Smoothie M2

The place to discuss your hardware and software/firmware modifications...
Post Reply
User avatar
Tim
Posts: 1205
Joined: Thu Apr 10, 2014 2:19 pm
Location: Poolesville, Maryland
Contact:

Smoothie M2

Post by Tim » Sun May 10, 2015 4:18 pm

So, by popular demand, I ordered my Smoothieboard from Hackaday:

http://store.hackaday.com/products/smoothieboard5xc

It arrived yesterday and spent all night on my doorstep (oh, WHY don't they ever ring the doorbell?!). Fortunately it didn't rain last night.

I unpacked it and was dismayed to find no micro SD card in the slot or in the box. Eventually, I discovered I was sitting on it. It pays to open the package carefully, although I thought I was doing so. I think that the micro SD card is supposed to be in the slot, but it looks like the board is designed in such a way that the SD card can get pushed in and disengage if the board is pressed completely against a flat surface, such as the side of the box it's shipped in.

I unplugged the USB from my RAMBo and plugged it into the Smoothieboard. I was delighted to find that it immediately came up under Linux with two recognized devices: The serial communications channel at /dev/ttyACM0, and the micro SD card at /dev/sde1, which I mounted onto /media/. This is the first of the several major advantages of the Smoothieboard. I can write directly to and read directly from the micro SD card from my host computer. Better yet, I can do the same over Ethernet, although I haven't tried that communication channel yet.
smoothie1.jpg
Smoothie board, unpacked, plugged in, and ready to talk!
smoothie1.jpg (198.57 KiB) Viewed 15396 times
I decided to start a new thread here dedicated to the conversion of an M2 from a RAMBo to a Smoothieboard. Note that I have a very old-generation M2, where the RAMBo is several revisions old. Among other issues is that the solid-state circuit breaker can't take 24V without a catastrophic and possibly explosive failure, and the stepper drivers are the older 8-microstep version, not the newer 16-microstep version. I solved the circuit breaker problem by soldering in an automotive fuse. The Smoothieboard is maybe 50% or so larger than the RAMBo, but given that if you position it in the M2 such that all the ports are in front, then there is the same amount of clearance as on the RAMBo; the longer part of the board is toward the back, where on the M2 there is empty space. So that part seems all well and good.
rambo1.jpg
My existing RAMBo, a mess of wires, including extra ones for the LEDs.
rambo1.jpg (202.72 KiB) Viewed 15396 times
The first thing to think about when considering whether or not to do a Smoothieboard upgrade (other than waiting for my complete set of posts to see if I was successful or not!) is whether it is worth the price. Sure, the Smoothieboard has some great features, but the RAMBo is no sloucher. RAMBos shipped for the past (approximately) two years support 24V operation, have the 16-microstep drivers, and have the same I/Os as the Smoothie (enough for a dual-extruder M2 setup). It remains to be seen what advantages the GRBL-based, non-Marlin firmware has to offer. I have already mentioned the ability to remote-mount the micro SD card, and the Ethernet port. There's also the advantage that you can talk about the board without conjuring up unwanted images of Sylvester Stallone. Although it may make you hungry for a frozen blended fruit drink.

As I continue posting to this thread, I will mostly be following instructions on http://smoothieware.org/3d-printer-guide. I have read through them, and they seem simple enough.

jsc
Posts: 1864
Joined: Thu Apr 10, 2014 4:00 am

Re: Smoothie M2

Post by jsc » Sun May 10, 2015 4:36 pm

Sweet! I am excitedly awaiting your first test print photos. It will be interesting to see if the better acceleration handling tones down vibration-induced ringing artifacts around sharp corners. I suggest a box at 100 mm/s or higher, and maybe my dimpled cylinder ringing test: http://www.thingiverse.com/thing:247162

Sorry for leaping ahead....

I noted in an earlier Smoothieboard thread that the MOSFET outputs were rated at 12A, vs. 15A on the RAMBo. Wondering how that affects bed heat times.

User avatar
Jules
Posts: 3144
Joined: Wed Jan 21, 2015 1:36 am

Re: Smoothie M2

Post by Jules » Sun May 10, 2015 4:41 pm

We await your verdict! :D

User avatar
Tim
Posts: 1205
Joined: Thu Apr 10, 2014 2:19 pm
Location: Poolesville, Maryland
Contact:

Re: Smoothie M2

Post by Tim » Sun May 10, 2015 7:19 pm

I have many things on my plate today (it's Mother's Day, for one thing!) but hopefully I can keep working on it through the week.

I am somewhat skeptical that it will stop the ringing patterns, since the documentation seems to show the path planning having the same trapezoidal shape to the velocity. That's just a picture, though, and I haven't looked at their algorithms.

