Dovetail myth regarding pin dimensions.

[Parlo]

Here is a vido link demonstrating this simple process -

This technique can also be adopted during the design process to make machining easier without special pins and fewer dimensions.

 

[mcostello]

Gee, I take 2 transfer punches from the set that are about the same thickness as the depth of the dovetail, and measure the resulting dimension. Cut the dovetail to the same dimension and it works the same as the origional. I have only cut 3 tool holders with this method and all 3 fit the same as the bought ones.

 

[Parlo]

Gee, I take 2 transfer punches from the set that are about the same thickness as the depth of the dovetail, and measure the resulting dimension. Cut the dovetail to the same dimension and it works the same as the origional. I have only cut 3 tool holders with this method and all 3 fit the same as the bought ones.

You are replicating a male with a male or a female with a female which will work with those size pins. Making a mating pair will have different dimensions over the pins for the male and female unless the correct size pins are used.

 

[Parlo]

Here is a formula to calculate the size of the pins for any dovetail angle.

( Cos angle divided by [ 1 + Cos angle ] ) x depth = common pin diameter.

Use the depth of the deepest part.

 

[Bruce Billett]

Question. As you are measuring a triangle by using the center of the pin straight down and across the bottom with the angle known. What is the reason for using round pins?
Won't square pins work as well? They are easier to machine, making one LONG square pin is pretty simple and cutting it in half is easy enough pretty much ensuring the critical dimensions are the same. And frankly I see nothing in the equations that require any sort of circular numbers other than to figure out where the triangle is inside the pin.
Just don't see the point?

 

[Parlo]

Thanks for the post. If you watch the video in the original post you can see how involved the traditional equations are. I looked at the problem to make precise mating dovetails from the perspective of using the same diameter pins for the male and female. The simple formula of using a pin diameter 1/3 of the depth allows the exact same dimension to be used for both parts. This allows one part to be made to an approximate size without having to hold to a tight tolerance. Then the mating part will theoretically be a size for size fit when machined to the same dimension.
This method ONLY works with the correctly calculated size pins. The simple 1/3rd formula is by far the easiest compared to all the faffing about in the original video. I urge you to watch it again or see the calcs in the machinery handbook.

-----------

Question. As you are measuring a triangle by using the center of the pin straight down and across the bottom with the angle known. What is the reason for using round pins?
Won't square pins work as well? They are easier to machine, making one LONG square pin is pretty simple and cutting it in half is easy enough pretty much ensuring the critical dimensions are the same. And frankly I see nothing in the equations that require any sort of circular numbers other than to figure out where the triangle is inside the pin.

It is traditional to use round pins as the tangent points on the dovetail angles are easily found.
I'm not sure how square pins will work in a 60 degree groove.
A 30 - 60 - 90 right triangle section would work with the height 1/2 the dovetail depth but would be far more difficult to make than using gauge pins.

 

[Bruce Billett]

Thank you for answering so quickly,
However the question remains, What is so magical about using pins rather than a square or even rectangle?

 

[Parlo]

Thank you for answering so quickly,
However the question remains, What is so magical about using pins rather than a square or even rectangle?

Measuring angles and tapers are often accurately measured using precision pins and balls. Using other shapes can work, but they have corners which need to be precise as they will locate on them. It's just so much easier to rest a pin in an angled feature and the centre can very easily be calculated.

 

[Bruce Billett]

Thank you!

 

[whitmore]

I'd just like to mention that the dovetail might NOT be critically dimensioned by dowels packed in the socket,
if the those surfaces (big-end and slant side) aren't those that determine the position of the fill item; you don't
care about the depth (and big-end face) of the socket if the dovetail mates with contact of the outer/upper face of
the socket and the sidewall.

Indeed, if you don't make the sidewalls parallel-aligned, you get a wedge dovetail (sometimes used
for quick-connect on photographic tripods) that mates on sidewall and big-end rather than sidewall and
outer/upper face. I've pondered making a plinth/compound lathe carriage that accepts multiple
hats on such a wedge dovetail. Biggest problem is that the entry direction has to tolerate
cutting forces without loosening (probably wants to face the drive spindle).