New CNC Mill on the horizon

[fvfdrums]

Ok,

so a bit more research, how about the cold air gun approach? I have a decent compressor with plenty CFM output. If the temp is right would the aluminium cut well with a cold air system? this would remove the need for coolant.

The machine will primarily be cutting ally plate I would imagine but it will be used to cut hardwoods as well and I know of the evil wood snot you speak of when mixing wood and coolant as I have had it before on my lathe! its worse than concrete.

[JAZZCNC]

Ok, so a bit more research, how about the cold air gun approach? I have a decent compressor with plenty CFM output. If the temp is right would the aluminium cut well with a cold air system? this would remove the need for coolant.

Never the twain Shall meet IMO but if you must cut wood then I wouldn't use coolant but instead plenty of Air and put up with the Mess and lesser quality finish.
Failing that then think about system that use's coolant that evaporates quickly like ethanol, good but expensive.!

One problem with coolant or should say small amounts of coolant is that it Can shock tooling and cause micro cracks in certain tool materials like Carbide. So it's often considered better to have No cooloing at all or Go large and have full flood cooling.

With Aluminium I find Air alone isn't enough and it needs just a small amount of lube to help with any hard Sticky spots and melting to tool, which happen often in lesser grade Ali like 5 series.
This is why I like the idea of John's system in the Mister thread because it would mean having full control of the squirting but mainly use air for clearing chips. This would work well if your mixing materials.!

One day I will get round to finishing a directional Air & coolant squarting ring around the base of spindle which follows the inverse of Axis movement and squirts air or coolant as required, Bit like Datron system does.! . . . . Just not enough hours in a life time.!!

[Gotty101]

Gotta.... I searched cable trunking. If you search chalon components on Google go to their website and then look under the cable management section its in there.

Thats great thanks for that.

Trev

[fvfdrums]

Dean,

Thanks for that, I love the datron machines they are great, I have not looked closely at the accessories but the coolant head you mentioned looks cool I just checked it out.

I had a word with my friend who is building this with me who also designs all the electronic wizardry in our other projects and we are going to have a crack at something similar. I have a bit of background in pneumatics so I am hoping if we put our heads together on this we could come up with a neat solution that maybe others could use as well.

so in terms of the fogless mister am I right in thinking there is an air supply that can run independantly and the air supply also feeds a pressurised container (regulated) full of coolant which has its own control valve to allow the oil to be introduced into the flow?

[Boyan Silyavski]

Air supply pressurizes at 2 bar the container which is just about right for the air flow needed, so no need for super precise separate regulator at the end, one normal regulator with filter before the container is enough. There is a second precise valve that regulates the coolant flow in the nozzle. 4mm OD tubes each are enough for the flow both for air and coolant. 0.8-1.5 mm nozzle at the end depending what exactly you want to do, i would say 1mm is perfect flow wise and control wise. Polycarbonate containers are rated 6 bars usually. the precise air flow formation depends on the outside angle + inside angle combination /of the nozzle tip/.

Hope that helps and clears the basic principles of the design.

[fvfdrums]

Thanks Silyavski, that certainly does.

I like the idea of using a system like this but the worry is still even if we are really over the top at cleaning the machine after its been used for the odd bit of hardwood are we likely to still get a build up of "woodcrete" as I shall refer to it as from now or is it likely to be far less due to there not being flood levels of coolant involved?

Certainly I will be trying to deveklop something that directs the air flow relevant to the direction the cut is going that should not be a problem at all given the projects we (the guy building with me/my business partner) have going on I believe we already have something we can just reuse for this anyway

[Boyan Silyavski]

I like the idea of using a system like this but the worry is still even if we are really over the top at cleaning the machine after its been used for the odd bit of hardwood are we likely to still get a build up of "woodcrete" as I shall refer to it as from now or is it likely to be far less due to there not being flood levels of coolant involved?

Here is a video from yesterday when we were finishing my new machine Z plate. We were going to use fly cutter but at the end decided to get lazy and drink coffee so CNC-ed it on the other mill, as its the back side so it would not be seen. Ignore the dull cutter.

Until sec 15 on the video you could see nothing, but the flow is there just enough to cool it. Then i decided to give more flow so it could be captured on the video.

For the sake of using the camera the nozzle was pointing from right to left and as you could guess cooling the back instead all, whencut was going left-right , but even like that it worked. How i know? We did not break a bit and the cut was good. Ideal is when the flow comes from our point of view, when we are in front of a mill.

