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

To measure the radius in the same way as the gauge Bill described you could use a square without a notch in the corner and try putting different thicknesses of wire in the corner of the square to find the largest that doesn't make the square rock against the test piece, then draw it up in a large scale (or in CAD) to determine the radius. The catch is that for radii between 0.1mm and 0.5mm you would need to source a selection of wires between 0.0344mm and 0.176mm.

Sorry about the rotten drawing, I have only just switched to 4M CAD and haven't worked out how to add diameter to the dimensioning toolbar (or how to dimension manually)

Brian G

As the OP has advised the component has a variable (hand made) radius, what exactly is being measured? A simple set of radius gauges viewed against a light source will produce an indication of the radius and allow acceptance or rejection of the component. Anything else appears to be a waste of time and money.

I hope we will all be able to retire on our share of the Royalties from this device when it comes to market.

Robert Butler

Just stone the edge with a fine oilstone, that is bound to be within 'limits'

Tony

am i missing something ? if i read this right this is a handmade radius , file ? than i would check it by eye

@Dave
> 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?

Both.
i.e. I want to create some test radii by hand to see what feels right and works right, so that I can formally specify them with a view to out sourcing at some point. 

And then, having measured what I think works correctly, I want to apply the radii consistently across multiple parts to make sure that I am making them consistently.

[To get clear, for functional reasons I need the radii to broadly be as small as possible, this is because I need to surfaces next to them to be as large as possible, HOWEVER if they are too small then the part starts to become dangerously sharp. So some sort of compromise 'happy middle ground' has to be found by experiment. ]

Yes, the radii will be machined by hand mostly using a deburring wheel. Possibly I may use a hand file (or files) but I am trying to avoid that.

Various edges will need to be measured along quite long parts.
i.e. it will be much more like this:


And never like this:


==> So I can't see how a light source would help me.
 

@peak4
Yes, now that I see how it works, I can see that that in many ways the "Helios Preisser H0597305" is the ultimate answer(!). One down-side is that it is assuming that the radius is perfectly circular, which given that the radius will be made by hand that will be a rash assumption. Maybe I could find a much cheaper one somewhere.

Where I've got to so far
I think where I've got to is that I need to try to buy a few different sizes of simple looks that have a L shaped corner and broadly look somewhat like this Starrett:



...If that is, anyone actually manufactures such a thing in smaller dimensions
e.g. say: 0.4, 0.3, 0.2 or even 0.1mm ??

But it's starting to look like nobody makes such a tool.

J
 

Edited By John Smith 47 on 30/12/2021 20:53:06

Could this not be made? Turn several pieces of small rod to known diameters, on the lathe, and glue each to a stick?

Yes in effect that would do the same job as the 211 euro Helios Preisser thing.

But it would also be nice to be able to see into the corner to see how how much light was getting paste the radius, to see how circular my hand-made fillets are. 
 

Edited By John Smith 47 on 30/12/2021 20:59:44

See P1-3 of this catalogue from FujiTool
https://www.fujitool.co.jp/pdf/catalogue_vol218.pdf

Or FujiTool's own web site
https://www.fujitool.co.jp/eng/products/measuring_tools.html

The exact style you seek (L300M2) only start at 0.2mm,
but 272MAA 01-408 are available in a set from 0.1mm-1mm

They are available from Hogotex, who seem to have branches in Germany, Netherlands, and Belgium.
https://www.hogetex.com/radiusmeters-type-272-serie

If you're prepared to work in increments of 5 thou, rather than metric TTC in the US make some
https://www.travers.com/product/ttc-grg-28-radius-gage-set-57-025-025

Bill

Edited By peak4 on 30/12/2021 21:31:12

Just knowing what it is and why it needs such fiddly to produce parts would be enough, from the limited information we've been allowed.

In these circumstance a radius gauge can only serve as a comparator given the use of a deburring wheel or hand file to produce the radius. Therefore simple radius gauges should be more than adequate due to the variability of manufacture. Radius gauges have a flat datum surface to bear on the sample and assuming the sample is flat the approximate radius (which is all that can be hoped for) established.

Robert Butler

I was under the impression commercial projects were a no no on this forum, this project appears to be 100% about making money and 0% about someone learning workshop skills as a hobby.

Whilst what you say is correct, the real difficulty would seem to be finding radius gauges to measure the radii requested by John.
As soon as you get down to a minimum radius of 0.1mm, and thus similar increments, the price goes up considerably and the number of manufacturers decreases.

I did find some to the pattern and tolerance requested, but they weren't obviously available to order in the UK or US, but are available from Japan for 66,500 Yen. They were also out of stock from this retailer, but whether that is a similar situation to eBay stores here which just increase the price of OOS items, I don't know; they also show the wrong photo for the named product listed.

I'll let you do the price conversion yourself, but I suggest it will be beyond the means of the OP.

Bill

Edited By peak4 on 30/12/2021 23:21:07

To be honest, to measure stuff sub 1mm you are looking in to optical equipment to get the best idea of what you are looking at. All what I could suggest has already been put forward.

My first inclination was shadowgraph with appropriate screen/mask, the measuring loupe would be the next best thing (cheaper but one would struggle with wide parts the ends only being measurable)

Then there is a Mitutoyo contracer: https://www.mitutoyo.com/products/form-measurement-machine/contracer/

That would be the most accurate and easiest method and also the most expensive.

