Tension for belts questions?

[MythN7]

When using a guitar tuner to get parallel belts matching for their tension level, does each printer model of printer require a different value or is the Sweet Spot value. Generally the same all around?

I have an Ender 5S1 The two belts that came installed seem to be way too tight and were noticeably different tuner values than each other. So what would I want to use when I don’t quite got the experience on what feels right by touch.

I was also thinking about putting a string over the middle of the top part of the belt, and then adding weight to it until it pulls down. Just barely touching the bottom portion like pinching them together so that I could use the same string of weight on the parallel belt to get an exact match on tension.

But again The issue is still revolving around the trying to figure out the sweet spot for how tight it should be to begin with.

Also would like to ask, I’ve learned order matters significantly in more places than you can count for what to tighten first and last,.
So which would I want to do first for the best results?
1: Set my eccentric nuts so the wheels are balanced, with loose belts.
or
2: set the eccentric nuts in the loosest possible position and then tension the belts accordingly before setting the nuts intended position.

Thank you.

[Matthew]

Hey there,

Yeah each model of printer usually has a sweet spot which is dependant on the length of the belts. For the 5 S1 you will want to try to move the toolhead as close to the center as you can so that there is even spacing on all of the belts. I don’t remember what frequency you want the belts to be at, but you can usually find this info online. Maybe someone else here knows what it should be at.

In terms of your order of operation, I would loosen the wheels, tune the belts, and then tighten the wheels again. Although the wheels and the belts are not super dependant on each other, so this step is really not very important. You can do it either way!

[MythN7]

So I’m not sure if this is overkill or a good method for someone who desires a way to check and measure every setting for consistency completely making the based on feels right as redundant as possible.

Side note.
About the belt being in the center so it’s even like you said. I actually did notice when it came as its brand new state after putting it together.

Just on the running it, on the leveling setup and then the auto leveling about 25% of the way in to my first test print a benchee boat, I had a lot of wheel dust already built up and I couldn’t figure out why this was.

After getting through the filament that was supplied with the printer, I wanted to try and fix it as it had other issues I won’t get into here, so the POM wheels already I discovered that on the top were really ground apart from the outside and the wheels on the bottom were ground apart on the inside. So I replaced them with new ones.

And then I noticed when I pulled the printhead all the way to the front, the left corner was about 8 mm away from the edge compared to the other one.

The way I attempted fixing that was to loosen the set screws on the left right rod connecting the two belts so that the two gears driving the belts could spin independently and as soon as I loosen that set screw, you could actually feel the hot Head Gantry snap into place a lot better like how a diving board oscillates after you slam it, Even then, I noticed that the belt had to be taken right off the gears and adjusted so that it was 3 more teeth back.

That’s just a side parallel observation that caused me to take the printer right apart. Even farther down than the way it arrived to try and get it aligned up better.

So the process I did today and currently where I am is.

1: in green I put t braces I made sure that they were square to the front frame, and the stopper blocks on their ends were exactly the same distance away from the frame as each other.
2: I took the bottom wheels off that had the eccentric nuts completely for now, as it seemed gravity holding it down on the top appears to be very close to what I understand the amount of fiction they have when turning them on the spot.
3: I put two little meccano wall things on the bottom of the Gantry with T-Nuts and I took the dial knobs on the bed completely off. So again it’s just gravity holding it without the springs engaged.
4: with the belts loose I could drive the Gantry forward and back and there was no hitching as well as the left and right mechano things were according to the naked eye perfectly level to each other along the two back corners and the two front corners at the same time.
5: I added the purple making sure that they were the exact measurement from the back as each other holding the Gantry tight so it won’t slide back and forth, assuring that it was Square on all three axes to the frame.
6: the two belts spining freely, I started tightening the belts until it sounded the same when I plucked it with my finger. Whatever Hertz it would be when it actually starts to sound like a guitar just barely.

