Some advices for my CNC router build = Appreciating!

[Letomoto]

Hello group , this Zone is a massive ocean of informations !!! Incredible knowledge ressource !!!

But ! , i find it hard to collect all the data needed to be usefull to my own project type , there is about 3000 different builds so everything is in many directions ! I am a hobbyist and i am not running an industrial company and not having very huge budget for this build .


Here is my build i am planning . Cartesian CNC router table to work on aluminum , maybe some steel and wood for sure . I am writing a list of the categories i am searching and have some opinions to confirm if i am searching for good targets ., would help !



--HGR20mm linear rail guides and ballscrews SFU-1605 to build a 1000mm X 1600mm *************** are on their way right now**************



--Shopping for a 3-4kW ATC spindle and VFD , i reallllly want automatic tool changer , now what i find wont ever cost under 2000$USD , this hurts , but wondering how many watts should i look for milling aluminum ? 2.2 - 2.5 - 3 - 4 - 5 - 6kW ???



--The step drivers , direct driving the SFU-1605 ballscrews , now thinking about NEMA 34 6N.m 850ozf.in Closed loop , would it be enough power or too much ? is the accuracy good enough direct driving a sfu1605 ? would it worth using geared 3:1 to add the steps/mm to add precision of the movements and get more smoothness of the curves BUTwill slow down the whole traveling , acceleration , more high pitch motor noisy and rise the price . If geared would be usefull could we take a big power NEMA23 geared that would output probably as much torque of NEMA34 but more steps per mm ? i am having trouble to see the best choices , i also see steppers in different installations ways , Direct , Pulleys timing belts Ratios , what would be the good choices ? Right now planning the direct driven ballscrews and NEMA34 , opinions ?




- The motion controller ! Confused about a few different types ! I am now using my Craftex Mill/Drill i converted to CNC with Mach3 and Smoothstepper to control the drivers. I see there is an interesting controller but i have lots of questions ....

[COLOR=var(--ytd-video-primary-info-renderer-title-color, var(--yt-spec-text-primary))]CNC Controller DDCSV1.1[/COLOR]
https://www.youtube.com/watch?v=0ivxcOnEBBM&index=19&t=0s&list=PLSEbLVUjHRwBqkkV0paAGyAwqDple5Yag

Controller Questions:

1- Is it doing the whole 100% job from the USB drive toolpath G-Code but maybe can't connect to a computer to get the g-code files directly from the computer and you have to take the USB drive from the machine , shove it in the computer , save the toolpath , remove from the computer , shove it in the controller , open , select file.... sounds like its been developped 20 years ago ?

2- Is there a way to connect this to a computer and combine the physical DDCS controller and controlling with a computer software like running quick wizards like in Mach3 to do anything like holes patterns , bore sized hole , manual G-Code , quick automatic zeroing tool changes , monitoring everything and playing with tons of settings options or it is 100% enclosed and can't fiddle with anything?

3- I can see there is a bunch of other controllers ... Centroid Acorn CNC , MASSO , Smoothstepper Mach3 , Pokeys , UCCNC , etc , even Arduino .... i am now using Smoothstepper ESS (ethernet) on my small mill/drill CNCed . The Smoothstepper cost me 200$ and is connected to the cheap breakout board that came with the steppers/drivers kit from ebay . I can see the prices range go from 30$ to 2000$ , it is hard to know what would be easy to use , to setup , behavior , potentially do a digitizing probing , get enough switches and add-ons connected and controlled by it AND satisfaction.... And for software i only tried MACH3 in my life so i dont know how the other ones are better of worst and if MACH3 can control any controllers ..


