I have had a lot of user feedback in the last few weeks and it appears that the tape mounting method works for some people and doesn't work for others - and I think it's because the back panel of the machines vary in flatness. Some back panels don't sit flat on the machine - I think it's the foam gasket inside that pushes the back panel out unevenly - and the mounting method is only really suitable for attaching 2 flat things together, otherwise the tape lifts off after some time. An additional bracket will stop the filter from falling off but it will not stop the tape lifting off. I will need to figure out another method of attaching it while retaining a relatively good seal. I will come back to this to design a better mounting method. Sorry for the wait!
Purpose
The ideal fume extraction setup should capture all fumes and either completely filter it clean or send it outside through ducts. Even though the P1S and X1C are enclosed, the enclosure has many holes and gaps, the entire printer should be enclosed in a cabinet or zip-up tent with ducting and an additional fan to push the air outside. The fan should provide enough flowrate to ensure the electronics do not get too hot and that, if printing PLA, the chamber doesn't get too hot to cause extruder jams through heat creep. However, depending on your situation, such a setup may be undesirable or unachievable for several reasons.
Perhaps you can't put your printer next to your window and you don't want to run ducting across the room.
Perhaps it's too difficult to seal the window gaps after routing a duct through it. If you live in a hot/cold/noisy area, window sealing could be important.
Perhaps you consider such an ideal setup is overkill for the occasional hobby use.
Perhaps your mom will yell at you.
The NTK Filter is adds a simple VOC and particulate filter behind your P1S or X1C to provide some additional filtration without taking much space or additional electronics. It also has an additional benefit of reducing the sound level of the fan, especially in the more annoying higher frequencies. This is a part of a larger project (Noise and Thermal Kit) but can be used independently.
Performance
Before assessing the performance, let's consider some physics:
A bigger filter creates less head loss (or “back-pressure”). Not only is there more “vent area” for air to pass through, air will move slower across the filter. Typically for an object moving through air, drag is proportional to the square of the air velocity. This means that doubling the filter size will more than halve the head loss.
Slower air movement through a filter increases efficiency of a filter for very fine particles and VOCs. The slower air gives very fine particles and VOCs more time to collide with the filter media through Brownian motion (ultra fine particles are captured through static charge, not by mechanical means).
The total head loss in a system is largely determined by the most restrictive element/s.
Putting a HEPA filter before the VOC filter increases the life of the VOC filter as particulates don't clog the pores of the VOC filter media.
These are some considerations of the P1S and X1C machines:
There is no built-in HEPA filter.
The OEM carbon filter is small and very restrictive but the chamber fan still works sufficiently as it is designed for the filter.
For printing PLA, if you don't want to open the door/lid due to the smell and/or noise, air flowrate is important to keep the chamber temperature down, as noted in this Bambu Lab Wiki Article.
The printers have a lot of gaps and holes where chamber air can escape before being filtered.
With these in mind, the following changes are made to the filtration system to try to increase filtration performance.
The OEM carbon filter (VOC filter) is replaced with a refillable carbon filter cartridge approximately 3.5x the size with a similar thickness for lower head loss and increased filter efficiency.
A HEPA filter stage is added to capture particulates. This is also very large to minimise head loss introduced by this stage.
The HEPA filter is placed before the carbon filter to increase longevity of the carbon filter.
From a couple of days of testing, I have noted the following results.
Can adequate flowrate be maintained to keep a cool chamber temperature when printing PLA for a long time without special build-plates (such as Supertack)?
Yes (with ambient room temperature ~20 deg. C, further testing to be done in summer)
Can the VOC filter efficiency be improved from the OEM setup?
ABS Print: YES - significant improvement when printing ABS with 20% chamber fan according to my sniff-o-meter. However, the ABS smell is still easily noticeable.
PLA Print: YES - completely unnoticeable (this is a big deal to me as my wife and I usually find the smell super annoying!)
I do not have an air quality meter currently. Testing is ongoing.
Here is a simple airflow test video to illustrate the extraction performance:
Limitations
While the NTK Filter improves on the stock filtration system, it is definitely not perfect. VOCs and particulates may still leak out through gaps and holes, you may open the chamber door while printing or the carbon filter stage can lose efficiency. If you are printing some of the nastier materials, you should do as many of these things as you can:
Install a recirculating chamber air filter to supplement the filtration (I'm working on something for this, see below!),
Keep the room well ventilated during (and for some time after) printing
Use a room air filter
Stay out of the printer room during (and for some time after) printing
Print in a room with an extraction fan (perhaps your laundry or an unused bathroom)
Print Material and Settings
Use PETG or a material with better thermal properties with the included print profile. The profile is tuned for Bambu PETG HF and you may need to change the filament profile. PLA is likely to be perfectly fine but I have not tested it.
Clean the plate and use glue!