This is where I stop right now.
My idea going forward still is to figure out what tuner Hertz I should try to use, and then set them to that.
And then with the drive rod still not forcing them to be in sync. See if it still rolls smooth and stays parallel in all directions.

If that checks out I’ll lock it back in the front clamp tighten set screws on the drive rod, and then put in the bottom wheels and set the eccentric nuts to have the same tension that the top wheels currently have.

Hopefully this will allow me to have a perfectly even bed so that each spring on the four corners will have the same amount of tension on it to maintain its levelness.

Is there anything in my process for overkill precision desire that I’m not seeming to account for that will cause the system to implode on itself when it’s going under the insane speed that it likes to home at?

Thank you for taking the time to read this.

[Loosenut]

@Matthew

Yeah each model of printer usually has a sweet spot which is dependant on the length of the belts. For the 5 S1 you will want to try to move the toolhead as close to the center as you can so that there is even spacing on all of the belts. I don’t remember what frequency you want the belts to be at, but you can usually find this info online. Maybe someone else here knows what it should be at.

And to makes thing even more confusing the X and Y belts are not always the same.

[Matthew]

Yes you are absolutely correct! I can’t remember off the top of my head what the formula was to find the optimal belt tension but I will take a look and see if I can figure it out. Unfortunately most of that is still theoretical, so the values you get may not match what you will want in real life.

[Matthew]

Just about to step out for the day so I will take a look at this tomorrow, from a quick glance it looks to me like you have most of the big things covered. I will take a closer look first thing tomorrow though!

[MythN7]

Got it all together. Finally balanced perfectly with the clamps off the eccentric nuts are all in the loosest possible position, but it still feels a bit tight, unless It’s supposed to be that tight.
And the belts I was able to get perfectly matched it seems like around 83hz feels good on the belt itself when you try and pull it down, but when you move the Gantry back and forth it feels like there 's too much muscle needed to move it and around 60 HZ feels about right for the effort to push it, but the belts feel quite loose so it’s hard to say.

And the two screws on each belt tensioner to lock it in place after you adjust the spacing screw once you tighten those the Hz value when plucking the belt increases by about 30. On one side and a value on the other side with the motor. That’s inconsistent. So after tightening these the guitar tuner no longer shows them as matching but they feel the same.

I also noticed having the Gantry against the stopper blocks to measure its squareness once I remove the stoppers The back side it’s incredibly easy to make it go out of square just by turning one of the belt’s tension. Even a quarter of a turn in extra tightness. I can actually see that side pull off Square.

Is it normal for the Gantry to be like a hair trigger level of delicate being able to pull the frame off Square by tightening the belt by like a quarter of a millimeter?

Seems like it being that fragile to pull out of alignment. It’s an accident waiting to happen because I’m sure after use one of the belts could stretch by more than that in no time throwing the squareness way out.

I do know that with the guitar tuner the tension on the belts When it arrived from the factory, we’re around 120 and 110
One of them them is not accurate because I know that they were at least 15 HZ off each other.

For now I’m not trying to find the perfect sweet spot of HZ I’m trying to find out a general range that is considered definitely too loose and definitely too high and try to find the spot in between as tweaking later.

[Matthew]

@MythN7

60 HZ feels about right for the effort to push it, but the belts feel quite loose so it’s hard to say.

The belts shouldn’t be too tight otherwise they will wear and can cause skipped steps. The best way I can describe the tension is similar to how an elastic band feels when it is wrapped around a bundle of newspapers, does that make sense?

@MythN7

Is it normal for the Gantry to be like a hair trigger level of delicate being able to pull the frame off Square by tightening the belt by like a quarter of a millimeter?

You can try and tighten up the frame, but unfortunately the frame on this printer was fairly rickety. You can tighten it up using some corner brackets and stuff, that’s about the best you will be able to do.

