[RobbyA]

Hi All,

I am inspired to build my own bench top cnc mill, but I am concerned it may not be a realistic proposition. I want I high precision machine with the capabilities of cutting steel if need be. I already own a HF45 clone (Hafco). I considered initially to convert this machine but I'm put off by the dovetail slides, variable build quality, stability etc.
I mainly want to machine aluminium, but the odd bit of steel. My desire is to make go kart type parts like bearing blocks, spindle parts etc.
Is it more realistic to just buy a Minitech 3/4 (or equivalent) platform and add parts as I can afford them. I have a friend who is on this forum that can assist with all the control systems. I don't feel like dropping $3k + on ground plate, rail, ball screws etc only not to realistically achieve better than 0.001" tolerances. So, can it realistically be done or am I better off just to save up and buy something?

Many thanks in advance.
Rob

[MichaelHenry]

only not to realistically achieve better than 0.001" tolerances

How much better than 0.001" do you need to go and on what sort of parts or features?

For example, on a bearing block, I would think that only the bored pocket for the bearing would benefit from that sort of accuracy and the rest of the part could be made to much looser tolerances.

[BAMCNC.COM]

I agree, get a boring bar and a good dial test indicator. Spend money on good tooling and fixturing, learn how to adjust your mill, or modify it to be stronger.

[SCzEngrgGroup]

There have been countless threads on here asking exactly the same question. I'm not going to re-start those never-ending discussions, but you would be well advised to find those threads, and understand the MANY trade-offs that MUST be made, and understand how difficult it is to achieve a true 0.001" accuracy with these machines. Suffice it to say, there is a reason why commercial machines with verifiable specs in that range are FAR more expensive than hobby machines.

Regards,
Ray L.

[RobbyA]

Thank you guys for your wonderful response. I'm still trawling through ALL the wonderful threads regarding buying vs building.
Michael, you are exactly right in that generally it is only the bearing surfaces that nead the finer precision. Reflecting on this, a lesser machine with some fetteling might suffice. Part of my reasoning for a more precise machine is I would prefer not to do secondary operations, say place the bearing carrier in a 4 jaw setup lathe. I have read since that any machine is only as good as it's operators setup. I have underestimated the importance of this.
To achive a 0.001" interferance bearing fit for example, is my life going to be much simpler using a precision mini mill/bench top machine.
Another part I would like to create is ~6" dia detailed aluminium wheel centers (mini sprint). Is this a job for a 4th axis or is it just as practical to do it on the 3 axis?
Sorry for all the newbie questions. I don't have much cash to splash, so I hope I can make the best informed decision about future purchases.
Is a larger Minitech type machine suitable for my application?

Many thanks in advance
Rob

[hndswthtshdws]

Are you going to mass produce these parts or are these for you (and your team)?
Unless you are mass producing parts, just use a boring head for your critical hole dimensions. 10 or 15 thou is probably a fine tolerance for all your other dimensions. 5 thou is achievable repeatably on a well converted machine.

Could you post a pic of the mini sprint wheel centers?

[Fuelrush]

Not to high jack your thread and it may help you too but when we say a benchtop machine could hold .005" or any other dimension (I've seen this in many other post too.) What kind of geometry are we talking about? Possibly a .005" linear error in all axis? Over what kind of distances? I've seen some people say over a foot, some over inches. Is that compounded as you move from say a series of holes? Thanks.

[wizard]

Any of the inexpensive Chinese mills will require hand work or rebuilding to get them up to par. Why not work to get the one you have in acceptable condition?

Thank you guys for your wonderful response. I'm still trawling through ALL the wonderful threads regarding buying vs building.
Michael, you are exactly right in that generally it is only the bearing surfaces that nead the finer precision. Reflecting on this, a lesser machine with some fetteling might suffice. Part of my reasoning for a more precise machine is I would prefer not to do secondary operations, say place the bearing carrier in a 4 jaw setup lathe.

There might be a language barrier problem here but a CNC machine does not automatically mean you have eliminated secondary operations.

I have read since that any machine is only as good as it's operators setup. I have underestimated the importance of this.
To achive a 0.001" interferance bearing fit for example, is my life going to be much simpler using a precision mini mill/bench top machine.

What precision bench top machine? I mean really there is a whole range of precision desktop machines that will only do well within a certain range or capacity. Both TAIG and Sherline make what are considered precision machines but are worthless for boring an automobile engine.

Another part I would like to create is ~6" dia detailed aluminium wheel centers (mini sprint). Is this a job for a 4th axis or is it just as practical to do it on the 3 axis?

I have no idea as I don't even no what you are talking about.

Sorry for all the newbie questions. I don't have much cash to splash, so I hope I can make the best informed decision about future purchases.

Convert the mill you have! Tune it up for proper operation

Is a larger Minitech type machine suitable for my application?

I seriously doubt it. Realize though that I can't grasp what you are after. As I see it your shortest route to a CNC machine would be a conversion of the one you already have. It is likely you lowest cost option.

Many thanks in advance
Rob

[underthetire]

Accuracy is the eye of the beholder. . .001 over 12" or even 20" is a sloppy mess in my world. When we start talking sub micron levels wake me up .

My point is, you need to determine what you need.

[RobbyA]

Thank you Hndswthtshdws, wonderful insight into repeatability of converted machines. I'm only looking at doing small low volume runs for myself. I'm trying to work out a way of posting a picture from my iPhone as I am on site at the minute. Just for a reference the wheel center I want to make is similar to the Weld Wheels 4 pin. In hind sight it may have on OD of 8" (rim is to suit 10" tires).
I would love to hear more about repeatability and tolerances that people are achieving with there own machines.

