Tin's 1100x900mm Build

Hi all, my name is Martin and I’m from Argentina. I’ve finally started printing the parts for this incredible machine. I’m enjoying a lot building this CNC.

My MPCN will be of 1100x900mm with a working area of approximately 1070x870mm and a Z rail of 350mm. EMT conduit can’t be used here, because they have the text engraved, making an uneven surface for the bearings. So I’ve been trying to source 25mm stainless tubes like mad, and luckily found some right where I work… They were storaged for ages like 10 meters of it.

This app was used to calculate the desired working area:
Thank you stevecd!

I have 5 rolls of black PLA, but to make it less boring I’ve bought a roll of green PLA from Verbatim. First time I use their material and to be honest it’s awesome. A bit pricy here, but totally worth it.

Less words more pics… Rollers done!


I’m going to use 2,54mm 4 pin JST connectors, so I designed a simple mount with holes just for zip ties. http://www.thingiverse.com/thing:1454274
These are the connectors if anyone is interested:
4 Pin JST

and the middle is done too:


If you are going to use allen type hardware, make sure to use flat headed screws or low-profile ones for the middle part.
If not, the head will interfere with the tool mount.

The table is almost finished.
Will post some pictures any time soon.

Flickr Album

Looks nice, I like your colors.

Your right about the Z axis, the heads of the bolts are close to the threaded rod, so you will need to watch that.

Wow, your build is going to to be super clean! I really like those flat allen screws.

Your pics are amazing how do you get them so nice? I thought it was a render at first!

Thanks for the kinds words guys.
I take the pictures using a white sheet of paper as background, then I clean the edges with photoshop. Brightness is increased by +/-40 and saturation reduced by +/-10. Auto tone, auto contrast and auto color do most of the trick fixing the colors.

This is the original pic:

Oh and middle assembly is done! :smiley:
I have to say that I never thought it would be like this. It feels very rigid and tough.
I’m really impressed with the attention to detail you put into this design Ryan.

I’ve used 4 aluminum spacers to improve rigidity. But they actually don’t help that much, they ended up being just an aesthetic resource.

and here is the table. I had some 4x2’’ wood planks so I used them.
The problem is that it was an overkill to use 2 of them as legs…

Now I need to cut the removable board that fits in the center and make some holes so that the legs of the MPCNC can be regulated easily.

Well, I’ve been cracking my head to design a belt tension system that can work with the default corner blocks, but it’s hard. So I approached the issue the other way. First, design the belt tensioner and then adapt the corner block to it.
The corner block ended up being a part stiff and easy to print, with a bulky look that I like, but the parts of the tensioner are more tricky to print. The teethes of the clip should match perfectly the timing belt. Also, the belt should be held tight between the walls even without tension applied from the screw.

The army is ready.
Notice the Vicious tribute :smiley:
Very hard to take good pictures of them, they look better live.

Corner Block with M3 plastic inserts.

It’s a bit hard to put it inside de tensioner. This was the easiest way to do it.

First push the timing belt and then the clip. The idea is that the timing belt ends up as close as possible to the corner block.


I still need to make some diagrams of the assembly and clean the parts a bit. But as soon as they are ok, they will be uploaded to thingiverse with step files if someone wants to mod them.

Nice, that is something I haven’t seen before, and that logo…something about it looks so…awesome. Take it easy on the tension though, these really are super loose on my machines. I just cleaned one up and put on some new zipties and mine are just past the point of the ziptie bounce. I need to make a good belt tension gauge or something, but really loose.

Good to know Ryan, I might have tighten them up too much to avoid skipping steps. Will loose them a bit.

Btw… I’ve just uploaded the files. Just the 25mm version, the 23.5 maybe later today.
But step files are up too, so feel free to mode them.

Hope it helps.

I’ve read somewhere in the forum, that to calculate the current limit to each stepper, I need to calculate the amps when fed 12v.

I hate the electronics part, and I think I’m a bit confused.
Please correct me if this is wrong.

For the X and Y I’m using 42BYGHW804 nema 17.

1.2A x 3.6V = 4.32W
4.32W / 12V = 0.36A
For two steppers 0.36A x 2 = 0.72A

So with the DRV8825:
1000mA ----> 500mV
720mA -----> x

I need to set the driver to 360mV.
Is this ok?

Turns out it way more complicated than that, but the formula still kinda works.

Those numbers are fine, start with them. To really fine tune it, slice up a large spiral vase 3d print, let the printer run for a while, then feel the steppers. They are doing pretty good when they are warm but not hot enough to melt the PLA. You need a file that keeps the motors constantly engaged, and moving, spiral vase is nice for that.

That is the super scientific method, wear a lab coat while doing it.

Isn’t it easier? For DRV8825, supplied max current is Vref * 2 (from the datasheet).

So for coils rated at 1.2A, max Vref is 0.6V , or 1.2V for 2 motors in parallel. I wouldn’t go that high – about 70% of the values should be enough.

You don’t need to do any complicated calculations.

Proof: V=IR
Resistance of coil = 3 ohms. This doesn’t change (outside of heat effects)
Max current we set at the driver: 1.2A (ideally around 70% of this)
==> V = 3 * 1.2 = 3.6 [which is correct; the rated coil voltage]

No need to worry about 12V anywhere.

Everything was going really smooth, but the other day I fried my Arduino Mega. I don’t know if I did something wrong or if it’s a hardware issue.
I will try to explain what I did. Sorry if it’s not proper English.