Another thing i learned yesterday is that to clamp the goose neck at the dial hole on the mill is good and easy, but best would be a thin collar clamped directly at the lower part of the arbor/??/ , so it moves always with the bit and always cools where necessary.

As you see there is no mess, we were using the bigger 1.5mm nozzle at minimum adjustment. According to Jeff he could cut say 1/3 faster at least and he is using typical HSS cutters/not sharp to my standard :hysterical:/ next time i will bring him brand new USA made micro grain carbide cutters that are ultrasharp and polished, so we could see how much we can push that thing. Anyway i soon am finishing my machine so i hope to see what happens with cooling and me taking 16k rpm, deep, fast passes in ali, checking my setup rigidity.

PS. After reading about the pulsing, yesterday had to test how good my setup leans chips. I will make more test and video but the size chips you see on the video, it blows them easily away,without any problem . tested it on the bed where there were mountains of them, 15cm away which is a longer distance than usually. So no worry there, i believe i will have no need for pulsing. Will have to check for steel.


https://www.youtube.com/watch?v=c07M_-QUVUE

[fvfdrums]

Thanks for that video it looks good im going to look into it further.

In the meant time I've got the y axis moving now, it's great in one direction so with the flanged edge of the ballnut leading. However in reverse back end of lead screw leading it sounds like a wood pecker. Click click click every rotation. Any ideas? It's all greased up with lithium grease, not enough grease? Crap in the nut? Don't get it on the z axis

[fvfdrums]

Wood pecker issues seems to have been relegated to a random peck every now and again. I fill the ball nut with some more grease and ran it up and down so I'll keep going and see what happens.

My next question is about series or parallel connection of the x axis motor? Which should I do? It's a nema 34 motor with 8 wires rather than 4 as per the nema 23. I've read about the two but just want people opinion as to what's best for this machine?

Thanks

[JAZZCNC]

No brainer go with Parallel.!! . . . .In series you'll need a lots of volts for any kind of speed and torque. Parallel gives you more torque at higher Rpm.
Series is good if your not going too fast Ie: on Mill as it gives more torque lower down but soon as Rpm start to rise then torque drops rapidly.

[fvfdrums]

Cheers Dean!

[fvfdrums]

Ok so thats all three axis working now great stuff! very happy with that, just need to get the limit and home switches wired in and the spindle housing made up and then I think its on to the bed and some fine tuning of the machine.

I am thinking about getting a full sheet of material for the base of the bed and then assembling aluminium strips on top to form t slots with a wall around the outside to contain some material however it won't be very high because we still need to be able to move the bed in the future so maybe 150mm hopefully it should contain a small percentage of the shrapnel I fully expect to end up on the workshop floor!

what thickness should the base piece be.... or should I just mount them direct to the top of the steel bed.

[JAZZCNC]

I am thinking about getting a full sheet of material for the base of the bed and then assembling aluminium strips on top to form t slots with a wall around the outside to contain some material however it won't be very high because we still need to be able to move the bed in the future so maybe 150mm hopefully it should contain a small percentage of the shrapnel I fully expect to end up on the workshop floor!

Put some Thick Plastic sheet round the sides so then you can raise or lower and still contain everything. Thick along lines of bellows but all around edges.

How to have Bed.? . . .Well there's million dollar Question.!! . . . . . . So many options and not one will suit everything you do on the machine.?
What I mean by this is that you will always find that some Job won't suit how you have the Bed.?

My bed is Aluminum pieces arranged to make a T-slot but I find most the time it's covered with a Spoil board that I screw into rather than using the T-slots.?
Reason being is that 95% of my work is milling aluminium that I drill thru, Cut thru or Pocket thru which makes a bloody mess of the shiny aluminium bed.
The work piece always falls so that you can't reach the bloody T-slots or I need scaffolding to reach them for secure clamping. 90% of the time most jobs have one or two holes and if not then usually no problem making some around waste edges so screwing down is easy and quick.

On the machine's I build I find the best setup is Solid stable base board, Aluminium or Palstic if using coolant with a Matrix of tapped holes Drilled or in router case Marine Plywood with Laminate face and threaded inserts. Then a sacrificle surface on the top with Same Matrix of holes drilled for Access for clamping. The top board can be screwed into if needed, which is often the case. It can cut when profiling, pocketing or drilling and be re-surfaced when becomes too chopped up.
It can be removed easily when more accurate work is required like engraving, which often needs a very flat surface and work sticking down.

