1×5.25" Optical Disk Drive Bay to 4×2.5" up to 15mm thick

This print is a conversion from now mostly useless optical drive bays (discs/DVD/CD bays), AKA 5.25" drives. The adapter will fit up to 4 2.5" drives instead and can fit all standard formats (so each of the 7mm, 9.5mm, and 15mm thick variants). Essentially, you can turn a single 5.25" bay into up to 4 of the highest performance 2.5" HDDs or SSDs for all your NAS needs by using any of your old and useless cases.

Looking around, this also seems to be a very high density package, as 3 5.25" bays could give you space for 12 15mm thick drives (ie intel's commercial grade SSDs).

I made this as I'm planning to turn this old Dell Optiplex into a simple server, I had 2.5" drives lying around, and I lost Dell's stupid storage drive tray adapter because of course it's proprietary and nothing else will fit.

TL;DR: Turn any old CD or DVD or empty 5.25" drive bays into a high density 2.5" drive array.

HARDWARE REQUIRED

4 × 2.5" drive screws per drive you plan to install. This means 16 screws to fill the bay. These screws usually come with computer cases when you purchase them, but if you're planning to install 2.5" drives, you should have them anyway.

That's it for hardware…

IMPORTANT INFORMATION

• The adapter is meant to be inserted from the front of the case, like most 5.25" drives.
• The 2.5" drives are mounted using their 4 bottom screw mounts. The holes inside the print aren't threaded, the screws will simply sandwich the print.
• High performance HDDs may be able to go above 50°C, at which point PLA will start to soften. I don't know about SSDs either, but PLA might not be the best choice, but could still work (especially if the case has good ventilation).
• This model was made for a slim tower from an old Dell Optiplex. You might need to edit the STL in Blender if anything is likely to get in the way.
• Power cables are particularly tight in specific configurations (depends on case) but can still be fitted with a little encouragement.
• This adapter is held in place by friction, not screws. Please be careful during transportation 

PRINT INSTRUCTIONS

• 2 walls using 0.4mm wide lines/nozzle.
• 0% infill, else the print will take forever and go into vibro mode every layer…
• 4 floor and 4 ceilings (or at least 3).
• No supports, and use bridge options if possible
  • If your printer can't bridge very well, enable supports. That being said, other than for bridging, no supports should be required.
• Print with the thin front place of the adapter facing the print bed
• Don't worry about “weird behaviour”, it's normal :)

For me, this printed in 2hrs using a heavily modded E3P. Not uploading the file because someone will absolutely break their printer with it, but if you happen to have an E3P running highly tunned Klipper, a SpeedDriver Direct Drive, low tension filament support, coated glass bed, and belt tensioners, ask me the gcode and i'll happily give it out.

DESIGN METHOD

This model was designed by taking into account that a 2.5" drive seemed to fit at least 4 times inside a 5.25" drive bay's volume. Actually, if only using 7mm thick ones, you can easily fit 6, but this was designed taking into account old computer cases that need this many drives are likely low end NAS boxes, so space was left for beefier 2.5" drives up to 15mm thick (aka the chunkiest out there).

I used Fusion360 and based myself off of official measurements for 5.25" and 2.5" drives, as well as measurements I took by myself. In the end I used a reference 2.5" drive from GrabCAD to have something more precise.

The configuration is 2 upright drives at the bottom, and 2 upside down drives at the top. This allows to attach the drives using their bottom screw mounts, meaning I didn't need to worry about how to access the interior side-screws if there were 2 drives side by side. The tradeoff is that if the 5.25" drive bay is inside a tight enclosure, it's extremely hard to connect power to the drives in most if not all configurations, but it's still possible (from experience ;3).

Then, I printed it a few times and fixed what was wrong or what I broke in a previous fix. I ended up with this design, which is sound as far as I know. The only “hard” thing about it is that hard drives can reach temperatures above PLA's glass transition temperature (when it becomes malleable/prone to deform), meaning PLA shouldn't be used if possible. PLA is only 10°C under “max” operating temperatures under heavy load for HDDs, but this is not going to cut it for a server/NAS. I recommend using ABS or PETG, but as I don't have those materials, I can't comment on anything other than that they can withstand the heat of the drives. The front ventilation holes are there because of this.

CHANGE LOG:

Revision 1: Fixed an issue with ventilation and clearance

Revision 2: Drastically increased ventilation potential and fit as well as installation tolerance

Revision 3: Fixed mesh errors. Fixed clearance errors. Increased screw clearance and ease of installation for the drives.

Revision 4: Fixed remaining mesh errors that were caused by lazy modelling early on. Likely final.