What tool do I need? I need to measure the radius of a tiny fillet on a 90° edge.

Hello

I need a tool to enable me to measure the radius of a small convex fillet where two flat faces meet at 90 degrees.

The radius will be very small - somewhere between 0.1 and 0.5mm.

What tool do I need?

J

A very small radius gauge!

Robert Butler

"Fillet " normally means an internal (concave) radius, as the root of hot-rolled steel angle, and if its end is accessible it can be gauged with a twist-drill shank.

For a radius along an edge, you'd need a radius-gauge - but lacking a set, you could make one by drilling the right-size hole through a slip of sheet-metal, near one corner, and carefully cutting a 90º sector from it.

Hi John where are you as it may be possible to borrow or use some ones gauges. I personally have sets of Starrett radius gauges in meteoric and imperial from when I did toolmaking at work before retirement.

David

> "Fillet " normally means an internal (concave) radius

1. This appears to be out-of-date.
As Wikipedia says: "Autodesk Inventor, AutoCAD, CATIA, FreeCAD, Solidworks and Vectorworks refer to both concave and convex rounded edges as fillets"

2. I have been very explicit that I only wish to measure convex radiuses.

3. I do not wish to make a tool, I wish to purchase a precision tool.

4. In particular I need something that is designed to work on 90° convex edges... 

OK, I shall further research "radius gauges". However their sets seem to come in various flavours, many of which are measuring concave rather than convex radiuses.. or both!

Thanks

PS I am trying to avoid the dreaded Amazon. Who else would you recommend trying? (But must do a rapid delivery service!)

Edited By John Smith 47 on 30/12/2021 14:19:30

Most radius gauge sets seem to include blades with measure convex and concave.
For the larger radii, it tends to be on the same blade, but smaller ones use separate blades.
The smallest I've seen, start at 0.3mm, but that doesn't mean they are the smallest available.
I include this link to Amazon, though they are available via eBay from foreign shores.
https://www.amazon.co.uk/R0-3-1-5mm-Measure-Leaves-Metal-Radius/dp/B008MJTJWQ

I suspect higher quality tool merchants may also sell them, but at a price.
RS do a set, but it starts at 0.75mm and goes up in o.25mm steps, so too large and too coarse.
Mitutoyo only seem to start at 0.4mm

For smaller, maybe you would need access to a shadowgraph/Profile projector, but that certainly won't come cheap, even for second hand.

There are certainly vintage magnifying overhead projectors available, but I'm not sure they would suit your needs.
https://www.etsy.com/uk/listing/230507873/vintage-magnajector-enlarging-projector

I wonder if a conventional overhead projector may work.
i.e. use a measured disk of say 1cm next to your workpiece; project both adjacent images onto a wall simultaneously and scale the measurement from the known calibrated size.

Another possibility, I've not fully though through, would be a magnifying loupe, with an appropriate reticle.
I'll include this link, to give you some idea, even though it's from over the water.
I'm sure similar things will be available from UP suppliers
https://www.peakoptics.com/index.php?main_page=product_info&cPath=66&products_id=10

Just found a UK supplier, so maybe worth investigating, but reading increments of 0.1mm might be tricky
https://www.theloupestore.co.uk/Scale-Measuring-Magnifiers/

Bill

Edited By peak4 on 30/12/2021 14:51:57

When |I was working (toolmaker) this was a daily task. Made easy with a shadograph with resolutions from 10 to 1 up to 100 to 1. Where are you located ? There may be a local company willing to let you use their kit or measure for you for a tea fund donation.

Yes, unfortunately it seems hard to find anything smaller than 0.3mm.
e.g. Starrett only seem to go down ton 0.5mm.

To get clear, what I am planning on is to find a gauge that has two 90 degree edges that make a concave space, and that has known radius whether they meet.

My plan would be to start with smaller radius gauges, and then work my way up in sizes. ==> And at some point the square edges would no longer fit around the square faces of my part, and the gauge start to 'rattle'.

i.e. This would happen once the rounded 'fillet' on the my part starts to interfere with the radius in the corner of the gauge.

So in this way, by measuring at what size of gauge the 90 degree edges no longer fit snugly, I would not need to visually see exactly what is going on in the corner with any accuracy. Instead I would 'feel' the fit.

