OneUp V2 Build Log.

[mr_intensity]

Assembling the Base

This thing needs some handles.

So, upon completing this entry, I will finally be caught up on everything that has happened.  I have been 1-2 weeks behind on everything.  The events in this log entry actually happened last Monday.  It is Saturday as I write this.  I have been spending the week catching up the log, and have not really done anything with my OneUp.

I bought the L-brackets at Lowes, the same time I bought the paint.

Jeez.  I always thought L-brackets were standard "hardware," and you might have two or three sizes available.  First, I could not find L-brackets amongst the "Hardware."  I could only find it amongst screen door hardware, and then my size was limited to "how many holes."  If I knew there was only one size bracket (large), I might have cut the base larger.  I have other reasons for regretting such a small base, but I'll discuss those later.

The L-brackets came with their own screws, which I judged to be too long.  I purchased some #6 by half inch (12.7 mm) screws.  I bought two 13-packs.  They offered a 30-pack for everything except #6 by half inch.

   

For the wooden risers, I chose 3-hole brackets.  For the OneUp itself, I was going to use 2-hole brackets, but I ended up using 1-hole brackets.  I changed for two reasons:  1) I was actually considering putting the brackets on backwards, so that they "bent under" the OneUp, due to the small base.  There was a protrusion between the holes (that strengthens the angle) that prohibited me from installing this way.  The 1-hole brackets did not have this protrusion, so I could (if I wished) put them on backwards.  2) I cut the number of OneUp attachment screws in half.

   

Assembly went without incident.  Well, almost.  My friends talked me out of turning the OneUp brackets around, and they overhang the edge of the base a little.

   

As for the almost without incident, I painted myself into a corner.  My order of assembly went something like this:  1) Screw riser #1's L-bracket into the base.  2) Screw riser #1 to L-bracket.  3) Line up the electronics board in the grooves, holding riser #2 and the electronics board together on the base with the attached riser #1 in order to figure out where riser #2's L-bracket goes.  4) Screw riser #2's L-bracket to the base.  5) Realize that riser #1 is now blocking the screwdriver from screwing in riser #2.

I had to remove riser #2's L-bracket from the base.  Screw it to riser #2 and then screw it back in to the base.

Lesson learned:  Work through the entire assembly process in your head before you start screwing stuff down.

Connecting the OneUp to the base went without incident.  I did, in my haste however, fail to route the Y-axis motor cable under the OneUp to the electronics board before bolting the OneUp down.  We had to get some stiff welder's rod, and fish the Y-Axis motor wire under the OneUp after the fact.

Lesson learned:  Make sure you know how you are routing your wires before you screw your base down.

   

The risers work perfectly.  The angle brackets are not exactly 90 degrees, and actually bow in slightly.  This provides clamping force to help hold the electronics board in place.  I could not have asked for a better setup.

Lesson learned: It's a world of difference screwing in a machine screw into a laser cut hole in melamine, as opposed to force-screwing a wood screw into the stuff.  I don't think these screws hold too well in the melamine.

During design, my friends thought that I should not be handling the entire thing by the handle.  I have to agree with them.  This plywood is heavy, and it puts a lot more stress on the handle.  Plus, I am not completely confident with my connection of the OneUp to the base.  Paul (who loaned me the Tiny Tim saw when I was re-working my 3D-printed SIDEBRACEs) suggested that I put handles on the base, and carry it like a serving tray.

   

I wish I would have made the base bigger to accommodate such handles.  During the week, Jake texted me, and said that he had some handles that he thought would fit.  I disbelieve, but we will see at Monday's build night.

The lack of handles, however, shall not prevent me from messing with the electronics.

[mr_intensity]

Software

Wasting an SD card.

I run Knoppix at home.  I tried all kinds of other distributions, and settled on Klaus Knopper's distribution.

A year ago (pretty much to the date), I created a knoppix 7.7.1 bootable USB flash drive in order to do "Arduino Stuff."  I picked up an Arduino MEGA 2650 knockoff, and an electronics parts kit, to play around with Arduino, and possibly to mate with a RAMPS board at some point.

Anyway, since that time, Knoppix 8.1 has come out, and it is warped toward 3D printing.  So, I figured I would do a new install for my OneUp.  This version would be installed on an SD memory card, as I rarely use SD cards in my laptops, so it would free up a USB port as I 3D-print.

