Newbie looking at 6040 (China CNC) for polypropylene 15mm sheet

I have no idea why you are tumbling parts like this. It's very non standard practice for machined parts of this nature.

Quote:

---

Originally Posted by spluppit

Again it's not working as you think it should, but the controller is not the problem in this case.

---

Not necessarily.? Mach3 by default will let you Jog away between tool changes then when cycle start then go to next operation.
However, when using external controller some functions are taken away from M3 and handled by the controller and I'm pretty sure Jogging is one of them.
So if the controller doesn't handle this correctly when it hands control back to mach3 things can get screwed up.
This why Neale was seeing an issue when probing with Cslabs controller. In Cslabs case it's actually certain versions of Mach3 which changed how they worked at fault not the controller. Works fine with some not with others.

I think this could be the case here. I don't know which controller your using but might be worth trying few different versions of Mach3. I find older versions to be more stable, like neale I use 0.28.

I nearly put in that i have zero experience of Mach of any kind. I was just drawing from my general experience of commercial controllers and especially ones with manual tool change like knee mills as an example. Some do accommodate out of position moves some don't. Mach will only be based on commercial controllers features. Looking at Jogaras reply from the mach forums it seems this was the case from what he said.

I guess what I'm saying is, why not program a simple G00 move to combat having to move out of position for a tool change manually. The controller will know where it is all the then. No reason for any errors if you do that. As you know Jazz, its a very simple and standard thing to do when required.

Quote:

---

Originally Posted by spluppit

I nearly put in that i have zero experience of Mach of any kind. I was just drawing from my general experience of commercial controllers and especially ones with manual tool change like knee mills as an example. Some do accommodate out of position moves some don't. Mach will only be based on commercial controllers features. Looking at Jogaras reply from the mach forums it seems this was the case from what he said.

---

Whether or not Mach3, or its predecessor which then became LinuxCNC via a different development path, is based on commercial systems isn't really the issue - it supports the M6 toolchange command, and allows the user to add in a couple of macros that are called automatically before and after the toolchange itself. When it's working, it also allows things like tool-height setting using a touchplate during the toolchange, which is really useful on amateur-class machines which do not have preset calibrated toolholders and the like. It supports that way of working as well, but is a bit more flexible. However, as mentioned, when you are using an external motion controller, it can get confused if the motion controller does things Mach3 doesn't know about. I would certainly suggest trying an earlier version and general reports agree that .028 is a good bet.

Quote:

---

Originally Posted by spluppit

I nearly put in that i have zero experience of Mach of any kind. I was just drawing from my general experience of commercial controllers and especially ones with manual tool change like knee mills as an example. Some do accommodate out of position moves some don't. Mach will only be based on commercial controllers features. Looking at Jogaras reply from the mach forums it seems this was the case from what he said.

---

Actually Mach3 isn't based on commercial controllers which is exactly why it became so popular because it did it's own thing without any of the restrictions commercial controllers imposed. However this eventually became it's downfall and why it as so many strange and wonderful habbits today.
As people requested more and more features for more and more oddball setups then because Art fennerty was such an obliging bloke it became a monster.
Each version that got released to take account of someone's request would break something for someone else. On top of this Art wasn't actually a programmer by trade just a genius who decided to have go because couldn't find a controller that did what he wanted. This meant the code was lets say less than easy to fathom for developers.
This is why so many external motion controllers fail because they don't have artsofts support and the code is mine field which tried to be All things to All machines and in lots ways won but in others failed badly.

All this said the fact remains that Mach3 DOES allow jogging in tool changes so it should work. If it doesn't then it's most likely down to the controllers plug-in.
Yes there are always work arounds but shouldn't have to in this case if the controller plug-in was working correctly.

Not sure where I am going wrong here.

Putting my 2 parts together and finding that one looks like it is a rhombus though it is meant to be oblong.
The long sides are parallel and flat to the touch. But on both ends, the opposite corners are not parallel.

https://s9.postimg.cc/busq3t5nz/IMG...814983_HDR.jpg

Looks like one of your axes isn't square

Great.

Well I am fucking done with this shit now tbh
I have wasted a good 6 months fucking around with it and problem after problem and now £150 in stock wasted.

IDK why this has come up now. It was fine before.

Just looking at some parts made prior to moving workshop and they are showing a slight wonk to one side on the same axis.
Minute compared to this 0.3-0.5mm we are now seeing.

Any ideas on how to fix the machine?

Close to calling it quits on this and going a different route. Though I would have to sell everything to afford said different route.
Anyone want to buy this machine?

Quote:

---

Originally Posted by JOGARA

Close to calling it quits on this and going a different route. Though I would have to sell everything to afford said different route.
Anyone want to buy this machine?

