[Fatamorgana]

Hello!

I got a supre for stress releaving/heat treatment today, 1800$ not very cheap so I have to look around other places aswell.

Im located in Norway.

The thing is that I want to use the materials I have laying around, and that is

6m 150x150x10mm
12m 100x100x6mm
60x60x800 (2pcs of steelbar)

I have made som changes to the initial design.

I will weld in the 60x60 steelbar and they will be used as base for linear rails.









Sent from my iPhone using Tapatalk

[joeavaerage]

Hi,
OK, so you have a couple of costs so far, the ballscrews at $1400 and heat treating could be as much as $1800.

I understand wanting to use the materials you have on hand, but that is forcing you into compromises which you do not need to make.
Lets be honest here, if you went and bought all the steel for this job brand new it would still be less than the cost of the ballscrews.
If you want to economise don't do it on the cheapest part of the build.

The good news is that Norway is hugely industrialised so anything you could possibly want is already available.

May I suggest pricing three bits of plasma cut steel, say 50mm or 60mm even 75mm thick, say 800 long and 300wide. Now put that in your mill and machine a channel
down the middle. If you don't weld anything to the steel slabs then you wont have to stress relieve. You might be surprised how cheap slabs of steel are.

Given that you are in the EU you'll have good access to big chunks of continuous cast iron. Price it....you might be surprised.

As peteeng says start at the Z axis (ie the spindle carrying axis) and then work backwards.

My own design was to make three near identical axis beds complete with ballscrews, rails, servos and spare no expense in making them as good as I could. Then bolt those axis beds
to a constructed steel frame. In my case the cast iron axis beds when completed including ballscrews etc were about $16000NZD. The steel frame consists largely of two 'L' shaped pieces
of plasma cut 32mm steel. They were 'only' $2000NZD. So the frame is in some respects the cheapest part.

Even if at some stage I replace the frame with a different design, or make it stiffer or whatever, the same axis beds would still be used. You might say a modular design.

Craig

[peteeng]

Hi Lars - I have leftovers too. Occasionally I muster them and say I'll design something that uses them up. But then I find it's a compromise in some way. So they go back on the shelf or I give stuff away. Why not design the best machine you can, then see if the leftovers can be used vs starting with a compromise which will further infiltrate your machine in ways you won't like? Forget about cost and design a great machine. Especially in what's called Round 1. You have just emerged from what's called "The fuzzy front end" of the project. You are gaining clarity and direction on the machine. The aim of R1 is to realise a functional general arrangement that is at minimum functional design. MFD is defined by if you take that part away does it work? answer NO so that part is a primary element to the design eg the buttresses on your prior design take them away and it still works, so don't have them there in R1. Boil down the design to absolute basics... Once you have a good GA you move to R2 which covers structural requirements and secondary elements. As Craig says material is the cheapest part of the project.... Peter

[joeavaerage]

Hi,
this outfit used to have an excellent range of material, but they have turned off the stock list, maybe you'll have more luck with them:

https://www.aco-eurobar.com/en/continuous-casting

Craig

[peteeng]

Hi Lars - Looking at your current design. It has possibilities. You have placed a rail foundation in the middle of an SHS. Move it to the section edge so it does not vibrate. Plus cap the ends of hollow sections. The ends are half the stiffness of the tube as they can deflect at the free edge. Plus it stops the hollow acting as an organ pipe making odd sounds. Now since you are machining the "machine" you can take the punt and not TSR the frame. You have to leave a good machining allowance to do this. You weld it out as best as possible, stitch it symmetric welding, a bit here a bit there. Do a little and let it cool vs welding it all out. May take some time but it's worth it. Then you set it up and take light cuts and see how the frame is going ie release the part and measure it and see if it stays straight. If it's stays straight your good. If it deflects a bit compensate in the new set up and machine some more. This will settle the lands in.

In regards to heat treat was the price a job price or based on a $/kg price? Often you can ask for the job to be done with another job to reduce the cost. They may already try to wait until they have a full oven to do the run. Or the expensive price might be based on your job alone. Ask them how to reduce the cost and it may surprise you. They may have a regular job that it can be done with. But the main thing at the moment is that you can have it done and budget around it... Plus the hollows can't be closed out if you are to TSR there must be breathers in the tubes. Ask the heat treater what size holes they need. Peter

[Fatamorgana]

I quickly did some drawings when I got back home now, are we speaking the same language here with something like this?

80mm slabs.

I got some good prices on steel so that wont be any challenge budgetwise.