I am having LOTS of questions at the start of this project , opinions would me very appreciated !!

thanks for reading



Similar Threads:

• Build Thread- Router home build Australia. Solid steel build
• Need Help!- I have a 4x8 CNC router and I'd like to build a portable benchtop router (24" max)
• Newbie - To build or not to build Router/Plasma Table
• appreciating good TIR

[routalot]

Forget about a router for steel.By the time you have reduced the spindle speed to something suitable there won't be any torque.Its a simple calculation torqueXspeed =power.On the other hand you seem to be specifying colossal amounts of torque for you axis drives-will you be moving the table or gantry using ballscrews,rack and pinion or other means?Because in all these systems there are factors that might multiply the torque inputs.There is even a phone app to calculate the force generated.

[Letomoto]

well , i am not totally knowing yet how much weight is gonna be in the gantry but this build here is very exciting me and this is the type of category i am preparing the plans . Zeeflyboy's build looks like an amazing build !!! such Por-no!! It is looking pretty overkill build and he is overskilled ! i am reading lots of stuff but in hiw thread but it is impressive and he seems like he have a great budget


https://www.cnczone.com/forums/diy-c...28752-cnc.html

[wizard]

This machining of steel on a router is wishful thinking in my opinion. You will never come close to even an import mill drill much less a Bridgeport. Yes it is possible to do some machining and engraving of steel but you will be so limited in capability that it isn't even worth considering in my mind. Even high quality engraving is hard to do well on most DIY router builds.
i
As for the rest of your post, in my opinion you are going at this a bit backwards. It is far better to define the machines size and basic mechanical structure first and then work on the controls. For one we don't even know if you are doing moving table or moving gantry here. If the machine isn't too large there are some advantages to a moving table design for a machine focused on aluminum and other more difficult material. Once you have a good idea of what you are moving you can concentrate on fight sizing the servos or steppers.

As for the motion controllers well that is an area of rapid flux and as such I can't really tell you what to go with. Probably the most professional solution is Centroids followed closely by LinuxCNC. The rest are under development. In the end what you want to achieve are significant factors in your choice.

By the way running larger router spindles requires a significant AC supply, you can't plug those larger units into a 20 amp wall outlet.

[Letomoto]

Well , steel is not priority for me , i still have the Mill/Drill that is probably more rigid than a router and can keep the lower rpm range for it.

The size of the machine is gonna follow the lenghts of the rails/ballscrews so 1000x1600mm but i already know the working surface is a bit smaller .

I am starting this build to be able to do production of aluminum parts i am designing and it is the main goal. My mill/drill is limited to a 6"x18" table surface and there is a whole lot of things i cant produce.

Yes my questions can be deep but this would help me orient my building well before hitting troubles of planning.


Soooo , if it is direct driven 1605 ballscrews would it bring more precision with geared stepper or it wouldnt help anyway? Probably pros and cons

[ger21]

--The step drivers , direct driving the SFU-1605 ballscrews , now thinking about NEMA 34 6N.m 850ozf.in Closed loop , would it be enough power or too much ?

You've already made a big mistake by using 1605 screws. They require high rpm to get high speeds, which is difficult with steppers. And the larger the stepper, the slower they'll be. You'll probably get the best performance from 1605 screws with small Nema 34 motors, ~450oz-in, with a rating of ±6 amps.

Soooo , if it is direct driven 1605 ballscrews would it bring more precision with geared stepper or it wouldnt help anyway? Probably pros and cons

1605 will give you plenty of precision. Geared steppers are not a good choice, as they have a lot of backlash. Driving the screw with a belt is a much better option. IF anything, you want to gear it so you have less resolution, and more speed, by spinning the screw faster, and the motor slower.

Controller Questions: (DDCSV1.1)

1) The purpose of a controller like this is to eliminate the need for the PC. Many people see this as a benefit, even though it is much more limiting Imo.
2) No, it's a standalone control, no PC. If you need the features a PC provides, then there's no reason to use a control like this.

3) Choice of control software is a matter of personal preference, and feature requirements. The only way to know which you prefer is to try them. Unfortunately, you can't try most without purchasing them.
Mach3 can only use controllers specifically designed for Mach3. The ESS is one of the best for Mach3, and is by far the most popular.