Knoppix 8.1 has a new feature:  Remaster Knoppix.  If I understand the feature correctly, when it creates the compressed filesystem for install to the flash drive, it actually goes out and updates all the packages on-the-fly, so what you install is the latest-and-greatest software.  I chose not to do this, as it would take too much time.  Build night was the next day, and I needed to get the software in order for when I got the electronics wired up.

During the install, I found the notes for my Arduino USB flash.  I created it "For Arduino and 3D Printing."  So, I could just use the K771 version, and skip K81.  But, I wanted the latest Operating System, so I forged ahead anyway.

It was a disaster.

The K81 install to SD actually went fine.  Klaus and his team do an excellent job with their distro, and its install CD.  The problem is that neither of the old, used HP netbooks that I plan to use as a 3D-print controller can boot an OS from the SD card.  Flash drives, no problem.  SD card, problem.

It was now well after midnight, and I needed to go to bed.  I would take the K771 USB flash to Build Night and use that.

The next day, I checked to see if Repetier-Host was installed on K771.  No dice.  It was not even in the repositories.

SAY WHAT???

I told K771 to refresh its knowledge of the repositories (remember this thing was installed a year ago).  Ah, THERE is Repetier-Host.

Upon install, I was told that there was something like 36 packages that either needed installed or upgraded.  Most of these were "Mono" packages.  If I am not mistaken, Mono is sort of like Wine.  It is a Windows-compatibility library.  Hm.  Repetier-Host must be a Windows application.

Long story short:  I installed/upgraded everything without a hitch, and trundled off to build night.

Oh, yeah.  Before shutting down, I did a check on Slic3r.  It was there, as well.

NOW, I trundle off to build night...

[mr_intensity]

Electronics II

Damn!  I forgot my small screwdrivers.

Something I forgot to mention in last week's stand build:  After I put the stand together, I did one last thing before calling it a night.  I got some masking tape and made labels for all the wires, so I know what is what.

   

Well, okay, I labeled everything with a plug on it.  That meant that the extruder heater and the extruder fan went unlabeled.  However, these are four wires that are very different.  So I know what is what.

Anyway, I got to build night early.  Good!  The first thing I wanted to do was get the "factory settings" for the potentiometers on the Pololu A4988 drivers.

   

I set my cheap-arse Cen-Tech multimeter to 20V, and had at it.  Using Q3D's wiring diagram, I knew which A4988 went with which axis.  And I took notes:

+--------+----------------+
|Driver  | Reading (Volts)|
+--------+----------------+
|extruder|    0.66 V      |
|Xaxis   |    0.64 V      |
|Yaxis   |    0.66 V      |
|Zaxis   |    0.68 V      |
+--------+----------------+

Right!  With that out of the way, unplug the thing from the wall, and start plugging motors and stuff in to the RAMPS board.

I plugged in the Y-Axis motor first.  Then, I thought about it.

Maybe I should put this thing in its place in the stand before I put all the wires in.  I tried to do this with the Y-Axis motor plugged in.  That didn't work too well.  For starters, the fan fell off the board as I started to contort it to get it in place with the Y-Axis motor cable still connected.

Out came the Krazy Glue™ pen to make the motor mount permanent.  Also, out came the Y-Axis motor plug to make the electonics board mount easier.

With the electronics board properly mounted, plugging in the wires went without incident.  Then, it came time to do the screw terminals.

"Damn!  I forgot my small screwdrivers."

Luckily, my chapter had some small screwdrivers, so I went at it.  First step:  Strip some of the insulation off the extruder heater wire, and screw it into the terminals.  It was actually a pain in the tukkus to get the wires to stay in the hole.  But that was nothing compared to the extruder fan wires.

There was nothing in the wiring diagram about the fan.  So, I just treated it like the cooling fan on the electronics board.  I hooked it in to the screw terminals on the RAMPS board at one of the 12V rails.  Big mistake.

Tip:  Once you have the power rails wired up to the green connector on the RAMPS board, ***NEVER*** touch those screws unless you absolutely have to.

I could not get the damned screws to hold the wires after I pulled them to twist them with the fan Wires.  Something that should have been very simple was a battle royal.

   

Smoke Test II:  Okay, now that I plugged a bunch of stuff in to RAMPS, and since I messed with the power connections, it was time to run a second smoke test.  Plug the thing into the wall, and...

