'Otter'. Build log.

[Blazing Black Beard]

The 'Otter' CNC machine I am building is based on rigidity for harder materials for smaller jobs and versatility for larger jobs with softer materials.

I have collected a small amount of essential parts to build my machine. I will continue to post images as I progress the 3D design, procure more items for the build and as I make parts for the machine.

I am a beginner as the last workshop I was actively part of was about 25 years ago and plenty of skills have been lost over time since then. :-(

So here goes, any criticism is welcome to help improve my build.

Current parts list:

8x 100x50x3x1250 cold finished hollow section Strongbox 235 EN 10219:2006 grade S235JRH BSEN10204 2.2CERT

4x NEMA 23 3Nm 3Amp stepper motors

6x TB6600 micro-stepping drivers (2x redundant spares) Amazon lost them before delivery

Still to buy:

Bed plate
Linear rails and bearings
Ballscrews
Cable tray and chain
Stepper motor drivers
VFD
Motion control unit
Power supply unit
Workpiece coolant mister system
Cooling pump for spindle
ATC spindle
Vaccuum and compressed air cleaning system

Future upgrades:
Auto tool changer built with above Otter CNC

I'm currently researching Hiwin linear rails and bearings but they are hellishly expensive on the web sites I have found. I don't want to cheap-out on rails and bearings as I need a fair amount of accuracy and would like to machine harder materials, minimising vibration. Any help finding cheaper but still good quality linear rails and bearings would be welcome.

I'm also not sure about the TB6600 stepper motor drivers as they seem fairly cheap and I have absolutely no idea what would be required of a more capable machine. They seem to cover the requirements of the NEMA 23 motors, with micro-stepping and current limiting. Are there any problems with these and if so are there better alternatives? Amazon failed to deliver and there are much better alternatives.

I've been messing around with GRBL but have not settled with a motion control solution yet. I am not familiar with GRBL's capabilities and could quite easily be encouraged to research a more established standard of motion control, as long as it's reasonably economical to implement. This appears to be the smallest challenge considering that the physical build will be much more demanding as far as calibration and physical tweaking are concerned.

I will be starting to build a frame for the table with the steel box section next weekend. I have zero welding skills so I'm considering drilling precision holes and using substantial bolts to fit brackets that will secure the frame both outside and inside. This should be just as good, if not better, than any weld for rigidity.

That's about it for now. Any comments and suggestions would be appreciated.

BBB

[mekanik]

Hi B3
Welcome to the forum
The first rule of any build is not to buy any of the electrics until you have everything designed in CAD/Sketchup. then offer it up for the guys to look over, they will put you right regarding things you might have overlooked. Most of the guys tend to buy ther rails and ballscrews from Fred, He will supply Hiwin and the cheaper generic stuff, best to get it all in one shipment, you will have to pay import duty ect, but he has a good reputation on the forum and will rectify any problems with gear thats not as described.
If i (or some other kind soul) can find a link to Freds shop i will post it.
Good luck with the build.
regards
Mike
Just a couple of links i had saved not sure if the first one is Fred's


https://www.aliexpress.com/store/product/Anti-Backlash-Ballscrew-RM1605-L1000mm-SFU1605-Ballnut-BK12-BF12-End-Support-1605-Nut-Housing-Bracket-6/314742_1639768705.html?spm=2114.12010612.0.0.PXFXn 5

https://www.aliexpress.com/item/HIWIN-Taiwan-made-2pcs-HGR20-L-1000mm-15mm-linear-guide-rail-with-4pcs-HGH15CA-narrow-carriage/1852980416.html?spm=2114.13010208.99999999.263.UMY DXN

[Blazing Black Beard]

