Thread: Magnify small parts

Would this help?

http://www.ebay.com.au/itm/USB-10X-5...item2ed6414370

Include something of a known size in the image and scale off that?

Originally Posted by neksmerj

I've just been asked to draw up some very small parts. They are brass plates approx 5mm wide x 12mm long. They have curved ends, cut-outs in the centre and protrusions out from each of the flat sides.

My problem is trying to measure them. What I need is a device that will greatly magnify them, say 10 times, and be able to take dimensions off the enlarged image.

Any clues?

Ken

I personally use one of these:

F+C.jpg
http://www.fc-optic.com/en/SCALE_MAGNIFIERS.htm

It magnifies 8 times, has a field of view of 25mm, one graduation on the graticule is 0.1mm wide (and is so clearly visible and sharp, that one can very confidently interpolate 2 times better than that, eg to 0.05mm). The thing is beautyfully finished (Swiss made). Mine is now nearly 35 years old, it was not cheap, but I would buy it again. The downside is that it only measures on a flat plane (it has a glass bottom). The things are originally used to measure fine maps and films.

Either that or I have one of these:
http://www.jaycar.com.au/productView.asp?ID=QM3539

I have one of these and I find the optics pretty good as it's a glass lens which surprised me. It also has some scales on the inside in both metric and imperial. They are OK but for fine work I find the markings are too thick. If I need some thing a little better then I lay a steel rule with 0.5mm markings as a backdrop.

I'm not sure as to what level of accuracy you need to measure, anything finer than about 0.5mm or so then you may need to look at other options. An optical comparator? Some people report to using an overhead projector, projecting the part onto the wall. I like that idea, sounds like a great idea.

Simon

CBA's looks good. I'd get one of them!

Simon

I have a few things to choose from

To give you an idea of what master splinter has linked to. I have one much the same.
It doesnt "zoom", though it does have two magnifications. It also comes with software to making measurements of length, angle and diameter.


Stuart

Is that your USB camera Stuart?

If so then I can vouch for it. I was impressed when you showed me.

Simon

Unless it is automated Or very exspensive I don't think you will find anything that will enlarge and then take dimensions.Then again if you find something that will give dimensions why would you want to enlarge it.
May be I missed something but your question was very straight forward

Originally Posted by pipeclay

Unless it is automated Or very exspensive I don't think you will find anything that will enlarge and then take dimensions.Then again if you find something that will give dimensions why would you want to enlarge it.
May be I missed something but your question was very straight forward

To take dimensions, you need to place something like a ruler directly onto the object.

To take very small dimensions, you obviously need a small and finely graduated ruler. Something like a ruler with with 1/10mm or 1/20mm subdivisions.

One Problem with such fine rulers is, that you cannot clearly read it with naked eyes. If you do not magnify the ruler, the fine subdivisions look more like a grey band to your eye. Some sort magnifying lens is needed, preferably some optics that do not distort the object you want to measure.

Another problem is, your ruler needs to be very close to the object being measured. The further away it is, the greater the resulting parallax error. In order to be very close to the object, the ruler needs to be extremely thin. in the case of the example magnifier I linked to, this is done by evaporating a metal scale onto a glass surface. A similar process like the anti reflective thin metal coating on a camera lens, just in the form of fine lines and numbers. Also, this scale is printed as a mirror image to the UNDERSIDE of the glass surface plate, thereby reducing the distance of the scale to the measured object to zero: the scale itself now touches the object to be measured.

Here I am abusing my smartphone camera to look through the F+C measuring magnifier looking at the beam of a common Mitutoyo caliper. This gives an idea what sort of resolution to expect. The phone camera is of course far from ideal, looking by eye directly it looks of course much much sharper and clearer. The white band is a thin layer of paint printed on half of the "ruler", this allows to use the device on white objects, as well as on black objects.
tt1.jpg

Originally Posted by simonl

Is that your USB camera Stuart?

It is. (well one of them lol)

Originally Posted by pipeclay

Unless it is automated Or very exspensive I don't think you will find anything that will enlarge and then take dimensions.

Its neither but it can. You would want to take a little more care in the set up than I have done for the pictures here.
The powers are 20 and 200 times. You can tweak those numbers in the software when set up against a known length if needed.

An optical comparator would(should) give more accurate results over a larger range......... but it didnt cost $30.

Depends on just what accuracy Ken is after and just what the part looks like.

Stuart

http://en.wikipedia.org/wiki/Opaque_projector

Cba's seem the go

a manufacturers web site
https://www.emsdiasum.com/microscopy...measuring.aspx

available in Aus from here
https://proscitech.com/?navaction=sh...ge=2#o22-07-01

or the fowler one on amazon...it seems to have various scales included
http://www.amazon.com/Fowler-52-665-.../dp/B00B5HK8HM

might get one my self..lol

Once you have the image it is fairly easy to load it onto a CAD screen, then scale it to the correct size. You can then measure or draw over the top of it.
This is Stuart's scale in Tubocad

Neil

If you want to use a microscope to measure, you can use an eyepiece with an inbuilt cross hair with graduated graticule. Before first use, you must calibrate this for every objective/eyepiece combination. To calibrate, you put a precision scale onto the stage and focus it. Then you compare it with the eyepiece graticule, and calculate the correction factor. This correction factor you note down for future use. It is fairly accurate in the central region, but may be a bit off at the fringes of the visible field. Lenses and eyepieces can have 5% tolerance in focal length, so the correction factor is only valid for the particular lenses is was acquired with.

The same is true with a microscope with electronic camera, or indeed a cheap USB microscope. You also need a fine scale to calibrate it, a scale with at least 1/10mm subdivisions. Just a steel ruler with 1mm or 0.5mm lines is not going to work -- well, I mean it does not make any practical sense, since our eye can read such a ruler with the naked eye. And if you can read it with your eyes, why use an USB microscope at all? Also, the simple lens that comes with the cheaper USB microscopes may have some substantial geometric distortion, which may cause easily 10-20% measurement error towards the fringes of the viewing field.

Originally Posted by cba_melbourne

The same is true with a microscope with electronic camera, or indeed a cheap USB microscope. You also need a fine scale to calibrate it, a scale with at least 1/10mm subdivisions.

You dont need a scale to calibrate, just a "known length". It can be the length of the part he wants to measure

Originally Posted by cba_melbourne

Just a steel ruler with 1mm or 0.5mm lines is not going to work -- well, I mean it does not make any practical sense, since our eye can read such a ruler with the naked eye. And if you can read it with your eyes, why use an USB microscope at all?

The steel rule pictures where used A. Because I had them. B.Everyone knows what a 0.5mm rule looks like.

Originally Posted by cba_melbourne

Also, the simple lens that comes with the cheaper USB microscopes may have some substantial geometric distortion, which may cause easily 10-20% measurement error towards the fringes of the viewing field.

Easily 10-20%? mine shows nothing like that.

But as I said it depends what accuracy Ken is after. I still have no idea

Stuart