If I calculate tolerance on the holes, it’s anywhere between 3.3 to 6.1%. Is this an expected tolerance for the Snapmaker? If not, is there a way to improve this?
Filament is laid down from 3D printers ideally into an obround shape. Dimensional tolerance is sensitive to under and overextrusion (controlled by the calibration of esteps as well as the slicer’s flow multiplier). Deviations from the correct amount of extrusion affect the width of walls, which will make holes smaller or wider as a result.
Out of the box the snapmaker has unacceptable dimensional tolerance, 3% or worse is typical.
With proper esteps and flow calibration that can be brought to tenths of a percent.
There are other calibration tools, these are what I use.
If you are doing dimensionally critical parts I’d strongly suggest using Cura, Simplify3D, or PrusaSlicer - Luban does not expose enough settings for fine tuning when required.
There are additional settings, like infill overlap, that can affect dimensional accuracy.
Thanks Brent! I plan to do parts which require much better tolerance than 3-6% which is what I’m seeing on my printer out of the box. Specifically, I plan to build a gimbal for a digital SLR. I’m concerned that some of the printed gears may not function properly without getting the tolerance more in line.
I appreciate the links and I’ll have to dig in to them tomorrow. For tonight, I’ll just drink some craft beer and imagine the possibilities.
BTW, I realize I’m probably preaching to the choir here, but I must say this … if better tolerance levels than 3-6% are achievable with the Snapmaker 2, why the hell are the machines flying off the production line without finer calibration? Is it an issue of time for Snapmaker? Seems to me that a ton of questions regarding the printer’s capabilities could be abated if the company just put a little more time into fine tuning the devices before they are sent out. The price point is such that the expectations from the end-user are high.
Having said this, I’m a complete newcomer to 3D printing. So, inaccurate devices may be the norm for 3D printing at this point in history? I may be equating 3D printing with machining because they seem to share a similar background.
because $$$. This isn’t an ultimaker. This is a (for the price point) cheap 3-in-1 that works pretty well at all 3 things. You can easily pay the same or triple for a single-function 3D printer.
You’ll also find changes to the extruder, nozzles, and occasionally filaments, you’ll have to recalibrate. It’s not a set-and-forget system.
As @ElloryJaye alluded to in the other thread, you’re going to be continually disappointed at the level of 3D printer advancement as long as you’re expecting everything to work first try - the level of technological advancement just isn’t there yet.
I added to my post above. Admittedly, I’m new to this so I don’t understand the history as well as I should. I started using vector based printing devices in the 90s, but I was just cutting vinyl. There are a million more factors involved when the 3rd dimension along with various materials from various vendors are added. I suspect that I need to lower my expectations slightly? I thought I let the industry age long enough so that the majority of issues would be addressed, yet here we are. Nonetheless, I’m going to work through the issues because I see the possibilities.
Part of the ultimaker appeal is people pay (way) more for a system that approaches “it just works” levels. But you have to buy their machine, their parts, and their filament for that promise to mean anything.
Interoperability isn’t there industry wide yet. Too much competition for price that drives down quality to just barely acceptable levels. I saw @WilliamBosacker mention this to you in another thread - you have to learn which filaments are acceptable as many have too much dimensional variance and too much moisture to be considered even barely adequate.
This machine was sent flying out the door as the firmware was still being written. Here we are a year after launch and there’s still 66 open issues with just 1 part of the 3 firmware systems that have to work together (Issues · Snapmaker/Snapmaker2-Controller · GitHub) (not even including Luban, a 4th system).
There are many MANY things that could/should be better, but I think they don’t have enough manpower to do it, so they are prioritizing where they can. And factory calibration was apparently not one of those ¯\_(ツ)_/¯
One of the things they did get right: the construction and engineering of the metal components, the geometry of it, it’s pretty good for the intended application. It won’t CNC metal like many people have tried, and occasionally things inside break, but overall it’s pretty decent.
I just threw that brand out for no other reason than it’s expensive, and occasionally regarded well. I have a friend who ran a manufacturing research program using 15 of them and had issues all the time with them not working as they should. Even very expensive machines don’t always work as they are promised.
I’ve been using Hatchbox PLA because I read somewhere that it’s a good product. But I have no frame of reference other than the Snapmaker spool that shipped with the printer. So far, I’ve had better results from the Hatchbox spools than I’ve had from the Snapmaker spool. Which brand(s) do you prefer? I suppose this varies from material to material type, but let’s assume PLA because it’s the low hanging fruit for a someone new to 3D printing.
Addemdum: I’m printing some down and dirty stuff right now so tolerance is not an issue. Frankly I don’t even care about stringing, banding, or anything else right now because I’m just printing fastener (eg. screws, bolts, nuts) organizer boxes that are going to live in a drawer. I have everything on Fast Print and I’m bumping up the work speed to 150%. Just as long as the boxes don’t shatter, I’m good.
I use hatchbox and have had good luck with it. There’s plenty of recommendations from people on here that all seem reasonable (matterhackers, prusament, etc).
The snapmaker spool that came with mine worked. Many more people, however, received wet batches that do not print well at all. I wouldn’t recommend it as it has terrible QA QC and cannot be trusted to be good.
A big part of it is software that isn’t ready to do what has been promised. It’s either beta or in some cases even alpha. Firmware updates are rushed out and not well tested and tend to cause as many problems as they solve.
But I think the biggest problem is that they’re determined to market it as being ‘plug and play’ and super simple for beginners and it isn’t. They’re trying to sell it as doing everything automatically right out of the box and it just isn’t the case. If they would just admit that there are certain things that need to be calibrated manually, then many peoples expectations would be adjusted and they’d have a much better experience. If you look at the user manual for the Prusa, which is a reasonably priced, but very well regarded 3d printer, the first few pages after assembly instructions are a dozen or so calibration tests to help dial in the printer.
-S
Welcome to the world of 3D printing. There is a bit of messing around to get exactly what you want. It’s not SM2.0, it’s just about any. Part of the reason is that filaments have different behaviors. Brand to brand, or even same brand different batches can impact it. Extrusion setting can be a big deal for tolerance. Also, make sure the filament is dry. Moisture will create poorer results and impact tolerance and wear.
If you are printing gears or parts that will experience wear or stress, PLA is not the best. PETG filament is resistant to high temperatures and water, prints stable dimensions, and no shrinkages. PETG combines ABS durability and PLA printability. Little harder to print than PLA but much better for your use case, at least for gears.
Also, from my experience Overture is really consistent in its quality.
@brent113@VaughnDTaylor I think I’ve mentioned it before, but I have an Ultimaker S5 Pro Bundle (came with the filament cabinet that doubles as a table for the printer) the thing is amazing, but 2 things of notable mention.
*Touchless calibration, 100% auto, no input needed. Sure, that’s awesome. Does it justify its price? Hell no. Replacement parts extremely expensive? Hell yes.
*Downside? For its price, it’s print size is frustratingly small.
If it wasn’t for the fact that it was gifted to me, I would have never had one. Are the benefits of entirely automated calibration worth the price tag? Absolutely not. I’m actually quite annoyed by its small bed size as well. I love it but would I ever spend this much money on a printer just for a No-Calibration setup? Absolutely not. While it prints absolutely amazingly, the benefits in my honest opinion do not justify its price. What I was gifted was the best of the best from the current Ultimaker lineup. $10k worth. Personally I think Ultimakers pricing is jack assingly high. The cabinet alone is over $3k
@WilliamBosacker after @VaughnDTaylor posted this, I went to go test mine between all my printers. My S5 was of course the most accurate for hole size, but what shocked me was the fact that my IDEX, which has a severely warped heated bed, was more accurate than my Snapmaker. So I tuned it up and it’s now on point.
