Enclosures and Ventilation (of the machine, not the print)

Short Background:
I recently convinced my employeer that we needed a Taz 5. It has been printing nearly daily since we got it–building parts for an enclosure in which it will reside. The enclosure includes a filter for scrubbing the nanoparticles and odors and a blower to do the work. Inlets will (for now) simply be “leaks” where the panels meet the frame, but it should still maintain some nominal negative pressure. If not, it will be upgraded to more filter flow capacity.

Moving on to the point!

The box fits snugly (not TOO snugly) around the Taz 5, but currently does not have any external holes for the electronics enclosure (which clearly has a fan and vent holes). There is probably a 3/4" gap between those holes and the external wall.

Question 1: Will this be enough to satisfy the ventilation (cooling?) requirements for the electronics during a print job? Or will things get too toasty? Those of you who’ve built hot boxes will probably know the answer. If I recall, I’ve seen some with ducts to the electronics fan. Temp in my office just inside one of the non-fan holes is about 83F/28C.

Question 2: Can I shove the power supply in the box too? It’s got a substantially loud fan on it and this would quiet it down a bunch. But since it has a fan, I assume it gets warm. Too warm? Quick thermocouple measurement says it’s dumping hot air at around 90F/32C or so which is pretty toasty.

Add the two above to a heated bed and warm lump of plastic and it might be too much. What do you all think?


You need to duct cold air into the electronics enclosure. If you don’t have cool air flowing to that box, it will cook the $150 rambo board.

You do not want the power supply inside the heated enclosure for any reason ever. It will melt. Build an acoustic foam noise baffle box separate for it if you need to, but don’t run hot air through a power supply unless you like buying new power supplies.

a filter for scrubbing the nanoparticles and odors

This is not usually a problem, even with multiple 3D printers operating indoors. Are you operating the Taz in a cleanroom?

  1. Duct the electronics. Check.
  2. PSU stays outside. Check.

It’s going to be in my office. I’m taking the better-safe-than-sorry approach to filtration.

Some research I did a while back strongly suggested toxic emissions from 3D printers is negligible.


Looking at “Acrylonitrile Butadiene Styrene Copolymers (ABS): Pyrolysis
and Combustion Products and their Toxicity - A Review of the
Literature”, it seems styrene is the dominant organic volatile,
occurring as <10mg per gram of material pyrolized. Supposing 10% of our
plastic pyrolyzed into vapor as 1kg was printed, approximately 1g of
styrene vapor would be produced. If this vapor remained within a 27m^3
area, and the air had a mass of 1kg per meter, styrene gas would amount
to 1 part per 27000 parts by weight, or about 37ppm. The Permissible
Exposure Limit for styrene vapor appears to be 50ppm.

These figures represent a significant overestimation. For starters, we
are highly unlikely to print 1kg of ABS before ambient air circulates
well beyond 27m^3, and the pyrolysis rate is almost certainly far below
10%. Further, styrene gas is heavier than air. Finally, other organic
volatiles appear to be produced at rates one or two orders of magnitudes
lower, pushing them deeper into the very low ppm range.

Disclaimer: Unprofessional estimate, not tested, don’t sue.

Understood. I have seen a few opinions leaning both ways as well as the research done at IIT in Chicago and while it might be harmless, it will still be harmless if filtered.

Right now the smell of hot PLA is not particularly bothersome, but it is still nice to have fresh air.

Ok. In that case, since you will be building the encousre, be sure to include a temperature-controlled fan. Ambient temperatures around 30-40degC are supposed to be ideal for printing.

I agree with Piercet on both accounts. If the fan is really bothersome I would highly recommend you change it to a Noctua brand fan, they are incredibly quiet and reliable. Every single fan in my office equipment is getting switched over as they die (4 so far) to the Noctua fans.

I have pretty good hearing and I have to get within 4 inches of the fans to hear anything while the old fans I can hear from across the room.

The noise is not that bad. Considering we have a large vent hood with an obnoxiously loud fan in the suspended ceiling in addition to a clean bench with another loud fan, this thing isn’t really an issue.

I should mention the box is pretty much complete at this point. I added the small duct to get air to the electronics and as a quick test, I put a TC about 2 inches from the center top of the interior of the box and heated up the bed and extruder and printed a partial layer. With the filter fan on at low duty, the TC read about 80F / 27C. I suspect that’ll increase some after more time but it’s a reasonable start.

Pictures. You can see the hack filter/blower in the rear left and RAMBo duct if you look close.

And one bonus shot that shows how I shoehorned the printer into a box made from standard 2 ft dimension parts from McMaster-Carr:

the Y axis bed sticks out the door a little… still totally contained when the door is closed and everything has full range of motion.

oh and regarding the above photos, all the black parts used on the box were printed except the handle. It was a lot of fun getting the printer set up and running to make useful parts for its own box. The learning curve on the Taz was very reasonable for me. It does help that I found a Taz 4 STEP assembly and that McMaster-Carr provides most of their stuff with 3D models. 20+ years experience in 3D CAD also makes it pretty easy to get things done.

2 weeks ago I would not have been able to discover I needed a fan duct and be able to post pictures of it a couple hours later. I’m in love.

Hi y8s,
Would it be possible for you to share more details on the case you made(materials, size’s, process)?

iv been considering a Taz 5 for a while now but also wanted a enclosure but there are very few encloses similar to your but not many good looking for the recent Taz Printer.

if you are unable to or do not wish to share details i will understand.
thank you,

I can definitely write up the materials list, but it may take a while. I’m pretty slammed at work. The short version is that I started out with www.mcmaster.com 1 inch square extrusion and the 3-way corner connectors and some 1/4" thick PETG 24" x 24" panels. The latches are magnetic and the hinge is a 24" long piece of piano hinge.

I made most of the rest on the printer. I have 3D files, though they aren’t my best designs. When I get back to this topic, I’ll post everything I can.

thank you that sounds great, i look forward to it!

Here’s my variation on y8s’s enclosure design. I used one translucent side panel, one clear front panel, and the rest are black panels. The main improvements are the lighting and the use of lift-off style door hinges. The enclosure makes it much easier to successfully print ABS without having air drafts cause warping problems.

Here’s the BOM…
2 (McMaster) 1151A83 Surface-Mount Lift-Off Hinge with Holes; Left Hand, 304 Stainless Steel, 3" Leaf Height, 2" Width
4 (McMaster) 47065T242 External Connector, 1/4" Thread Size for 1" High Aluminum T-Slotted Framing Extrusion
8 (McMaster) 47065T244 3-Way External Connector, 1/4" Thread Size for 1" Aluminum T-Slotted Framing Extrusion
1 (McMaster) 47065T164 Handle for 1" High Aluminum T-Slotted Framing Extrusion
16 (McMaster) 47065T101 Aluminum T-Slotted Framing Extrusion; Single Profile, 1" Size, Solid, 2’ Long
1 (McMaster) pack (4) 47065T139 Compact End-Feed Fastener, 1/4"-20 Thread for Aluminum T-Slotted Framing Extrusion
2 (McMaster) packs (4) 47065T147 Dual End-Feed Fastener for 1" Single & 2" Quad Aluminum T-Slotted Framing Extrusion
1 (McMaster) box (50) 91255A269 Alloy Steel Button-Head Socket Cap Screw; Black-Oxide, 10-32 Thread, 3/4" Long
1 (McMaster) box (50) 91255A267 Black-Oxide, 10-32 Thread, 5/8" Long; Alloy Steel Button-Head Socket Cap Screw
1 (McMaster) pack (100) 90730A411 Type 18-8 Stainless Steel Narrow Hex Nut 10-32 Thread Size, 5/16" Wide, 7/64" High
1 (McMaster) pack (100) 96765A125 Black-Oxide 18-8 Stainless Steel Flat Washer Number 10 Screw Size, 0.203" ID, 0.438" OD
1 (McMaster) pack (25) 91255A522 Alloy Steel Button-Head Socket Cap Screw Black-Oxide, 1/4"-20 Thread, 5/16" Long
1 (McMaster) pack (4) 5537T161 Compact End-Feed Fastener, M5 Thread Size for Aluminum T-Slotted Framing Extrusion
1 (generic) .110" clear acrylic 2’x2’
1 (generic) .110" gray translucent acrylic 2’x2’
1 (generic) .250" black acrylic 2’x2’
2 (generic) black plastic panel 2’x2’
1 (eBay) LED Flexible Light Bed Table Desk Lamp 30cm Tube Style High Power 5W Goose Neck
1 (eBay) 3D Printer Filament Spool Holder Stand - Ball Bearing - Universal
1 (eBay) 5m Super Bright 12V 5630 SMD 300 Leds Lamp Flexible Strip
1 (eBay) 12V 4A 48W Power Supply AC to DC Adapter for Flexible LED Strip Light
1 (eBay) RF IR Wireless Remote Switch Controller Dimmer for LED Strip Light
4 (K&J Magnetics) D48-N52 1/4" dia. x 1/2" thick Nickel Plated Axially Magnetized
2 (Home Depot) weatherproofing strip
2 (custom printed) wire clips
1 (custom printed) stabilizer bracket
44 (custom printed) .110" panel brackets
12 (custom printed) .250" panel brackets
4 (custom printed) door magnet brackets
1 (custom printed) lamp bracket