[MythN7]

I’m not sure if this helps in terms of a visual that can be used as a fixed concept for tension.
I have a roller hanging off the belt in the one picture and I attached the entire direct drive hot head assembly to a piece of metal attached to it. So the entire weight hanging off it.
The piece of metal the hot end It’s attached to is almost the same weight so whatever the hot end direct drive weighs at about 50 to 70% of its weight.

And the amount it’s pulling the belt down from gravity is about 50% the distance it needs to go before making contact with the bottom belt.

One would think if you could measure how much mass it takes to move the belt, a measurable amount of distance you could use this.

To get it to move this much under that much weight is with the guitar tuner reading approximately 80 HZ with the belt tensioner unlocked, and around 110 with it locked.
Not sure why it changes so much from not tightening the belt but just putting in the screws that hold it in place more.

The other two pictures, are just before I took the original wheels off after amount of use equal to just the filament that came with the printer. I was having to wipe dust off the rails along that track from noticeable buildup just in the amount of movement auto leveling does.

I have the frame put together pretty solid. I used machinist clamps and squares to put it together along with plates to keep everything flush before tightening in any of the screws.
And I also used loctite on the bolts.
So the frame appears to be pretty solid, unless you’re talking about the ease of actually making the metal torque and deform under running force.

If anyone has this printer and and knows the belt tension in a somewhat proper state, maybe you could do something that’s idiot proof anyone can reproduce from a picture alone such as the fixed amount of weight, hanging off the belt showing its tension level that way.

I think the hardest part is not having stores with printers set up by professionals so you can actually touch them to see what this. “FEEL’S RIGHT” actually means.

There’s way too much instructional videos where the steps are until it feels right as the instruction.

Here’s a bad analogy, but if you’ve never experienced it or have properly tuned version to sample that’s like trying to learn how to read Braille you some grains of rice glued the paper and a YouTube video absolutely impossible.

I guess I’m just a bit frustrated, that I have 3/4 of a mechanical engineering degree, an advanced diploma in 3D modeling and animation. But yet I can’t get this printer to even work remotely like it’s supposed to.

[MythN7]

After setting it all up and crossing my fingers I had a bunch of bad prints on the bed plate that had some gouges in it causing the rebuild in the first place, putting on a brand new plate and turning the print speed down to 25% on the printer override with the rabbit that came with I was able to get this in an hour and 20 minutes.

Then trying to print at 50% speed and 100% speed. I got this.

I’m not sure if you can tell If my belts or anything are too tight or loose from the parts that got printed okay. But around the same time they all started being dragged around.

Then I took the night off in agitation, in the morning. I put a the replacement nozzle that came with the printer in as the previous one is the one that gouged up the plate.

I also put the printer on The kitchen table that’s a lot more sturdy than an end table I was using.
I also leveled it using a Feeler gauge at 0.1 mm, and set the print speed to 10% so the line on the left would draw really slow and I use the live adjustment to put the nozzle even closer to the bed so that when the plastic came out it looked like it was being flattened on the top and not just dropping a line down to cool, so the overall height from the nozzle is something like 0.06 mm which really scares me that close cuz I can’t afford to buy a new build plate.

I’m not sure if the rabbit or boat that comes with the printer has auto leveling engaged.

Restoring the print speed to the default 100% after this live Z tweak, I was able to get these two prints out back to back using this filament at 60 bed 205 hot end.

And I was able to get these out at the default print speed.
So the rabbit took about 23 minutes and the boat took an hour and 20 minutes.

This replies are a bit long-winded. I’m sorry but I wanted to explain all my steps and results in case it matters, but I’m actually pretty pumped. I got two prints out back to back at the default print speed.

Now I just got to be able to figure out how to. According to this topic title tweak The Sweet Spot for tension to make it look smoother. Unless it’s perfect and I look I need to look at other adjustments instead.

To try and nail these fine tune adjustments, I think I’m going to be printing off a crap ton of gridfinity so I can actually make use of the plastic while I’m calibrating and trying to learn how to read a print quality.

Thank you all for your support.