[Fatamorgana]

Edit: the left supportplate was misplaced


Sent from my iPhone using Tapatalk

[joeavaerage]

Hi Lars,
yes that is very much like my first mini-mill.

If the material is thick enough, and the channel is not too deep then you don't need the ribs underneath the Y axis. I would try to make the axis beds out of the thickest material you can
get at reasonable cost. Here my local profile cutter stocks 75mm med tensile sheet.

You will need a gap between the Y axis and the column or you will lose a lot of travel. The bigger the gap between column and the Y axis give you room to put decent size parts on the X axis table,
but the greater the distance the more compliant (flexible) the structure becomes. There is a balance to be struck.

I would in the first instance make enquiries of steel suppliers in your area. It will become apparent very quickly whether this approach is economic.

Here I pay about $6.00NZD /kg for profile cut steel. If I wanted three pieces say 800mm x 300mm x 75mm then the total weight would be 432kg or $2592NZD ($1550USD).
I would call that economic, but your opinion may differ.

Craig

[peteeng]

Hi Lars - This billet style of build is very good. Some details to figure out but far superior to a hollow section build. Peter

[joeavaerage]

Hi Lars,
the difference in height between your proposed ballscrews and the height of the linear rail cars determines the depth of the channel that you would have to mill.
The idea is basically to have the channel as a minimum and thereby leave as much material at the base of the channel and retain the maximum rigidity.

You may well have noted that many ballscrews are advertised as 'cut flange' ballnuts. This is to reduce the overall height of the ballscrew and aid in the design of low profile machines
of maximum stiffness.

While I was lucky to get the ballscrews I did, they were not cut flange. I also bought from dy-global a set (six rails 12 cars) of 20mm HSR linear rails by THK as new old stock. I would
have preferred 25mm or even 30mm rails but that's what I could get at the time. The consequence is that the difference in height is quite large and therefore the channel quite deep, 70mm.
Given that I was casting the axis beds I could make the base as thick as I wanted, but would have been a poor choice for a solid slab of steel as I have recommended to you. Even a 75mm
slab with a 70mm channel would leave no material to have any rigidity.

I would suggest you consider the linear rails and the ballscrew as a combination, because that will in large part determine whether milling a channel in a thick piece of steel is feasible.
I personally would like to retain 40mm-50mm thickness in the base suggesting the channel be 25mm-35mm deep with 75mm slabs. Of course steel can be had in much thicker slabs
but the cost creeps up. There again there is a balance to be struck.

Craig

[joeavaerage]

Hi Lars,
I imagine that given Norway's huge oil, gas and marine industries that you have a very competitive market for steel which will benefit you. I also understand that Norway is a world leading
producer of aluminum. If there is any price advantage to buying local product then aluminum should be considered also. Here in New Zealand aluminum is moderately pricy. For a structure
of given stiffness aluminum is about 25%-40% more than steel. These are the conditions that apply here....but maybe different in your home market.

Just as a matter of interest New Zealand produces over 90% of our electricity by hydro and geothermal. One of the biggest users of electricity is the aluminum smelter at Bluff, and you might imagine
we'd get 'home team advantage'...but no. I would personally like to line up all the RioTinto bosses and f****k them up the arse with a stake!

Craig

[Fatamorgana]

I just got quoted 1050,- dollar for 3 pcs of 300x800x80 and 1250,- for 300x800x100


Sent from my iPhone using Tapatalk

[Fatamorgana]

What does this chart tell me? Seller says its a preload nut but chart says «no preload»





Sent from my iPhone using Tapatalk

[joeavaerage]

Hi,

I just got quoted 1050,- dollar for 3 pcs of 300x800x80 and 1250,- for 300x800x100

That is very competitive and you could make a very good machine with such materials. I am envious!

Cannot say which is correct, namely no-preload vs preload. It is not a cut flange type and the diameter of the flange is 85mm. Thus from the underside of the saddle to the lower plane of the channel must be at
least 85mm, or 90mm as a practical matter. The height of 25mm linear rails/cars is 36mm. therefore the depth of the channel must be 90-36 =54mm. Therefore even 80mm tick material would be down to
26mm thick in the channel, which I think is too low.

If you used 30mm linear rails/cars, the height is 45mm. Thus, the channel would be 45mm deep and the thickness remaining with 80mm thick material is 35mm. Better, but still marginal.