---

Sorry but what did you expect.? It's cheap chinese Router which was never designed to cut hard materials to any degree of precision.! Your expecting too much from it.

Quote:

---

Originally Posted by JOGARA

Any ideas on how to fix the machine?

---

1. Identify the axis which is out of alignment
2. Identify what should put it/keep it in alignment
3. Post photos and ask for suggestions

It sounds like you've knocked it out of alignment by hammering it beyond it's capability, you may be able to hammer it back.

Quote:

---

Originally Posted by JOGARA

Anyone want to buy this machine?

---

It's a great idea to fix it, it's probably something that way, when asked what broken machinery is worth I always quote the Rag and Bone Man's opening gambit to children - Tuppence and a Balloon :D (I'd actually give about £50 as I could use some of the parts for a 3D printer build I'm on with ;-) )

There is always the bodge option.
Machine the parts so that the out of squareness matches up.

Quote:

---

Originally Posted by m_c

There is always the bodge option.
Machine the parts so that the out of squareness matches up.

---

Been thinking about doing that. Does not impact the internals of the case as we built in tolerance to the PCBs.

As you know I have the same machine. I have wasted too many hours hammering the gantry back and forth, shimming it, squaring the spindle etc.

When alls said and done you have to realise this is a very entry point hobby machine that we are pushing to its limits cutting aluminium. Being made out of pieces of extruded aluminium bolted together, it moves. It moves with different atmospheric conditions, it moves under load/vibration from cutting aluminium.

In my case my gantry was skewed radially and actually by about 2mm and 1mm respectively IIRC. I've had mine in pieces more times than I care to count trying to make it as good as I can, opening up holes, retapping threads, shimming it. It never stays square for long. In the end I've given up trying to keep everything square as it's a waste of time unless you have an hour to kill every time you set the machine up, and just accept that it is what it is and it makes what it makes.

In your case you can solve the problem of your parts not matching by cutting one side the other way round, then the rhombus will match that of the other half, currently you have them opposing each other. Or you could cut them oversize initially then bolt them together and finish. The ones you've already made could be saved this way.

And in regards to the conversation about controllers, unless OMIO changed controller between my purchase and his it does allow for jogging the spindle anywhere between tool changes, I've never had a problem.

If you got the machine squared up you could put less load on it with a single flute cutter and a ramped profile cut where the cut depth is considerably less but the feed speed is higher.

To fix the squareness the easiest way I found was to drill two holes for precision dowels in the Y-axis, then using a precision edge run a dial indicator along the X-axis with the precision edge clamped against the dowels. I then loosen bolts on the gantry and gently tap it in the direction required before re-tightening the bolts and re-checking. Repeat a few times and you will get it fairly close to square (although you will come to realise that your gantry is probably not flat, but bowed slightly leading to it being impossible to square out completely - I just tweak until the dial indicator reads the same at 1/4 and 3/4 travel).

I ended up just having it as square as I could get it over the centre of the bed where I do most of my work, the squareness varies a little as you move around the bed due to the lack of straightness inherent in extrusions. I found it generally holds the square ok-ish after messing with it, but if you move it then you'll have to re-do.

Ultimately I do think its good for the money as an entry point, but for really precise work you have to work within the limitations of the machine and spend a lot of time tweaking. The best option in terms of value for money is to take what you've learned from the machine (plus all your tooling and associated equipment) and use it when designing/following a design on here for building your own machine. The conclusion I came to is that while I've managed to make some really good stuff on it, it simply isn't worth fixing up or modding extensively as you end up chasing your own tail.

That said I have managed to make some pretty precise parts with it, and it is still leagues ahead of the unsupported rail type 6040 machines.

The road to go with this is very simple- buy a CNC mini mill. Fit a DDCSV controller and you will never have any problems. I have had 0 problems from the time i stopped using Mach3 for anything at all.

You could start changing the controller anyway. And as they told you- using single flute mills. And reinforce all as much as possible as cheap as possible.

Been a few months. Thought I would update this car crash of a thread with a little something.

As some of you have seen already, I farmed out the work. Glad to say that it has gone really well and has actually save a lot of my time and was totaly worth it.

Went with Scott from Ward Engineering.

Helped me a lot with creating better models for faster and more efficient machining and general things like adding some margin for the powder coat and tolerances.