The built-in support can be broken off then cleaned up with the Bambu scraper or a sharp knife. Any white stress marks from the cutting can be removed using a heat gun or a lighter.
Feel free to add your own print profiles (and keep the profile points!) but please add a good description to help people choose. I've uploaded the STL file of an un-pre-supported main body in case you want to try to support the large overhang in a different way.
Other Materials Required
3x HEPA filters - These are the same filters as on some iLife vacuum robots (~80mmx40mmx15mm) as well as the popular “BentoBox” mod. They are available on Amazon (example from Amazon.de) and Aliexpress as well as many other stores.
Approximately 250mL of activated carbon pellets or granules designed for purifying air (the weight will vary depending on pellet shape and size). They are available on Amazon as well as many other stores. They do not need to be “acid free” as this is not a recirculating air filter - however, do consider getting carbon that's not “acid washed” so you can use the same filter media for a recirculating filter. I bought ones that look like chips/shards but the cylindrical pellet ones will introduce less air resistance.
The carbon I bought was not the Nevermore branded carbon but had these 3 lines in the description which were promising: "Highly effective and exclusively water vapor-activated coal. Not impregnated. Not washed", "The entire production takes place without chemical additives. Not treated with acid, therefore no oxidation", "Preferably also used as a recirculation cleaner in 3D printer housings. With pH value of 9-11, no more rust on bare metal surface". I'll add a link to it once I verify that it's OK for internal recirculating filter use.
1mm thick VHB tape ("nano tape" and “mounting tape” are good search terms) . They are clear acrylic tapes and can be re-positioned and removed cleanly. There are many cheap unbranded options available on Amazon (example from Amazon.de)
2x M3x12 machine screws (can be M3x16 but these take longer to screw in)
Note: if you use an AMS Hub, you will need to reposition it a little upwards as it is very close to the exhaust grille in its default position.
Remove the OEM carbon filter. If this is fairly new, you can cut this open and use the pellets in the new filter. If it's old, throw it away.
Attach the VHB/nano tape to the manifold body in a rectangular pattern as shown in the photo. Ensure that the tape is continuous around the rectangle to get a good seal and do not cover the small hole as these give access to the case screws.
Remove the tape backing, align it the best you can and press firmly.
Install the 2x M3 nuts into the nut pockets.
Install the HEPA filters.
Fill the carbon cartridge with carbon media. Fill as much as you can without having the lid bulge out.
Close the carbon cartridge lid and ensure that the holes line up.
Attach the carbon cartridge to the manifold using the 2x M3x12 screws.
Attach a cardboard flap to the poop chute and tape over some of the large holes and gaps to make it easier to maintain negative pressure in the print chamber.
Adjust your filament start g-code as appropriate to ensure that the chamber fan turns on automatically. This is an example of my ABS-GF filament start g-code: M106 P3 S51 ; S51 = 20%
Frequently Asked Questions
I saw a video where someone said HEPA filters are marketing gimmicks and MERV 14 filters are the best! Why HEPA?
For recirculating filters such as room air filtration, a less restrictive (not High Efficiency) filters are often better because a given fan can generate a higher flowrate and filter faster. Many off-the-shelf room filters use HEPA filters and this is a poor filter choice and definitely a marketing trick. People that are sensitive to air particulates often make their own filters using MERV 14 filter panels used in the HVAC industry. This is not the case for a single-pass filter such as this one - we don't need a huge flowrate and very little of the exhausted air will come back through the filter after getting into your room so HEPA is the correct choice in this case.
In in-chamber recirculating filters (such as the popular BentoBox), even though HEPA may not be the best choice technically due to the lower flowrate, lower resistance filters may not be available in the form factor suitable for inside a 3D printer. Having a filter you can source easily in many countries is a good design choice.
I'll be adding more to this as I continue to get feedback and questions!
Support
If this model helped you, please leave a thumbs up and a boost. If you would like to donate or sell these commercially, please join a membership!
Boost Me (for free)
Boost button is right here!
Membership
Donation and commercial license options are in here!
Join
And stay tuned for “AUX Pro” - an auxiliary fan duct with a filter module option to turn it into a slicer/app controlled high flow chamber filter without any added electronics.
Okay. I previously tried to explain why your design isn’t ideal (bad idea) on reddit, but since you’re confident in your approach, let’s discuss this professionally.
First thing first, you cheated in the airflow demonstration video by blocking the purge chute port and turn off the aux fan. Without blocking the purge chute, and set the aux fan on at 100% which is at 50 CFM, how's the airflow?
Anyway, after having your latest design analyzed using professional airflow stimulation software, here are some insights provided by an HVAC engineer (Whom is my friend):
Your explanation about static pressure and airflow reduction does not fully consider the chamber fan's realistic capabilities. The Bambu Lab chamber fan (P1S, X1C) is a small axial fan rated at around 92–100 CFM only under ideal, no-load conditions. In practice, adding multiple small HEPA filters combined with activated carbon granules and also with multiple supporting structures blocking the airflow will dramatically increase static pressure, significantly reducing actual airflow. This real-world performance limitation isn't fully addressed in your design.
