MakerFarm 12" Prusa I3V - my build, usage, and mods

I feel like an expectant father waiting for the kit.

My intent here is to blog about the building of the kit, calibrating it, using it, and modding it into something more than what it is.

The wait begins....

One of the more interesting things that I have found is that most small, independent companies/operations with limited resources have efficient online/web ordering processes. MakerFarm fits into this category. The unit was purchased using their website and there were no issues as I went from point A, to B, to C... I have dealt with larger organizations that force me to keep my hair short so I can't pull it out dealing with their online ordering processes.

Now waiting on UPS to deliver the unit and yes, I do have another set of words to fit the abbreviation and it has to do with how they handle boxes/packages being transported by them.

So what printer are you getting?

I am in the same boat...a trans-atlantic boat..

The box is at the UPS station. Now to pick it up tomorrow and see what kind of shipping damage it has sustained.

The box survived intact. One side was crushed a tad but was inconsequential. MakerFarm needs to hold a Master's Class with Amazon, and a few others, on how to properly pack a box for shipping. Everything arrived just fine. A well appreciated touch was to open the box and see first thing, on top of the copious amount of the pink packing peanuts, the shipping list, face up, and not manhandled. This build looks to be as fun as building a HeathKit (Oops....dated myself there). More to follow.

Note to forum: I am not a bot. Your algorithms are stereotypical fail.

Quote
itchytweed
Note to forum: I am not a bot. Your algorithms are stereotypical fail.

Yes, I to feel your pain. Only seems to happen when i try to post from my phone though

Look forward to your build log

Chris

Edited 1 time(s). Last edit at 05/02/2015 08:54AM by lokiwizz.

That's why I am collecting extra parts to build another larger 3d printer. My Folgertech Prussa I3 is just a learning journey for me. I too will be following this.

Quote
lokiwizz

Quote
itchytweed
Note to forum: I am not a bot. Your algorithms are stereotypical fail.

Yes, I to feel your pain. Only seems to happen when i try to post from my phone though

Look forward to your build log

Chris

The post was from my phone over 4G-LTE. This is over wy-fy. ...no message.

Edited 1 time(s). Last edit at 05/02/2015 09:11PM by itchytweed.

The monstah......It Lives!!!!! (with apologies to Dr. Frankenstein)

Film at 11.

Word to the wise... the heated bed on this beast draws 34 amps at 14.2 volts. Luckily, I have a 50 amp supply. Finished up making the wiring neat and some wire tie manufacturer is rubbing their hands just a bit more...

I have it printing in PLA. Not the best print but good for a first try on an unknown file.

Quote
itchytweed
Word to the wise... the heated bed on this beast draws 34 amps at 14.2 volts. Luckily, I have a 50 amp supply. Finished up making the wiring neat and some wire tie manufacturer is rubbing their hands just a bit more...

Do you connect the heatbed right through the Ramps board, or do you use some kind of (solid-state-) relais?

Now you can print a mrs. monster
Olaf

No RAMPS board, using RUMBA board. There is an interposing mechanical relay between the RUMBA and the heat bed doing the heavy lifting. If this was through the RUMBA, I am quite sure that it would be a crispy critter. I may go so far as to ditch the relay and put on an external MOSFET transistor to do the heavy switching but so far it does not seem to be needing it.

OK, Im glad to hear that.
If it is a solid state Relais, in fact it is a semiconductor already.

If it is a mechanical relais, be aware, that they tend to "stick" when they actually should switch "open".
I hope you`re using Marlin with the "thermal runaway" option enabled...
Olaf

Quote
o_lampe
OK, Im glad to hear that.
If it is a solid state Relais, in fact it is a semiconductor already.

If it is a mechanical relais, be aware, that they tend to "stick" when they actually should switch "open".
I hope you`re using Marlin with the "thermal runaway" option enabled...
Olaf

There are SSR's in the necessary ratings that would work, providing that they can respond quickly and don't need a load of heatsinking. I will have to check into that vs. a cascaded MOSFET. Thermal Runaway has been turned on and I would think that it would be really slacking and incredibly stupid to NOT use that feature.

Sticking is caused by the contacts being pitted from the arc that generated on opening when dealing with inductive loads. I don't think that will be a huge issue here but a diode snubber is in the plans.

A shout out to Triffid Hunter and his calibration guide. Good going on it.

Just had an interesting failure dialing in the printing. I got a MINTEMP fault. Location of failure was that the thermistor wire broke right where it was soldered to the wire. That issue is now fixed so back to getting this working.

Two items:

1) Elmer's Washable School Glue Stick is my new best friend (the purple one) and a single blade scraper for cleanup.

2) Since I had a devil of a time finding a copy of Marlin/Configuration.h to work from for this printer, I am attaching it as a starting point for the community.

Followup on the z-axis - I replaced the vinyl tubing couplers. I went to HomeDepot and purchased the following:

2) Everbilt 595402 1" nylon coupling
4) Everbilt 574453 M4-.7x6mm setscrews

I used a 1/8" bit to drill two holes in each to make a coupler. The M5 shaft and the NEMA 17 shaft are a perfect fit. Put the coupler on the stepper first, tightened up its setscrew, Then I put in the leadscrew until it bottomed on the stepper shaft and tighten its setscrew. No more slippage and it looks a bit more professional. I realize that this make the operation a bit more rigid and may put a bit more stress on the nut trap follower but the slippage issue, even with wire ties was disconcerting and that is no fault of MakerFarm here at all. It is just my particular random situation.

If you think this "blog" is disjointed, you should see my notebook....

I printed out a set of gears for the extruder as the ones I have are powdering. So, I put a couple of drops of Tri-Flow lube on them and the dusting has settled down. But the gears pointed out to something about the frame. There is a torsional issue that affects the frame and its stability. High move rates can excite this and cause shifting and it was apparent in the gears by looking at the layers.