I really don't see the point to all of this, its really only a break-edge from what I can make out from what John has described so doesn't need to be quantified. The works drawings I work too have break all edges in the notes section which is enough for us employee to work out that they don't want huge great chamfers(would be specified if that was required)

As Robert Butler has said through out this post, a set of rad gauges along the Moore and Wright pattern would suffice if one has to be really pedantic, although as others have said the size required doesn't really exist. They could be lazer/water jet cut but the kerf has to be thought about.

Just my thoughts that will no doubt be disregarded

William

Damm I wasn't going to comment further on this thread but you are 100% correct.

Tony

It strikes me that this is putting the 'cart before the horse' Why make something and then try to measure what has been made? The usual sequence is, design the component and then make it to the design. I get the feeling that the OP has little or no knowledge of basic engineering principles. If this fillet was hand made, by whatever method, it tells me that it does not need to be that accurate, why make a simple job difficult? What is being asked is virtually impossible with the kit most model engineers have or could afford.

I have a micrometer barrel that was saved from a tool that was being scrapped, like a depth gauge without the foot. I would make a tool that held it at 45° to the edge and could be adjusted to read zero on a sharp corner. Then anything less sharp can be measured and the reading converted to a reasonable idea of the radius with simple geometry. You may also decide that a simple 45° chamfer is all that is required. It would certainly be good enough to decide what was a good value to go for. Imagine the centre finding slide of a combination set with a micrometer barrel instead of a sliding rule.

Martin C

Actually it is the promotion of commercial products or services that is not allowed, not asking about a project that may be commercial. From the T&Cs

"Upload, post or otherwise transmit any unsolicited or unauthorized advertising, promotional materials, or any other form of solicitation."

Edited By JasonB on 31/12/2021 10:11:27

much more interesting.

In the world of "Pro Bodgery" we would make up a small punch tool (chisel shape) with a short length of annealed copper wire or solder and tap it into the radius. You now have a copy of the radius you can examine away from the piece under investigation or blue to test against another location on the part.

Edited By Bob n About on 31/12/2021 12:50:37

Been away. Catching up.

This is rather overwhelming but I shall try to respond to each person, but thank you for all your contributions.

@MikeK - in principle, yes, along the lines of what Brian G says.
The L-shaped Starrett design would do me fine, however they don't make them small enough. TBH, If I can't find some fairly simple tool to do the job I just abondon ship. To get clear the reason why I want to know the radius of a fillet that I have made is so that I can outsource some of the work to someone else.

@peak4
Yes, that Fujitool "L300M2" looks perfect. Although I would only want the first 3 or 4 tools in the set. Any idea where I could buy one?
[ No, I don't like the look of L300M2. Nor the "Radiusmeters Type 272 serie" from Hogotex I want a 90 degree L shape (with a radius in the concave corner). 0.010inches is 0.254mm which is already slightly large for my purposes.]

@Robert I simply don't understand your comment. My apologies.

@Pete As I have said elsewhere this project is currently a very long way from being commercial.

@peak4
I couldn't undertand that page, but 66,500 Yen would be £427 which would be an INSANE amount of money to pay. Particularly for what would only be 3 or 4 small pieces of metal that if pushed I'm pretty sure that I could get laser cut at reasonable accuracy for a lot less money.
But seriously, this mission/odyssey is definitely not worth spending more than say £150 on.

@William
You are correct that it is roughly speaking a "break-edge" situation, but there are competing ergonomic issues that need to be balanced by experiment. And I then need to quantify the radiuses. I am now confused about what you mean by the "Moore & Wright pattern".

@Tony Pratt
No, I am a very long way from making money on this project.
Yes, I am keen to learn about workshop skills & any suitable tools that anyone can suggest that could help with the measurements.

@Dusty
> Why make something and then try to measure what has been made?
It's called a prototype.
The key discipline is ergonmics.
If it turns out that no such tools exist of the size I seek, fine let's all move on. At this point that seems unproven.

@Martin
Yes, if I can't get nicely rounded fillets, then as you imply maybe 45 degree chamfers would do.
Wait, do simple "L"-shaped little tools exist which look broadly like the Starrett that I found exist which have chamfers not radiuses?

@JasonB
I hope it's 100% clear to everyone that I am not promoting any specific product, and am merely seeking buying advice.


tldr; / SUMMARY:
If the tool I seek doesn't exist at a half-reasonable price (ideally sub £100) then let's all abandon this thread.

RECAP:
What I'm seeking is a simple tool(s) to measure the radius of convex fillets (i.e. the radius on "break-edges"?) created on 90 degree edges.

Something similar to that Helios Preisser H0597305 would do the job, but 211 Euros is GBP177, and worse post Brexit there will now probably be a hefty import tax on it... ==> over £200?

So I was hoping to buy some very simple L shaped pieces of steel with precision radiuses inside corners, with which to make quick checks.

THOUGHT:
If, as some of you have suggested, I just assume circular radius and just focus on measuring the equivalent 45° chamfer, what about buying something like this?
**LINK** for c. £56


Or a digital equivalent for c. £76?
**LINK**



Or similar... (?)

J