coordinate measurement machine for hobby usage

[PeterKing]

I'm interested to make some parts for radio control model car. I would draw and improvise my idea improvement on the parts then pass on machine shop for production . Drawing a part is not difficult for me but making sure the right dimension or angle is a challenge. Most of the parts that I need to be make must refer to the original part precise dimension. So, I'm looking for a measuring tool that can help to gather the right coordinate. Want to know is there a low cost CMM that will fit my need. The biggest part size I need to measure is 12x12inch. Thanks a lot

 

[dgowans]

NextEngine 3D scanner, perhaps?

 

[MiketheEngineer]

I am sure an e-bay search for micrometers and calipers would give you what you need. These are not cheap - expect to pay a bit for them

Good Luck

 

[KENAT]

Not sure you'll need your own CMM, and they are very expensive and I believe you'd need quite a bit of training to use properly.

A 3rd party measurement house may be a better option.

http://zcorp.com/documents/112_ZScanner-singsheet-v08.pdf

(large file takes a minute)

Or conventional methods as Mike suggests.

KENAT, probably the least qualified checker you'll ever meet...

 

[PeterKing]

Mike, Yes. I already have the micrometer and caliper. The problem is I do not know how to take the measure of an angle using those tools. Please see the attachment for example. How do I measure the angle of the slope? If it possible?

I believe the following CMM will work for my need:

NextEngine 3D scanner

https://www.nextengine.com/indexSecure.htm

ZScanner

http://zcorp.com/documents/112_ZScanner-singsheet-v08.pdf

Faro Gage

http://www.faro.com/content.aspx?ct=di&content=pro&item=4

Those 3D scanner is easier to use but less accuracy. But I think 0.005 accuracy of 3D scanner is good enough for my application but of course the higher accuracy is better.

My question is which way should I go for a quick jump start? Learn to use conventional methods? 3D scanner? Or Faro Gage arm? Which method is best suit for my application? Which one is the easiest to learn? What which one is the most cost efficient? What is your recommendation for me? Thanks a lot!

 

[PeterKing]

this is my part

 

[dgowans]

If you can measure the diameters at two different known elevations on your slope it is a trivial exercise to find the angle, assuming that the slope is constant.

If the parts you need to reverse engineer are similar to what you've shown you won't need too much in the way of high-end measuring equipment. I'd consider adding a good height gauge to what you currently have.

 

[ajack1]

How accurate does the angle have to be? Here are some cheap ideas.

Not very- use a protractor.
Fairly- measure across the intersection points both ways using a vernier, simply trigonometry from there.
Very- Put the part on a sine bar (basically a flat plate with two dowels at known centres) and put slip blocks under one end, increase or decrease until flat with a DTI, again simple trigonometry is needed.

If you do not have the gear to do this any machine shop/ toolroom would do this probably for the price of a pint as against thousands of pounds for a 3D scanner or tens of thousands for a Faro arm.

Even if you get the info from a 3D scanner it will only be a point cloud or very crude surfaces, you will still need a 3D modelling system to get anything usable out of it.

 

[tygerdawg]

Why buy the equipment and measure yourself? Send the part to a good machine shop with their own equipment, pay them an hourly fee to generate the data for you. They would (presumably) have people qualified to render good measurements. It's gotta be cheaper than buying your own equipment.

TygerDawg
Blue Technik LLC
Virtuoso Robotics Engineering

http://www.bluetechnik.com

 

[PeterKing]

dgowans: Thank you for your reply. yea, that is the problem. I have no clue no how to measure the upper elevation.

ajack: The measurement must be very accurate cuz I will pass my drawing for production. The reasons why I want to buy a CMM is:

1. Once I got the CMM I will work with machine shop to produce more parts. So, keep on doing this for fun and for some business. Actually, I will hire a CAD drafter on doing this but I need to know how does the process work in advance. That's why I took classes for the solidworks. I don't need to be advance on the skill but I want to understand the concept first in order to work with machine shop, CAD drafter more smoothly.