So my Vote is either Aluminium plate with matrix of holes or Marine ply with laminate surface. Both using surface board.!

[fvfdrums]

Once again thanks Dean, I will weigh up the pros and cons for what we are likely to be machining in the near future.

So I spent yesterday dismantling the z axis as I knew there were a few things I needed to take care of before I was happy to say that was done. One of them being some proper mounts for the spindle being as the ones on there were temporary polycarb ones.

I have drilled the plates for the pinch bolt just have not put the slot in it yet so I will be doing that shortly however I wanted to get all the covers made up whilst it was off the machine and easy to access.

I have also put in place the holes for the internal pipes I am putting in for the coolant / mist / blower system - more on this to come but there will be 4 small heads which will direct the air/mist relevant to the path of the cut as the machine moves around. Similar to the datron system Dean pointed out earlier in the thread. Currently waiting on some solenoid valves to turn up for this so I thought I would prep the pieces ready.
1508715088150891509115090

[JAZZCNC]

Francis is that a cover your putting over the spindle. If so why.? . . . . Covering the whole Z axis is a good idea as it keeps debris out of screws and rails etc but I wouldn't cover spindle as it traps in heat causng cooling/pump to work harder and potential for heat expansion.

[Boyan Silyavski]

Francis is that a cover your putting over the spindle. If so why.? . . . . Covering the whole Z axis is a good idea as it keeps debris out of screws and rails etc but I wouldn't cover spindle as it traps in heat causng cooling/pump to work harder and potential for heat expansion.

Especially true if you forget to switch on the pump or if hardwired like mine, it switches off for some reason and you don't see it. I have made a habit to check temp with hand.
Normal work will not f^^k the spindle, it heats a bit /50-55C/ but that's all. But with that cover...

[JAZZCNC]

Especially true if you forget to switch on the pump or if hardwired like mine, it switches off for some reason and you don't see it. I have made a habit to check temp with hand.
Normal work will not f^^k the spindle, it heats a bit /50-55C/ but that's all. But with that cover...

Just been experimenting with cooling of these spindles and I've got it down to sealed system with no external water tank etc which all fits inside my spindle covers. Lovely and neat running lovely and cool in bench testing @ 24K going to give it a serious thrashing cutting over weekend to see if can fry it.!!

[routercnc]

Just been experimenting with cooling of these spindles and I've got it down to sealed system with no external water tank etc which all fits inside my spindle covers. Lovely and neat running lovely and cool in bench testing @ 24K going to give it a serious thrashing cutting over weekend to see if can fry it.!!

Liking the sound of that Jazz. Always looking for ways to tidy up the overall look including the plumbing. Any chance of some pictures if it works? Or even if it doesn't . . . .

[fvfdrums]

Thanks for the advice once again. My intention is that the top section above the spindle is going to house the "controls" for the cooling/misting system. There will be vents in the housing once finished covered with a mesh to allow air in and out but due to the "other" bits in there I wanted to try and keep it clean.

I will however keep an eye on it all and monitor the temp of the unit. I was already planning a digital temp readout on the control panel so one for the spindle is no additional problem.

Another issue encountered. These linear carriages and the infernal little ball bearings that keep falling out every time I have to remove the rails!!!!!!

So I got on the 3D printer and made some rail sections to insert in when the rails not in there. They came out pretty perfect. I'm assuming new carriages are usually shipped with something like this?

Anyway I've still had a million of the little buggers fall out. What's the best way to keep them in as your feeding the rails? Ive tried grease but they still come out its the last "2" that are hardest to fit.

Patience? - I'm all out of that and there's none on back order :p

http://tapatalk.imageshack.com/v2/15/04/02/ce9f4919ce2bd4ec0aaf3cba9813bd56.jpghttp://tapatalk.imageshack.com/v2/15/04/02/d3298acae15be37d8b38ff038148a466.jpghttp://tapatalk.imageshack.com/v2/15/04/02/3432d37be59448d1054d14069caa5b75.jpghttp://tapatalk.imageshack.com/v2/15/04/02/0c6b99015b16902aa47105e918e2a3bb.jpg

[JAZZCNC]

Anyway I've still had a million of the little buggers fall out. What's the best way to keep them in as your feeding the rails? Ive tried grease but they still come out its the last "2" that are hardest to fit.

just get a piece of foam slightly larger than rail and insert into slot then let the rail push the foam out as you insert rail.
Your correct that when new they come with inserts to keep balls falling out but they are not anything as fancy as your 3D print often just rectangle piece of plastic with chamfered edges.