[Letomoto]

oh ! this is good opinions and explanations ! Thank you it is helping !

so SFU-1605 are enough precise so no need at all for geared steps , one question cleared :-)

so for nemas 34 , every oz.in values rising makes them stronger but slower everytime because the core is heavier built to give more torque ? Thanks for the oz.in suggestion i was not really knowing what the table would need , helps me aim to the correct choice ! so around 450oz.in ! second cleared question :-)

So , allright will keep the ESS and Mach3 ,will not ask me to learn every damn new bits , no standalone controller limited . 3rd question cleared !

Thanks , i never built a full machine so i was into an ocean of choices with no experience in this build. i did convert my benchtop Mill/Drill but is not having any linear bearings (actually added HGR20 rails on my router table parts order to convert it to bearing linears! ) ,now very big flat mechanical slides and is running nema34 1250 oz-in . This is my first cnc conversion and i maybe took too much power just thinking it would never fail but didnt knew they would probably slow the motions and being louder .

[Letomoto]

i am having a small question on the Steppers choice ... i understand the torque i got suggested 450oz-in would be good enough but should i take 450oz-in NEMA 23 instead of NEMA 34 ? is there a difference in between them? if no difference , probbly a NEMA23 kit is probably gonna be cheaper and less weight and lower shipping fees , am i right?

[ger21]

I would go with the Nema 34, like these.
https://www.automationtechnologiesin...or-dual-shaft/

For good performance, you need motors with a high current rating, like these, at 6 amps. It's important to use high quality drives to get the best performance.

[Letomoto]

well , the 465oz-in and the 1200oz-in are at 6 amps , what is the 6amps rate telling ?


here in the Nema23 closed loop 570oz-in is telling 5A and it is the same price of the Nema34 465oz-in 6A
https://www.automationtechnologiesin...0-50-4b-cable/

[Letomoto]

Allright !! linear motion parts got in !!!






there is 10 rails . using 6 rails for the table and i ordered a small set to upgrade my mill/drill X-Y













i am actually playing with quotations software with different aluminum extrusion profiles and it is in USD$ , i can see the price of the whole frame is totally changing fast and i am not totally sure i need what kind of profile , stiffness i need , light or heavy-duty etc...
i am now thinking of building a steel table top who would be stiff and straight who would cost a lot less money but i am a bit scared of building a warped plane table because welding pulls a bit on the angles . Would add extrusions on the steel table but would need smaller and light profiles .

And yes probably someone is gonna find the Y axis longer than the X axis , making a long carriage but will be more accessible working surface for me

About 1500$ 80x160 and 80x80 light profile




About 1000$ 80x160 C-Beam , 40x80mm carriage/gantry and 40x40 for the surface


About 700$ for 80x80 , 40x160 for the gantry , 40x40 table surface holder

[ger21]

For a given motor size, the motor with higher current will usually be faster.
Say you have three 450 oz motors.
1 Rated 2 amps, 1 rated 4 amps, and one rated 6 amps. The 6 amp motor will have the most torque at higher speeds, which will make your machine faster.

Now, if you have three motors all rated 6 amps, 1 at 450oz, 1 at 960oz, and one at 1200 oz, then the smallest motor will be the fastest.

Look at the inductance. The lower the inductance, the better. The 465 oz Nema 34 ha an inductance of 1.65mH. The 570oz Nema 23 has an inductance of 2.5mH.
With the 570oz motor, the torque drops off rapidly above 400 rpm. With the 465oz, it drops off much more gradually, with the 465oz motor being more powerful once you get over 500RPM.

[Letomoto]

oh wow , it would look like flow charts charts of turbo turbines ,

so the lowest inductance with let the motor keep his torque chart more evenly equally thru a wide rpm range because the low Inductance is more free flowing current , more efficient. this is related with the amps rating specs

Seriousy Gerry , you are really helpfull . Your knowledge is really shared nicely and understandable all the listings of specs on the sellers product pages are confusing a lot when not knowing everything !