[Fuelrush]

Accuracy is the eye of the beholder. . .001 over 12" or even 20" is a sloppy mess in my world. When we start talking sub micron levels wake me up .

My point is, you need to determine what you need.

LOL, that must be some fancy stuff you make. I work on helicopter parts and I guess most things we make are a sloppy mess. :stickpoke:

[RobbyA]

Sorry, I should clarify what I mean by secondary (or even primary) operations. I want to know if I could successfully make something round, like a really small car wheel, with and/or without using a 4th axis to rotate the wheel as it is machined. My point being am I just dreaming contemplating that a nice bench top machine (Tormak, Benchmen Minimill etc) could make something as round as say a $4k Taiwanese manual lathe? Is it more realistic to have a blank turned up on a cnc lathe and pocket out the spokes myself, or would the nice bench mill get acceptable results?
I apologise for any confusion.

[wizard]

Sorry, I should clarify what I mean by secondary (or even primary) operations. I want to know if I could successfully make something round, like a really small car wheel, with and/or without using a 4th axis to rotate the wheel as it is machined. My point being am I just dreaming contemplating that a nice bench top machine (Tormak, Benchmen Minimill etc) could make something as round as say a $4k Taiwanese manual lathe? Is it more realistic to have a blank turned up on a cnc lathe and pocket out the spokes myself, or would the nice bench mill get acceptable results?
I apologise for any confusion.

First let me say Tormach has a nice little Mill, I've seen them in action. The problem with your question is that it depends on the wheel. In the end it would probably be easier to split the work between a lathe and mill. If you search YouTube you can find a few videos of full size wheels being made in a few different ways. So even in industry there are multiple approaches. In your case I'd probably do the mill work first and then the lathe work.

As to the Tormach it isn't exactly a bench top machine. That is probably a good thing because most likely you would want a fairly large machine to try to do everything on the mill. I'm thinking Y axis travel would be a significant limitation for six inch or so wheels. I'm thinking ten or more inches of Y travel to even try to mill the whole wheel.

Speaking of wheels, I never heard them called centers. Centers are something that you rotate a part between. It is that common language thing going again.

[RobbyA]

Thanks Wizard. A machine with 9-10" on the y would be a must. The term wheel center (or "centre" in the Queen's English) is often used by American to describe the inner part. They call the the whole assembly as 3 piece wheels, as you use 2 spun aluminum ("aluminium" in my part of the world!) rim halfs to complete the assembly.
So to machine center bores and OD's a mill is probably not the best, even on a rotating axis? Could I expect 0.003" or better repeatability in a Tormak or Syil in the real world?

[MichaelHenry]

So to machine center bores and OD's a mill is probably not the best, even on a rotating axis? Could I expect 0.003" or better repeatability in a Tormak or Syil in the real world?

Using a rotary table with an end mill to mill the periphery of a wheel into a circle is probably doable, but would be far faster on a lathe.

You could profile the periphery of the wheel with an accuracy of 0.003" or better with suitable care and a mill that is rigid enough and has enough X/Y travel. Little things can make a difference here, though, such as proper care to setup and the way the tool is mounted and measured. For example, a 0.001" runout on your cutter would give you a 0.002" error in your wheel OD.

You might be surprised how much slop can be tolerated in many parts.

By the way, it's "Tormach", not "Tormak", just in case you are Googling the company.

Mike

[Fuelrush]

You might be surprised how much slop can be tolerated in many parts.

This is good information here. Just like many things can and will be over engineered. You can spend a lot of time trying to over machine some dimension on a part that won't matter in the least. On a wheel your lathe work will be more important than the mill work especially if its a lug-less wheel. Your spokes won't mater if they are .005 to wide etc. If your just doing the center of the wheel and not the rim/hoop I would bore the blank and mount it on a matching mandrel and turn the outside and face it/make the hub. Then mount it in the mill, indicate the center, and then CNC the design of the spokes. Not sure if thats exactly what your going for but it probably will be a variation of that I would think.

[RobbyA]

Fuelrush, Wizard and Michael, your wisdom is very much appreciated. I'm learning fast! Thank you to everyone for their responses. The systematic processes that were recommended seem very practical, realistic and down right clever. I now have a lot of confidence that I can achieve what I want to achieve in a realistic manner. I picked up pretty quickly that I misspelt 'Tormach'! I have requested a quote from my local distributor for a PCNC 1100. The physical size of the machine and cost would probably be my only setbacks.

[bebob1]

You mentioned in the OP that you dont like dovetails slides? Just for your current mill or dovetails in general? Since the tormach also uses dovetails FYI.

[RobbyA]

My friend who has built several cnc machines recommended linear guides for cnc applications. His recommendation was that a dovetail machine like my Chinese made Hafco 46 would be a pain in the backside to keep the lash under control whilst not imposing a lot of load on the steppers/servos. His comments were made in relation to precision machining steel on a regular basis. Would it be safe to assume that companies like Tormach, Syil (I noticed Syil had dovetails on the y&x) etc would have superior dovetails to my Chinese machine?

[bebob1]

Yea hes right, dovetails are normally considered not as ideal as linear guides. If this is important to you then the tormach may not be a good fit since they also use dovetails. But realistically, its normally about good machining skills to create good parts anyway. If you are skillful you can get away with a less ideal machine. Likewise you can have the best machine and not know how to use it to its potential and still do worse. For all practical purposes the tormach should be pretty good for most folks.

Im not sure if any company can be said to be better than any other company unless a proper scientific study was done to compare results and show all the statistical variations. Im not sure such data exists, but there is a general belief that products from certain countries are higher quality than products from other countries.

Regardless, its probably best to state your maximum workspace you require for your mill (wts the maximum part size you will ever intend to make), then state your budget and people can offer you other options.