No smoke.

Whew!  Now all that is left is to plug in the netbook, and fire up Knoppix...

[mr_intensity]

NOVA

It don't go.

For our Spanish-speaking friends out there, you already know what I am saying.  For everybody else, an urban myth:

In the 1960's and 1970's, Chevrolet sold a car called the NOVA.  This is not to be confused with the NOVA that was sold in the mid to late 1980s, which eventually became the Geo Metro.

Anyway, The urban myth is that GM had a marketing problem in Spanish-speaking countries.  "Va" is a conjugation of the Spanish verb Ir, which means "to go."  Literally, "No va" translates to: "It don't go."  The urban myth then goes on to explain that nobody in Spanish-speaking countries would buy the "Nova" because "it didn't go."

Snopes has disproven this urban myth.  Although "no va" literally does mean "no go" or "it doesn't go."

In any event, I named this entry NOVA, because I connected my OneUp to my netbook, and fired up Repetier-Host.

But it didn't go.

Nothing.  The extruder did not even get hot.  I have several theories as to what is going on:

• First, the version of Repetier-Host is not the same as Q3D's documentation shows.  There are options that the instructions are asking me to set that are not even in my version of the software.  So, something could be set wrong.
  
• Second, there was a (windows executable?) program on Q3D's website.  This program is called setupRepetierHost_1_6_0.exe, and I cannot get it to run under Mono.  DosBox tells me that it has to run under Win32, and Wine wants me to load a bunch more Mono and Gecko packages.  So, I have not run it, which again means that if it is configuring stuff, there may be something not configured.
  
• Third, (and I think this is what is REALLY going on) according to Q3D's own SoftwareAndElectronics.pdf instructions, the firmware comes pre-installed on Arduino/RAMPS and is ready to run out of the box.  I'm willing to bet that there is no firmware on the thing, and I have to install it.  I have the metal FDG firmware.  This is good, as I have a metal FDG.
  

It has been, like, a year since I loaded anything into an Arduino.  So, I will have to go and look that stuff up again.

It is now Thursday, as I finish writing this account of Monday's build night activities.  I still have not got the OneUp out of my car, as I have been sending out job applications and handling phone interviews (and writing Build Log entries).  Maybe I will get to work on it this weekend.

Or maybe it will wait until build night.

[mr_intensity]

Firmware

Well, at least some progress.

So, I had all the software and firmware downloaded to my main PC.  However, I had a completely different laptop/netbook configured to do Arduino/RAMPS.  I had to get the firmware on to the correct machine.

It was actually easier to boot the "3D-printer server," connect to the net, and simply re-download the firmware from q3dprinter.com.  I once again downloaded the StainlessFDG firmware, and actually sought out the old firmware (for BrassFDG) and downloaded it as well, just to have it.

After unzipping the .zip file to the Marlin directory in my home directory (~/Marlin), I fired up the Arduino IDE and opened Marlin.ino.  I compiled/uploaded the firmware, unplugged the printer from the wall, and plugged it back in to cycle power.

Aside One:  I need to mount an off/on switch on this thing.

Aside Two:  With our speculation of q3dprinter.com biting the dust in April, should we download all files off of the site?  Once the firmware, documentation, and design files are gone, they are gone for good.

Anyway, I tried to run repetier-host, and still nothing but communication errors.  I killed repetier-host (it would not gracefully exit), and fired up minicom.  The minicom software only allowed certain speeds, and 250000 was not one of them.  Ironically, they did have 2500000 (one extra zero) available, but this setting was not helpful.  I had read on a "loading your Marlin firmware" tutorial online, that the default speed setting of 250000 causes problems with some setups, and a good option is to modify the firmware (Configuration.h file) to set the Arduino speed to 115200.  This is actually the default setting in Linux for serial communication on /dev/ttyACM0 (which is how the Arduino/RAMPS board shows up in my Linux distribution).

I did the speed setting, recompiled, uploaded, and then fired up minicom.  Upon connection, minicom spit out a bunch of jibberish, with stuff like "Marlin 1.0.2" in there.  I knew instantly that I had communication with the printer's electronics.  Upon firing up repetier-host, it told me that the printer was connected, and it was sending commands to the printer.  I still did not get the printer to do anything.