On antoher note I mentioned something to neil earlier about home and limit switches which may interest you because I don't think you've done those yet have you.?

The Csmio control doesn't let you share home and limit switches in the same way mach3 normally does it.?
Normally with Mach you can wire switches in series and connect to one input then when homing mach sets all axis using that one input.
Csmio will let you share home and limit for one axis but each axis must be on own input. You can't wire all switches in series for every axis.

[fvfdrums]

I wasn't planning on sharing home and limit switches. If by this you mean using the same switch for both home and limits?

I have proximity sensors on the home and industrial microswitches on the limits.

Does this still apply?

[JAZZCNC]

I wasn't planning on sharing home and limit switches. If by this you mean using the same switch for both home and limits?

I have proximity sensors on the home and industrial microswitches on the limits.

Does this still apply?

Well yes and No.!! . . . No you can't wire all home switches in series they need to be on separate inputs, one for each axis.
But Yes you can wire all Limits in series and just use one input for all axis.

So in your case you'll need 4 inputs. 3 for homes and 1 for limits if wired in series.

[fvfdrums]

Thanks Dean!

Well been a bit quiet for a week or so due to work however have had a productive saturday. All home switches wired in and working well and all limit switches wired up and stops in place and they all activate well. Looking at the limit override switch is this something I should configure in mach? seeing as there is an option and I have plenty of inputs of the CSMIO left or can it simply be a switch in the circuit to override the limits?

managed to do a bit of testing and some very very basic cuts just coding on screen in mach to test things out. All axis are pretty smooth though there is still a sort of rattly vibration which I think is coming from one of the two x axis balloonist. Is it worth taking it apart cleaning and reassembling? I cannot be 100% sure thats where it is coming from but its roughly when the machine hits a certain speed in the middle of the x travel.

Still need to do some work on tuning the motors have them set at 160 steps per mm at the moment which I think is right but need to do a bit of work on the acc and vel I am not wanting to run this near its limit but the access from the home point is soooooo slow. Should I expect it to be a bit quicker or is that normal? once up to speed it seems to be fine.

Next on the list is getting the bed material sorted out and ordered in now we have a definite working area and there is a hydrovane compressor on the way to power the mister system/cold air gun (not bought specifically for this as I do all sorts with air however I have been needing an excuse to buy a machine of a decent size!)

Also need to sit down and do some fine adjustments I think the x axis is ever so slightly on the wonk! not by much but squares corner to corner are about 0.1mm out of square at the moment which I dont consider too bad seeing as we have not really spent a lot of time fine tuning yet and also it has plenty of adjustment in the machine. so I am reasonably confident we can get it producing a square enough square for what we need!
15166

[fvfdrums]

Just noticed when trying to get over 12k rpm in Mach I get an error saying something about pulley size incorrect? Any idea what this is all about?

[JAZZCNC]

Looking at the limit override switch is this something I should configure in mach? seeing as there is an option and I have plenty of inputs of the CSMIO left or can it simply be a switch in the circuit to override the limits?

Personally i just use the Limit overide inside Mach. But if you want an override then just wire a Momentry button in the circuit (don't use a Switch). Only time i use an Override button is if the Limits are wired in the Safety circuit and knockout Power Etc Which I don't do any more for limits.


managed to do a bit of testing and some very very basic cuts just coding on screen in mach to test things out. All axis are pretty smooth though there is still a sort of rattly vibration which I think is coming from one of the two x axis balloonist. Is it worth taking it apart cleaning and reassembling? I cannot be 100% sure thats where it is coming from but its roughly when the machine hits a certain speed in the middle of the x travel.

Yes but not yet.! Run it slowly for a while and tweak alignment of screws. You'll be surprised how much just little missalignment can make to how machine runs and sounds. Then if it doesn't go away then think about stripping down after a few weeks when things have settled down.



Still need to do some work on tuning the motors have them set at 160 steps per mm at the moment which I think is right but need to do a bit of work on the acc and vel I am not wanting to run this near its limit but the access from the home point is soooooo slow. Should I expect it to be a bit quicker or is that normal? once up to speed it seems to be fine.

If 10mm pitch screws and drive Micro stepping is 1600 then yes 160 is correct. Safe velocity for testing would be 4 or 5000mm/min with Accelleration of 500-600. With the aim of reaching around 10,000mm/min and 1000-12000s.

Homing is slower and will be a percentage of the Velocity which is set in Homing & limits. Faster you home the less accurate the repeatabilty. 20% is a good figure when at 10mtr/min.

