Prusa Mk2S to 2.5 Bear

If we have to fix it ……

My Prusa i3 Mk2 has been a solid performer as a printer. There have been some issues with the electrics however. The power supply that came with it shorted out. Twice. In two different spots.

Loose and shorted input voltage selector
Arcing caused this solder joint to fail a month later

I replaced that with a genuine Meanwell 350. Fixed.

The next problem was a “thermal runaway” error that turned out to be a burnt connector for the heat bed. Internet search popped hundreds of instances of this particular problem along with the most bulletproof fix, hard solder the connection to the board. I decided to go a step further and replaced the wiring with silicone jacketed high strand wire and added abrasion resistance and an XT60 connector to make heat bed removal easier since it would have been hard soldered at both ends otherwise.

For a company to ship with this KNOWN issue boggles my mind. Could have ended very very badly.
Custom cable installed on the Rambo end, heat bed got the same treatment

When the MK2.5 kit was released it seemed like a good way to get the new extruder, print surface etc. at a reasonable cost so I ordered it. Many many months later news trickled in that the kits would be shipped but a key component would not be part of the advertised bundle. Prusa cannot manufacture the powder coated print surfaces, at all…complete vaporware. They issued a 10$ voucher or some other pathetic fix and shipped the steel sheet surface with regular stick on PEI sheets. Whatever..”fool me once….”  and all that.

My printer had been performing well since the heatbed fix and I did not want to pull it apart JUST to apply an upgrade. I decided to wait until another issue required maintenance, and would then do the 2 to 2.5 upgrade. In the meantime I found a frame kit that replaces the threaded rod and plate on all 2.x Prusas with extruded aluminium members. This kit is called the Bear upgrade , I purchased the required stock and hardware for that and will do a custom upgrade from a mostly stock MK2S up to a MK2.5 Bear.  Today I got another “thermal runaway error”, I have confidence in my soldering so I fear the board may be compromised…lets find out. First step is to disassemble the existing configuration.

Disassembly is complete enough for me…

Step 01

Link to Step 01 assembly instructions

The Bear kit comes with excellent instructions. The first step is to cut the extrusions to length for the new frame. I bought a 100t blade since I cut aluminium occasionally and have been destroying my regular chop saw blades and leaving a horrible burr on the cut parts. Some lubrication on the blade is recommended, I used Brute Lube multi-purpose cutting wax. This setup in a regular miter saw worked great, did not even need to file the edges of the cut.

To relieve the edges of the 90° Joining Plates , I used a Fredrik Dick Equaling File #2 cut for the rough removal then 500 and 1200 Emory Cloth for a reasonable finish. This combination took about an hour including the  full relief on plates in the steps to come.

Set up for edges after initial test
F Dick file from RioGrand
Edge relief close up


Pay attention to the exact location of the screws and retaining nuts. You should have the 4 outermost corner holes empty when the next two steps are complete.

Frame rough assembly (partial)

Step 02

Link to step 02 assembly instructions

This step adds the fourth side to the frame, dimensions and squares the assembly. 

Retaining nuts installed and frame rough assembly

I used some 3-2-1 blocks to verify that the frame was square. Using the Z axis beam to determine spacing was an excellent design choice by the designer.

3-2-1 blocks to verify square corners

Step 03

Link to step 03 assembly instructions

Step 3 is an easy one. Add the feet to the bottom corners and add some retaining nuts for future use.  If you followed the illustrations properly (I did not)  you will have the four outermost holes in the 90° Joining Plates available for this step.

Step 04

Link to step 04 assembly instructions

In the last step we made sure the frame was square, now we will make it level. Do not assume your workbench or table is perfectly flat use an iron table saw top or in a perfect scenario a surface plate. 

Use caution when handling the frame at this point the loose T-nuts will slide out of the tracks and scatter under every nearby object, some never to be seen again.

I used a table saw as my reference plate, my surface plate is too small

Step 05

Link to step 05 assembly instructions

Tapping one end of each upright is a pretty straight forward operation. I used a standard hand tapping handle with a little tap magic cutting oil. I went about 12mm deep which is the length of the longest M5 screw I think this build calls for.

I tapped the hole since I had the tools. The self tapping screw option is just as good.

Step 06

Link to step 06 assembly instructions

Like in the first step the diagonal edges should be relieved to prevent wire chafing. In addition on 4 of the plates , relieving ALL of the edges is recommended. I did this step even though I plan on running all wiring through a nylon braided conduit for abrasion resistance and a clean appearance. 

Close up of one of the edges after filing

After completing the assembly I realized that my tapped holes in the uprights were facing down. I only tapped one end of each upright. I figured that Murphy’s Law would show I missed an instruction asking that the tapped ends be facing up. To my surprise there is no indication in step 06. I also checked and the tapping instructions also do not mention tapping 4 holes or 8. Since this is complete and I have tightened it after squaring, Im not going to do anything at this point, if the tapped ends need to be up in a later step Ill just tap those in place.

Step 07

Link to step 07 assembly instructions

Here is where it starts to look like a thing. Using the pre-installed t-nuts its a simple matter of bolting the upright loosely in place.

Build helper in place 

My build helper was .1 mm under at 95.9 so I added 2 layers of masking tape to one end and brought it to 96 mm. 3-2-1 blocks were used again to make sure everything was square once tightened down.

Z axis installed and verified

The final diagram for this step shows a measurement from the top of the bottom rail to the underside of the z axis top gantry ( 316 mm). Unless you have a way to offset or translate one of those points this is going to be hard to measure accurately with common rulers and such. I brought the assembly back to the table saw and measured from the table to the underside. I then subtracted about the height of the bottom rail and it was about right and more importantly it was consistent side to side.

Step 08

Link to step 08 assembly instructions


Today is the Tuesday before Thanksgiving and I have hit a show stopper….

The makerspace I belong to had all of its windows replaced this month and during the process some of the cubicles got windblown. Mine was one of them and some boxes got knocked off the shelves scattering the parts around. I came in and cleaned up the mess, a little annoying but no real harm done.

In this step of the build I need 4 rod holders, I have one. Maybe they got swept up before I got here, more likely I forgot to print enough. Either way once the MK2 went down I was doomed to hit this step. I should have done an immediate inventory on the 2.5 upgrade/bear parts , I would have had time to print any missing parts before it flaked out.

Either way , here we are.

The plan is to finish disassembly of the MK2 down to the bones and put it to one side. I may reassemble it ( going to cost me a day or two) and try and diagnose the “thermal runaway”. If I need to order a new mini-rambo then the sooner I know the better anyway, can I figure out how to upgrade to the Einsy I have….dont know. Next step is to get everything I have for parts sorted, the MK2s broken down into components and my mental; state to regroup. 

Yes we have 3d printers here in the makerspace but all my parts are in white PETG and all we have for community use is 3 mm PLA. I’ll work this out even if it takes another full printer build to do it.

May 2019

Well I’ve been trying to rebuild the Prusa so that I can print the parts needed for the upgrade to continue. Have been running into one issue after another with the latest being a clogged? extruder? In the middle of running some first layer calibrations the extruder began to miss steps and grind on the filament. I have completely disassembled the hot-end and will reassemble on Monday. (*sigh*)

Author: Joe