Hi B3
Welcome to the forum
The first rule of any build is not to buy any of the electrics until you have everything designed in CAD/Sketchup. then offer it up for the guys to look over, they will put you right regarding things you might have overlooked. Most of the guys tend to buy ther rails and ballscrews from Fred, He will supply Hiwin and the cheaper generic stuff, best to get it all in one shipment, you will have to pay import duty ect, but he has a good reputation on the forum and will rectify any problems with gear thats not as described.
If i (or some other kind soul) can find a link to Freds shop i will post it.
Good luck with the build.
regards
Mike
Just a couple of links i had saved not sure if the first one is Fred's


https://www.aliexpress.com/store/product/Anti-Backlash-Ballscrew-RM1605-L1000mm-SFU1605-Ballnut-BK12-BF12-End-Support-1605-Nut-Housing-Bracket-6/314742_1639768705.html?spm=2114.12010612.0.0.PXFXn 5

https://www.aliexpress.com/item/HIWIN-Taiwan-made-2pcs-HGR20-L-1000mm-15mm-linear-guide-rail-with-4pcs-HGH15CA-narrow-carriage/1852980416.html?spm=2114.13010208.99999999.263.UMY DXNThanks for that mekanik. After reading a few builds the motors I chose seemed to be what everyone else is using. Would you say I made a bit of a rushed decision to get NEMA 23 3NM motors and drivers?

I got the steel box section as I thought it would be more rigid and provide the extra weight needed to machine harder materials and minimise vibration. I didn't think architectural construction steel (which is what this seems to be) would be any less suitable than other types as it's already way more rigid than aluminium extrusion. Is there anything you or anyone else can offer regarding this choice or alternatives?

I will be sending images of my initial design for you guys to look at and see if I'm doing OK or missing some fundemental considerations.

Thanks again for the links. I'll check them out and compare prices and as always, I'm looking to shave anything off the costs of this build as long as it doesn't take away too much quality or precision. I want to get it right :-)

[Neale]

Black Beard - take a look at my "Avor" build (http://www.mycncuk.com/threads/11617-AVOR-–-a-steel-framed-medium-size-router). I wouldn't necessarily recommend that you (or anyone else) tries to copy it, but you might get a feel for the kinds of things that go into a machine like this. For example, those stepper drivers will very significantly limit performance.

As suggested, ask more questions before buying too much more, but welded steel box, Hiwin-style rails and ballscrews can give a very capable machine - I've recently been profile cutting small steel components with mine. Good luck with the build!

[Blazing Black Beard]

Black Beard - take a look at my "Avor" build (http://www.mycncuk.com/threads/11617-AVOR-–-a-steel-framed-medium-size-router). I wouldn't necessarily recommend that you (or anyone else) tries to copy it, but you might get a feel for the kinds of things that go into a machine like this. For example, those stepper drivers will very significantly limit performance.

As suggested, ask more questions before buying too much more, but welded steel box, Hiwin-style rails and ballscrews can give a very capable machine - I've recently been profile cutting small steel components with mine. Good luck with the build!Thanks for the info. I'm checking out your build now. I don't have any experience with any CNC stepper drivers. If they are going to severely limit the machine they can be easily replaced with something more capable. I'm open to suggestions and any other information about stepper drivers that may help. Is there anything in particular I should be looking out for with better drivers?

[jarjar]

8x 100x50x3x1250 cold finished hollow section Strongbox 235 EN 10219:2006 grade S235JRH BSEN10204 2.2CERT


Ah, Tata Steel product! Can I ask where did you get it?
I would have chosen their Celsius 355 product instead, which is already stress relieved.

Also, you can't just bolt steel tubes and expect a good joint. The contact surface will be poor. You need to first bond the tubes with something like a steel-filled epoxy, and then bolt.

[Blazing Black Beard]

Ah, Tata Steel product! Can I ask where did you get it?
I would have chosen their Celsius 355 product instead, which is already stress relieved.

Also, you can't just bolt steel tubes and expect a good joint. The contact surface will be poor. You need to first bond the tubes with something like a steel-filled epoxy, and then bolt.Hello jarjar. Welcome. To my build log.

As my first post suggests, total novice here. Stress relieved? I'm not sure what that means. I chose this steel box section because it was half the price of anything else I could get my hands on and seemed a much more suitable and rigid material than aluminium. I didn't know different types of box section steel would be more suited.

What is their Celsius 355 product? I'm afraid I did not do much research into the types of steel box section available apart from dimensions. Maybe a bad move on my part? Any comments on this would be appreciated.

Neale, I took a good look at your posts before starting this log and this is where I got a lot of my rigidity ideas from. Although now it seems I have got myself some steel that may not be as good as yours, we will see. I hope the steel box section I have is not going to be a major drawback. If it is then please let me know.