[JAZZCNC]

Just noticed when trying to get over 12k rpm in Mach I get an error saying something about pulley size incorrect? Any idea what this is all about?

Ye go to Spindle Pulleys on the Config menu and set speed to 24000 for which ever pulley your using.

[fvfdrums]

Once again thanks a lot Dean :)

Homing set at about 15% at the moment but the machines not running that fast anyway.'good to know what numbers to possibly aim for once comfortable.

I'm pretty sure the screw in question thats making the noise was the one with a slight bend in it. I would imagine this could cause that? I have straightened it but can't get it perfect. I will keep an eye on it and see how it all goes.

[fvfdrums]

Been a month or so since I could get anything done on the machine but am aiming to have another full on weekend in about a fortnight so want to get some bits ready for the machine.

I am still stuck on the best way to create the bed surface for the machine. I would like to be able to catch most of the chips that sit from falling underneath for example if I just put aluminium strips running perpendicular to the steel current bed. I suppose a thin sheet of anything attached to the bottom side of the bed would do this.

As for the main material lets assume I use aluminium flat bar, how thick and wide does this need to be given I plan on using t nuts to secure down any spoil board or at least some similar sort of system.

fingers crossed I can get it cutting some material soon rather than the basic square and circle combo I have to keep replaying just to prove to people it works!

[JAZZCNC]

As for the main material lets assume I use aluminium flat bar, how thick and wide does this need to be given I plan on using t nuts to secure down any spoil board or at least some similar sort of system.

On my Old machine I used similair setup with approx 35 x 16mm strips but I'd suggest thicker material 20mm min depending on support spacing.
Now this worked OK but in all honesty I found it to be mostly waste of money as 2/3rds was covered with some kind of spoil board which meant couldn't access T-sots so material was screwed down etc.
The area that wasn't covered was only used when I needed better accurecy on say things like engraving etc and I didn't want to surface the whole bed.

The type of work your mostly doing will determine which suits best but I found that while mostly cutting aluminium and mostly 2D profiling that a Good stable spoil board with a Matrix of holes using threaded inserts worked best. It gave good clamping options and most flexabilty for work fastening.
Very rarely did I need to remove the spoil board for acceess to more accurate aluminium base.

If you need accurecy but not over a large area then I'd suggest buying piece of Machine tooling plate and drilling matrix of holes. Then cover rest of bed with stable spoil board material and use T-slot track or Matrix of holes again.!

Just T-slots I find restrictive when using as general all-round machine cutting mixture of materials and Job types.!!

[fvfdrums]

Thanks as ever Jazz :)

What would you consider a reliable spoil board material?

[JAZZCNC]

Thanks as ever Jazz :)

What would you consider a reliable spoil board material?

NOT MDF. . Lol

Phenolic Resin fibre board type material works very good but can be costly. Can get Plywood with Phenolic face which is resonably stable.
Good quality Ply wood with a liminate face can work well.
HDPE but again expensive.

[Boyan Silyavski]

Here #165 (http://www.mycncuk.com/threads/6619-Quite-an-Unusual-one?p=68612#post68612) i make some comparison of different materials for bed. Have to agree with Dean that the best material i found is phenolic sheet/ for sacrificial bed/.

Furthermore if your machine is well done, you say- cut through deep to 0mm and it cuts through but without even scratching the phenol. Not so cutting aluminum though, where you have to find the sweet spot.

[fvfdrums]

Right. So we decided on the aluminium sheet with a matrix of holes in the end. After looking at parts we are likely to produce chances are we too will be using spoil board 75% of the time. So that's turning up next week. Went for 20mm thick sheet.

So all I really need to sort now before the machine can start properly cutting is the water cooling. Any suggestions on a water pump? I've found a bunch on eBay but some seem too low powered 5w solar type and others seem too high pressure ? 80psi or there abouts. Any pointers to specific pumps used recently the older posts links are now dead.

Finally. Cutters.

What's a good cutter to start with using in aluminium on a 3kw spindle? Single flute? Just want something to experiment with really

And as far as machining hardwoods I have a bunch of high quality router cutters would these be best for this application?

[fvfdrums]

Just had some problems with the x axis locking up when going at a reasonable speed. Convinced we still have a problem with the ball nuts I have taken them apart only to find 53 balls which is 18/18/17 in terms of 3 tracks.

Am I right in thinking 17 balls per track and that having too many could cause this binding at a higher speed?