The custom printed part files are attached.
custom printed parts.zip (259 KB)

Thanks for BOM Printhead.

Is the electronics enclosure able to draw in outside air similar to y8s’s approach? If so, was it a manual cut in the acrylic, or a laser?

I have a Raspberry Pi integrated into my enclosure that drives the printer using OctoPi, so I’m very interested in keeping things cool in there.


Hey that’s awesome. The mood lighting and tint are pretty sweet.

I ended up trying and failing to make a few different spool holders and ultimately the spool sits just where yours does. I put it in a small Akro Bin style bin box to keep it in one place.

I loosened the screws holding the Lulzbot control box and raised it a few mm so that it clears the 1" enclosure rail (be careful doing this, as one of the screws will interfere with the Z-axis if the control box is raised; I had to remove the screw). So the control box sticks out through an opening in the side panel and no duct is needed. The side and back panels are 1/16" black ABS, which can be cut by hand with a utility knife. Here’s a picture of the control box peeking out of the side…

The enclosure temps stabilize in the 90’s so a RasPi inside there will probably be fine.

Very interesting.

My Taz5 fit between the walls. I also tried to raise the control box with the same issue. I managed to get it to a point where it all squeezes in nicely.

I did have to cut an ugly hole and make a short duct though.

It Looks like Lulzbot may be putting out information maybe even a “Kit” for making a box, though it is labled as a cat gaurd but still would work for most people. though personal i would makes ure to have the bottom panel as well so it is a fully enclosed box and add the air filters.

I dont know if this is complete since it doesn’t seam to have a proper link but looks like it was designed to sit on top a Taz 4/5 printer

Link: https://ohai.lulzbot.com/project/cat_guard/

how much would you pay for a kit like this? 400, 450, 500?

The Cat Guard is not a particularly clever design, and I’m surprised that there are no printed parts used. It does have a clean look due to the slide-in panels, but the slide-in panels make the T-slots unusable for other purposes. You’ll need a lot of space for the enclosure and to swing that door. The door design is poor, so it’s hard to make fit well and there are large gaps at the top and bottom. Should have used lift-off hinges for the door (but then you’d have to specify if you want a left or right handed door). Much nicer magnetic door latches can be made; they’re printable parts and they’re cheap to make. There’s little provision for running cables out of the box; it needs a grommet, which is another printable part. Ideally an enclosure will have a mechanism to anchor the Lulzbot printer to the enclosure so they both stay in place. Considering it’s possible to build a really nice one-off enclosure with all these features plus lights, fans, etc. for about $500, a mass produced Cat Guard shouldn’t cost more than $200.