[jackjr-123]

I think you should finalize your design before buying more random parts and trying to make them fit

But yes you can buy bars of Blanchard-ground steel and bolt them. This is my go-to supplier: https://en.stahlnetz.de/a/67-Halbzeu...eziplan-1.1730

[Fatamorgana]

I think you should finalize your design before buying more random parts and trying to make them fit

But yes you can buy bars of Blanchard-ground steel and bolt them. This is my go-to supplier: https://en.stahlnetz.de/a/67-Halbzeu...eziplan-1.1730

Thanks, will check it out!

The build on paper(cad) is pretty much at the final stage, but the «as is» measurements Are hard to to until you Get the real parts in your hands. I have been struggling to Get second hand good quality parts for ballscrews/rails so Ideally is not the choices it would be of the shelf, so designing is also around the parts I got sourced and not the other way around, thanks for the tip anyway.

Also the illustrated model was something i quickly drawed up like a dummy

Sent from my iPhone using Tapatalk

[joeavaerage]

Hi Lars,

Did some measurements, my gutter needs to be 37 something mm with my sourced parts, thats to much.

Thats rubbish!!!! If you have a 40mm gutter that means there is 60mm undisturbed steel underneath.....surely that is enough!!!! The rail idea works but inevitably they'll rock sideways.

I think you are trying to make a 'perfect' machine, stop being so bloody stupid. 'Perfect is the mortal enemy of good enough' Don't try to build a perfect machine,
you'll never actually succeed but worse it will delay if not stop (due to cost) you from building a machine that is good enough. Been there, done that.

I had my axis beds cast so I did not machine the gutter exactly......but the 'gutter' is 70mm and the thickness of the base is 50mm. They are fine.

You are correct a solid undisturbed piece of 100mm thick steel is as stiff as hell and machining a gutter into it causes to flex more.......but you need that gutter, end of story, so machine it in there.
What's more I think you should do it yourself. The Bridgeport clone on which I made my first mini-mill was a little undersized so I had to get creative and do multiple setups on the larger
parts to do it all. It was a PITA, but it worked OK and I learned heaps. If I had to pay someone to do it I could never have afforded it and that little machine would never have been built. Was it
perfect?....no bloody way was it perfect, and yet I used it for eight years.

At the start of this thread you said that you had a Russian mill and one of your goals in this project was to use it to make your own machine and learn about milling. That is a laudable goal,
don't give up on it. Unless you are some sort of 'mechanical genius' it is highly probable that doing it yourself will result in inaccuracies and other quirks.....but how do you expect to learn otherwise?.
Prior to me using the Bridgeport clone to make my mini-mill I had perhaps 100 hours experience milling, and most of that on various drill-mills. Once I had finished my mini-mill I had something
like 600hours experience milling.

Stand back and take a survey of the threads on this forum. How many hundreds of threads where someone is trying to design the perfect machine....and then realize how few of them
actually end up with ANY useable hardware. I would guess one in ten or fewer makes it from the drawing board to actuality. It comes down to 'I can't make a machine as good as I really wanted....so
I didn't build one at all'. Don't be one of those nine out of ten.

Just as a matter of interest you might try asking contributors to your thread to post a pic of the machine they have built, it might surprise you to find that few have, and those who have, how
imperfect their machine is. Here's the rub though, a machine that works, no matter how imperfect is still better than no machine at all.

I think you should finalize your design before buying more random parts and trying to make them fit

That more rubbish. The start of this thread was about the difference between buying new or buying second hand. The problem is new C5 screws are just too expensive.

The problem with second hand is you have to take what you can get.

The quality of the 32mm C5 screws Lars has actually got would cost $5000USD or more if he were to buy new which would have stopped the project right there.
So Lars bought second hand for a small fraction of new....the downside is that the sizes are fixed.....thats life. Lars has spent $2000 but has parts worth $7000....
that is a good step in the right direction. That makes the project he has in mind affordable whereas new parts would stop it cold.