   

While the manual control button screen seemed to work, it also did not seem to work.  Previously, I could not interact with the buttons at all, as they were grayed out.  Now, they were lit up, and if I pressed on a section of the arrow, it would show me Y would increase by 1mm or 10mm or 0.1 mm (depending on where I clicked on the arrow), but I still could not get the printer to do anything.  There was a text bar at the top of the window for entering a GCODE command, and a SEND button.  I would think that clicking on the arrows (or home buttons) would have "built" a command up there, that you would then send to the printer.  However, that did not appear to be the case.

Repetier-Host reported the extruder at 350 degrees C.  However, the heating element was as cold as ice.

I shut down at this point, and declared a minor victory.  I now have communication between my PC and the printer electronics.

Next step is to get the printer to do something.

[MisterAcoustic]

Hi mr_intensity,

I just wanted to jump in here to mention that you are having potentially the exact same issues I had trying to get my TwoUp to run under Linux (Ubuntu Mate). Unfortunately, I never got it to work, even though I have seen many claims from people who say they have. I am a complete amateur at Linux, so my failure wasn't totally shocking. The machine I have been using with my TwoUp is dual booting linux and Windows XP. The intention was to never run XP again, except in case of emergency. I finally booted into Windows, and the TwoUp worked fine (well, as fine as it can considering how broken it is) - let's say that at least all the axes move properly.

I hope you can figure it out, because I will switch if I can. I wanted you to know, since you haven't made the printer work yet, that's it's quite likely that the printer is fine, and we are missing something about the Linux setup.

Wish I could be more help.

Stan

[MisterAcoustic]

I wanted to add that I believe this is a driver issue, so you might try looking for an older version  - things have changed, but I _think_ it's the unsupported driver at the bottom of this page:

http://www.ftdichip.com/Drivers/VCP.htm

That page now claims that support for their devices are now supported directly by the kernel. So maybe we should uninstall the driver instead.

Lastly, there could be something weird going on with Repetier Host - so you could try Cura instead.

Stan

EDIT: A belated edit on this to mention that the above really applies to old Printrboard clone hardware. The symptoms are so similar though, that I'm wondering if it's not a general problem.

[mr_intensity]

Hi Mr. A,

I have not given up on the OneUp, it is just that things got busy. Then, I lost my operating system.

No, really, I physically lost it.

I run Knoppix, which was designed to be a "portable" operating system. You install it to a flash drive, not a hard drive. Then, you can take the flash to any computer and boot from it.

I use flash drives with a small physical form factor (called a "plug-and-stay configuration"). Check out pictures of the SanDisk Cruzer Fit for an example of what I am talking about.

Anyway, I physically lost the small Cruzer Fit drive, and just recently I found it.

Thank you for your words of support, and the link to the older drivers. I have come up with some progress (I got the motors moving) with the "current" drivers. I am thinking at this moment that I have a hardware problem. I will be posting the results of my troubleshoot shortly. I need to bust out the multimeter and do some readings first.

Also, I did not use Cura in my troubleshoot, but rather pronterface. It does a better job of logging error messages than does Repetier-Host.

Also, I **believe** that R.H. is more like Windows software running on Linux, while pronterface is Linux software running on Linux. Don't hold me to that, though.

EDIT - I just followed the link. I plug FTDI serial devices into Linux all the time with no problems whatsoever. If the issue were the serial drivers, I would not get the printer spewing "Marlin jibberish" at minicom. So, I think the drivers are OK. The problem at this point APPEARS to be a faulty thermistor. But, let's wait till I get the post written, and we can brainstorm.

[mr_intensity]

Tweaking the firmware

There isn't much thermistor wiring to check.

So, I had a long layoff from the OneUp. As mentioned in a previous post, part of the problem was that I lost my operating system...literally.  For about a month, my OneUp resided in the back seat of my car.  However, I found my missing OS, and was back on the war path.

There has been movement on both the hardware and software front, but first let's review a boo-boo from a previous post:

Tip:  Once you have the power rails wired up to the green connector on the RAMPS board, ***NEVER*** touch those screws unless you absolutely have to.