You can see how critical it is to reduce the flange height, either by cutting the flange yourself or buying ballscrews that already have 'cut flange'. For instance the NSK FA range a 25mm ballscrew is 48mm
across the cut flange, and giving you a little clearance say 55mm. With a 25mm linear rail/car and 36mm height the channel depth is 19mm and with 30mm linear rail/car of 45mm height the depth of the channel need only be
10mm deep.

Note how the cut flange style goes a very long way to matching the height of the linear rail/car and therefore the depth of the channel can be much lower. This means that for an axis bed of a given stiffness
the thickness of the material can be reduced as can the machining time to cut the channel in the first place.

With a channel depth of 19mm in an 80mm thick slab leaves 61mm thick at the bottom of the channel. Even were you to mill 20mm deep T slots into the channel as I did for the X axis there is still 40mm of material
undisturbed between the bottom of the channel and the T slot. This is the great advantage of cut flange ballscrews. Despite the cost of new NSK FA ballscrews the very favorable dimensions may make the design
and build that much better that the cost is not only justified but rewarded by savings made elsewhere.

Craig

[Fatamorgana]

Craig - there is some high valuable learned leasing right there, my humble thanks for your time and feedback on this matter.

The hunt for the right ballscrews w/nuts continue. Still waiting on reply from bst.

But most diserable is from dy-global through the price is higher.

My steel supplier Is of gold worth as they push insane amount of steel to the local industry, so the prices are bcs of quantity very low. Im just a person enjoying my hobby and they still take my buisness, thats something to appreciate.


Sent from my iPhone using Tapatalk

[peteeng]

Hi Lars - As an aside you can drive the table on the side. This can side step the gutter issue as then you don't have a gutter. But large rails/cars and a "smaller" screw will give you a small stack height. Plus use a cut flange nut. The search continues. Speaking of cars you will need medium or high preload carriages so look those up. Peter

[TMToronto]

In designing my Z axis assembly for my hobby CNC router, I also tried to pay attention to the heights of the various components.

Although my ball screw is smaller than those currently being discussed at 15 mm, but when I planned my support bracket purchases from SYK, I chose the low version motor/ball screw support. I also chose 20 mm rails (mostly because I did not want to use M4 bolts) but also to get the roller guide height higher. I chose a cut flange nut model for my ball screw assembly which also kept this height lower.

When I looked at the SYK catalogue it looks like the 'low' version components are available up to a 25 mm ball screw.

[Fatamorgana]

Evening citizens

I have now gotten hold of two hiwin grounded 5mm ballscrews with preloaded nuts, quickly did some drawing tonight based on the steelslabs we were talking about earlier. The ballscrews are for x/y and are equally 750mm long 32mm dia.

I have turned the rails for xaxis the other way around after alot of reading here at the forum these last weeks, but cant just figure out what will be the right spacing between the cars to get the right travel etc.

I will take all of your feedback with appreciation.

[pippin88]

Always check how bolts will work

You will likely want flanged linear rail cars - they can be bolted from below

[joeavaerage]

Hi,
yes, that design would work, but I think you can do better.

In particular I see you have one of your 'slabs' in the center sitting on top of the Y axis? That is conventionally called the Y axis saddle.
The issue is that you require two set ups. You need to drill the bottom side to mount the cars and the ballnut mount, and then from the topside you need to drill for the
X axis rail cars. I can tell you from experience that having two setups that are exactly 90 degrees to each other is a f*****kup waiting to happen.
This was basically what I did with my first mini-mill.

My new build I had decided I needed a MUCH better solution.

The X axis is constructed similarly with an identical saddle. The X axis saddle is lowered upside down onto the Y axis saddle and through bolted. This allows you to tweak the two axes to be exactly orthogonal.
It also means that the now upperside of the X axis 'slab', that is to say the opposite side to the channel (or gutter using peteengs term) can now be milled with T slots.
What I did was make a steel saddle for the Y axis (275mm x 275mm x 20mm) and drilled for the Y axis cars and ballnut mount.

This arrangement is very much simpler. Firstly it allows you to bolt the ballnut mount and cars to their respective saddles WITHOUT those same cars and mount interferreing with the assembly
of the X axis cars and mount on the top of a single piece saddle.

With reference to the pic, see that the 'slab', actually a casting in my machine, is upside down with the gutter underneath and T slots milled into the top. That was why I am so set on having the
lowest possible gutter and thereby have as much material thickness left so you can mill T slots without loss of stiffness. My Y axis is all covered over with way covers and I'm not going to disassemble
them just to take a pic.

Craig