The goal now shifts to making the best use of the parts he has, and not falling into the trap of 'perfection' but realizing that building a machine requires compromises
and that no machine is ever perfect, at yet a machine that is 'good enough' still puts him in the one out of ten category.

Craig

[jackjr-123]

That more rubbish. The start of this thread was about the difference between buying new or buying second hand. The problem is new C5 screws are just too expensive.

I cannot disagree more. I've ordered a bunch of ballscrews over the years. All C5 or C3 ground. The last one I bought a couple of weeks ago was a TBI 1000mm 2005 C5 ground with custom end machining for $250.
You don't have to order a ballscrew from your local THK or Steinmeyer dealer quoting you a ridiculous price for a one-off.

I would not even consider buying a ballscrew second-hand. Between the right length, diameter, pitch, ends machining, preload, .... you'll simply never find the perfect match.

[joeavaerage]

Hi

TBI are not ground, they are precision rolled and supposedly are to JIS C5 standard, but they are not ground....good for the price but they are not ground.

I would not even consider buying a ballscrew second-hand. Between the right length, diameter, pitch, ends machining, preload, .... you'll simply never find the perfect match.

You are correct, its near impossible to get a perfect match but you can get close. I got three 32mm C5 BNFN double nut THK screws and six THK FK25 fixed end mounts for $1000USD including
shipping to New Zealand. Perfect, not quite, but pretty damned good. Would have cost $10,000 new.

You might not consider second hand because you cannot get the perfect dimensions but then you have to take the quality hit to get parts you can afford which are dimensionally perfect. Your choice.

I chose otherwise. Lars has chosen similarly.

Craig

[jackjr-123]

TBI are not ground, they are precision rolled and supposedly are to JIS C5 standard, but they are not ground....good for the price but they are not ground.

Sorry but you have no idea what you are talking about.
I still can tell what is ground and what is not.


Also absolute pitch accuracy is not a big deal. The ballscrew can be mapped. The main advantage of ground ballscrews for machine tools is the consistent preload along its length.

[joeavaerage]

Hi Jackjr,


BS, TBI roll their ballscrews, always have.

Also pitch accuracy is not a big deal.

????? The real killer is cyclic variation, also called 2PI error. That is near impossible to map and short of ultra high bandwidth servos and linear scales impossible to correct. That TBI can roll screws
and achieve C5 is good, very good even, if you believe it.

Craig

[jackjr-123]

BS, TBI roll their ballscrews, always have.

Seems like you know better than TBI themselves, so no point in arguing anymore.

I meant absolute pitch, I edited my previous message. The pitch consistency is what you get with ground screws, and indeed what is important.

[jackjr-123]

Also, about to ordrer some linear guides/cars. Got a great offer from our favourite shop in Korea. What you think of this Brand?

I might be a bit late, but avoid if you can. Those IKO LM guides are old designs with 2 rows bearings. State of the art bearings are 4 rows in O arrangement for much better stiffness.
Even Hiwin HGR are better than these.

Also do you know what the preload is?

[pippin88]

Craig you are incorrect

TBI make ground ball screws, as well as rolled (like every manufacturer). Loon at their catalog.

I have three ground TBI ball screws in my shed.
The difference between a ground and rolled screw is very obvious. There can be no doubt.

[joeavaerage]

Hi,
like Bosch Rexroth the screws are rolled and then finish ground. TBI claim C5 and C3 grades, Bosch Rexroth claim close to C5...but not quite.
The argument has always been that rolled then ground is cheaper....although Bosch Rexroth conveniently forgot that.

If you believe that TBI can manufacture and sell a ground ballscrew for $250 when all the other leading manufacturers sell them for $1500 and more....then go for your life. I call BS.

I made my choiceas has Lars. You make your own.

Craig

[pippin88]

Craig there are no physical signs of rolling. They do not have the little grooves in the crest (major diameter) that rolled screws do.