I'm hoping to make a start to the 3D design over the next couple of days and I will post some images of my progress.

[Blazing Black Beard]

I know this is not something I should really be considering to buy so early on in the design stage but I could not help but take a look around at spindles. There's a mind boggling amount available out there and one that especially caught my eye was this:

https://youtu.be/pWNeW-lH9RA

Any comments regarding suitability or alternatives would be appreciated.

[Neale]

Don't worry about the steel quality - I didn't! I just bought something from my local steel supplier, probably generic "construction" steel. Doubt if it's stress-relieved or anything else. It is described as "Grade 43a", whatever that means. Welds and cuts OK, and I knew from the outset that I would need to build in adjustment/levelling, etc. Being fussy about steel quality is OK if you can actually buy the quality you are after in small enough quantities (a lot of suppliers will charge more in carriage than the cost of a small quantity of steel - that's not criticism, just a reflection of commercial reality) and if your design is actually sensitive to the steel quality. My design, such as it is, could be beefed up significantly by using, say, 4mm wall thickness rather than 3mm, and for the static bed and frame components, that might not have been a bad idea. However, 3mm works fine for the gantry and does keep mass down.

Don't think you've said how big this is going to be, or what you want to cut on it. This drives some aspects of the design. For example, my machine has about 1.5x0.75m cutting area. For that, I have used 2005 ballscrews. If I were only cutting wood, then 2010 would have been better - faster rapids and cutting speeds. However, I end up doing quite a lot of small fiddly bits and the speed isn't too critical so 2005 works well for me. Similarly, I use NEMA23 3Nm motors; for faster speeds I could have gone to 4Nm for a bit more torque but I'm OK at the moment. However, stepper motors need volts to get performance out of them; don't be confused by spec sheets that say things like "4.2V". That's a red herring. Most of these size motors will work best from about a power supply giving about 68V via a suitable stepper driver. There are plenty of threads that look at different drivers and how to build power supplies.

I'm using a "standard" 2.2KW spindle and matching VFD. Don't know much about ATC spindles apart from the fact that a decent one will cost big bucks.

[Blazing Black Beard]

Don't worry about the steel quality - I didn't! I just bought something from my local steel supplier, probably generic "construction" steel. Doubt if it's stress-relieved or anything else. It is described as "Grade 43a", whatever that means. Welds and cuts OK, and I knew from the outset that I would need to build in adjustment/levelling, etc. Being fussy about steel quality is OK if you can actually buy the quality you are after in small enough quantities (a lot of suppliers will charge more in carriage than the cost of a small quantity of steel - that's not criticism, just a reflection of commercial reality) and if your design is actually sensitive to the steel quality. My design, such as it is, could be beefed up significantly by using, say, 4mm wall thickness rather than 3mm, and for the static bed and frame components, that might not have been a bad idea. However, 3mm works fine for the gantry and does keep mass down.

Don't think you've said how big this is going to be, or what you want to cut on it. This drives some aspects of the design. For example, my machine has about 1.5x0.75m cutting area. For that, I have used 2005 ballscrews. If I were only cutting wood, then 2010 would have been better - faster rapids and cutting speeds. However, I end up doing quite a lot of small fiddly bits and the speed isn't too critical so 2005 works well for me. Similarly, I use NEMA23 3Nm motors; for faster speeds I could have gone to 4Nm for a bit more torque but I'm OK at the moment. However, stepper motors need volts to get performance out of them; don't be confused by spec sheets that say things like "4.2V". That's a red herring. Most of these size motors will work best from about a power supply giving about 68V via a suitable stepper driver. There are plenty of threads that look at different drivers and how to build power supplies.

I'm using a "standard" 2.2KW spindle and matching VFD. Don't know much about ATC spindles apart from the fact that a decent one will cost big bucks.Neale again thanks for all the info. My gantry will be somewhat like yours I assume so my 3mm thick 100x50x1250 box section should do fine for the X axis (left to right on gantry and table ends).

I will be cutting small aluminium parts with high accuracy and superior finish, larger aluminium plate with accurate hole cutting but still relatively small compared to the size of the machine which will be approximately 1000mm x 1500mm cutting area footprint. I also aim to easily cut steel sheet and small accurate steel parts from time to time.