Overall I could not be any happier. They turned out great and I think I have nearly 40 of this style of case.
Currently have close to 80 more cases being made right now and hopefully around 25 of another part
I should note we did save a few pennies on tooling as these are being powder coated.


https://s15.postimg.cc/uibv5zb8b/IMG_1812.jpg

https://s15.postimg.cc/yep71z3xn/IMG_1813.jpg

https://s15.postimg.cc/sqiwb1257/IMG_1814.jpg

https://s15.postimg.cc/sdri4wwqz/IMG_1816.jpg

https://s15.postimg.cc/3x9caedzv/IMG_1817.jpg

https://s15.postimg.cc/3khy4hydn/IMG_1819.jpg


Powder coated in house...

https://s15.postimg.cc/p48hknltn/IMG_1807.jpg

https://s15.postimg.cc/6rchnt5vf/IMG_1808.jpg

https://s15.postimg.cc/uv39c5lsb/IMG_1809.jpg

https://s15.postimg.cc/7gva06e4r/IMG_1810.jpg


I am broke now though so if anyone wants to buy the 80mm spindle LED ring lights from me, either as a blank PCB or as a kit or fully assembled let me know :p
https://postimg.cc/image/6z04kkomf/
https://postimg.cc/image/l5fvftucn/

Nice work Scott. And that's the difference between Hobby or not.

Thanks Chaz. Sometimes the right job needs the right equipment to get the consistency and overall quality. These units are also now made in the correct material for the job, so anodising is now a realistic option should that path be looked at again.

Jogara should be given a lot of credit to be frank. He has done pretty darn well considering nearly everything has been stacked against him. 10 out of 10 for bravery and perseverance.


This the next lot being machined.

Attachment 24564Attachment 24565

Quote:

---

Originally Posted by JOGARA

Been a few months. Thought I would update this car crash of a thread with a little something.

As some of you have seen already, I farmed out the work. Glad to say that it has gone really well and has actually save a lot of my time and was totaly worth it.

---

@JOGARA, a big thank you from downunder where I have thoroughly enjoyed following your thread, and have learnt loads from it.

Cheers,
Andy

Usually these things are die cast but I guess volumes are low?

Anyway beautiful work spluppit ! Can’t beat shiny fresh cut aluminium.

Any ideas on how to square the gantry?
I need to do some pretty accurate circles in a week or so. Just cut a 22mm hole for a carbon rod and the circle is oval...

Sure that’s not due to backlash?

Other than improving the backlash situation, the only answer to that is backlash compensation in the control software.

To square the gantry the best I found was to drill 2 holes for precision dowels in the y axis and butt a precision square up against it. I then run a dial indicator along the edge in X axis, and then loosen the gantry bolts and smack it with a mallet until it’s as square as I can get it. You won’t get it perfect as the gantry isn’t completely flat.

Quote:

---

Originally Posted by Zeeflyboy

Sure that’s not due to backlash?

Other than improving the backlash situation, the only answer to that is backlash compensation in the control software.

To square the gantry the best I found was to drill 2 holes for precision dowels in the y axis and butt a precision square up against it. I then run a dial indicator along the edge in X axis, and then loosen the gantry bolts and smack it with a mallet until it’s as square as I can get it. You won’t get it perfect as the gantry isn’t completely flat.

---

Could well be backlash.
I just measured it again and my 22mm diameter is 0.04 under in the Y and 0.14 under in the X.
It is pretty close but next week I need to do a 190mm diameter hole and close fit a bearing in it.
Yes I am using a caliper... Need a bore gauge.

It is not as off as I thought but if I can get it close as possible, would be less nervous about the big part.

What Zee said although also indicate along the Y also to make sure you have the square square.

That’s a big bore and a tight tolerance for a machine like that. Good luck.

Quote:

---

Originally Posted by JOGARA

Could well be backlash.
I just measured it again and my 22mm diameter is 0.04 under in the Y and 0.14 under in the X.
It is pretty close but next week I need to do a 190mm diameter hole and close fit a bearing in it.
Yes I am using a caliper... Need a bore gauge.

It is not as off as I thought but if I can get it close as possible, would be less nervous about the big part.

---

You can't measure a bore with calipers and believe that 0.04mm measurement or even the 0.14mm for that matter..!! . . . Only a bore gauge can do that with any accuracy.

Also, it's not just the gantry squareness you need to worry about, the spindle tram and parallelism of the rails all come into play and I know for sure that these machines are not good enough in either of those areas to machine a bore that size to any decent accuracy.

Hmm

Built a new enclosure for the machine. This meant I disconnected the PC and moved it for a sort time while I built everything.
Put it all back together and now the CNC stutters. :joker:
https://www.youtube.com/watch?v=3GlH5yDGhRc

[edit]

It looks like turning the backlash toggle off stops this...

