required in the beginning was a link to the article. Now problem solved because of the article; rather than a lesson on what English prepositions mean.
Ahem. Are you sure it didn't help?
Would have been easier to simply say strum the top and bottom belts that go across the linear rail. Unless I'm missing something, that was never clear in the revised instructions.
I'm not going to argue, but your particular brand of gratitude for the efforts of those trying (and succeeding) to help you is a bit off-putting.
Yes, when I saw the lecture on what "top" and "bottom" mean I tuned it out. But you are arguing. I read the knowledge base article and it clearly explained the process. Simply linking it at the beginning of the thread would have been sufficient to clear up confusion on the part of several participants. Anyway, better to stop this discussion here and move on.
And yes, I'm a bit angry that I have a $1000 printer that I've been having to constantly tinker with to make it work.
RE: New VFA Tuning Method rocks!
But you are arguing.
No I'm not!
RE: New VFA Tuning Method rocks!
Ok, so I tried this new method initially prior to the knowledgebase being updated with new instructions to tune. I ended up with a very skewed gantry after an hour of tinkering. Annoyed I loosened everything, re-squared it and just went back to the old method. Printing yesterday I started hitting those annoying resonances again. So back looking for a solution, today I saw this thread and seeing some of the positive feedback said I'd give it another go.
Knowledge base now updated with some lovely pictures, I followed the new instructions. Key here was that I had time and more patience. I tightened both belts in very small increments uniformly. In about 15mins I hit the resonances, input shaper and phase stepping calibration complete. I just completed the noisy star test from printables... the differences was something else. I went back to work at my desk and had to check if the printer had started!
I was not bothered by VFA's before but decided to throw that test on too... just finished and it's much better. Was it bad before? I didn't really notice an issue as it was already miles better than my ankermake M5 for VFAs, but now they are barely noticeable on the shining black PETG I just printed. Are they gone? No but, I mean I have to look for them.
It is frustrating to have to tinker with the printer but at least Prusa are looking for solutions. I am not aware of other retail orientated manufacturers being as active in this front.
RE: New VFA Tuning Method rocks!
Thanks for your reply, that's exactly what I did, setup belt. But after correctly calibrating at 98 and 93 Hz, the frame is no longer square, and the Y test passes fine. However, the calibration at the start of the print is endless.
I have to redo everything and adjust by ear to get the frame square, and it works, but the frequencies are around 105 Hz and 87 Hz.
So after spending a few hours on it, I hesitated to throw the printer out the window. This type of adjustment is still not very professional.
There is a definite design flaw, due to the lack of frame rigidity.
The MK3/S MK4 and MK4S was a success, and I never had this kind of problem. The Core One is much less reliable; the slightest change renders it inoperable. Too bad...
This isn't an evolution, but a regression.
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!
Thanks for your reply, that's exactly what I did, setup belt. But after correctly calibrating at 98 and 93 Hz, the frame is no longer square, and the Y test passes fine. However, the calibration at the start of the print is endless.
I have to redo everything and adjust by ear to get the frame square, and it works, but the frequencies are around 105 Hz and 87 Hz.
So after spending a few hours on it, I hesitated to throw the printer out the window. This type of adjustment is still not very professional.
There is a definite design flaw, due to the lack of frame rigidity.
The MK3/S MK4 and MK4S was a success, and I never had this kind of problem. The Core One is much less reliable; the slightest change renders it inoperable. Too bad...
This isn't an evolution, but a regression.
Totally agree. I've had the printer for 6 weeks now and the tinkering with it to get to function is endless:
RE: New VFA Tuning Method rocks!
Opinions may differ on this. If you find yourself with the urge to throw it out of the window again, stop, take a deep breath, then PM me for my postal address. I'll gladly take it off your hands. 🙂
I find my CORE One to be so much better than either of my MK4Ss (by virtue of the smaller enclosure and the higher temperatures), that I almost never use my MK4Ss now.
Thanks for your reply, that's exactly what I did, setup belt. But after correctly calibrating at 98 and 93 Hz, the frame is no longer square, and the Y test passes fine. However, the calibration at the start of the print is endless.
I have to redo everything and adjust by ear to get the frame square, and it works, but the frequencies are around 105 Hz and 87 Hz.
So after spending a few hours on it, I hesitated to throw the printer out the window. This type of adjustment is still not very professional.
There is a definite design flaw, due to the lack of frame rigidity.
The MK3/S MK4 and MK4S was a success, and I never had this kind of problem. The Core One is much less reliable; the slightest change renders it inoperable. Too bad...
This isn't an evolution, but a regression.
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.
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!
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.
Haha - yes, I figured I wasn't about to score a free CORE One! Haven't we all been there though - wanting to hurl stuff through the window?
I do have a couple of MK4Ss here, both in Prusa enclosures, which are now surplus to requirements, not to mention that I also have a Ston Wolf on order and need to make room. But if anyone (in the UK) felt disillusioned after converting a MK4S to a CORE One, and wanted to go back, perhaps one of my MK4S+Enclosures might be a quick and easy way to do so, and might be considered a fair swap 🙂