2. It is my hobby, I see this as one of my toy.

Thank you tygerdawg and ajack1 for the suggestion that I can pass the part to machine shop for the measurement then return me the drawing. This is very good suggestion. I think I can go to the easy way for starting. But again, I think it is fun to know on how to measure the part by myself.

Thank you everybody.

 

[btrueblood]

Put it on a small optical comparator, and measure the angle directly from the angle gage on the comparator screen (typical accuracy well below 1 degree). Most good machine shops have one of those lying around in the tool room or inspection area.

 

[KENAT]

Like I said "A 3rd party measurement house may be a better option. " this may be a machine shop or a place that specializes in metrology.

The Zscanner link should have been on another line.

Let's play this the other way, how much are you willing to spend?

KENAT, probably the least qualified checker you'll ever meet...

 

[PeterKing]

Kenat: I'm willing to spend around $5000 on getting the machine. You think zscanner is the way to go?

 

[dgowans]

I saw a demo of the ZCorp scanner. Impressive, but you won't touch one for 5K. The NextEngine is in your price range, although I can't comment on how well it works. I'd bet you can finesse it its output enough to be able to generate a decent model.

 

[MikeHalloran]

We bought a "low cost" CMM a few years ago. It cost $10,000, which is very, very little as CMMs go. I insisted on the "high accuracy model", supposedly good for +/-.010".

You know what? After you power the machine up and calibrate it, it ... just sits there and does nothing. You have to touch the probe to points of interest and press a button once for each point. Which doesn't sound very taxing, but you have to touch _exactly_ the right points to get meaningful data about a regular feature like a circular boss, and you have to touch a _lot_ of points to capture an arbitrary surface. The worst part is trying to digitize an edge, or any male feature, by touching along it with a pointed male probe... which keeps slipping off. WHile at the same time supporting the floppy arm and moving the probe in such a way that you don't bang into a travel stop or disturb the workpiece location. Speaking of which, if you get the workpiece in the wrong part of the arm's envelope, you can't touch all the parts in one setup... and good luck correlating data from two sessions with the arm and part in different relative positions.

Using a CMM is tedious, mind- numbing work, that no one will do for more than an hour without protesting or finding something else to do or demanding a raise.

You'd be better off to buy a nice surface plate, height gage, angle gages, gage pins, etc.,, and do your measurements in more traditional ways, and you'd still have some beer money left over.



Mike Halloran
Pembroke Pines, FL, USA

 

[ajack1]

Peter the Zscanner will not do what you want.

We have had demos on both this and the next engine scanner as one of the services we offer is reverse engineering parts for historic and vintage cars and we were looking for an affordable (to me, small company owner) way of measuring various parts.

Both are very good, for the price, at getting some kind of data in on shaped surfaces; imagine the mouse in your hand or an exhaust manifold, what neither do very well is accurately measure holes or even surfaces and all laser scanners struggle with pockets. So if for example the taper on your part sits in a “valve seat” you would not scan in accurately enough. As I said in my previous post even with this data as a point cloud or a stl file there is still a good amount of modelling required to get workable data, although with next engine there is a direct read into solidworks, as you mentioned this earlier.

For complicated work Reinshaw do some fantastic stuff that incorporates both probing (high accuracy) and scanning (basic surfaces) that can be merged however you are looking at around £130,000 ($260,000), they also offer a service in the UK at £800/day ($1,600).

For the part you show any CNC with a probe could locate the centre move out a set amount and probe the tapered surface, move out a further amount and probe again and that is all the information you would require, probably less than 15 minutes work and I would guess any machine shop would do this for “free” if they had the manufacturing of the parts. Failing that again as I said earlier a simple set up on a sine bar would do the job, again sub 15 minutes work.

In the UK at least next engine is sold through solidworks resellers and the Zscanner through Zcorp resellers, we have had no problems getting demos and even the loan of a machine, this is the best way to evaluate anything IMO.