Not exactly true about the power MOSFETs. The MOSFETs themselves (the big ones) are rated at 20V but the connectors are rated 12V. There isn't any mention about installing a heftier connector. There are, however, comments about combining MOSFETs, and using solid-state relays. I'll take a closer look at that when I have everything else under control.

(edit) Pffft. Good catch, Insta. 20 AMPS, 12 AMPS. What was I thinking?? Clearly, I have way too much on my plate today, and shouldn't even try posting.
Last edited by Tim on Mon May 11, 2015 3:25 am, edited 1 time in total.

User avatar
insta
Posts: 2007
Joined: Tue Sep 16, 2014 3:59 am

Re: Smoothie M2

Post by insta » Sun May 10, 2015 8:53 pm

Tim wrote:I have many things on my plate today (it's Mother's Day, for one thing!) but hopefully I can keep working on it through the week.

I am somewhat skeptical that it will stop the ringing patterns, since the documentation seems to show the path planning having the same trapezoidal shape to the velocity. That's just a picture, though, and I haven't looked at their algorithms.

Not exactly true about the power MOSFETs. The MOSFETs themselves (the big ones) are rated at 20V but the connectors are rated 12V. There isn't any mention about installing a heftier connector. There are, however, comments about combining MOSFETs, and using solid-state relays. I'll take a closer look at that when I have everything else under control.
For connector voltage, anything rated 12 volts is also rated up to about 300 volts (unless you meant amps...?). This is not the case for relays or solid state electronics, but is going to be the case for plugs.

Also, a 20A MOSFET is a pathetically small one. The channel itself might be rated 20A, but the FET is probably rated 70-200A.
Custom 3D printing for you or your business -- quote [at] pingring.org

jsc
Posts: 1864
Joined: Thu Apr 10, 2014 4:00 am

Re: Smoothie M2

Post by jsc » Sun May 10, 2015 10:15 pm

You're right, it looks like the Smoothie firmware is constant acceleration, although they do a ton of lookahead, and they do seem to have done something clever with cornering: https://onehossshay.wordpress.com/2011/ ... algorithm/

TinyG seems to be the guys doing the superduper motion planning, here is a thread comparing the two:
http://smoothieware.org/forum/t-839730/ ... s-smoothie

User avatar
insta
Posts: 2007
Joined: Tue Sep 16, 2014 3:59 am

Re: Smoothie M2

Post by insta » Sun May 10, 2015 11:07 pm

TinyG's constant jerk > everything else.

I just wish they'd make a version that's actually for a damn 3D printer and not CNC cutters!
Custom 3D printing for you or your business -- quote [at] pingring.org

User avatar
Tim
Posts: 1205
Joined: Thu Apr 10, 2014 2:19 pm
Location: Poolesville, Maryland
Contact:

Re: Smoothie M2

Post by Tim » Mon May 11, 2015 3:33 am

The main thing is that I think the Smoothieboard has the capacity to do much more complicated things like continuity through a few more orders of derivatives, even if nobody has gotten around to coding it yet. I'm not sure the RAMBo can manage that, although there's an awful lot that can be done with clever algorithms and careful coding.

Looks like the first headache, though, is going to be sorting out the power distribution. The Smoothieboard has three power terminals like the RAMBo, but they are in an odd assortment of arrangements.

LonV
Posts: 44
Joined: Wed Apr 09, 2014 8:40 pm

Re: Smoothie M2

Post by LonV » Mon May 11, 2015 4:45 am

I've seen this print crazy fast. I'm assuming that quality is still a factor when going so fast?

-Lon

User avatar
Tim
Posts: 1205
Joined: Thu Apr 10, 2014 2:19 pm
Location: Poolesville, Maryland
Contact:

Re: Smoothie M2

Post by Tim » Mon May 11, 2015 2:48 pm

jsc wrote: I noted in an earlier Smoothieboard thread that the MOSFET outputs were rated at 12A, vs. 15A on the RAMBo. Wondering how that affects bed heat times.
Shouldn't affect it at all. I measured the resistance across the bed heater, and it's 3.1 ohms. So at 24V, the bed heater draws approximately 8 amps. So, too much for one of the small power MOSFETs, but no issue for one of the big ones. The extruder heaters are 15.8 ohms, so 1.5 amps at 24V, which matches the value in the Smoothieboard configuration, so I guess that's pretty standard for extruder heaters.

The MakerGear power block is a Mean Well SE-450-24 and delivers just under 19 amps at 24V. It's got enough juice for the job, even with a few extras like LED lights, but it's not overkill. The output is split into three channels, and the Smoothieboard has three power channels input (not including the 5V input, which I can't use with the 24V supply, which only has 24V outputs). Unfortunately, the Smoothieboard has the three power inputs in different places on the board, not all in a row like on the RAMBo, so I am going to have to remove the nice power connector and replace it with three individual ones (which, by the way, come with the Smoothieboard in a small bag of parts).

Post Reply