[Letomoto]

i am tripping balls right now !

[wizard]

oh wow , it would look like flow charts charts of turbo turbines ,

so the lowest inductance with let the motor keep his torque chart more evenly equally thru a wide rpm range because the low Inductance is more free flowing current , more efficient. this is related with the amps rating specs

Sorta. The first thing to realize is that steppers are not really motors in the sense of a brushed DC or induction AC motor. Steppers have electromagnets that turn on to orient rotor poles to the electro magnet. Once the coil is turned the rotor steps into position and is held in that position for as long as that coil is turned on.

Inductance affects how fast the current can rise to the rated value, this along with mechanical factors in the rotor affect how fast the rotor can turn. To get the quickest turn on time possible the voltage applied to the coils is very high relative to the amount of current that needs to flow. Thus to prevent the stepper from catching on fire the electronics in the stepper drive switches that voltage off when the rated current is achieved. The drive can do this real fast keeping the coils saturated at the rated current.

This is profoundly different than other motors where current is largly the function of back emf. It is also the reason why stepper drives should be set as closely as possible to the rated current of the stepper. Torque is a function of current and the drive regulates the current precisely, so you need correct driver setup to realize manufactures operating performance.

Seriousy Gerry , you are really helpfull . Your knowledge is really shared nicely and understandable all the listings of specs on the sellers product pages are confusing a lot when not knowing everything !

Gerry is the man!

I’m hoping I didn’t confuse things much with my explanation above but I think it is important to understand what a modern driver does when connected to a stepper. In simple terms the driver is responsible for current regulation.

[Letomoto]

yep i am now looking at Closed Loop stepper and Hybrid driver 3N.M but it is looking pretty more expensive than the basic steppers , i read quite a lot of things and it seems like a very good choice but it is 2X price ... my project is already raising the expected budget !

i set up all the mechanical motion parts lenghts that i received today in the Fusion 360 , i need to design the gantry plates , exact mounting points of the linear rails , adjust the table widths to align gantry plates and have correct clearance to fit the motion components , clearance under the grantry beam etc .... it is a pretty hefty job , it is actually my first machine design but i did played a lot for 3D printing a second 3D printer but it wasnt the same level !

[wizard]

I would not bother with closed loop steppers. Instead look into AC servos if you want something “better” than steppers.

I quoted the word better because we need to remember what we are making here. In my opinion it takes far more effort to build a machine that can really justify something other than steppers. Even if your only goal is speed you still need a well designed machine to operate well at high speeds.

yep i am now looking at Closed Loop stepper and Hybrid driver 3N.M but it is looking pretty more expensive than the basic steppers , i read quite a lot of things and it seems like a very good choice but it is 2X price ... my project is already raising the expected budget !

i set up all the mechanical motion parts lenghts that i received today in the Fusion 360 , i need to design the gantry plates , exact mounting points of the linear rails , adjust the table widths to align gantry plates and have correct clearance to fit the motion components , clearance under the grantry beam etc .... it is a pretty hefty job , it is actually my first machine design but i did played a lot for 3D printing a second 3D printer but it wasnt the same level !

Yes this is something that is very often underestimated. Doing a machine where 100% of the design work has been accomplished befor the first tool hits metal takes time. This even more so if you Do the mechanical analyst to verify that the machine will operate within well defined specs. This is why many with buy plans or wing a build. Building on the fly is doable if you have the machine building skills but maybe shouldn’t be suggested for everyone.

In the end routers, at least most of the builds discussed here, are light weight machine tools. As such your expectations have to match the machines capability.