Using activated carbon granules directly behind the HEPA filters, as your design suggests, is problematic. Granular activated carbon (if not acid-washed) inherently produces fine carbon dust, especially when airflow directly impacts the granules. This configuration would inevitably release carbon dust into the airflow downstream, potentially contaminating the clean air stream and the room. So either change it to acid-washed granular activated carbon or put HEPA filter after the granular activated carbon. He understands this placement will likely increasing clogging of the HEPA filter. But the primary propose of this air filtration system is to protect human health. A shorter HEPA filter life and more frequent filter replacement is a necessary trade off.
In the airflow stimulation test, the poor duct initial transition in your design significantly reduces the effective of the larger cross-sectional area immediately following the fan outlet. This substantial area reduction causes unnecessary airflow restriction and increased turbulence, resulting in higher static pressure and reduced efficiency. Additionally, the internal support structures inside your duct further compound this issue by obstructing airflow and increasing turbulence.Your theoretical assumptions regarding airflow and filter efficiency don't align closely enough with the practical limitations imposed by the specific hardware (fan and duct geometry) you're using.
Here is my comment:
Open to different opinions. The current design is not ideal and also may cause potential damage to the chamber fan. Add an additional boost fan in your design may help with the airflow issue.
Hi, thanks for your comment and taking time out of your day to use professional software to simulate the setup. I understand that your HVAC friend may be concerned as what he has to think a lot about in a recirculating air environment is "air changes per hour" and flowrate is very important. In such an environment, getting the most out of your fan and reducing head loss at each stage is very important.
However, maximum airflow is not the goal here. This is a single-pass system. The goal is to generate some negative pressure to reduce unfiltered air getting out of the gaps, and optimising for fan efficiency or flowrate is not what I'm trying to do. I'm not filtering a room. The goal is to improve on the stock system to work well-enough for PLA, run with 10-30% fan with other materials, be compact, doesn't require additional electronics and be easy to install - I have achieved all of this, and outlined the limitations for people that may want a better system.
The load on the chamber axial fan is OK for the reasons I've already given. And it sounds like you're saying adding "multiple" HEPA filters is a bad thing - but putting "multiple" in parallel as I have, like resistors in electronics, reduces losses - maybe this fact got lost in transmission when you were talking to your HVAC friend.
I don't care about a little bit of barely noticeable carbon dust coming out in the first run - this is harmeless - just give the filter cartridge a vacuum if you're worried. To whoever's reading, I don't suggest people get acid-washed carbon as the commenter suggest so the same carbon can be used for a recirculating filter (acid-washed carbon will cause the steel parts in the printer to rust rapidly).
You are right in that the design is not "ideal" from a pure filtration point of view, and that's why the model description STARTS with a description of an ideal filtration setup. However, in engineering, there are many interests to balance and different designs meet different interests better. This setup (along with the new recirculating filter I'm working on) may be the ideal setup for many people.
Hi, can you take a video showing the actual airflow without blocking the purge chute and set the aux fan at 100%?
And I don't know where did you get the idea about negative pressure. The term “negative pressure,” in practical HVAC and filtration contexts, refers to an enclosed (or mostly sealed) space where air is actively extracted at a rate greater than it enters, causing a slight vacuum or pressure below the ambient air pressure outside. It is created only when airflow extraction exceeds incoming air through limited openings, maintaining pressure below ambient. Since a 3D printer enclosure typically has multiple unsealed openings (unless like what you did, by blocking the purge chute and turn off the aux fan), achieving meaningful negative pressure isn’t practical. Your system is essentially creating directional airflow, not negative pressure. The internal enclosure pressure remains nearly equal to the outside atmosphere, meaning airflow containment relies purely on consistent directional airflow, not an actual negative pressure condition. If you still disagree, I strongly recommend verifying this with a digital differential manometer—we performed such testing during our design process for the P1S exhaust adapter and confirmed these conditions clearly.
Regarding activated carbon, acid-washed activated carbon granules typically produce minimal or no dust, making them ideal for use outside the printer chamber. The potential for corrosion you mentioned applies primarily when acid-washed carbon is used directly inside the printer chamber, not externally.
About the multiple HEPA filers, what he means is that your current design, which uses multiple small HEPA filters and activated carbon granules in a two-stage filtration arrangement, inherently requires additional internal support structures. These structures, which are not optimized, significantly obstruct airflow and increase turbulence, resulting in a notable increase in static pressure. During the test, even with the chamber fan at maximum speed, practical airflow could drop significantly, likely into the 25–45 CFM range, depending on the real world performance of your HEPA filters.