The machine earlier did lose a load of steps though and fked up my part -_-

[edit 2]
Now getting that "Is the Xpod inserted OK? it has ceased to respond..." error again god dam it

Surely if you are looking at a bearing fit , then bore micrometers are what is needed.

So. Did you do it?

Did it work?

Hi I have a 6040 machine, and it works very well, someone earlier said acrylic melts, if that's the case then your setting are all wrong, I use a lot of acrylic in my projects and have never had melting. My settings are, for a 3mm bit, !0,000 rpm, 254 feed and 80 plunge.

Well it seems you have to be lucky, I made a set square yesterday, and it turned out perfect, all I have every done it tram the spindle.

Quote:

---

Originally Posted by erniehatt

Hi I have a 6040 machine, and it works very well, someone earlier said acrylic melts, if that's the case then your setting are all wrong, I use a lot of acrylic in my projects and have never had melting. My settings are, for a 3mm bit, !0,000 rpm, 254 feed and 80 plunge.

---

What units is the 254 value in.? ie: inch/min, mm/sec, or mm/min because if it's either of the first two then I very much doubt you are reaching those feeds? Every Chinese 6040 machine I've ever come across uses 5mm pitch ball-screws and struggles to reach much past 5000mm/min. Your 254 figure if in imperial units would be cutting at 6500mm/min, If it's mm/sec then it's even worse because that would be 15,240mm/min which it couldn't reach if you threw it out the top floor window.!

I see this all the time with Chinese machines in that people enter feeds n speeds in CAM, often the default settings for tool size in software like Vcarve pro, etc, and because Mach3 or whatever control software they are using doesn't complain if it's higher than the max velocity settings they think they are reaching those feeds.!.. They are NOT, and can't possibly because the machine isn't tuned for or even capable of those feeds.

Regards Melting acrylic then it could be because of the type of acrylic, Cast acrylic machines so much nicer than the other stuff which I've forgotten the name of.! The other stuff can be a B'##'#rd to machine and melts for fun IME.!

Thanks for the feedback.
I only go by what the program I use tells me, I use Carbide create, and it just gives a figure that I use, I don't profess to be an expert
The Acrylic I use is stuff I get from eBay, so it could be anything.
Everything is in MM' So what figures am I really using.

Quote:

---

Originally Posted by erniehatt

Thanks for the feedback.
I only go by what the program I use tells me, I use Carbide create, and it just gives a figure that I use, I don't profess to be an expert
The Acrylic I use is stuff I get from eBay, so it could be anything.
Everything is in MM' So what figures am I really using.

---

Ok, that seems quite slow if mm/min, but it does depend on the tool your using. ie: # of flutes, length, HSS or Carbide, coated or uncoated, etc.
I don't cut acrylic very often but the last time I did it was with 1.5mm dia tool single flute carbide cutter in cast acrylic and I'm pretty sure I was cutting much faster than that.!

For a 3mm single flute cutter even if HSS steel, I'd be looking towards double that feed for a starting point and with double the RPM.!

Thanks for the reply and helpful info.
Igot me to experimenting, today I cut a pattern in 19 mm plywood, using the following setting.
15,000 rpm, 508 feed, and 203 plunge, with a 3 mm 2 flute bit, it cut very very well.

Quote:

---

Originally Posted by erniehatt

Thanks for the reply and helpful info.
Igot me to experimenting, today I cut a pattern in 19 mm plywood, using the following setting.
15,000 rpm, 508 feed, and 203 plunge, with a 3 mm 2 flute bit, it cut very very well.

---

You are still cutting well below what you probably could. Don't be afraid to experiment, you'll be plesantly surprised how much faster or deeper you can cut.

You don't mention the tool material HSS or carbide.? or the DOC.

If tool is HSS then try 2mm DOC 900mm/min 20,000rpm, you could probably easily cut at 3mm DOC but 2mm will be safer start point.

If tool is carbide then try 1250mm/min 24,000rpm 3mm DOC and work up from there. Carbide tooling works best when it's being pushed hard so don't be afraid to give it some welly.
The machine will be the weak link when it comes to carbide tools as you push harder because it will vibrate and carbide doesn't like vibrations or chatter as it's quite brittle, the chatter cause it to chip the cutting edges which then leads to excess heat and if bad chatter will snap the cutter.

HSS is more forgiving but can't be pushed hard like Carbide can and wears more quickly losing it's cutting edges faster.

Just remember the deeper you cut the more important chip evactuation becomes so make sure you have dust extraction fitted.

Thank you.
I am not sure what material the bits are, they came with the machine, probably cheap Chinese.The DOC was 1.5 ...
I will be experimenting with them,I will be able to push them some, they really didn't cost anything.
Thanks again.