[Letomoto]

this is also a big project for me to get experience points , learning a lot of things and developping skills further . Challenging me alot and exciting ! For steppers i am not looking for the fastest speed i could reach but i would like to never skip a step somewhere and wreck a big stock where a production of maybe 20 parts was started and the toolpath got offset half way in the job and not telling it .

I do not have a lot of knowledge on closed loops and zero knowledge for AC Servos . i was thinking the closed loops were normal steppers but acting like a servo with their encoder and driver. A way to reduce the price of a servo but if you tell that the closed loop doesnt worth much i will dig more on their comparisons




Edit : I am searching/reading the comparisons and i am reading a reply from Gerry !

Re: Are Chinese servos a gamble? Servos vs Leadshine closed loop

I've heard that those chinese servo's work well, if you know how to set them up. I've heard that documentation is severely lacking.
I opted for DMM servos, as the price is only slightly higher, and they have better documentation and support. They would probably be much more expensive in Australia.

Comparing an AC Servo to the Leadshine "Easy Servos" is comparing apples and oranges. The AC servos are much more powerful. The Easy Servos are still stepper motors, with a torque curve that drops off as rpm's increase. AC servos have a flat torque curve, up to 3000 RPM

Gerry


Went to search on Aliexpress for 3N.M servo and ouch it doubled or tripled the 3NM closed loops prices


I also found there is an expansion board for the Smoothstepper ESS to add encoders input . and also dual shafts steppers ... if i get simple steppers and having troubles with steps at some point maybe i could upgrade them with encoder if needed . Right now the price i found for the closed loops 3n.m on Aliexpress it would cost me almost 600$usd for 4-axis but simple steppers 4-axis with BOB , psus 273$usd shipped

And Gerry suggested a 450oz-in for my setup and the Nema34 are rarely under 890oz-in in the market , So should i focus searching for 23 or 34 ? sorry i am a bit ADD


Here for Nema23 425oz-in 4 axis 4.2A rate

https://www.aliexpress.com/item/NEW-Free-shpping-motor-CNC-Router-4-Axis-kit-4pcs-TB6600-Stepper-motor-driver-4pcs-Nema23/32854609719.html?spm=2114.search0104.3.49.15ef1885 DwzVZX&ws_ab_test=searchweb0_0,searchweb201602_2_10065_10 130_10068_10547_319_317_10548_10696_453_10084_454_ 10083_10618_10139_10307_538_537_536_10059_10884_10 887_100031_321_322_10103,searchweb201603_51,ppcSwi tch_0&algo_expid=a9df58a0-dde4-4243-bee5-88ba0f936251-7&algo_pvid=a9df58a0-dde4-4243-bee5-88ba0f936251

then a kit of 890oz-in Nema34 it jumps up to 455

https://www.aliexpress.com/item/Good-quality-Wantai-4Axis-Nema34-Stepper-Motor-85BYGH450D-007-890oz-in-Driver-DQ860MA-7-8A-80V/32821205632.html?spm=2114.search0104.3.79.2f234b4e zifqYA&ws_ab_test=searchweb0_0,searchweb201602_2_10065_10 130_10068_10547_319_317_10548_10696_453_10084_454_ 10083_10618_10139_10307_538_537_536_10059_10884_10 887_100031_321_322_10103,searchweb201603_51,ppcSwi tch_0&algo_expid=5a56a75d-da1f-4d8f-ac8a-e6c0c4aca240-13&algo_pvid=5a56a75d-da1f-4d8f-ac8a-e6c0c4aca240

[ger21]

I already gave my advise.
Most of those kits perform very poorly, especially the ones with larger motors.
there are a lot of threads here from users who bought cheap kit with big motors, and found that they are much slower than small motors.

[Letomoto]

Yes i understand this part , having a hard time finding small oz-in 34s but i found 23s at the oz-in you suggested 425-450 are available so compare a few different amps rates and select the high rate .But you suggested a 34 would be good.

So am i wrong in the logic or a high force 23 would be slower than a small power 34 ?