1). Mounted the Ramps on top of the Mega.
2). Put the jumpers, all 9 of them to get 32 steps.
3). Connected the Ramps+Mega to the PC with the USB port (12v PSU still not connected).
4). Loaded the firmware.
5). Unplugged the USB and placed the DRV8825 drivers.
6). Turned on the Ramps+Mega using just the 12v PSU and then calibrated the drivers.
7). After that, I plugged the USB to the PC. At the exact moment I’m doing this, the notebook turns off.

The Mega is no longer recognized by the PC.
The 2560 chip was extremely hot, to the point of burning when touching.
* I can’t recall, but I think that the USB port of the Megas was hot too.

I’ve already bought another Mega, but I’m scared of connecting it, before knowing what went wrong with the other one.

Stuff checked with the new Mega:
A). The output voltage of the 12v 20a PSU is 12,23vDC with no load.
B). With Ramps+mega connected to the PSU, but unplugged from the PC, the output voltage of the Mega’s USB is 6,28vDC. Shouldn’t this be 5vDC? If I raise the voltage of the PSU, the USB also rises.
C). I measured the ground to ground voltage between the PSU and the Mega USB and it’s zero. So there is no voltage difference.
D). Then I checked for continuity between the -V of the PSU and the ground (earth) of the mains cable and it doesn’t “beep”, neither does the +V so there is no continuity.
I read that to fix this -V line floating, I need to bridge one the -V to the ground terminal.
Is this the right thing to do? I know nothing John Snow.


Sounds like that mega might have had a bad voltage regulator. Not sure if there is an easy way to test that.

That is why mine cost so much, It takes a long time to assemble program and test them, and on top of all that you get some duds, once and a while.

Just a shot in the dark, check the FETs on the top of the RAMPS board. The metal heat sinks are live, and they shouldn’t be touching. A friend of mine toasted one with that mistake. There are two of them that can touch with a little bending (seems risky enough to me to deserve some tape).

These are what I’m talking about:

I would love to buy everything from your web Ryan, with the kit there are less variables to worry about. It sure makes the assembly of the MPCNC much more enjoyable.
But ATM my country doesn’t allow more than 2 shipments per year, and I’ve already exceeded that. =(

Thanks Jeff, I will check that when I’m home.

Doesn’t bother me at all, I completely understand. I was just using that as reference I go though hundreds of these and some are just bad. That is why I have those ramps replacement parts. To fix the funky ones. The voltage regulator is a bit to hard to solder nicely so I will sell those super cheap one day as proto boards.

But it could be anything, like @jeff said, mosfets touching, stray solder, unplugging a motor while it’s powered up, static, ect.

Hi guys,
Fets look ok. I was just reading this:

"Overheating 5V regulator on the Arduino Mega

Unless you provide external 5V power or provide 5V through USB cable. the regulator on the Arduino supplies 5V power to the Arduino, the RAMPS (which uses very little) and anything else connected to it. With nothing else powered from the RAMPS, the voltage regulator will run quite warm but not overheat (in tests, I was even able to turn up the input voltage from 12V to 15V without overheating it). With a 20x4 LCD connected, it’s still OK with 12V input. However, if you power a servo or a graphic LCD from the RAMPS, then you will almost certainly overheat the voltage regulator. The usual symptom is that the system will not work unless it is connected to a PC via USB. Or the system may work for a few minutes, hours or days, then fail.

Workarounds include (a) removing D1 on the RAMPS and providing external 5V power to the Arduino/RAMPS; (b) driving the backlight of the graphical LCD from 12V through a series resistor instead of from 5V (a 120 ohm 1W resistor is about right for 12864-type displays), © using an Arduino variant with a more power full voltage regulator (e.g. Taurino)."

I understand half of it… but I think that what I should do is add another power supply to the Mega.
Maybe a cellphone charger? it’s 5vDC, and connect it to the Mega though the jack input.

I’m still reluctant to make a bridge between the -V and the Ground of the 12v power supply.
Mixing DC and AC doesn’t seem right.

It shouldn’'t hurt to connect the DC ground to the earth ground, but I don’t think you should. I don’t see how that would help.

I think the jack (round connector) on the mega is the same as the Vcc (12VDC) on the RAMPS board. You don’t want to connect that to 5VDC.

What they are saying is that you can connect the USB to 5V and the Vcc to 12V, that will make the arduino and the screen get powered from the USB and not the regulator on the arduino.

I don’t think that was your issue either though, because it failed when you plugged it into the computer…

Got it, thank you for explaining it better.
I was looking for possible explanations to my burnt mega, and thought that the floating and/or the missing 5v load was the issue.

But… I grew some balls today. Connected the ramps to the notebook with the 12v PSU and it is working!
I did use an old ATX power supply instead of the switching one, so I still don’t know if the problem was due to the PSU or just a faulty Arduino Mega.

Thanks you both again for your feedback.
Now it’s time to cut the threaded rod and finish the Z assembly.

Posting so that I get notifications from this thread. Very cool build so far. Hope I don’t fry my mega. Sounds like good to have an extra one in stock.

Hi, I actually finished the build and did a pen plotter test. But couldn’t update this thread due to lack of time.
It’s currently in stand by mode, I’m saving for the DWP611 router which is one of the best you can get over here, but a very expensive toy :stuck_out_tongue: