I thought I'd start a separate thread specific to modifications to my SainSmart 3018 Prover V2. I welcome comments, but I wish I could reserve multiple first messages and edit them retrospectively so I can keep a post per upgrade and edit as I do it/learn from the experience to keep it all in one place. never mind!
Anyway, my machine is a prebuilt 3018 style CNC router. You just have to bolt the gantry and connect up a few wires – all the hard work is done for you.
Upgrades Proposed:
A 3d touch probe
The X axis gantry bearings
The spindle motor
The Z axis (the whole assembly)
A laser cutter/engraver with air assist
Possibly moving the Y bearings and rails
1) A 3d touch probe so I can physically scan objects and/or engrave/laser/mill curved surfaces and zero stock/centre find etc. Using the machine as a scanner was the primary reason for buying it.
2) The X axis gantry runs on two round bars, and it flexes a lot, it might be ok for engraving and light or steady cnc routing but the rest of machine frame is really quite solid, it can be significantly improved with a minor tweak and about £60. This invloves fitting a pair of MGN type rails directly to the existing 2020 vee channel seen behind the existing round bars – the round bars and bearings go in the parts bin. A new saddle will be designed and 3D printed, and if/when more serious work takes place, something more metallic might emerge from scraps.
3) Stock motor details are few and far between but I think its rated about about 10k, 1.5A 12v…. I'm not sure how these numbers add up but it runs at 24v 10krpm apparently, independent tests online confirm it hits about 9krmp @ 1.6A 24v no load. There are multiple "upgrades" on the manufacturers website and I happen to have a European made 20k motor that exactly matches the specification of the Chinese 20k spindle upgrade they advertise but mine has a far superior brush assembly, shorter body but higher weight, its a direct fitment.
I've tested it already and it's got significantly more welly at the same output but appears to be turning at roughly the same speed, someone has borrowed my RPM meter and I'm not sure how these PWM speed controllers really work, so its up in the air a bit. The spindle was set to the same rpm and I just assumed fitting a motor with twice the rpm would result in, double the RPM…. as I said, I'm not sure if that's how DC PWM speed control works.
4) The Z axis ….is plastic but seems solid enough, however it only has very limited travel and I can see that being more of an inconvenience than anything, the first thing I wanted to test engrave (a piece of 11mm board) couldn't be reached, I had to stack a couple and redo the clamps. The saddle that carries the Z assembly and rides on the X rails, will be no use when I change the X bearings and their layout. I will 3D print a new one and mount the spindle as close to the new X rails as physically possible, I noticed quite a distance between spindle centre and X axis centre which increases leverage and promotes twisting in the X bearings, any reduction here will further improve rigidity. I will also widen the bearing spacing, reducing my X travel, but increasing rigidity especially if/when milling!.
5)Laser…. I've always wanted one…… the PCB has 2 headers for a laser and air assist and the manufacturer does their own 5.5w concentrated spot laser with a fixed focal length (I guess that's 5.5w input power not output but I don't know, the header supports upto 2 amp)
6) Possibly moving the Y bearings and maybe even rails back to the positioning on previous version of this machine, it has less travel, but should be more stable for light milling work, I was unhappy with the saddle as it was narrower, and then I noticed the bearings and rails under the bed were quite narrow too, and then I spotted the specification change, increasing travel on all axes, I double checked a few frame dimensions and it's the same, so they must've narrowed the bearings. The controller on the V2 is a significant upgrade, featuring headers for the lasers, better drivers IIRC, a better 32bit arm GRBL controller vs the previous (presumably arduino) 8bit GRBL controller and limit switches preinstalled on all axes + 2 spare….. now 1 spare…. whoops, I don't know my own strength…
No time scale for the above. I bought the machine purely to digitise some parts that are practically impossible to measure and/or have bizare contours and stuff, I need accurate 3D models to realise another dream, to run CFD on heat transfer and fluid flow though various parts. I will probably use the machine to make balsa dummies for test fitting, if it can one day actually machine soft aluminium, I will give it a shot but I suspect I'll be paying someone else to machine the finished parts properly, as the softness of the alloy makes it quite tricky to machine without damaging.
I've managed to resist buying one of these cheap 'engravers' Russ as I'm not sure of it's practical value if you are already set up with manual machines (at least for most of what I do).
But then I didn't originally think I really needed a 3D Printer and I've found a lot of uses for that. Learning 3D CAD made a big difference in my case. It wouldn't be nearly as useful if I was just printing Thingiverse designs….
So I'll be interested in hearing how you progress.
Ballscrews might be a good upgrade if not already fitted. For the X (I assume you mean left-right) axis, it looks like the electronics are bolted to the back of the rear rails. Removing that and fitting a piece of 6mm plate right across, bolted into the edges of the side "cheeks" would be a lot more rigid. A Z axis assembly fabricated from alloy plate would be more rigid than plastic if you want to do routing.
Even on a proper mill pure soft aluminium is horrid to machine, much better to use a machinable alloy (6082?). A material which is a delight to machine is Corian, get offcuts from an up-market kitchen fitter. Good for making precise parts that don't need to be very hard or strong. There is a material not unlike MDF called Valchromat which uses a melamine resin rather than urea formaldehyde. It routes very well, doesn't blunt cutters like MDF, is through-coloured, and not nearly so unfriendly to breath the dust. Great for things like spoil boards, small sacrificial mounting plates etc. Expensive to buy an 8×4 sheet so try to get offcuts. I actually got a sun-bleached sheet almost free that had been in a showroom, I also made a clock case from new stuff (blue), I had the components CNC routed and took away all the offcuts too, I seem to have a lifetime supply at least for small projects.
Basically the rpm of a typical DC motor depends on the number of turns, the magnetic circuit design, and the applied voltage. So a motor that runs at 10,000rpm on a 24v supply will very nearly be generating 24v back emf with no load. At 12v the same motor would run at 5,000rpm. Controlling by PWM, you take the "duty cycle" which is the ratio of on time to on+off time and multiply the supply voltage by that fraction to find the effective voltage. So a 24v 10,000rpm motor supplied at 24v with 50% duty cycle will run at 5000rpm.
I got one of these in kit form a few years ago just to play with (ie. not the slick SainSmart machine.). The original electronics was very probably arduino based and had the name "Woodpecker" inscribed on the PCB. The electronics were very prone to electrical interference, crashing at odd moments. In the original SainSmart lit. they add capacitors to the motor stop this, I tried this with my primitive setup and it smoked the capacitors so I moved on to attachable ferrite rings and a big diode the "wrong" way round to trap the back emf of the motor. The SainSmart box of electronics seems more stable although I did manage to wreck a MosFet controlling the motor on one.
Adding MGN type rails in addition to the rods on the x-axis makes a lot of difference to the rigidity (somewhere out there an Australian user describes his mods).
The z-axis is made from 3D printed parts but seems relatively OK. Thingiverse has recipes for improved and larger capacity mounts but a bigger spindle would bring larger loadings and strains. Where do you you stop?
I tinkered with the anti-backlash nuts on the x-axis drive in that I added screws to push off the brass to positively take up the slack rather than rely on the spring between the two brass fittings on the drive screw..
At the end of the day the 3018 is a learning tool, an engraver. What I need is a source of easy material to machine to play with and not be too upset when it lands in the bin.
I can recommend steering clear of this style of spindle motor (ubiquitous on Amazon, Aliexpress, etc.) on grounds of both electrical safety and mechanical shortcomings.
My ultimately unsuccessful attempt to make a silk purse out of a sow's ear:
I’ve managed to resist buying one of these cheap ‘engravers’ Russ as I’m not sure of it’s practical value if you are already set up with manual machines (at least for most of what I do).
But then I didn’t originally think I really needed a 3D Printer and I’ve found a lot of uses for that. Learning 3D CAD made a big difference in my case. It wouldn’t be nearly as useful if I was just printing Thingiverse designs….. So I’ll be interested in hearing how you progress.
Regards,
IanT
Was having a look around to see how things were going (e.g working) here now and spotted my earlier post above wrt 3018 CNCs… 🙂
Well, I’ve been looking at a high-spec 3020 CNC router (from Fox Alien) that comes with linear rails and ballscrews as standard. I’ve watched quite a few ‘Upgrade 3018’ videos and also considered a DIY/CNC conversion approach but finally decided that I probably had (for a change) more money than time and needed to spend a bit more than the standard 3018-type router usually costs. A ‘Black Friday’ special brought it just within my ‘top’ price and it was just a bit too tempting to resist. So one of these WM3020s is on it’s way and once I’ve figured our how to make it actually work and I’ve got something useful to show, I’ll report back…
They’re a bit weedy for anything serious but the GRBL makes the CAD/CAM part a lot easier to learn and they are fabby CAD learning tools
There’s always something else to learn… ugh
There’s always something else to buy…ugh
If you use a program like candle 1.1.7 you can dig about in the “locked” part of your GRBL settings via the console and sort out your limits for various movements
The initial settings are often very conservative
use $$ to get the data stream which is often a single huge line (yeh great thanks!)
Yes Jason, having watched a few other 3018-type YT reviews, I was very impressed with that aluminium part the reviewer cut with it – straight out of the box. It’s the only ‘3018’ type CNC I’ve seen do something like that with ease.
I have my larger (manual) machines for the heavier work I do Ady and they are fine for my tool making and any ‘restoration’ work required. My machines are a little antique in nature and I spend far too much time on them I’m afraid but I do quite enjoy it. I was surprised how useful 3D printing was in the workshop and the WM3020 is really for my Gauge 3 work, where I hope to cut model parts in ply, hardwood and (non-ferrous) metals. Most of this will be small, fairly simple 2D stuff but if I can get the machine to do more complex 3D work then that would be great too.
I’ve already been looking at the software required to make the machine work and have downloaded several things already, including Deskpronto (DXF to GCode) and UGS (GC Sender). From the video, it seems Fox Alien supply Candle with the machine and it looks easy to zero the tool etc with it. On my 3D Printer I decided to just learn one slicer (Cura) and stick with it and I’ll try to take the same approach with the CNC machine. So I’d be interested in other’s folks experience with regards UGS versus Candle?
I design everything in Solid Edge these days and can produce DXFs for any parts I want to CNC. I guess I could still use Fusion for my CAM but I’m not sure I really need that complexity – hence Deskpronto.
Thank you for the link to the GRBL G-Code lists btw Ady. They will certainly help with understanding what I’m looking at in UGS/Candle. All helps to keep my aging grey cells active too! 🙂
That was what looked good to me, many videos I have seen of these small machines just nibble away at aluminium and produce dust, he was making good size chips.
I find my CNC makes far less mess than the manual mills and although I only have a 5000rpm spindle it is also quieter than the X3 as there is no gear noise from the head. As the 30x 18/20 machines are so small it is not a major job to put them in an enclosure if needed.
The machine Ian links to has a 10K spindle which is not too bad, it is the ones with 20K and upwards that can wine combined with the less rigid ones where high pitch chatter becomes a problem but the one in the video does not seem to suffer from that.
I have two workshops in practice- my large machines are all in my ‘Outside’ shop (which is unheated) but I also have an ‘Inside’ shop (small room) where I have an EW lathe, hand shaper and other ‘modelling’ type tools – as well as the 3D printer now. Space is at a premium in both but I spend much more time in the small shop in Winter than the outside one. SWMBO seems to think going to the other shop when it’s cold isn’t good for me (which is how I managed to get permission for the inside room 🙂 )
I haven’t built an enclosure for the 3D Printer and have stuck to PLA+ to minimise any fumes/VOCs that I might get using other kinds of filament. I’ve also decided (for the same reasons) that Resin printing might not be good for me.
I’ve just seen Jason’s post, which is encouraging from a noise pov – but as I intend to rout wood/ply parts as well, dust will be clearly be an issue. So I think an enclosure will be mandated once I start to explore that side of things… .
Having had a CNC life for a few years I would say the stepper motors are an insignificant item as far as noise. The main cause is the actual cutting action and the second the spindle/motor combination.
I’m afraid there’s a limit on how much I’m willing to spend Ady.
I looked at a ‘conversion’ kit and it was all adding up very quickly – NEMA 23s, Controllers, Ballscrews and of course – the time to actually do the work. A few years back, I might have been tempted but I’m beginning to realise how precious time is now. Another factor is the ‘downsizing’ issue, the time when we will have to move and perhaps I can no longer house my larger/heavier machines. I will then have to downsize my ‘engineering’ activities to match I’m afraid. Fortunately, modelling in Gauge 3 provides scope to build nice big models without needing too much room to do so. Maybe a carriage or two? 🙂
Actually, a number of my G3S colleagues have been using CNC for a while now and producing quite outstanding results. Most recently this GWR Armstrong has been completed and it really is quite special!
I have a feeling that Mike mentioned having access to a Syil CNC. When I checked what a ‘Syil’ was though, it was a little bit more than I could afford, so the WM3020 will have to do! 🙂
I’ll be very interested to see how the 3020 compares to the 3018. I have the latter (sitting on a shelf). Apart from the noise level the major problem was getting the table aligned. There is no formal alignment method. It’s just a question of shifting it on the mounting screws. Then as you tighten a screw it shifts again.
The only reliable method seems to be to make a somewhat thicker table, tighten the screws and pin it. Then put a milling cutter in and plane the surface. I was about at that stage when “other things” intervened a few years ago and I’ve never go back to it since.
From the pictures, it looks like the 3020 might be better.
Fairly standard practice if you are mounting work where depth needs to be fairly accurate, I often skim the top of a mounting block if held in the vice or a backing board if fixed to the table. When you are setting your machining heights to 0.1mm increments or less then you need the work and what it is fixed to to be flat. particularly if you ar estarting with stock that is already the finished thickness, not such an issue if it is thicker as you just skim the top of the work and often the bottom as a second op.
Good example here where I set the depth of the outer profiling cut to be just a bit more than the material thickness and you can see I have cut through but not gone as far as cutting into the board below. So within two thicknesses of tape.
Can you tell me how you got the letters engraved please?
Im trying to make some labels out of two colour plastic.
When I import the DXF into Desk Proto it wont create the cutter path around the letters because they are to small. I just need the letters to be a single line but they appear in the DXF like this so it wants to create two cutter paths
I have a basic 3018 cnc that was given to me, it runs a Developower v2.0 controller. All I need it to do is some basic engraving on two colour plastic
I used Desk proto as it was free and mentioned up-thread also Candle, should I be using something else.
I think a full enclosure and vacuum connection would be first on my list if in the same room as a 3D printer – well as anything really.
One of the reasons the dedicated laser machines are so ‘flimsy’ is to keep the weight down to be able to move fast. You will probably find you have to dial down the laser a lot to avoid burning even with air assist just because it can’t move fast enough. Oh and another reason to make that enclosure to take the smoke away – vacuum needs exhaust hose to outside.