There are six pages of stepper motors, several NEMA 17 and several NEMA 23, both unipolar and bipolar.
I personally prefer 8-wire unipolar motors wired as bipolar, but that is just my preference. I prefer 8-wire unipolar motors so that either a unipolar or bipolar drive can be used to drive the motor.
EDITED: But keep in mind, that if you go with 8-wire unipolar, instead of 4-wire bipolar, depending upon your wiring methods, you may spend considerably more on jumper wires and connectors.
There's a lot going on in this thread! I'm very new to this CNC stuff, but I've just taken the plunge and bought a bunch of stuff:
From Amazon.com for $487.70 total:
Zen 7"x7" CNC kit
Zen spindle
Zen Proximity switches (3X)
Zen E-stop button
Directly from Zen (to save $10 in shipping for 3D kit) for $147.95 total:
3D conversion kit
limit switches (6x)
1/8" endmill
1 mm endmill
ballnose mill
From Ebay:
4-axis TB6560 stepper driver (from goodlucksell) for $69.87 total
12V, 10A power supply (from dashop) for $17.99 total
From Makerbot for $283.12 total:
MK6 Stepstruder kit
1.75mm natural ABS (1 lb)
1.75mm black ABS (1 lb)
3mm PLA (1 lb)
I didn't realize till today that the Stepstruder comes with it's own stepper controller, so I could have just gotten the Zen 3-axis board.
But, I think it will be simpler with the 4-axis board anyway.
So, for $1006.63 I think I've got everything to do either CNC or 3D print.
It'll take a few weeks for everything to arrive, then we'll see what I've gotten myself into
Here are some unanswered questions that my have changed my purchase some:
Can I go back and forth from the Zen regular 7"x7" and the 3D print version?
If not, maybe I should have gotten the lower cost 3D print build directly from Zen...
Can the stepstruder go back and forth between 3mm and 1.75mm plastic?
One other big question is whether or not these stepper controller boards can be driver directly from the 3.3V Prop. Seems they're intended for a PC parallel port, which I think is 5V...
Oops, forgot one thing: Power for the Spindle!
From Ebay (zahangzixing2008) for $18.98 total, a DC-DC converter that can take the 12V supply input and spit out a variable voltage (up to 50v, 3A)
ZEN have 3x 1.3A = 3.6A already. If You add one more Stepper You need add this Current to. To that You need add at last 20% more current to Power supply TO be sure that it is enough.
IF Heating plastic will be from same supply YOU need add that Current to.
Oops, forgot one thing: Power for the Spindle!
From Ebay (zahangzixing2008) for $18.98 total, a DC-DC converter that can take the 12V supply input and spit out a variable voltage (up to 50v, 3A)
Saphieha, the DC-DC converter was just for the spindle, which is rumored to draw 1A at ~28 V. That should correspond to maybe 3A from the 12VDC supply.
I think the Steppers are 1.7A, and I'll only need 3 for CNC mode. So I think I'll be OK with the 10A, 12VDC main supply.
PS: Wow, all my Zen stuff (both direct and from Amazon) shipped already! Should get it all next Thursday...
What do you guys think about the "Automated Build Platform with Relay Board Kit" for 3D printing?
I was thinking it's a waste of money, $160 just to print multiple parts.
But now, I'm not so sure... I'm not even sure what kind of surface I need to print on...
What do you guys think about the "Automated Build Platform with Relay Board Kit" for 3D printing?
I was thinking it's a waste of money, $160 just to print multiple parts.
But now, I'm not so sure... I'm not even sure what kind of surface I need to print on...
Definitely cool. Question is, do you need to produce a continuous supply of parts? I don't have any large orders yet, and I would probably have my face pressed up to the glass while it was printing, at least for the first few dozen parts; so I would probably be on hand to tend the machine. But something like this would be handy if I ever need a comtinuous process.
The Automated Build Platform is a god send!!! We're using it to print the internal parts for our Wingmans. We just start printing at the end of the day and come back in the morning, voila 50 parts made!
