I had hoped to get a post up a lot sooner than now, but the last few weeks have been extraordinarily busy. I had been flying and driving all over the US for work, starting a new relationship (should you care), and even had a parrot for a week. Add that onto a million other things and my time for updating this site went right down the tubes. SO, this post is kind of a big one, heheh.
Aluminum Bed and Carriage Upgrade
I started my upgrading spree by installing an aluminum carriage instead. I had been rocking a big, heavy, slightly warped wooden carriage and bed platform for the last few years and that just had to go. The new aluminum bed and carriage are much sleeker and flatter which adds to the ease of printing significantly. The only real downside is that the aluminum mars rather easily when trying to pry the prints off the bed.
The switch to the aluminum bed was fairly painless. I flipped the smooth-rod mounts "upside down" to raise the rods up and allow the carriage to clear the frame. Given the large thermal mass, the bed took about a half an hour to heat up to 80C. I stuffed the region between the heater and the aluminum build plate (bed) with a torn up cotton t-shirt. It certainly worked as I reach temp in about 10 mins and the bed does not need to be powered constantly to keep it there.
I designed and printed out an extended y-endstop for this new bed. The reason being that the Prusa Mendel endstops were designed to work with a much smaller carriage foot print in addition to having more clearance between the smooth rods and the bottom of the y-carriage build plate. When I installed the new bed, the clearance dropped WAY down and the foot print increased which meant I needed to move the endstop from underneath the carriage to the side where it would not get in the way. You can find the files for that here: http://www.thingiverse.com/thing:1161643
E3D Lite6 Hotend
One of the issues I ran into rather frequently was my hotend jamming. I had been using a Makergear V2 hotend for a while and could not get a decent amount of reliability from it.The problem that particular hotend suffers from is that it has a very long, wide, metal tube that the heater mounts onto. When the heater turns on to melt the plastic, it does so throughout the entire metal tube. While this sounds great, if the molten plastic away from the heater cools ever so slightly, the entire hotend jams and the extruder either grinds into the filament or buckles it, ruining your print.
I did a bit o' research and saw that the hotends from E3D were very highly regarded as some of the best ones you can buy. I discovered they had a relatively recent and inexpensive version called the Lite6 and ordered one from Matter Hackers and decided to try it. While the Lite6 cannot be run very quickly at all (the ABS starts to bead instead of laying down smoothly), it prints EXTREMELY reliably and cleanly. I have not had a jam yet and I have printed more in the last couple months than I have in the last three years.
I did purchase an extra "bowden" connector (ie, and airline connector) so that I would have the option to run my setup in a "bowden" configuration. Looking at the speed limitations, I may simply avoid that as I cannot run at speeds fast enough to actually warrant that change.
Given that I was trying to upgrade the hotend onto an ancient printer, I did have a few issues. The main problem was that the x-carriage did not have the geometries needed to fit the new hotend very easily. To fix this, I designed a new carriage and got it installed. Works like a charm - you can find the files here: http://www.thingiverse.com/thing:1161560
New Stepper Drivers for Independent Z-Axis drives
One of the things that would drive me up the wall was that I had a very limited Z speed. If I tried to move too quickly, I would end up skipping steps on one or both Z-axis steppers. I had tried turning up the current on the drivers and greasing the threaded rod with a dry PTFE lubricant, but neither worked. So, I decided maybe I just can't get enough current with the stock RAMPS stepper drivers.
To try to solve this I purchased 5x DRV8825 stepper drivers off of Amazon. These are obviously Chinese knockoffs, but they got decent reviews and I don't really need to push them to their max loads. The idea here is that I am going to take the unused spot on my RAMPS 1.4 controller board and use it to control an extra driver. Then, instead of using one driver to control two motors, both Z-axis motors would have their own supply.
This both worked and didn't work. On one hand, the drivers fixed the problem. On the other hand, they pull WAY too much current. Holy leakage Batman. I had originally installed all 5 drivers but had to revert all but the Z steppers back to the old A4988's. I kept blowing my PTC's (resettable fuses) due to the massive current draw even after calibrating them. I tried placing a PC fan directly against the PTC's to keep them cool, but it only saved me 10 seconds before they blew.
The driver boards that I picked up cost $17 at the time for five of them. Add another $7 for thermal epoxy to attach the heatsinks.
Auto-Leveling and Marlin
When I got my printer, I was running a very old version of Pronterface with the Sprinter firmware when I first built my printer. While I'm sure newer versions are better, I had nothing but problems back then. So, I went searching for a better option and found Repetier Host and then Repetier Firmware shortly thereafter. My printer was exclusively a Repetier machine for a long time and it served me fairly well.
And then I discovered some of the newer auto-leveling that companies like Lulzbot and Printrbot were doing on their flagship models. I did some research and found that this can absolutely be done on my printer and fairly easily but it required Marlin. I had heard good things about Marlin for a good while but held off because I like the ability to tweak things in the EEPROM from Repetier Host. Given the extra feature set, I decided to bite the bullet and swapped over. The move was fairly painless. I threw in my settings from my last firmware, had to reverse the directions of a couple motors, and TADA! Done. I still need to add a bit of code for the Z-axis upgrades, but that shouldn't take more than an hour or two with testing.
Now to the hardware itself - Since I had the aluminum bed, I decided to go with the Printrbot idea and use an inductive sensor. I purchased one of the cheapy ones from Amazon. I have not had a chance to implement it yet as I haven't designed or printed out the new carriage yet (as I mention before).
Edit - now that I have a very flat bed, I might forego this upgrade. The Z endstop seems to serve as everything I need without excess complications. I may end up doing this upgrade later as a side project, but it is no longer on my list of current updates.
Upcoming Upgrades and Project Sneak Peaks
I have a few things I'd like to in the next couple weeks. This includes finishing up the changes to Marlin for the autoleveling and separate Z-axis motors, designing and printing a new carriage to better mount the E3D Lite6 and the inductive sensor, and designing and printing a new endstop switch holder.
After I get all that done and functional, I'd like to print out two new extruders - one direct drive and the other "bowden" (though, I may buy a milled aluminum one instead). If I design my carriage well, I will be able to swap between them easily so I can do fast prints in addition to some more exotic, flexible filaments.
And then after all that, I think I will be done on this printer. Then it will be time to do a few upgrades in hardware and software to the CNC machine I built a couple years ago to let me mill PCB's more easily. I may end up writing my own host and firmware as I am not currently happy with anything that is out there for Open Source CNC machines too. And after THAT is all done, well, let's just say I am doing all of this in order to support yet ANOTHER project....that involves high powered lasers....