So, I’ve opened PAHT-CF, dried it on X1C bed for more than 10h with recommended 90 C. Then went to print some stuff while using drybox with 10% humidity where PAHT-CF spool was located. Prints came out OK (left model in pic). I swapped filament to PET-CF then and left PAHT-CF in drybox for another 2 weeks.
Today, I decided to print some PAHT-CF again. Dried spool again for way more than 16h on X1C bed at 90 C. After making first print, models were garbage. It is clearly visible, that there is humidity inside still (middle model in pic) - it kinda resembles more a model made out of PET-CF (right model in pic).
I don’t get it. Why didn’t the spool dried like the first time? Or does this filament have some expiration date, once you open, you need to use it in few days, because it gets totally bad after? For a moment I thought I mixed the filaments and I was printing PET-CF, but I swear, it was PAHT-CF.
I think I may have figured it out! When the spool was new, the end piece of filament was nicely tucked inside. But the second time I was drying, it slightly protruded out of the spool, enough to raise the cardboard I was using, so the cardboard wasn’t completely flat with the bed. It must have been that, I just cannot think of any other detail that would have caused such a big difference.
Moisture in the filament isn’t a “yes” or “no” kind of thing. How much moisture the filament has depends on the environmental conditions present while it was absorbing moisture, both temperature and humidity and probably atmospheric pressure all dictate how quickly the filament will absorb moisture and by corollary, how quickly the filament will dry out.
In very humid ambient conditions, the time it takes Nylon to absorb moisture can be very short, hours. And using just heat, with the same ambient humidity, the time it will take to drive that moisture off will be much, much longer.
So every time you dry your filament it’s going to need something different unless it was exposed to the exact same environmental conditions this time vs. last time. You know what wet filament looks like, if you dried your filament and it still looks wet, odds are good that you simply didn’t dry your filament long enough.
My personal opinion is that especially for Nylon, any drying time less than 24 hours is probably too short unless it’s dead of winter and ambient humidity has been low for a while.
From what you said and photos does not look like a Wet filament issue- still possible
Filament can sit for months and years open not a problem especially if in dry box with silica
Time frame very close to a firmware update which was released recently
Have you done firmware update last few weeks?
Quite a few people with unexplained print issues including my self had problem which was solved after the following procedure:
Factory reset , machine calibration ( i think it asks you after factory reset ) if not from the device tab button calibration , or from the touch second menu -Utilities
and then back to normal , do not ask why no idea
If not that the flow looks terrible , the PA is bad , hope you are not using textured PEI plate and autoflow together ( does not work) . Also check you have not changed some how speeds, temperatures and cooling( Unlikely )
EDIT:
The cardboard theory is highly unlikely again nothing impossible
Great, just do not forget the machine calibration it calls lidar calibration , again i think it will ask you after factory reset . and please update if that was the issue , intend to send an ‘angry’ support request to update the firmware update procedure at least , as every time i update firmware some issues
Here’s my setup and I have been happy with the results.
Bambu Labs X1C with AMS “dryer mods” using activated alumina. I’m using Sensor Push sensors which can measure humidity levels from 0 to 90 plus. The sensors alert me whenever the humidity levels in the AMS go above 5%.
Both my PA-CF and PAHT-CF I’m currently using has been stored in their respective AMS units for 3+ months.
With this setup, I’ve never had to dry any of my filament before printing.
Here are some pictures showing the setup along with some test prints.
As an aside, I calibrated one the Sensor Push humidity sensors using 32% calibration test packs. I then calibrated the other sensors using the calibrated one. All sensors were within 1 to 2% of calibrated sensor and the app allows you to add the calibration adjustment if wanted. I also tried Govee and Tempi.fi sensors which worked well, but the Govee sensor embeds ads in their app, and the Tempi.fi app was good but not as a good as the Sensor Push one.
I stopped using the sensors everyone typically uses (i.e. rectangle or round ones). All are rated at best down 10%, although a surprising large number of them won’t read much below 20%. Bad news is the all the connected sensors are relatively expensive. If I recall $75+ USD. Maybe not bad for 1 AMS unit, but if you have 8 of them it adds up.
I also found some 5 to 15% humidity test cards which allows me to sanity check that the sensors were in the ballpark. These cards are typically used in electronics packaging and are available through large electronics distributors but in a pinch can be ordered from Amazon (US) as well.
I’m super happy with the setup. My only real wish would be that changing out the activated alumina was easier. In a perfect world you wouldn’t have to remove the spools to change activated alumina which for 8 AMS units ends up taking a chunk of time. Toying with the idea of creating interchangeable desiccant packs that can be plugged in and removed from the front of the AMS of units. This though would require a fair amount of AMS “dremel mods” though so yeah … probably not happening any time soon. Although maybe BL will see this and make updated AMS units … PLEASE PLEASE PLEASE!
