Re: [CAD_CAM_EDM_DRO] Hi :) - making parts
Posted by
CNC 6-axis Designs
on 2010-11-26 01:29:43 UTC
You got good advice from Jon and David, two well respected members.
IMO, the most important Q you need to define is what is "well" for you ?
It is very possible to make small numbers of parts, slowly, on any
chinese mini-mill or lathe (cnc OR manual).
Here, we talk about a few units / day.
If your item needs more than a few, your needs are different.
It the parts need to be precise, again your needs are different.
Precise might be optical parts, parts that need precision bearings fit
into them, or parts that will go into or over someone elses parts, such
that they have a specified tolerance.
Then there are the vast majority of "just parts", say adaptors for
micrometers, cameras etc, which might need say an unusual thread size,
but the thread only needs to work well, rather than be a qualified
thread which has a tolerance for error and an acceptable range.
In general, making parts, if these are made in any quantity, will
require a significant investment.
Something on the order of 5-15.000 Us.
It does not matter weather the parts are steel or alu, what make of
machine you use, wheather the machine is old us industrial, new chinese,
or whatever.
Any and all of these can and will work, but there are a lot of variables
affecting the resulting speed, quality, look, cost and practicality of
what you want to do.
As an example, I made a tooling plate, in steel, on a little
high-precision 7x cnc lathe.
100 mm diameter, about 3 cm thick.
Less than 0.01 mm error from nominal, ie good commercial quality.
Took 3 hours.
The same plate, on my 12x lathe, takes 1/2 hour.
If I did "plates", it would take me about 2 days to make a set of jigs,
fixtures and setups, and I could produce plates at maybe 6 / hour.
If the plates needed to be better than about 0.01 mm, I might need to
finish them on a grinder, and the cost and time of each would be about
double.
If one of the surfaces needed to be near an optical flat, the cost of
each would about triple from above, ie 6 times basic or more.
CNC takes out the manual work from using the machine, but does not in
and of itself, solve "making parts".
The question you need to solve, is how many parts you will make, what is
the required quality and accuracy in each part, and then look at what
you need to do to get these parts made well and effectively.
If the parts are made in any quantity, you will probably require some
industrial components.
These components are often shockingly expensive.
They are also typically about 10x more productive than any other option,
and making any significant quantity or high quality of parts requires
some of these industrial components.
And example might be a go/no go gage.
Think of 2 precision ground components, that allow you to test, and
re-test if necessary, in a few seconds, whether the part is upto spec.
They gage might cost 1000$ (each) (or more).
However, it is often the best, easiest, cheapest way to measure a high
quality component.
A 100-500$ digital micrometer will maybe work, but may take 2 minutes
vs10 seconds to test, and great care must be taken by the operator, who
needs some experience.
Any one of us here, with some n.000EUR in existing tooling, will have
micrometers, cables, connectors, racks, wiring, tooling plates, jigs,
indicators, stands, DTIs, angle plates, vices, micrometers etc.
You need these, over time, and the better you want to make your parts,
the better you must know and be able to get the right quality/cost of
each tool.
This ancillary tooling and stuff will far exceed the cost of any machine
you buy upto the industrial lathe size.
IMO, the most important Q you need to define is what is "well" for you ?
It is very possible to make small numbers of parts, slowly, on any
chinese mini-mill or lathe (cnc OR manual).
Here, we talk about a few units / day.
If your item needs more than a few, your needs are different.
It the parts need to be precise, again your needs are different.
Precise might be optical parts, parts that need precision bearings fit
into them, or parts that will go into or over someone elses parts, such
that they have a specified tolerance.
Then there are the vast majority of "just parts", say adaptors for
micrometers, cameras etc, which might need say an unusual thread size,
but the thread only needs to work well, rather than be a qualified
thread which has a tolerance for error and an acceptable range.
In general, making parts, if these are made in any quantity, will
require a significant investment.
Something on the order of 5-15.000 Us.
It does not matter weather the parts are steel or alu, what make of
machine you use, wheather the machine is old us industrial, new chinese,
or whatever.
Any and all of these can and will work, but there are a lot of variables
affecting the resulting speed, quality, look, cost and practicality of
what you want to do.
As an example, I made a tooling plate, in steel, on a little
high-precision 7x cnc lathe.
100 mm diameter, about 3 cm thick.
Less than 0.01 mm error from nominal, ie good commercial quality.
Took 3 hours.
The same plate, on my 12x lathe, takes 1/2 hour.
If I did "plates", it would take me about 2 days to make a set of jigs,
fixtures and setups, and I could produce plates at maybe 6 / hour.
If the plates needed to be better than about 0.01 mm, I might need to
finish them on a grinder, and the cost and time of each would be about
double.
If one of the surfaces needed to be near an optical flat, the cost of
each would about triple from above, ie 6 times basic or more.
CNC takes out the manual work from using the machine, but does not in
and of itself, solve "making parts".
The question you need to solve, is how many parts you will make, what is
the required quality and accuracy in each part, and then look at what
you need to do to get these parts made well and effectively.
If the parts are made in any quantity, you will probably require some
industrial components.
These components are often shockingly expensive.
They are also typically about 10x more productive than any other option,
and making any significant quantity or high quality of parts requires
some of these industrial components.
And example might be a go/no go gage.
Think of 2 precision ground components, that allow you to test, and
re-test if necessary, in a few seconds, whether the part is upto spec.
They gage might cost 1000$ (each) (or more).
However, it is often the best, easiest, cheapest way to measure a high
quality component.
A 100-500$ digital micrometer will maybe work, but may take 2 minutes
vs10 seconds to test, and great care must be taken by the operator, who
needs some experience.
Any one of us here, with some n.000EUR in existing tooling, will have
micrometers, cables, connectors, racks, wiring, tooling plates, jigs,
indicators, stands, DTIs, angle plates, vices, micrometers etc.
You need these, over time, and the better you want to make your parts,
the better you must know and be able to get the right quality/cost of
each tool.
This ancillary tooling and stuff will far exceed the cost of any machine
you buy upto the industrial lathe size.
> Total newb here...I'm about to embark on making small parts and I'm[Non-text portions of this message have been removed]
> trying to figure out the least expensive way to do it...but do it well.
>
> Most parts will be no larger than 2"x3"x1", made of hardened aluminium,
> so I guess a basic mini-CNC-robot or CNC-mill will be the way to go.
>
> The parts will be flat on two sides...many with a cut 3/8" groove on
> the bottom side.
>
> So what should I start with?
>
> What pitfalls should I avoid?
>
> Should I take a college course in Cad/Cam/Machining?
>
>
Discussion Thread
bigshrek2005
2010-11-25 11:20:03 UTC
Hi :)
Jon Elson
2010-11-25 12:11:10 UTC
Re: [CAD_CAM_EDM_DRO] Hi :)
David G. LeVine
2010-11-25 23:22:10 UTC
Re: [CAD_CAM_EDM_DRO] Hi :)
CNC 6-axis Designs
2010-11-26 01:29:43 UTC
Re: [CAD_CAM_EDM_DRO] Hi :) - making parts
Stephen
2010-11-26 05:50:50 UTC
Re: Hi :) - making parts
ED MAISEY
2010-11-26 11:23:28 UTC
Re: [CAD_CAM_EDM_DRO] Hi :) - making parts
bigshrek2005
2010-11-30 01:03:38 UTC
Re: Hi :)
Jon Elson
2010-11-30 09:12:44 UTC
Re: [CAD_CAM_EDM_DRO] Re: Hi :)