Machining question.

[qumefox]

Not sure what section this goes in so i'll put it here.

I need to turn a few press fit heatsinks and I know there's an easier way to cut the diode pockets than with normal tooling.

So my question is about end mills. Is the sizing on end mills the resulting diameter or the tool diameter?

Meaning if I plunge mill the pocket with a 5.6mm end mill, will I get a 5.6mm hole, or a larger one? and should thusly use an undersize end mill?

 

[qumefox]

Bumpity Anyone?

Guess i'll just have to order an assortment of end mills and find out myself.. which kind of sucks because good ones aren't cheap. heh.

 

[bbshamsa]

The resulting hole diameter will be the same as the rated diameter of the tool you're cutting with.

So, yes, a 5.6mm dia end mill will cut a hole 5.6mm dia.

Although, if you want a press / transition fit you should aim for up to a millimeter smaller diameter.

So, I'd go for a 5mm dia end mill if I were you.

Otherwise it's likely you'll end up with a loose fit.

 

[qumefox]

Thanks. Though a .6mm undercut sounds kind of excessive to me, but I guess it just depends on the accuracy of the endmill and my lathe. Always better to remove too little material than too much. heh.

 

[Eudaimonium]

0.6 mm undercut IS excessive.

I cut the pocket for diode with a DIY tool made out of solid bar of HSS, it has about 1cm long 1-2mm thin protrusion, which is mounted in such a way onto the carriage that it enables me to finely cut the inner side of the heatsink where needed (such as diode pocket).

So if you can find a very thin knife, it'd be perfect.

Also, aim for 5.57-5.58 mm inner diameter. Might be difficult, but it takes practise like everything else.

EDIT- Oh, and best to my experience, the dimension given by tool description is actually the tool itself, resulting work always ends up bigger (example, 12.00mm declared drill bit always does 12.1 mm wide hole etc).

 

[123splat]

I believe that the sizing is the tool size, not the resulting diameter, as the resulting diameter is dependant upon your setup and the machine's runout.

How big a lathe are you using? something like an engine lathe, or a brake lathe, or something like a mini? You might be better off grinding a custom boring edge tool (easier to get a flat bottom that way). Might be a bit of a bitch for an engine lathe though.

 

[qumefox]

Yeah I make my own tooling so i've done similar. Though if I can just use tailstock mounted endmill and just plunge it and be done with it.. Well. That will make things easier and less tedious heh.

And I have an atlas 618 6" lathe.

 

[bbshamsa]

0.6 mm undercut IS excessive.

I cut the pocket for diode with a DIY tool made out of solid bar of HSS, it has about 1cm long 1-2mm thin protrusion, which is mounted in such a way onto the carriage that it enables me to finely cut the inner side of the heatsink where needed (such as diode pocket).

So if you can find a very thin knife, it'd be perfect.

Also, aim for 5.57-5.58 mm inner diameter. Might be difficult, but it takes practise like everything else.

EDIT- Oh, and best to my experience, the dimension given by tool description is actually the tool itself, resulting work always ends up bigger (example, 12.00mm declared drill bit always does 12.1 mm wide hole etc).

When you're looking for a transition fit between two parts, the goal in this case (heat-sinking) is to eliminate as much air-gap as possible.

Bearing in mind the inevitable lack of accuracy from hobby kit that the vast majority of home users have access to, a tolerance of +- 0.1 - 0.2mm is more than reasonable to expect.

Therefore a target tool dia of up to 0.6mm less than the desired finish dia is ok.

This would result in a finish dia of 5mm + 0.1 to 0.2mm...

... leaving a hole somewhere between 5.1 - 5.2mm, with a resulting ~0.4mm material to transition with.

We're talking about aluminium to aluminium here (diode to heatsink), and with the softness of aluminium a 0.2mm radial squeeze is shared between both fitting surfaces.

Resulting in a stress of only 0.1mm on contacting surfaces.

Well within the malleability of aluminium.

Finally resulting in a proper transition fit.

 

[Eudaimonium]

You seem to be forgetting the part after the diode part - the lens thread.

It translates into actually having maybe 2 milimeters deep diode pocket, and how wide is the "supporting" part.

Also, diode is not aluminium, it's brass.

Under these parameters, I'm afraid you might break entire diode's seat if you try to stuff 5.985mm wide diode into 5.2 mm wide, 2mm deep pocket made of aluminium.

Besides, +/- 0.1mm does not stand true - with that cutter, I spend a LOT of time getting the diode pocket as perfect as possible.

There's time. Take it.

 

[bbshamsa]

Without specific knowledge of the posters available tolerance limits and machining skill I think good advice should lean towards basic principles of machining; you can't replace material removed.

For a basic hobby lathe, tolerance of +-0.1mm is common as far as I'm concerned.

That's on the shaft, so up to double when drilling or milling.

5.985mm??

I just measured one of mine and it's a lot closer to 5.6 than that!

Also, it's not hard to remove a tenth of a mm or so by hand or tool whilst in the chuck to ensure a perfect fit.

 

[Eudaimonium]

God damn, I meant 5.585mm Sorry!

That's by the technical specification in Datasheets.

+0mm / - 0.015 mm

Not sure why I hit 9.

You know, one of those days when, even in spite of 12 hours of sleep you've had, you feel like you've been ran over by a truck in your sleep, and feel like sh*t whole day?

Yeah it's one of those days for me.



So let's leave it at that.