I'm getting ready to start printing the "hidden" parts of our Wingman cases with them as well. That should reduce our case cost from $30 each to around $7 each! With that kind of savings, the Cupcake CNC will pay for itself in just a few months! (Did I mention the labor savings? It'll shave at least an hour of the assembly time!)
I may get it still, I'm on the fence... I'm kinda with prof_braino that I'd probably be looking at it the whole time...
Plus, I don't actually have a real application yet, I'm just doing it for fun...
Also, the price seems a little steep for what it does...
Thanks for asking, I'm still not sure what to build or buy, there are too many alternatives and I don't a basis for decision yet.
So far, I've altered my highest level requirements in this direction:
Scalable design, so the machine can be built to physical size and rigidity requirements for intended use.
The unit for my braino-lab needs to fit in the 2' x 2' space allocated on my dsktop
The bnu (braino-lab unit) much be sufficiently rigid for some milling (not determined yet, shooting for steel but expecting wax)
The bnu must accept a printing head, I'm favoring extrusion AND UV developed resin from liquid
I have a wicked power supply which should be overkill for any rig I could possibly end up with.
I have four barely suitable nema 23 stepper motors which will have to do unit I can scrounge suitable nema 34's so the mounting will have to be crafted accordingly
The geko drivers look good, they should be able to handle the steppers I'm heading towards, but the cost is still high.
I think I found extruded aluminum T-slot stock and end cap screws from the surplus store, but did not purchase yet.
The E-channel(?) you suggested for cable management looks like a good idea, but I'm not far enough along to spec what I need yet.
I'm favoring your "game changer" design, but obviously I'm having problems figuring out deflection on harder materials, so I'm still messing with mechanicals.
It seems the biggest concern is keeping the mill head steady under load. I like the way the Sherline machines keep the head steady and only move in the Z-axis, and the table provides the X, Y and rotary motion. But those machines only have a 6" x 7" work area. I'm looking into design like the "chess board robot" where the head only accesses a quadrant at a time, and the whole bed rotates to get to the other areas. Makes me dizzy when I imagine the bed rotating.
So basically I'm having lots of fun, but so far I've only shifted junk from one pile to another, and I haven't bought or built yet.
NEMA 34's are a fairly good sized motor. Are you sure you want to go that big? Just remember, the larger you go the more it costs to get it going. If you are thinking about buying T-Slot, visit that link that I gave earlier to 80/20 Inc. and download their catalog. In that catalog you will find UHMWPE bearings. You might want to consider using these.
Bruce
P.S. Don't be afraid to call the manufacturers directly for information and design assitance, they are a great source of information and help.
Here are a few more links that you may find useful in designing your machine. These links are the cream of the crop Ahhh what the heck, I will put them all in one place for easy reference.
Theres a misleading photo on Amazon about the machine and a CNC 3D printer adapter for the little machine.
'
The photo looks like your getting the full machine.But not really.
'
Zen Tools is pretty cool about catching this mistake and will send you an E-Mail if your not a current customer and have not purchased a CNC machine from them in the past.
Got the first stage of the Zen CNC/3D Printer done today.
I think the size is actually perfect... Looks like it's going to be about a 12" cube with motors sticking out.
The quality decent. I was a little annoyed that all the screws are metric and I had to dig up metric allen wrenches.
Stefan, I think maybe you meant to ask me?
I'll answer anyway... I've just finished the second axis...
Here's a close up for you:
My current opinion (although I'm a newbie to this) is that the individual axis are very sturdy and well aligned. My major concern is how well the Y axis is aligned to the X axis. I think that will take a lot of tweaking to get to the sub-millimeter accuracy range over the entire work area...
I haven't got my controller from China yet to drive the steppers, but I think the anti-backlash nuts prevent the play you're asking about.
It appears to me that the anti-backlash nuts just add resistance to the movement.
thank you very much for the pictures. From the detail-picture now I'm pretty sure it is a standard metric thread with diameter 8mm or 10mm. To be really sure I would need a macro-picture where the thread
fills the complete picture. Trapeziodal threads look very similar to metric threads from looking at them
from a distance of 50cm. But I won't tease you with such special wishes.