The solution appears to be putting adjustable blocks/outriggers to tighten up the twist as it appears to be caused by the two wooden cross pieces that are parallel to the X axis. They present themselves as having a slight rocker to them on the surface that contacts the table. Easy fix: a piece of wood with two adjustment slots on either side anchored using wing nuts.

The glue stick works well but I just tried the blue painters tape and it's way better with less mess. I had a hell of a time getting the excess glue off. I didn't have a scraper though.

Quote
milkypostman
The glue stick works well but I just tried the blue painters tape and it's way better with less mess. I had a hell of a time getting the excess glue off. I didn't have a scraper though.

It takes so little to make the print stick down. I have found that I can get about 8-10 prints out of one application. I put on just enough so there is only a very slight haze. When I do clean up, I use a single edge razor blade to take off the old stuff, clean with isopropyl alcohol, reapply and continue printing.

Did a couple more calibration prints last night. One was a one nozzle diameter box and a 20mm square afterwards. The nozzle is 0.4mm and I ended up with 0.4mm walls after averaging a bunch of measurements. After this, my 20mm box came out to 20.02mm, a side error of 0.01mm, which is 0.000394 in. Can't complain at all.

Now I have my interior holes coming out just fine. PrintSettings > Advanced > Other > XY Size Compensation. I am using a 0.4mm nozzle on the Hexagon, so I set it at -0.2mm and my holes are coming out just fine.

Last night, I did a print run and shut the whole system down.

Came back this morning and powered it up. No USB serial comms and both the RX/TX lights were on solid. The laptop would not respond to the cable. So, a 16U2 crash. Started surfing looking for similar issues. Yes, there were and under the same conditions. So, first things first, reload firmware to the 16U2. It is far easier to do that than replace the board at this time.

Went to [reprap.org] and found how to reprogram the 16U2. Followed the procedure - really easy procedure. Back printing after rebooting. If this craps again, then I will consider a warranty. Need to let MakerFarm know this.

BTW, my version of the RUMBA board looks to have a separate COM port connection: RX, TX, GND and +5. Wonder if my FTDI USB to RS-232 will play on there? No drawings show the port and neither do the pictures on the wiki.

Quote
itchytweed


BTW, my version of the RUMBA board looks to have a separate COM port connection: RX, TX, GND and +5. Wonder if my FTDI USB to RS-232 will play on there? No drawings show the port and neither do the pictures on the wiki.

Have a look at the schematic of the Arduino Mega 2560 http://www.arduino.cc/en/uploads/Main/arduino-mega2560_R3-sch.pdf]found here[/url] or the RUMBA schematics http://www.reprap.org/mediawiki/images/2/24/RRD-RUMBA_SCHEMATICS.png]here[/url]
The RUMBA is really an Arduino-Mega2560 reference design with some addons on one board..
The 16U2 is just an 'interface' between the USb port and the ATMega, just like the FTDI is.. The serial data is then connected to TXD0 nd RXD0.
So if these pins are are indeed available your FTDI should work..

Quote
ReSiStAnCe

Quote
itchytweed


BTW, my version of the RUMBA board looks to have a separate COM port connection: RX, TX, GND and +5. Wonder if my FTDI USB to RS-232 will play on there? No drawings show the port and neither do the pictures on the wiki.

Have a look at the schematic of the Arduino Mega 2560 http://www.arduino.cc/en/uploads/Main/arduino-mega2560_R3-sch.pdf]found here[/url] or the RUMBA schematics http://www.reprap.org/mediawiki/images/2/24/RRD-RUMBA_SCHEMATICS.png]here[/url]
The RUMBA is really an Arduino-Mega2560 reference design with some addons on one board..
The 16U2 is just an 'interface' between the USb port and the ATMega, just like the FTDI is.. The serial data is then connected to TXD0 nd RXD0.
So if these pins are are indeed available your FTDI should work..

Nice to get a second opinion before I trash either the adapter, RUMBA, or both. My FTDI adapter will play on either 3.3 or 5 volt levels. I have this tendency to play it safe. Even though the cost of the adapter or board is not that much, the inconvenience of having to replace is not worth it. Professionally, I play with parts that cost 5-7 figures so I just transfer that care from my professional life to personal life.

I have now populated the RUMBA with all DRV8825's and split the z-axis steppers onto their own driver. Plus, left them at 32x microstepping. Next mod is to put in the + axes limit switches. I would just feel better to have them.

I am currently doing prints at 0.1 mm z axis stepping. Maybe I will get a tad outrageous and try 0.05 mm steps.....gotta let my freak flag fly.

Also, the 16U2 has been behaving since reflashing it. The RUMBA board that MakerFarm sent out has on it a four pin header that breaks out the RX/TX data pins between the ATMega and the 16U2. So if it decides to go goofy again, I can work it that way.

This is a set of three 20mm^3 cubes printed with a hole sized for an M5 screw on each axis. They are printed in PLA at 100 mm/sec. At 0.4mm layer thickness, the 0.4 mm Hexagon was running at significant pressure as at the end of the print, there was a big goober left on the part at the z-axis lift when it was done. The 0.2 print seem to have small gaps on print restart after filament retraction - that can be addressed. Even though the surface of the 0.05 cube is great, print time take forever...

So, 0.2 will be the working thickness. Now to clean up the small stuff.

Tried something new. I went and kicked up the accel rates big time (5000 mm/sec^2) just to wave my freak flag. It allowed me to see a good amount of flexure to the wooden end pieces that have the front foldover bearings and y-axis servo mounted to them when doing Hilbert curve infills. Considering replacing front and back pieces with V-slot. Methinks I am going to weigh the table and work the numbers from there. This is what three day weekends are for.