I still havent resolved the issue. My prints do not look like that first model on the left. It is not so bad as the first one, but I wanted to get what I achieved first time. No amount of drying, using SUNLU or printer dryer works. The best I can get is something like @Dragster43 posted (left model), but since I got much better quality first time, I know it is possible. I am starting to wonder if there is some slicer setting that I had the first time…Which slicer setting could affect it? What kind of temp do you use when printing? So far, I used default 290.
did you do factory reset ? and check flow and PA calibrations
EDIT: @djeZo888 if interested read this post , if you dried your filament for more than 20h that is not your problem , for some filaments it takes 20-30h , i use silica color change probe which put together inside the roll and checking the color that is back to normal , but again don’t think this is your problem
No - I did only full recalibration. This is def not PA. It is all surfaces. Drying made it better, but it is not as dry as it was after first dry-run. I am wondering whether spool was quite dry when new and I just dont have hardware to take it down to the level of dryness it came when packed. I wanted to open a new spool, but I am off to a holidays for next day and didn’t want to risk another spool getting bad.
But when I get back, I will try following:
New spool.
Factory restore + calibration.
New hotend.
If all of that doesnt help… well, then I can try to dry with help of vacuum+heat.
But to be honest, I am starting to think that model on the left is not PAHT-CF but PETG-CF… any ideas how I could check it out?
It’s been a journey … here’s about half of the batch from a prior “buying spree” when I was trying to find a set of 10 of them that would read down to 10% consistently. Probably just need to toss them but can’t bring myself to do it just yet.
For what it’s worth, I’m just using the default profile when printing.
For my test print (i.e., parts shown in the picture):
Printer Setting was 0.4mm, 0.20 Standard Quality, Textured PEI
Material Setting was BL PAHT-CF and BL PA-CF integrated profiles (no tweaks).
Printers used have the latest firmware.
Spools were BL’s PAHT-CF and PA-CF.
I used Orcaslicer 1.6.6 to slice/print via the cloud.
For what it’s worth, the one printed that looks “bumpy” is actually extremely smooth. Picture doesn’t do it justice at all. Both objects feel identical if you run your finger along the time and don’t have that bubble/wet feel/look in person.
Thinking out loud… assuming the Fuzzy Skin option has been activated for your project? If you want to share your project file… maybe someone will spot a key setting, etc. If nothing else, I’m sure someone would be willing to test print and that might provide some additional insight.
If anyone has suggestions how to reduce warping and improve dimensional stability right after print is done… some sources suggest using only 2 walls. Having lots of infill is not a good idea, I know that already. And I need to wait for the temp to fall down before even opening the door of the printer. If there are some more tricks, let me know.
Which are spec’d for 2% accuracy from 0% to 99.9% humidity. Simple I2C interface. Probably use an ESP32 with a bare-bones hosted web page to report real time temp/humidity and retrieved logged data.
Cost in parts, probably <$15. Just need a few spare free hours for the code.
Be aware that the DHT22 sensors do have issues with reading drift. I had to replace this sensors in my garage monitor a couple times as they start to read incorrectly until I moved to using a thermocouple.
I went with SHT40s. Seem to be working good. Wanted to try an e-Paper display so I bought a board from DFRobot that has a 250x122 2.13" display built on to an ESP32, so it also has BT and WiFi for remote access (later, when I write the code). As well as a SD Card slot so I could log for long periods there if I want, as well as maybe remembering across power cycles.
The sensor has a heater for the humidity sensor part, that can be run at a few different power/durations to do things like “cook off” any condensation in the sensor. The app notes suggest using it periodically to maximize accuracy and minimize drift, so long as the duty cycle isn’t > 5%, so I’m pulsing it for 100mS every few seconds since power consumption isn’t a consideration.
It grabs date and time over WiFi just for the sake of using the WiFi HW, totally not doing anything but displaying the info. My first “enhancement” will be a simple web server I can hit from my LAN to see the same stuff as what’s on the display.
It plots the RH in the chart in the center. The display is 250 pixels wide, so if I make the chart area 240 pixels wide, I can chart RH levels every 6 minutes for 24 hours total display time. Scrolls right to left (so newest reading is on the right side, oldest on the left side). Right now, it logs a value every time through the main loop, which takes about 1s due to the time query to NTP.POOL.ORG over WAN. I had just breathed heavy on it for 10 seconds or so before taking the picture so there’d be something on the graph (which is about 4 minutes of history at the current logging rate). Not very high vertical resolution. I’m scaling over a 60 pixel high region so each pixel represents about 1.6% RH. But in “real life” the trend will be obvious and that’s all that matters. There’s also a spare pushbutton on the board I’ll used as a “reset” and then it’ll capture min/max values separately from the graph and real-time readout data.
Although maybe BL will see this and make updated AMS units … PLEASE PLEASE PLEASE!
