I love it. I need to try that out soon. It seems as if Marlin is at a really stable place now and they have not touched linear advance in a while so no better time than now. Glad you are always experimenting!
Ok, so I made a few prints this weekend. Now I can confirm you 100%: this linear advance thing is awesome!
First, my wife asked me to design her some kind of thing where she could put little milk containers for her coffee. I tried to convince her to drink tea instead, given the fact that she is Chinese, but she insisted and told me I was an offensive racist. Which I knew already.
Anyway, I designed this little milk dispenser with onshape. I printed it fairly slow. It took 4 hours to print, but the main reason was the slicer. I bet it could have taken 2 hours and a half if it didn’t have to do so many useless travels to print ridiculously small infills. Unfortunately this is the only thing that I cannot tweak up in my slicer software, and the only defects I had on this part were due to this, otherwise it would have come 100% perfect.
My wife, waiting for her little milk dispenser to be done:
[attachment file=67505]
The thing after a bit of cleaning:
[attachment file=67506]
[attachment file=67507]
And now in its final spot:
[attachment file=67508]
This worked very well actually, there is almost no Z wobble visible and very few defects. The only visible defects I saw were due to not enough cooling on the right side (this will be fixed this evening) and stupid useless micro infills (unfortunately I can’t do anything about that, except to take really good care of designing my 3D parts with all dimensions being multiples of my nozzle diameter. Maybe I’ll have to experiment with other slicers.
Anyways, I wanted to see how great of an improvement linear advance was over what I previously printed.
So I re printed one of the arms of my laptop holder. The original one had issues during printing, I printed at a too low temp with too much cooling and it slightly delaminated in some places, so I had to reprint it. The thing wasn’t my best print, but it wasn’t my worst either, I’d say it was in the low average quality of stuff I was able to print before.
I’ll now finally shut up and let the pictures speak for themselves. Can you tell which is which?
[attachment file=67509]
[attachment file=67510]
[attachment file=67511]
[attachment file=67512]
So, what do you guys think, does linear advance makes a difference ? 
1 Like
Whoa, that is a huge difference. Corners and even over hangs look much improved. I have to try this ASAP.
Hello everyone,
It’s been quite a while since I didn’t work on some mods on my MPCNC. Now that it works flawlessly, I think it is time to make the print head a bit more stiff and compact. It is quite a challenge since I have so many things in there, but it does take too much space currently, it looks bad and the whole thing is a bit flimsy. Plus I have to be there to untighten a screw before anytime I do an autolevel and then tighten it again just before it prints, so it is not fully autonomous.
My plan is to try to make something more compact, to gain a bit of printing space, to do it stiffer to limit print defects and to find a way so that I don’t have to do any manual operation. I’d like also to be able to swap for a different print head in less than one minute.
First thing I had to do was to make it stiffer. The main reason the thing is not stiff is, I think, the way nozzles are traditionally attached to the rest of the printer. I’ve got an E3Dv6 volcano, and the thing is only connected at the top, which leaves a long lever arm of about 80mm. Plus I have this whole liquid cooling system I’ve done I had to attach somehow. It’s a bit like if you try to hold a pencil, you’ll be more precise by holding it close to the tip than far away from it.
So what I’ve done was to rethink the way I’ve done this cooling system. I made several different designs and finally found one that works, apparently. The major difficulty was of course to get the thing water tight, but it seems like I finally managed to do it.
So basically it looks like this:
[attachment file=73005]
One O-ring on the bottom:
[attachment file=73006]
One O-ring on the top:
[attachment file=73007]
The water comes from one of the coupling and goes out from the other. The whole thing is made out of 3 parts sandwiched and tightened together by 4 long screws
[attachment file=73008]
I’ve tested it for a while and it doesn’t leak, aside from a few little holes on the sides which are due to my Delta 3D printer not working very well these days. But it’s an easy fix, just heat the tip of a screwdriver and melt the hole spot to fill it.
[attachment file=73009]
I think the main concern of most of you would be “Will the plastic be able to handle the heat?”. Well I’m not 100% sure yet, I still have to test that, but I’m very confident it will. Reason being that it shouldn’t see any heat in the first place, because it will be constantly cooled by the water inside, and it won’t touch anything hot, there should always be at least one or two millimeters of gap. The fins inside will remain cold and even if they did get hot, they are not touching the plastic since there are the O-rings in between.
I’m pretty excited about this mod, since it would be an easy, affordable and most importantly doable by anyone water cooling mod, which not only adds water cooling but also add stiffness to the whole thing. Previous method was already easy and cheap, but I think that most people didn’t have the silicon tube or didn’t know where to find it. For me, having this mod means that I can build several print heads with different nozzle sizes, and swap them in a few seconds.
So, now that the whole nozzle assembly will be a lot stiffer, I can go ahead and design the rest of the print head.
2 Likes
Come on Dui you have a CNC, time to mill your own water block. You know it is going to happen so you might as well just go for it. Mill in a few channels, tap a few holes, clear plastic window, RTV sealant done. Take you an hour.
Hehe, that’s true, I could do that, but I’d like to invent the most affordable and easy system as possible and maybe contribute a little to the 3D printers community, so that anyone can do it, not only myself. This thing will be for people who own a 3D printer, obviously, but they don’t always own a CNC so it might help, who knows.
Plus I prefer the 3D printing process, because it’s quiet, it’s not messy, it works every time and once your slicer parameters are ok it’s basically a no brainer, you don’t have to figure out paths, tools, stuff like that. 
Yeah I know, that’s just a convoluted way to say I’m lazy XD
3 Likes
Well Dui, I made one of you previous silicone tube style cooling mods and tested it in the sink and it was fine but I never got round to mounting it on my anet a8 cause I didn’t want to break it. I have today bought a broken printed off ebay for £44+£12 so £56 all in which hopefully the only extra I will need is an extruder motor. I figure I will get this one working and use the water cool mod. This way I can have a fine detail printer with the normal 0.4mm or less nozzle and use the new beta printer as my large scale 1.2mm nozzle job and employ that linear advance too.
recently I got some 2mm galvanized sheet 220x220 to allow my 18m inductive sensor to see through the 3mm borosilicate glass. It worked but now my bed heat up time is stupid long. Also the next mods I need to do are making a temperature controlled enclosure that can both heat and cool. My garage office is getting too cold this time of year for decent dependable prints.
I also want to design a printer based on the piper printer v2 and wood 20mm square cross section for the top and bottom squares. I got some of the first ideas done in fusion and can lift the rest of the parts from the original piper printer. Wierdly I have an idea that Id like to see how much of it I can mill out of wood rather than print and still have it accurate. That way I have a real project which requires me to finish my mpcnc.
Loving your designs as always.
1 Like
Hehe, I do understand your fear, but I never had any issue of leaks with my ghetto silicon tube system, just had a few droplets once because I did a mistake, but really nothing else. But yeah, I understand it is a bit scary because it looks totally crappy.
I’m not sure the watercooling will be very useful if your printer is a small one, but it shouldn’t hurt anyways. It will run a bit quieter and give you a more stable extruder temperature though.
I suggest you to use linear advance on all of your printers, it will make a difference even with a small nozzle, you’ll get more accurate parts no matter what. I’ve printed quite a lot of stuff since I’ve implemented this function and seen no downside at all, it’s just better at everything.
That’s something I’m about to try myself. Currently my Z probe is the nozzle itself, which works great, but I just ordered an inductive sensor two days ago to, hopefully, replace the whole system and get a bit more rigidity. But the thing that worries me is exactly what you’re talking about: I’m not sure it will work on my mirror, so I’ll have to test.
Steel is a very poor heat conductor, so you might want to get a much thinner plate and possibly experience with some heat transmission goo. Just like the ones used on CPU cooling, you can try to apply that goo between your hot plate and the metal sheet. It won’t make miracles but it might speed things up a bit. You can get this thermal paste in big tubes since it is widely used for heatsinking mosfets, it is extremely cheap (maybe 2 or 3 bucks). Anyways, i bet it is still way faster than mine, since my heating bed takes literally 30-45 minutes to get to the right temp, even more when the garage is cold.
That looks like a sweet printer. Very similar to Ryan’s designs and totally scalable, I like it.
I installed a BL Touch a while ago, and its made printing fun again. Perfect first level every time, no fuss.
Why are you guys using non-touch and worrying about not detecting, when a touch sensor detects everything?
1 Like
The printer I have with an inductive sensor has PEI on top of aluminum heatbed. It gets very repeatable results, but the window is very small. It is higher than the nozzle, but sometimes collides with the print (if the print warps some) and it also sometimes fails to probe (because it doesn’t detect far enough down). It’s a small window, really. I highly doubt the mirror will work with the inductive sensor.
The BL touch I have works well. I had a knock off one and that was ok, but not great. It was inconsistent probing in the same place. IDK, maybe I broke it early in my experiments? The authentic BL touch works great but it’s close to $40. It’s worth it IMO, but it’s a big leap amd if you wire it backwards, it goes up in smoke.
Well, mostly because I don’t really know this thing and I’m not sure how it works. For instance, does it deploys itself or do you have to mount it on some kind of servo motorized support? An other issue is that you can’t probe directly at the spot your nozzle is, it will necessarily be offset for a few centimeters, which isn’t that accurate in the end. So basically I built my own system which seems to be pretty similar, but should solve all these issues.
I’ve made a first assembly test this weekend to see if my design was ok. After a bit of massaging with a wood chisel it did fit right eventually.
[attachment file=73534]
BTW, I have a little suggestion/request for Ryan: could you add a little section in the forum or in the website for “verified” add-ons (things that have been designed for your machines by other people and are known to work fine) and some basic documentation for people like us who try to design some accessories for your machine? I’m asking this question because I had to design this thing from scratch, and it was a bit of a pain to retrieve the dimensions between the Z tubes as well as the correct angle for the support to attach with the tubes. I did it from the STL files but it took me a bit of time and it wasn’t perfectly accurate. It wasn’t a huge work to make it right, but some kind of basic plan with key dimensions would have make my life easier. Just a thought. Personnally I don’t really like to use the default tool mount because it introduces an offset for every tool. It’s not a big offset, but it is necessarily 20 or so millimeters, which adds unnecessary lever to the Z axis.
Anyways, the thing seems to work, I’ve let it heat at 220 degree for a few hours, the plastic is getting a bit too hot where the volcano heating block is protruding, but after putting a bit of insulation it seemed to be all right. No leaks, nothing seems to be wrong so far.
[attachment file=73535]
This thing also includes a Z probe. It uses the nozzle as itself as a Z probe, just like my previous version did, but this one should be way more sturdy. The whole nozzle assembly is secured on this platform which can pivot. It is pressed down by the two beefy springs you can see at the rear. When the nozzle will hit the table, it should move the assembly very slightly, enough for a screw to cut the IR beam of the sensor you can see between the springs. If the screw is set correctly, it will only need a tiny little bit of movement, less than .01 mm to trigger the thing. Currently I’m just worried that the spring might be a bit too stiff but I’ll have to experiment and see.
[attachment file=73536]
I still need to add the final thing, which will be an adjustment system for perpendicularity. Hopefully this thing should produce more repeatable and consistent results, it should be way stiffer and totally autonomous, in a much smaller package (which means a bigger print surface, I’ll gain around 120 mm on X and 50 on Y). Plus I’ll be able to switch between different nozzles in a few minutes. Next, if it works fine, I’ll just have to add a compact part cooling system.
1 Like
Yikes, who will do the verifying? As for the dims, just ask I will give them if I have time. Direct mounting means you have to make three versions or 2/3’s the users can’t use your part. I do understand the offset drawback though.
I think it could be a system like “if 2-3 or more people successfully used it then it can be considered as verified”. Or maybe a few “power users” who can test some of the interesting stuff if they are willing to.
It wouldn’t be for all the versions or all the items, but some items like endstop supports, cable chain systems, electronic boxes, some common tools like the needle cutters, drag knives, most common router brackets, vacuum systems, and whatever else I forgot. Things that most people will need at some point when they regularly use the machine and that you are not proposing to sell currently.
Of course you wouldn’t be responsible for the support, those items could be linked with the names of people who designed them so we’d just have to redirect people’s questions or complaints/demands to their respective designers.
Basically what I have in mind would be a page with different categories just like the ones I’ve mentionned above, a picture of the stuff, a little description about which version it can fit on, a link to the stl file, the name of the guy you should refer to for questions, or the corresponding thread and if possible leave the possibility for people to leave comments to describe how it worked for them.
I don’t know if this is too much workload for you, in which case just forget it, but I think it might be cool, convenient and reflect the dynamism of the MPCNC community as well as all the uses you can give to this machine. Kinda the same way the gallery does it. I remember I was a bit lost at the beginning and I had to go to thingiverse, try some random stuff and having no idea if it would work, I’d prefer to be sure before printing things that sometimes didn’t work well. Anyway, just a friendly suggestion, if you have other plans or view on it I can totally understand and I won’t insist.
1 Like
It’s a good idea. Maybe a voting system or something. Thumbs up, or ratings. Not sure what that would take. There was a page like that early on but we evolved so fast it was quickly outdated. Now it might work.
1 Like
MPcncThingiverse…?
A repository for all things MPCNC or lowrider or printer.
Hello guys,
It’s been a while since I didn’t update this thread.