Are they the same standard as a premium brand name? Probably not. That does not mean they are not ground ball screws.

[Fatamorgana]

Hello, and thanks for all of your replies.

Regarding to the question about the rails/cars from IKO I got a good offer for all of the axis that would fit my project rather good, they have been producing this type of hardware since 1950 something, so if its hardware from 2012 (wich mine is) or 2024 will not affect my project/machine very much. As Craig says the most important thing is that this is evolving against a finished product and not a stranded failure because of the costs.

Craig, thanks again for reminding me my initial goals, at one stage I just wanted to pay myself out of it, but what would the learning effect of that be, very limitied. Either way Im having some issues with my setup, my mill doesn`t got the travels I need for not having to do multiple settings. Im short of 300mm in x-direction. I have done some illustrations in fusion to show you my issues, I have come to the conclusion that I need to do it on my horisontal mill (got horisontal/vertifal mill combinated). And I also need to do it in to settings. Would greatly appreciate if someone could give me some hints in how the multiple settings would be done in the most accurate way.

[peteeng]

Hi Fatama - Usually on a job like that the party is set on machine jacks or spacers and shims. If you clamp down onto the bed like that you will bend the part slightly or bend the machine bed and this bend will show up as an inverse error in the part. I would machine small pockets into the part and use toe clamps or small clamps at the bottom vs using long bolts like you have drawn. I'm not a machinist and hopefully machinists will comment further. Not sure how the granite helps. You will carefully machine a reference side then indicate from that side for the other operations... The resulting squareness and flatness of the part will be up to the machines accuracy... Its very important to set up the part as a zero stressed set up. ie once set up using say 4 or more clamps on stands if you release a clamp the part should not move. Then do same at next clamp. If the part moves then it needs to be shimmed or if a jack adjusted until it does not move to ensure the part or the mill is not flexed in the set up. Peter

[joeavaerage]

Hi Lars,
the skill involved in machining parts in a machine which is marginally too small is quite a challenge, and yet it is one that every machinist must face. There are idea and techniques
that you will apply again and again whether it be a manual mill or a CNC mill. It is a learning opportunity not to be missed.

Peter is correct: if you clamp a part to the bed of a mill you will pull a warp/twist/bow out of the part to conform to the bed or alternately if the part is thick enough it will pull the bed to it.
Either way when the part is released it will spring back to its original twist/warp/bow. The trick is to try to get one surface as flat as you possibly can and thereafter that will become your reference surface.


You cannot afford to be too 'anal' about this, otherwise the task will exceed your skill, your machines capabilities and stop your progress. I would suggest that unless you know otherwise treat the bed of the mill
as perfectly flat. It probably was perfectly flat or nearly so when it was built and is probably still bloody good today. You've got to start somewhere.

When I machined the cast iron elevator weights for my first mini-mill I machined the bottom surface flat, or as flat as I could. I then drilled and tapped four holes down the center. To this I bolted a rail
which was a snug fit in the center T slot of the bed. That would mean that I could slide the part back and forth in the X direction (in the mill) and be assured that it was straight. This relies on you having
put the rail down straight and that the center T slot is straight, both of which require care and thought.....but thereafter removing the part from the mill and turning it end-for-end for example is no trouble.
The idea here is that you use a feature, in this case a rail, to provide repeatability easily.

I, also as Peter suggested, milled approx 1/2 inch pockets in three locations along each of the two vertical sides. This allowed me to use toe clamps to hold the part down without the top surface being obstructed.
Remember by this stage I had already machined the bottom surface 'flat' as I could, and I also assumed the mill bed is 'flat' so clamping the part at two or three locations along its length directly down to the bed
is permissible. It might well be argued that this is poor practice and an experienced machinist would use jacks/shims to ensure that no untoward bow be pulled into the part. The problem is that neither you or
I are experienced machinists, and so we make certain compromises to make the process achievable.