I did not realize how badly I screwed this up until I was reviewing my previous entries.  The fan cooling the RAMPS board does NOT connect to the RAMPS board green connector.  Rather, it connects directly to the power rails.  Thus, the following statement is false: "So, I just treated it like the cooling fan on the electronics board.  I hooked it in to the screw terminals on the RAMPS board at one of the 12V rails."  I should have connected it directly to the power supply rails and avoided the RAMPS connector completely.

As Denis Lemieux would say: "Two minutes, by yourself, you know and you feel shame, you know. And then you get free."

Hardware

I have not got around to adding handles to my 3D-Printer stand, but I did add a switch to the thing.  I picked up a toggle switch and some electrical wire from Lowes, and had Jake (the Boy Wonder, who is in Electrician School) do the honors of wiring everything up.

   

It was required for me to remove the Back panel of the stand, and drill two holes in it.  One hole was to run the electrical cable out, while the other was for mounting the switch.  I had to countersink a large area behind the switch-mounting hole, as the wood was too thick to mount the switch.

   

Jake stuffed the electrical wire in the hole, then wrapped the thing in electrical tape, to form a sort of strain relief, and then backed it out so the taped up area filled the hole. Then he wired in the switch, and used some of the extra wire to run everything to the power supply.

   

I left the front panel of the stand unmolested, as I plan to someday mount a panel controller there.

   


Software

Repetier-Host (RH) gave me no information in the logs.  As a matter of fact, it gave me no logs.  I decided to look at an alternative that might give me better logging.  In the Knoppix repositories is a package called printrun, and it's GUI console program is called pronterface.  It gave me no logs, either, which made no sense.  It gave me a running dialogue down the right hand side of the screen.  It might have even told me that the log directory did not exist (oops! my bad).  Anyway, I typo-ed the log directories in the configuration...on BOTH pronterface and RH.  Anyway, once logging was fixed, the magic error message that pronterface gave to me was...

[ERROR] Error:Printer stopped due to errors. Fix the error and use M999 to restart. (Temperature is reset. Set it after restarting) Please check thermistor wiring

There isn't really much to check.  The thermistor has two white wires that go from the thermistor to a plug that plugs onto the T0 posts.  It isn't rocket science.  It isn't even rocket engineering.  It is pretty much cut and dry.  Current will flow either way through the thing, so polarity is not an issue (even though I turned the plug over, and tried it the other way, anyway).

Googling the error message brought me to this discussion thread on RepRap.  To quote the post:

In Configuration.h, just change the TEMP_SENSOR_0 from 1 to 0. This will change it from being defined as a 100k thermistor to not used. Then you can run your motors without a temp. That's what I'm doing right now to test movement on my machine.

I was surprised, upon editing Configuration.h, to find that not only was TEMP_SENSOR set to 1, but also TEMP_SENSOR_BED was set to 1.  Waitaminute - I don't have a heated bed!  So, my first step is to set TEMP_SENSOR_BED to 0, and see if that fixes the problem.

Nope.  Same error.

Next, I set TEMP_SENSOR to 0, and was now able to move the motors.  And what did I learn?

I got everything backwards.

No, really.  All three axes were wired backwards.  When I told Y to come forward, it went back, when I told Z to go down it went up.  When I told X to go left, it went right.  Do you realize how much talent it took to screw up ALL THREE axes like that?  I mean, it is a simple fix (pull the wire out of the RAMPS board, flip it around the other way, and plug it back in).  But one would think I would have gotten at least ONE of them right.  I owe it all to clean living, and knowing something about electronics.

While the connectors are square, and really have no polarity, there is a very small arrow (triangle) raised on one side of the connector, and it points to "pin one."  The RAMPS board labels pin one on all of its connectors.  All I had to do was match pin one to pin one, and rely on the improving quality of Chinese industry.  Their repeatability and reproducibility allowed them to screw up the wiring on all three motors in exactly the same way.  In days of old, there would have been no rhyme nor reason to how they wired the motors.  Now they are consistent.

As of now, the only motor that is backward (i.e. pin one still goes to pin one) is the extruder motor.  Neither Pronterface nor RH would permit me to move the extruder motor with the extruder temp disabled.  So, it is the one motor I did not test.

More Hardware

Running my cheap but somewhat reliable Cen-Tech meter across the thermistor read a little over 36 ohms.  The longer I left the meter across the leads, the higher the reading climbed (well over 37 ohms).  Running the meter across the T0 leads yielded close to 5 volts (4.97, if I remember correctly).