I haven’t done much these past months, but I had the opportunity to test the watercooling system for a longer time. It works very well, I didn’t have any issue with it. I’ll continue to test it for a while and then I’ll release the files if I thing it is reliable enough.
Meanwhile, I finally motivated myself to create a thingiverse account, since I wanted to share a design with our beloved James Bruton so that, hopefully, he could use it to control his pneumatic stuff.
Anyway, I posted my design for the beefy MPCNC corners and feets. They’ve been installed on my machine for a while and so I can guarantee they work fine. I only made them for the 25mm version so I apologize to all of you living in weird countries who use crazy measurement systems. 
Anyway, here is the link:
Hope it will be useful!
Awesome, and I hope your part shows up in one of his videos.
1 Like
Hello everyone,
It’s been a while, I’ve been busy with many other projects, so didn’t have much time to play with the MPCNC.
But I did use it to print something useful for a friend: a big electric box so he can hide wires in his home’s wall he’s renovating.
So firstly I made the design of the box:
[attachment file=93008]
[attachment file=93009]
Then I printed it, using a 1mm nozzle. The walls of the box are 2 mm thick and each layer was 0.5mm. It took around 6 hours to print at fairly low speed because I wanted good adhesion so I also cranked up the temperatures to 225 degree C. This thing is quite strong. I ran out of black filament during the print so I had to replace it with red, hence this weird dual color. It will be painted in white anyway so it doesn’t really matter.
[attachment file=93010]
In the end it worked out pretty well:
[attachment file=93011]
[attachment file=93012]
[attachment file=93013]
I just had a tiny bit of warping on the bottom, probably because my heating plate didn’t quite reach the temperature I intended because the garage was very cold. On the next print I’ll add some flat circles in the design around the corners so it shall adhere better, then I can cut them off after it’s printed. Kinda like a brim, but only where it really matters.
Also, I forgot to check for squareness before printing so the thing is very slightly off, but it won’t matter much.
The good thing is that my fully 3D printed water cooling seems to work very well, even at 225 degreeC for 6 hours. I’m now pretty confident it might be reliable. I’ll have to print a second box so I’ll see if it still holds but so far it looks like it works!
4 Likes
I have been following your posts for quite a while now, so it’s always nice to get an update on what you have been doing
1 Like
Hey guys,
It’s been a long while without updates… reason being I had to move out of my house and find an other one, my landlord wanted it back.
So I found an other place and it took a while to move stuff, unpack and start organizing again. So far I haven’t got the time to restart my good ol’ MPCNC, but it is the next thing on my list. I just thought I’d share what I’m doing right now, since it can easily be applied to the MPCNC.
Basically, I’m building the control enclosure for my 6040 CNC. It uses almost exactly the same electronic setup as my MPCNC, the only difference being that I’m using TB6560 drivers instead of the tiny crappy DRV drivers. This way I could use 24V to feed the motors, so that’s a bit more speed and power.
I chose to use an old drawer I had laying around from an old useless furniture to make the enclosure. I thought it would be a good idea since it could be very easily accessible, yet discreetely hidden and protected under the CNC table. Table itself is not finished yet, it will be made of 4545 aluminum extrusion directly bolted on the concrete walls, so it will be extremely sturdy. I’m just waiting for my wooden planks to arrive. It will be fully enclosed and hopefully it should look very nice.
Anyway, Here is how I did the control box:
The components layout, two power supplies (48V for the spindle and 24V for the motors), a DCDC converter for the Arduino/ramps, the 3 stepper drivers and the brushless spindle controller:
[attachment file=117363]
I added a few controls in the front of the drawer: spindle control and an emergency switch. Of course I covered it with carbon wrap to improve the drawer racing performance:
[attachment file=117364]
[attachment file=117365]
So far I’m still using Marlin and will probably keep using Estlcam for the toolpaths. Maybe I’ll see later if I go for a more professional setup using Mach3, but I really like having the possibility to use a SD card so I’m not really sure yet I’ll go this way.
Here is how my new garage looks like, I was very lucky to get my hands on a lot of nice aluminum profiles for free, So I built a lot of stuff using them and plan to build a lot more. I’ve built this workbench/miter saw station, it took me less than a week from start to finish, working in the evenings. Aluminum profiles are super nice to work with, it’s really quick to put all that stuff together.
[attachment file=117366]
[attachment file=117367]
[attachment file=117368]
Hopefully I’ll be able to restart My MPCNC very soon, I have lots of stuff waiting to be printed so I’m quite impatient to get it back!
6 Likes