Many years ago I worked in a shop and we had a rather large but old English made lathe. It was pretty worn but I could make parts to 1 thou, maybe 2 thou no trouble, but beyond that the machine wear defeated me.
When we had a boat prop shaft taper to do the Boss would call in Bob, a retired fitter and turner of decades of experience. He would turn a taper to near perfection. It was always my job to lap the prop onto the taper
for perfect fit and I never had to spend more than half an hour doing it.....the taper was damn near perfect straight out of the lathe. Bob would spend all day doing it mind, consulting a little notepad he kept on this machine.
I asked him many times how he did it, but he would just smile and tap his nose....as if to say 'that's for you to find out'. Regrettably he passed away before I ever got but a small fraction of what he knew. Bloody shame really,
he was what I call a 'Master Craftsman', never let such a one out of your sight, you'll learn more from them in an afternoon than you'd learn in a month at trade school.

The point is that a Master Craftsman will produce parts which are dimensionally better than the machine that produced them. You and I cannot do that....yet. We have to rely on the integrity of the machine because
of a lack of experience. This is just a fact of life. If this idea does not appeal then you will have to pay someone else to do it, and presumably a larger and better machine than you have. If you are committed to
doing it yourself just be aware that there will be imperfections both as a result of your experience and/or machine.

My first mini-mill had lots of such imperfections, but the sense of satisfaction gained from doing it myself was and is still huge. Not only that I learned heaps, and those lessons stay with you. It is an experience not
to be missed. The saying is 'we learn by our mistakes' is as true with machining as anything ele.

Craig

[peteeng]

Hi Lars - I shall also add that you should rough the part out to say 1-2mm from final on all features as a test run for your plan. Then mistakes can be corrected. Plus this will allow the part to relieve itself if it has any stress in it. Once roughed leave it free for a week and then start your final machining phase with the knowledge gained. Metal does move and when I was involved in making large moulds (for plastic parts) I was surprised at how much a piece of metal can move once you rough it and cut gutters in it etc.. Peter

[joeavaerage]

Hi,
there is one complication about using a central rail on the underside of your axis bed. Most mills have a gutter surrounding the outer ends of the T slots to catch coolant.
That would foul with the rail, unless the rail itself were shortened, recalling the whole idea is to be able to slide the part back and forth so that the limited machine travel
can be used to make larger cuts than the travel would allow. The complication is to take to that part of the casting (of the mill) with an angle grinder and 'extend' the profile
of the T slot to accommodate the rail. The mill I have pictured is not mine, and it is in near new condition, and I would not presume to seek permission to treat it that way, but if its
your own mill....then???. The Bridgeport clone I used already had that relief ground into it....and indeed is what prompted me to the idea of having a rail bolted to the part for 'registration'
in the first place.

Craig

[joeavaerage]

Hi,
something you may want to consider is having one face of your material surface ground. I have little doubt you will find someone with a big enough surface grinder. They are pretty cheap to
operate, pretty much set it up and leave it to run, and often is pretty cheap to do as a result.

Any tool and die maker will know instantly what you want to achieve is a flat reference surface, and set it up appropriately.

One little trick is to make sure you remove any mill scale. A stiff wire brush in an angle grinder is a good solution. Mill scale is abrasive and tool and die makers are fussy about such things.

Craig

[peteeng]

Hi - I'd rough it first then grind. It may move once roughed. Peter

[joeavaerage]

Hi,
yes it might move....but you are getting too anal about it. If you were doing it 'properly' without regard to expense you'd stress relieve....then machine.

Hot rolled steel is pretty damned good, it does not have huge amount of stress left in it.

I'd say slap it in a surface grinder just to true up the lower face then mount your alignment rail. peteeng is correct it may move, in which case you have to go back and re-evaluate, but
more likely it will not move or do so on such a small scale that you can ignore it. Remember we are hobbyists, if you are going to pursue micron or sub-micron precision then I assume also you
are a millionaire.....Time to get real and do things which are achievable.

Craig