Request of additional technical data for filaments

Bambu Filament and Accesories
1

This is prompted by an email today about wet filaments. Knowing when one needs to dry filament is a basic question that should be answered by scientific data, not by guesses. To that end, I would suggest that filaments be tested and data like the following be collected to present as part of the technical data sheet (TDS).

Using BL PLA Basic as an example.

The test method should pass scientific rigor. For example, there should be enough sammple size (number of rolls and number of printers), randomization (i.e. filament rolls are randomly assigned to printers), and double-blind (i.e. the testers don’t know what the filament have gone through, the data analyzer gets the data in groups (A,B,C,D,E, etc.) but doesn’t know what the letters mean.

Filament: PLA Basic. Subject: Mechanical Property. Filaments store at 23C for 1 month.

Column 1 Column 2 Column 3 Column 4 E F G
Annealed and dried at 55 C for 8 hours Store in 20% RH, no drying 30% RH, no drying 40% RH, no drying 50% RH, no drying 60% RH, no drying
Young’s Modulus (X-Y)
Young’s Modulus (Z)
Tensile Strength (X-Y)
Tensile Strength (Z)
Breaking Elongation Rate (X-Y)
Breaking Elongation Rate (Z)
Bending Modulus (X-Y)
Bending Modulus (Z)
Bending Strength (X-Y)
Bending Strength (Z)
Impact Strength (X-Y)
Impact Strength (Z)
3

The equipment to do those tests is not cheap, or difficult to specify.

The tests must be the same across all participants.

This requires either designing a test and cheap machine to perform it - which will require everyone to purchase the exact same parts etc

Or using industrial tests like IZOD for which the machines to perform it cost more than an H2C.

4

@BambuLab already does the left most column “Dry at 50C for 8 hours” and have data for that in their TDS. The tests are done following industry standards such as GB/T- 1040 (which is the equivalent of ISO 527-1). So they alreaddy have the equipment, protocol, and know-how and this has been part of their routine testing of filaments.

To fullfil this request, all is needed is to do the same tests on 5 more sets of filaments. In my proposal, it will work like this: Dry enough rolls of filaments at 50 C for 8 hours. Then randomly assign the rolls into 6 groups.

One group go to print the test samples right after the drying, and go through the testing, which will yield the data for the group “Dry at 50 C for 8 Hours”. The other 5 groups go into 5 dehumidifiers setting the temp at 23 C (or 21 or 22 which is more common for a room in the US) and RH at 20%, 30%, 40%, 50%, 60% respectively and store for 30 days.

Then these stored filaments come out and the test specimens are printed, without further drying. Then these specimens are tested the same way as the “Dry at 50C for 8 Hours” group that did not go through storage.

Any additional cost is minimal. If they don’t already have dehumidifiers, buying them is less than pocket money for a company like BL.

GB/T-1040 does require the tests to have a minimum of 5 specimens for testing direction. So that meets the minimum sample size requirement.

It does not have the blind requirement. But that should be fine because the tests are mechanical and the results are number readings from a device. When I mentioned the double-blind requirement, I was also thinking of asking BL to rate the aesthetics aspects of the samples, but decided not to ask for too much in one-go :slight_smile:

5

As far as I know, this topic has already been discussed several times. The short conclusion was that estimating is sufficiently accurate for Bambu - especially considering the type of customers Bambu primarily appeals to.

While it may not directly answer the original request, it does explain why Bambu does not see the necessity. Providing a range or estimate is good enough for 95% of customers.

I’m not saying they shouldn’t do it; I’m just saying they probably won’t.

The few professional customers who actually need this, measure it themselves at their companies. They usually have the equipment lying around anyway - just have the trainee do it as a project for a few days.

Maybe it could be a good idea to reach out to the community - I’m sure there are people who’d be happy to volunteer and take some measurements. It wouldn’t be official Bambu data, of course, but it might still be a helpful approach… maybe?

6

I am working under the assumption that bambu wont do it so it needs a community effort, hence was discussing the equipment and procedures needed.

From art paint I know it’s difficult to keep matching colour and other properties across batches and I assume this is why manufacturers don’t want to give specifications that will lock them in.

I have seen this in filament where 2 rolls of the same colour from the same company don’t match.

Edit the imploacation from this also is also that the test needs to be repeated periodically.

7

A TDS is much more useful when the test conditions are explicit. For everyday buyers, I’d separate it into three layers: material data; handling and compatibility such as net mass, diameter tolerance, spool or refill dimensions, AMS fit, drying guidance, and expected profile; then batch and QA data such as lot, packaging moisture, test method, and sample size. A standard print coupon and a clear distinction between vendor data and community results would make comparisons easier to reproduce.

8

No, performing tests that conform to industry standards such as ISO 527 is beyond ordinary users. To detect if storing at a certain RH level affects the mechanical property of the print needs the accuracy and consistency that can only come from professional equipments and testers that are familiar with the standards.

Here are some of the reasons why this data should be available and why the TDS is the perfect place to include it.

First, why TDS is a great place for it. By presenting the data as TDS, it allows filament makers to keep their existing long-standing recommendations before they are ready to change while giving users data that they can use if they want to. If and how to use the TDS is up to the users, with caveats and disclaimers without explict promise from the makers.

Second, why these data should be available.

It will provide trust-worthy data to clear up the current state of confusions. Currently some dry their filaments before every print while some don’t dry at all. Some store their filaments at 20 % RH or even trying hard to push it lower while some others store them in the open. The data can clear this up.

For example, if it turns out that PLA storing at 30% RH (at 23 C) has no to little adverse effects on the mechanical property, then one doesn’t have to push it any lower. Users can also know that they don’t have to dry PLA after storing for 30 days at 30% RH before the print. This can avoid millions of unnecessary drying hours per year.

Even if BL decides to test and provide this data for the most commonly used filaments such as PLA, PETG, and TPU, or only PLA, the impact would still be huge considering the size of the 3D printing user base.

On the other hand, if storing at 20% RH makes the prints significantly weaker, than we should all be storing them at below 20% RH, even if the print surface may still looks as good as the freshly dried filament (the left most column in my mockup table).

@BambuLab (cc @Bambu_support :slight_smile: ) can take a lead in this and be the first company to provide this data, if only for the fact that BL is targeting the masses most of whom want to spend as little extra prepping time as possible when they want to print something out.