As for the ballscrews, I seem to keep reading that 20mm diameter x 5mm thread seems to be pretty common and you use them, so I think I may settle for these when I come to buy the rest of the components after I have finished the first phase of the 3D design.

I know virtually nothing about spindles or their VFDs. You mention 4Nm for the NEMA 23 motors. I'm assuming the holding force is better for heavier gantry arrangements that have more inertia, therefore need more force to ramp speeds without skipping position? Looking at motors and drivers is as mind boggling as spindles and the various (and sometimes exotic) materials that some people use in their builds.

I have been looking at a marble bed and gantry arrangement and I am having a hard time drawing myself away from that expensive venture (eek!). I don't think I will ever do it but it's hard to stop looking at these machines and what I am inclined to think about their accuracy. They must be extremely rigid and very heavy, not to mention unwieldy should one want to move the machine for any reason. I can also imagine that accidents on a marble bed/gantry machines can be costly.

Anyway, back to the real world where I think I will stick with a 100x50x1250 box section gantry. Would moving to 4mm-6mm thickness be of any benefit for my 1800mm sides? Is there going to be any benefit from using thicker box section on the top part of my bed frame? I'm guessing that the rigidity of the 4+mm box section is only of any benefit where there is an axis directly connected with linear rail and bearings and the thinner 3mm box section would be adequate to form the lower part of the frame and the supporting uprights and diagonals? Does that sound about right or am I missing something?

[Voicecoil]

There's a really useful motor/ballscrew calculator spreadsheet somewhere on this forum which I found really useful, though I can't seem to find it at the mo. - worth digging out and putting in your predicted mass/lengths/desired speed to see what motor/power supply suits.
On the subject of steel tubes I stumbled across this last week whilst looking for something else:
https://www.barrettsteel.com/products/tubes/erw-precision-tube/
other vendors seem to do similar products, it appears that the straightness spec. is better than normal stuff.

[Blazing Black Beard]

There's a really useful motor/ballscrew calculator spreadsheet somewhere on this forum which I found really useful, though I can't seem to find it at the mo. - worth digging out and putting in your predicted mass/lengths/desired speed to see what motor/power supply suits.
On the subject of steel tubes I stumbled across this last week whilst looking for something else:
https://www.barrettsteel.com/products/tubes/erw-precision-tube/
other vendors seem to do similar products, it appears that the straightness spec. is better than normal stuff.Voicecoil thanks for that. I'll try to dig out the spreadsheet.

Good point about the straightness spec. I may just use the 100x50x3 section as strengtheners in the lower part of the frame if it's not as straight as required. Is there any way I can check the straightness and rectify the rails should I need to? I can see lining up the rails might be an issue if I just use a gauge and the face of this box section should it not be adequately straight.

[Voicecoil]

I would say that a decent long straight edge and a large square are pretty essential when building a machine - unless you happen to have access to a large surface table of course!

[Blazing Black Beard]

Some NEMA 23 stepper motors came todayhttps://uploads.tapatalk-cdn.com/20181024/6936b694bfeaca856aa2184be26a6453.jpghttps://uploads.tapatalk-cdn.com/20181024/ec40f5b1deeae053b44315cbdf25b50c.jpghttps://uploads.tapatalk-cdn.com/20181024/37c46372f13daae14f3cd775fa5510b3.jpg

[Blazing Black Beard]

I would say that a decent long straight edge and a large square are pretty essential when building a machine - unless you happen to have access to a large surface table of course!So is there any way I can check straightness? Is there any way I can rectify it if it's not straight?

Is there any way I can reference another known straight face without too much expense?

Is there a way I can still use the lower grade steel box section, cut the the mounting points and then tweak the linear rails with something known to be straight?

I see people using the known good face of aluminium extrusion to set their first rail but this build doesn't use that for the structure.

[mekanik]

Most people just use the standard box section, if you are welding there will be distortion you can reduce this somewhat by just stitch welding it, doesn't have to be full weld. the distortion in the base (X) axis is is eliminated with epoxy grout levelling.
Regards
Mike

[Blazing Black Beard]

Most people just use the standard box section, if you are welding there will be distortion you can reduce this somewhat by just stitch welding it, doesn't have to be full weld. the distortion in the base (X) axis is is eliminated with epoxy grout levelling.
Regards
MikeHow can the face on the side of the box section be used to keep the rails parallel though? These can't be poured with epoxy as it would just run down and onto the floor.