We both highly suggest you conduct some studies to better understand and optimize these design decisions.
Here is a comment from an HVAC engineer: I can't think of a better way to improve the machine's source filtration system without adding extra electronics.
And a note for the engineer's friend: To generate airflow through the gaps of the printer, a negative pressure inside is absolutely necessary, even if it's minimal. And just because it's small doesn't mean it's insufficient.
Good job!
Been looking for an elegant solution to this. I’m limited where I can put my printer so this is near perfect for most of my printing needs. Profile was fine. The added support like another user said was a bit hard to remove, but easy enough to clean up (and I don’t plan to be looking at the back of my printer very much). Excellent directions. You can see the thought that went into the design. Easy to replace filters and carbon media. I’m super stoked on this print.
Excited to see how this works to supplement my Bento Box! My only recommendation is that the support piece (fin?) not be quite so embedded; I had to cut it off which left a bit of a blemish that I needed to touch up.
The whole thing printed very well. A truly beautiful design. Requires adding support under the filter (see photo), as it tends to detach. I installed the HEPA filters recommended by the author. Unfortunately, the air has great difficulty escaping the filter even after removing the original filter from the printer. This puts a lot of pressure on the fan's ball bearing, causing the bearing to seize and the fan to stop working. So be warned of the danger of ending up with a non-functional fan
Its absolutely not perfect and it could be improved, but for a V1 product its great. Havent tested performance yet, will need to be tested.
My assembly process was "ok". The designed supports are in my opinion filament waste and useless. Stock slicer supports are better.
Also, the tolerances on the carbon part is to tight. Maybe a little looser with magnets? Maybe a way to fasten it extra with the bolts in the printer would be great but double sided tape works for now.
Great V1 product, needs some work, thanks!
First, I love this print. Everything is tight and fits very well together. I filters fit great and the screw/nuts work well. (I would like a screw insert option as the nuts can fall out when changing the filters. I super glued them, but an insert would be nice). I went to install it on my P1S, but discovered the mounting holes don't line up with the printers OEM case screws. It looks like you designed the filter to be used exclusively with the wall you designed. I don't plan to print the wall, so was bummed when it didn't work. I thought just the adhesive tape seal could work, but the filter is too heavy. After reading comments and your write-up more carefully (great write-up BTW), it appears the oval shaft/through hole is for access to one of the OEM case screws. Suggestion: could you design a plug of some kind to fit in the shaft (maybe design a small shoulder in the shaft to catch the plug) that would adapt it to use the OEM case screw(s) so users like me could have the option to mount this filter without the wall?
Love the design and it printed easily. I just can't use it and I really want to. Boosted.
Hi there, sorry it's not working as intended! I found that mine stayed put even without using the 3 screws that attached to the side panel so I thought it should be fine. Your idea of using the OEM case screw is good but the screw is a self tapping sheet metal screw and it'd definitely need to be longer - that would be difficult to source I think, and a bit annoying to get a special screw for one spot. Instead, what do you think about a small secondary piece that sticks to the curved surface of the printer corner with the same tape, which then screws to the filter manifold to provide support for the weight? This way you won't need to reprint anything but just print an extra smaller piece. Hope that makes sense.
Thanks for the feedback. I think I figured out the reason why the tape works for some and doesn't work for others - the back panel of the machines aren't flat. Some back panels don't sit flat on the machine - I think it's the foam gasket inside that pushes the back panel out unevenly - and the mounting method is only really suitable for attaching 2 flat things together, otherwise the tape lifts off after some time. An additional bracket will stop the filter from falling off but it will not stop the tape lifting off. I will need to figure out another method of attaching it while retaining a relatively good seal. It would need to be something like strong magnets with some kind of gasket or using screws that go through the back panel.
Would it be possible to make a version of this filter that uses multiple standard Bambu charcoal filters, or better yet, the Voxel X-Filter (combined HEPA and charcoal and the same dimensions as the Bambu filter)? This would reduce the back pressure, increase filtration, and use standard filters.
it would be but I don’t think many people would use it on account of the cost… But maybe I’m just speaking for myself, did you look at the prices? Would you?
I'll be honest, I only print with PLA & PETG so will more than likely never print and use this..! BUT I admire the time and effort and write up (very interesting reading about air flow etc) therefore in my humble opinion you deserve a BOOST.
Thank you and I'm glad you think it's concise, even at the length that it is! I'm resisting the urge to get a LLM AI to tidy it up.
0
Reply
License
This user content is licensed under a Standard Digital File License.
You shall not share, sub-license, sell, rent, host, transfer, or distribute in any way the digital or 3D printed versions of this object, nor any other derivative work of this object in its digital or physical format (including - but not limited to - remixes of this object, and hosting on other digital platforms). The objects may not be used without permission in any way whatsoever in which you charge money, or collect fees.
Comment & Rating (63)