The anti backlash-nut with he spring is a pretty good idea to avoid play in the shaft-drive.
If I understand the function of the nut right the spring forces the nut to stay in touch always with the same side of the thread's flanks
for both directions. Without the spring the nut would change to the other flank of the thread if you reverse the rotation.
For aligning the X- to the Y-axis I have the idea to use a dial gauge like this
Mount the dial gauge on the Y-part then let the dial gauges header touch the x-part.
The aboslute measuring doesn't matter. The only interesting thing is how big is the difference.
Move the y-part from max left to max right position. Then you can see
how much it is mis-aligned and correct it until the pointer of the dial gauge stays within the same
0,01 mm tick of the dial gauge.
You have to buy this dial gauge. But it is not too expensive. And you can use it for aligning parts
that should be mounted on the x-table for milling too.
There are different kind of these dial gauges. The one linked above can measure some millimeters
this one http://www.hogetex.com/zwenktaster-0-0-8mm-0-01mm.html
can measure only in a range of 1 mm with 0,01 mm resolution but you can adjust the measuring-pin to
different angles to adjust to different measuring situations.
If you know a company near to you that has milling cnc-machines make a visit and ask if you can look
at there dial gauges to find out which type would be .best for your needs.
By the way: does the zen cnc-kit have a possability to adjust the aligning with threads?
I mean EXTRA threads not for mounting the parts together but for well defined adjusting the parts to each other by turning the threads?
Tightening the mounting threads "hand-warm" and then turn the adjusting threads to tweak the alignement. If the tweaking is finsihed
tightening the mounting threads to "stay hard in this position".
That is a nice looking machine. They seemed to have a hard time getting it off the platform...
Well, I've got all the parts for CNC now as my Chineese controller just showed up (was a little annoyed that they shipped signature required and I had to go to the Post Office to get it. Also, they put "Gift" on the customs, like they often do, which I think is immoral at best).
Anyway, here's a closer up of the drive for Stefan and some pictures of the controller:
Also, here's a shot of the "flexible drive coupling" which is actually a spring:
thank you very much for the closeup on the thread. Looks Indeed like a standard metric thread Size M8.
(the "hills" are symmetric triangles)
The "move-forward-distance per revolution is 1.25 mm. With a common stepper-motor 1.8° per step = 200 steps per revolution.
In half-step-mode (400 steps per revolution) this would result in a resolution of 1.25 mm / 400 = 3.125 µm which is pretty fine in my opinion.
What do you think about my suggestion for aligning X- to Y-axle using a gauge diameter?
Stefan, yes I found one at work I can borrow to do that test. I read on the Zen forum that somebody did this and had something like a 1mm deflection over the table range. My guess is that the alignment between stages is the weak point of this design... It's going to take a lot of careful measurements and tweaking to get it right...
I did such alignments on a lathe where I had to re-fix workpieces after releasing them. After two minutes I was at the level of 0,2 mm precision. After 4 minutes at 0,05mm which was enough for these workpieces.
I guess if I would have taken 15 minutes I would have come down on 0,01 mm precision just by "tapping" the workpiece while it is fixed "handwarm" so that it could still move a little bit after each tap.
For measuring it I didn't had to be careful. The dial gauge has such a great "translation" that it is very easy to see how many 0,01mm-ticks it is moving.
You just have to fix the dial gauge itself.
For the best alignement the construction should have something that you can move the relative position between X- and Y-axle part in a well defined way and in very small steps. If I understand the constrcution right the basic frame has "long"-holes. This means you can move the "upstanding" part
forward backward. If you do this only on one side y-axles turns around a little bit and will be in an angle of 89 degrees or 91 degrees to the x-axle.
Now if you fix one side that it really can't move forward/backward anymore but only turning and add
something that on the other side you can move it forward/backward in a well defined way (extra-thread)
I guess the thing can be aligned quite quickly.
Therefore you move the dial gauge from min to max position to find out in which direction you have to correct and how much.
Then you drive the gauge-pin back to the farest position with the gauge-pin still touching and measuring and start turning the adjusting thread
until the gauge measures the same value as in the opposite position.
Maybe by just writing this - it is hard to understand. Give me feedback on this.
Then you measure the difference again with driving the gauge from min to max to min. You do this for 2 or 3 iterations again and then it should
be aligned very precisely. Of course the ZEN-CNC-machine does not have adjusting threads. So you have to untighten the main-threads half, that keep all parts together
then adjust them a little tighten them and measure again. I guess this is quite uncomfortable compared to adjusting threads so I would start customising it by adding this.
The more I think about it the more I like the basic design of the ZEN-CNC and would like to improve it with these adjusting threads and make a copy out of "airplane-quality-aluminium"
(mechanical strength 300-500 N/mm²) or maybe steel. The mechanical strength of PVC is in the range of 30-60 N/mm²) E-modul of PVC less than 3500 N/mm² compared to 70.000 N/mm² of aluminium.
I don't understand why zen-cnc builds it from PVC. But I guess only a perfomance and precision-test
of the whole machine can show how good it is.
Another idea I have is to mount another plate on the main-plate and milling down 1 mm from the complete surface to "align" it just through this milling.
This would align to be exactly horizontal. (z-alignment)
Maybe somebody with more practical experience in CNC-milling can chime in and write his opinion about this idea.
I would also add parts like the red lines in the attached picture to make the whole thing more stiff.
The cost of aluminium and the machining of it would increase the cost of the machine significantly, that's why PVC, it's not perfect but the machine is not designed to do serious hard machining.
For larger beds it is common to machine them flat with the machine itself, for this machine I would not think it would be worth it, also if you can't mount a largish cutter you limit the table size to the area the machine can move over.
Comments
http://www.alltronics.com/cgi-bin/category/55
There are six pages of stepper motors, several NEMA 17 and several NEMA 23, both unipolar and bipolar.
I personally prefer 8-wire unipolar motors wired as bipolar, but that is just my preference. I prefer 8-wire unipolar motors so that either a unipolar or bipolar drive can be used to drive the motor.
EDITED: But keep in mind, that if you go with 8-wire unipolar, instead of 4-wire bipolar, depending upon your wiring methods, you may spend considerably more on jumper wires and connectors.
From Amazon.com for $487.70 total:
Zen 7"x7" CNC kit
Zen spindle
Zen Proximity switches (3X)
Zen E-stop button
Directly from Zen (to save $10 in shipping for 3D kit) for $147.95 total:
3D conversion kit
limit switches (6x)
1/8" endmill
1 mm endmill
ballnose mill
From Ebay:
4-axis TB6560 stepper driver (from goodlucksell) for $69.87 total
12V, 10A power supply (from dashop) for $17.99 total
From Makerbot for $283.12 total:
MK6 Stepstruder kit
1.75mm natural ABS (1 lb)
1.75mm black ABS (1 lb)
3mm PLA (1 lb)
I didn't realize till today that the Stepstruder comes with it's own stepper controller, so I could have just gotten the Zen 3-axis board.
But, I think it will be simpler with the 4-axis board anyway.
So, for $1006.63 I think I've got everything to do either CNC or 3D print.
It'll take a few weeks for everything to arrive, then we'll see what I've gotten myself into
Here are some unanswered questions that my have changed my purchase some:
Can I go back and forth from the Zen regular 7"x7" and the 3D print version?
If not, maybe I should have gotten the lower cost 3D print build directly from Zen...
Can the stepstruder go back and forth between 3mm and 1.75mm plastic?
One other big question is whether or not these stepper controller boards can be driver directly from the 3.3V Prop. Seems they're intended for a PC parallel port, which I think is 5V...
Oops, forgot one thing: Power for the Spindle!
From Ebay (zahangzixing2008) for $18.98 total, a DC-DC converter that can take the 12V supply input and spit out a variable voltage (up to 50v, 3A)
I am away until Sunday so enjoy the peace and quiet
3A are quite to low.
ZEN have 3x 1.3A = 3.6A already. If You add one more Stepper You need add this Current to. To that You need add at last 20% more current to Power supply TO be sure that it is enough.
IF Heating plastic will be from same supply YOU need add that Current to.
I think the Steppers are 1.7A, and I'll only need 3 for CNC mode. So I think I'll be OK with the 10A, 12VDC main supply.
PS: Wow, all my Zen stuff (both direct and from Amazon) shipped already! Should get it all next Thursday...
I was thinking it's a waste of money, $160 just to print multiple parts.
But now, I'm not so sure... I'm not even sure what kind of surface I need to print on...
Definitely cool. Question is, do you need to produce a continuous supply of parts? I don't have any large orders yet, and I would probably have my face pressed up to the glass while it was printing, at least for the first few dozen parts; so I would probably be on hand to tend the machine. But something like this would be handy if I ever need a comtinuous process.
So, what I may do is build the Zen 3D printer, print a reprap, build a reprap, move extruder head to reprap, then convert Zen back to CNC mill...
Sounds a little like robots cloning themselves.
Sure looks like you are in for some fun!
Cheers,
Peter (pjv)
How is your project going?
Bruce
The Automated Build Platform is a god send!!! We're using it to print the internal parts for our Wingmans. We just start printing at the end of the day and come back in the morning, voila 50 parts made!
I'm getting ready to start printing the "hidden" parts of our Wingman cases with them as well. That should reduce our case cost from $30 each to around $7 each! With that kind of savings, the Cupcake CNC will pay for itself in just a few months! (Did I mention the labor savings? It'll shave at least an hour of the assembly time!)
Bill
Plus, I don't actually have a real application yet, I'm just doing it for fun...
Also, the price seems a little steep for what it does...
Thanks for asking, I'm still not sure what to build or buy, there are too many alternatives and I don't a basis for decision yet.
So far, I've altered my highest level requirements in this direction:
Scalable design, so the machine can be built to physical size and rigidity requirements for intended use.
The unit for my braino-lab needs to fit in the 2' x 2' space allocated on my dsktop
The bnu (braino-lab unit) much be sufficiently rigid for some milling (not determined yet, shooting for steel but expecting wax)
The bnu must accept a printing head, I'm favoring extrusion AND UV developed resin from liquid
I have a wicked power supply which should be overkill for any rig I could possibly end up with.
I have four barely suitable nema 23 stepper motors which will have to do unit I can scrounge suitable nema 34's so the mounting will have to be crafted accordingly
The geko drivers look good, they should be able to handle the steppers I'm heading towards, but the cost is still high.
I think I found extruded aluminum T-slot stock and end cap screws from the surplus store, but did not purchase yet.
The E-channel(?) you suggested for cable management looks like a good idea, but I'm not far enough along to spec what I need yet.
I'm favoring your "game changer" design, but obviously I'm having problems figuring out deflection on harder materials, so I'm still messing with mechanicals.
It seems the biggest concern is keeping the mill head steady under load. I like the way the Sherline machines keep the head steady and only move in the Z-axis, and the table provides the X, Y and rotary motion. But those machines only have a 6" x 7" work area. I'm looking into design like the "chess board robot" where the head only accesses a quadrant at a time, and the whole bed rotates to get to the other areas. Makes me dizzy when I imagine the bed rotating.
So basically I'm having lots of fun, but so far I've only shifted junk from one pile to another, and I haven't bought or built yet.
NEMA 34's are a fairly good sized motor. Are you sure you want to go that big? Just remember, the larger you go the more it costs to get it going. If you are thinking about buying T-Slot, visit that link that I gave earlier to 80/20 Inc. and download their catalog. In that catalog you will find UHMWPE bearings. You might want to consider using these.
Bruce
P.S. Don't be afraid to call the manufacturers directly for information and design assitance, they are a great source of information and help.
Here are a few more links that you may find useful in designing your machine. These links are the cream of the crop Ahhh what the heck, I will put them all in one place for easy reference.
http://www.mcmaster.com/ McMaster-Carr
http://us.misumi-ec.com/ Misumi
http://sdp-si.com/ Stock Drive Products
http://www.onlinemetals.com/ Online Metals
http://www.qtcgears.com/ Quality Transmission Components
http://www.york-ind.com/ York Industries
http://www.applied-motion.com/ Applied Motion
http://www.igus.com/ Igus Inc.
http://www.hightechplastics.com/ High-Tech Plastics
http://www.smallparts.com/ Small Parts
https://nordex.com/ Nordex, Inc.
http://www.usdigital.com/ US Digital
http://www.nookindustries.com/ Nook Industries
http://www.bimba.com/ Bimba Manufacturing
http://www.smcusa.com/ SMC Corp.
http://www.8020.net/ 80/20 Inc.
http://www.antekinc.com/index.php Antek Inc.
http://geckodrive.net/ Gecko Drive
I suddenly have the feeling it's going to be a smaller machine than my mental picture...
best regards
Stefan
'
The photo looks like your getting the full machine.But not really.
'
Zen Tools is pretty cool about catching this mistake and will send you an E-Mail if your not a current customer and have not purchased a CNC machine from them in the past.
I think the size is actually perfect... Looks like it's going to be about a 12" cube with motors sticking out.
The quality decent. I was a little annoyed that all the screws are metric and I had to dig up metric allen wrenches.
Damn those round numbers!!
the mechanic seems to be quite OK. Round steel with linear bearings.
Can you take and post a detail picture of the drive shaft?
What type is it? Standard metric thread? trapezoidal? or something else?
Did you already made a first test about the play inside the drive shaft when reveresing the direction?
best regards
Stefan
I'll answer anyway... I've just finished the second axis...
Here's a close up for you:
My current opinion (although I'm a newbie to this) is that the individual axis are very sturdy and well aligned. My major concern is how well the Y axis is aligned to the X axis. I think that will take a lot of tweaking to get to the sub-millimeter accuracy range over the entire work area...
I haven't got my controller from China yet to drive the steppers, but I think the anti-backlash nuts prevent the play you're asking about.
It appears to me that the anti-backlash nuts just add resistance to the movement.
thank you very much for the pictures. From the detail-picture now I'm pretty sure it is a standard metric thread with diameter 8mm or 10mm. To be really sure I would need a macro-picture where the thread
fills the complete picture. Trapeziodal threads look very similar to metric threads from looking at them
from a distance of 50cm. But I won't tease you with such special wishes.
The anti backlash-nut with he spring is a pretty good idea to avoid play in the shaft-drive.
If I understand the function of the nut right the spring forces the nut to stay in touch always with the same side of the thread's flanks
for both directions. Without the spring the nut would change to the other flank of the thread if you reverse the rotation.
For aligning the X- to the Y-axis I have the idea to use a dial gauge like this
http://www.amazon.de/Optimum-Me%C3%9Fuhr/dp/B00365RIDS/ref=sr_1_2?ie=UTF8&qid=1299928332&sr=8-2
Mount the dial gauge on the Y-part then let the dial gauges header touch the x-part.
The aboslute measuring doesn't matter. The only interesting thing is how big is the difference.
Move the y-part from max left to max right position. Then you can see
how much it is mis-aligned and correct it until the pointer of the dial gauge stays within the same
0,01 mm tick of the dial gauge.
You have to buy this dial gauge. But it is not too expensive. And you can use it for aligning parts
that should be mounted on the x-table for milling too.
There are different kind of these dial gauges. The one linked above can measure some millimeters
this one
http://www.hogetex.com/zwenktaster-0-0-8mm-0-01mm.html
can measure only in a range of 1 mm with 0,01 mm resolution but you can adjust the measuring-pin to
different angles to adjust to different measuring situations.
If you know a company near to you that has milling cnc-machines make a visit and ask if you can look
at there dial gauges to find out which type would be .best for your needs.
By the way: does the zen cnc-kit have a possability to adjust the aligning with threads?
I mean EXTRA threads not for mounting the parts together but for well defined adjusting the parts to each other by turning the threads?
Tightening the mounting threads "hand-warm" and then turn the adjusting threads to tweak the alignement. If the tweaking is finsihed
tightening the mounting threads to "stay hard in this position".
best regards
Stefan
http://www.youtube.com/watch?v=GOEbZ98J1Ks
Well, I've got all the parts for CNC now as my Chineese controller just showed up (was a little annoyed that they shipped signature required and I had to go to the Post Office to get it. Also, they put "Gift" on the customs, like they often do, which I think is immoral at best).
Anyway, here's a closer up of the drive for Stefan and some pictures of the controller:
Also, here's a shot of the "flexible drive coupling" which is actually a spring:
thank you very much for the closeup on the thread. Looks Indeed like a standard metric thread Size M8.
(the "hills" are symmetric triangles)
The "move-forward-distance per revolution is 1.25 mm. With a common stepper-motor 1.8° per step = 200 steps per revolution.
In half-step-mode (400 steps per revolution) this would result in a resolution of 1.25 mm / 400 = 3.125 µm which is pretty fine in my opinion.
What do you think about my suggestion for aligning X- to Y-axle using a gauge diameter?
best regards
Stefan
Putting gift on the shipping label saves them nothing and you the duty so if you feel bad you can always put a cheque in the post to customs?
Looks like it will be a fun little machine.
Graham
I did such alignments on a lathe where I had to re-fix workpieces after releasing them. After two minutes I was at the level of 0,2 mm precision. After 4 minutes at 0,05mm which was enough for these workpieces.
I guess if I would have taken 15 minutes I would have come down on 0,01 mm precision just by "tapping" the workpiece while it is fixed "handwarm" so that it could still move a little bit after each tap.
For measuring it I didn't had to be careful. The dial gauge has such a great "translation" that it is very easy to see how many 0,01mm-ticks it is moving.
You just have to fix the dial gauge itself.
For the best alignement the construction should have something that you can move the relative position between X- and Y-axle part in a well defined way and in very small steps. If I understand the constrcution right the basic frame has "long"-holes. This means you can move the "upstanding" part
forward backward. If you do this only on one side y-axles turns around a little bit and will be in an angle of 89 degrees or 91 degrees to the x-axle.
Now if you fix one side that it really can't move forward/backward anymore but only turning and add
something that on the other side you can move it forward/backward in a well defined way (extra-thread)
I guess the thing can be aligned quite quickly.
Therefore you move the dial gauge from min to max position to find out in which direction you have to correct and how much.
Then you drive the gauge-pin back to the farest position with the gauge-pin still touching and measuring and start turning the adjusting thread
until the gauge measures the same value as in the opposite position.
Maybe by just writing this - it is hard to understand. Give me feedback on this.
Then you measure the difference again with driving the gauge from min to max to min. You do this for 2 or 3 iterations again and then it should
be aligned very precisely. Of course the ZEN-CNC-machine does not have adjusting threads. So you have to untighten the main-threads half, that keep all parts together
then adjust them a little tighten them and measure again. I guess this is quite uncomfortable compared to adjusting threads so I would start customising it by adding this.
The more I think about it the more I like the basic design of the ZEN-CNC and would like to improve it with these adjusting threads and make a copy out of "airplane-quality-aluminium"
(mechanical strength 300-500 N/mm²) or maybe steel. The mechanical strength of PVC is in the range of 30-60 N/mm²) E-modul of PVC less than 3500 N/mm² compared to 70.000 N/mm² of aluminium.
I don't understand why zen-cnc builds it from PVC. But I guess only a perfomance and precision-test
of the whole machine can show how good it is.
Another idea I have is to mount another plate on the main-plate and milling down 1 mm from the complete surface to "align" it just through this milling.
This would align to be exactly horizontal. (z-alignment)
Maybe somebody with more practical experience in CNC-milling can chime in and write his opinion about this idea.
I would also add parts like the red lines in the attached picture to make the whole thing more stiff.
best regards
Stefan
For larger beds it is common to machine them flat with the machine itself, for this machine I would not think it would be worth it, also if you can't mount a largish cutter you limit the table size to the area the machine can move over.
Graham