According to comments in Configuation.h, as well as in the post that pointed me to disabling the thermistor, the thermistor is supposed to be 100K ohms.  Hm.  What if I got a plug like the one on the end of the thermistor, and wired up a 100K resistor to it?  I could plug this thing in, in place of the thermistor, and fake out Marlin into thinking a working thermistor is plugged in.  If this idea works, I should be able to move the extruder motor, and feel if the print head is getting hot.  If it passes both tests, I can shut down before the thing gets too hot.  This will confirm that the problem is with the thermistor, and not RAMPS or Marlin.

What follows is a diagram of what I am talking about.  Since one has to be a forum member to view pictures, and I believe many people (both members and non-members) will be looking to troubleshoot this problem, I will diagram this with "ASCII Art:"

RAMPS Thermistor Test Plug wiring diagram:

+----------+      100K
|  Plug    |____/\/\/\/\___
|   To     |               |
|   T0     |---------------+
+----------+

Legaleese:  WARNING - Use of the above plug could cause your extruder to overheat, resulting in potential fire, destruction of property, or even injury or death.  Only use above plug sparingly, and only for testing the extruder motor/heating element in printhead.  DO NOT USE FOR 3D PRINTING.  You have been warned.

I might have a 100K resistor in my Arduino project pack.  The adventure will be finding a plug...

[mr_intensity]

The Long Layover

Cause I'm...Baaa-aaaa-AAAA-aaaa-ack...

NOTE:  The following was written last week.  I did not get a chance to post it until now.  Furthermore, it has an error in it, which I will correct in my next entry:

So, it has been almost a year since my last entry.  Durning that year, the OneUp sat dormant.

I pretty much ran out of money, as I was unemployed, and was desperately bouncing between potential jobs.

At 51 years old, Industry does not want me anymore.  The concept is called ageism, and it is pretty widespread.  It looks like I will never get another IT job again.

I finally ended up as a night watchman.  The job does not pay a lot, but there are opportunities for overtime, and I had a lot of back debt to work off.

While all of this was going on, I determined that both the thermistor and the heating element are deader than disco.  Think about it.  In my previous post I stated that a 100K thermistor read 37 ohms.  My Cen-Tech got no reading on the heating element, at all.  I never built the test plug.

With all my credit cards at or near their credit limits, I could not order replacement parts...until now.  I finally got enough breating room on my online card to place an order for some cooling hoses and clamps for my car.  I tacked the 3D printer parts on to that order.

I ordered Amazon ASIN B073TVBJVY, which was five NTC 3950 100K temperature sensors and five unnamed 12V/40W stainless steel heating elements.  I have since found out that another Amazon vendor sells these as three-and-three for less money (I only need ONE of each in working condition), so shop around.  Also, be careful as you shop, as some of the heaters are 12V and some are 24V.

Anyway, having received the goodies, the next step is to label them, so I know which is which.  Again, I bust out the Cen-Tech multimeter.  Results at different ohm settings are as follows:


12V, 40W, 304 Stainless Cartridge Heater (H1-H5)
NTC 3950 100K Thermsistor (TA-TE)

  |2000K| 200K| 20K | 2000|200|
--+-----+-----+-----+-----+---+
H1|  000| 00.0| 0.00|  005|5.2|
--+-----+-----+-----+-----+---+
H2|  000| 00.0| 0.00|  005|4.0|
--+-----+-----+-----+-----+---+
H3|  000| 00.0| 0.00|  005|4.4|
--+-----+-----+-----+-----+---+
H4|  000| 00.0| 0.00|  005|4.2|
--+-----+-----+-----+-----+---+
H5|  000| 00.0| 0.00|  005|4.2|
--+-----+-----+-----+-----+---+
TA|  116|116.5|     |     |   |
--+-----+-----+-----+-----+---+
TB|  112|112.2|     |     |   |
--+-----+-----+-----+-----+---+
TC|  112|112.8|     |     |   |
--+-----+-----+-----+-----+---+
TD|  114|114.5|     |     |   |
--+-----+-----+-----+-----+---+
TE|  119|119.5|     |     |   |
--+-----+-----+-----+-----+---+

Friday is my chapter's next build night, and I will be working on my car for that one.  Don't know if I'll have time for 3D Printer stuff.

Stay tuned...