==> i.e. In effect used a bit like a feeler gauge!

(If I did need to see what is really going on, then I have various high-powered clip-on reading glasses and if necessary I could even use my dissection microscope!)

Edited By John Smith 47 on 30/12/2021 15:12:20

To follow on from my last post, amateur and professional mycologists have the need to measure fungi dimensions under the microscope, and are often working out the sizes of individual spores.
There is software available which allows a USB camera image to be measured when it's imported to a computer.

For spore sized, you're looking at a specialist 100x oil immersion lens, but for larger measurements, such as gills etc, a cheap microscope with USB-camera might just do the job.
They come up for sale via Lidl/Aldi and thus appear in eBay when the owners get bored, often made by Bresser.
Just make sure the USB camera is supported by your operating system.

I'll include this page out of general interest, but mainly because of a link to Mycocam4 contained within.
https://www.first-nature.com/fungi/~microscopy.php

Bill

Edited By peak4 on 30/12/2021 15:15:40

This tool looks like it will measure the radius from 0-10mm in 0.1mm increments across right angles.

Radius gauge 0 - 10 mm, 0,1 mm (0597305)

**LINK**

Nice thinking Bill... however it seems to me that measuring along length of an edge that might be say 2 or 3 cm long, and have a variable (hand made) radius, would be best 'felt' (if technically possible) rather than inspected visually.

I think your link maybe broken, Is it this?
https://shop.kinexmeasuring.com/en/radius-gauge-0-10-mm-0-1-mm-0597305-p8900618c30c237/

Bill

Peter I can't get that link to work.

Bill has posted the correct link.

Looks cunning, but do any of you know how to use it? I can't work it out!

PS Also rather expensive at c. 211 Euros!

Edited By John Smith 47 on 30/12/2021 15:31:51

Can you clarify please John?

Do you want to:

1. measure arbitrary radii to know what they all are to some level of accuracy, or
2. to indicate that one radius is more-or-less correct?

The first finds out what a radius is, the second confirms a radius is within acceptable limits - perhaps a GO/NO GO Gauge.

Checking radii in the range 0.1 to 0.3mm is a tough requirement, especially if it has to be done accurately! You can look for light leaking around the corner of a thin plate placed into a gauge, but much harder to detect if a radius is correct along a lengthy edge.

Easy:

Gets harder with increasing length:

Radius gauges are easy enough to buy in larger sizes, but not many tiny ones about. Could be you need a loupe, shadowgraph or microscope.

Dave

Here you go, shown in use on P53 of their catalogue
https://metrology.mahr.com/fileadmin/catalogs/OnlineCatalogs/Helios-Preisser/index.php?catalog=HP-Metrology-EN&lang=en

Essentially, it requires knowing that your radius is actually a section of a circle.
It relies on three point contact of the two sizes and the circle radius>
The sliding vernier scale gives the radius measurement.
As you can see a 45° flat would also give a reading, but an erroneous one; compare it with the chamfer gauge at the top of the same page.

Bill

Edited By peak4 on 30/12/2021 16:05:15

The Helios Preisser H0597305 radius gauge works by offering the test piece to the gauge with the two relevant edges containing the radius against the gauge and the calibrated angular slide advanced until it touches the radius. A reading is then taken from the Vernier scale and compared to tables supplied? or calculated? I know not which. It all depends on the accuracy of the radius.

Robert Butler

Edited By Robert Butler on 30/12/2021 16:11:54

Some sort of shadowgraph would be my tool of choice. Watching this thread with interest to see where it leads. Are we to assume this radius will be machined rather than filed?

Tony

Edited By Tony Pratt 1 on 30/12/2021 17:26:15

the link sent by Peter H was caught by an oddity in the software used by ME* - it came out like this:

radius%20gauge%200%20-%2010%20mm%2C%200%2C1%20mm%20%280597305%29

Some typesetting and word processing programs use an extra 'hidden' symbol or two to tell the computer what it intended. Here the extra symbols include %, followed by a number, calling up a letter or symbol not in the 'standard alphabet' used in the days of 8-bit and 16-bit computers. I'm not sure what you or I can do about it, except wait for ME to update - or, * perhaps the fault lies within the computer used by Peter H. I wonder ...

Cheers, Tim

Edited By Tim Stevens on 30/12/2021 18:33:09