If you look at RoboCNC you'll see he just uses the straight vertical face on the side of the aluminium to keep his linear rails parallel and straight.https://youtu.be/YOVHtjXi07Q?t=2m02s

[mekanik]

The epoxy goes on the rail mounting surface, the guy in the video is using the ally profile as a reference, dont like this unless it's been machined to be flat, you should be using a precision straight edge to set the rails.
Regards
Mike

[Blazing Black Beard]

The epoxy goes on the rail mounting surface, the guy in the video is using the ally profile as a reference, dont like this unless it's been machined to be flat, you should be using a precision straight edge to set the rails.
Regards
MikeYes, that's what I thought. I need to find a precision straight edge then. Any suggestions?

As for the stepper motors I received earlier, they do not have the D shaped shafts. A friend of mine said it was a shame but they look like quality motors. I'm going to take an educated guess that round shafts aren't as good for this particular application, is this the case? Can anyone see any issues with them being round shafts as opposed to D shafts?

[Neale]

Unless you are using very fancy pulleys with D-shaped holes, then the flat is only for a grub screw to bear against. It's not critical and you could always put a flat on with file (if soft enough) or small grinding point. That's what I did with the ends of my ballscrews which did not have a flat as supplied.

[Blazing Black Beard]

Unless you are using very fancy pulleys with D-shaped holes, then the flat is only for a grub screw to bear against. It's not critical and you could always put a flat on with file (if soft enough) or small grinding point. That's what I did with the ends of my ballscrews which did not have a flat as supplied.Ah well I'm worrying for nothing then. I'll keep these on the shelf until the physical build is done. Still waiting for the steel to arrive, then hopefully I will have some idea what I'm doing with it.

With the top of the steel box section being the bed for my linear rails, would skimming a very tiny amount (possibly 0.1 mm or less) from them with an end mill give a flat enough surface to mount on so the epoxy can be skipped altogether?

[mekanik]

I am not being disrespectfull but i think you really need to take the time to read some of the build logs, these will give you an idea of the various build methods and proceedures for machine alignment.
Regards
Mike

[Voicecoil]

Yes, that's what I thought. I need to find a precision straight edge then. Any suggestions?

Vogel is a brand I've heard bandied about a bit - made in Germany so likely good kit. We ought to work out some way on this forum where such tools can get passed from person to person as they do builds.

[Blazing Black Beard]

I am not being disrespectfull but i think you really need to take the time to read some of the build logs, these will give you an idea of the various build methods and proceedures for machine alignment.
Regards
MikeYes I have been looking at a few and there seem to be some discrepancies between them, hence my suggestion of machining the top surface. I will continue to take a look through more and see if there's anything I'm not accounting for. Thanks again for the pointers. Any help is appreciated as always :-)

[Blazing Black Beard]

Vogel is a brand I've heard bandied about a bit - made in Germany so likely good kit. We ought to work out some way on this forum where such tools can get passed from person to person as they do builds.Wow, sounds like a plan. I'd be up for it even if it involved tracked delivery, smart contracts and whatnot. You know, the blockchain and escrow or something similar.

Tool and instrument decentralised repository maybe? T.I.D.R. Sounds cool. [emoji3447][emoji2923][emoji2425][emoji366][emoji423][emoji424][emoji3537][emoji375][emoji373]

[Neale]

With the top of the steel box section being the bed for my linear rails, would skimming a very tiny amount (possibly 0.1 mm or less) from them with an end mill give a flat enough surface to mount on so the epoxy can be skipped altogether?

My 1.8m long box sections had a dip in the middle of about 2mm one side and 3mm the other. Tricky to machine flat when the wall thickness is 3mm to start with! However, I don't know if this was "as supplied" or welding distortion but it is why I went the epoxy levelling route. Plenty of discussion on the forum about that. I did add doubler strips inside the box section to give a better bite for the rail holddown bolts

[mekanik]

Hi Buddy
Have a read of this,
http://www.mycncuk.com/threads/6546-Help-needed-with-understanding-accuracy-in-measuring-tools-I-want-to-go-straight-%29
Regards
Mike