Then you have to worry about stress making that flat surface look round. Too much torque can do that.
greg
Then you have to worry about stress making that flat surface look round. Too much torque can do that.
greg
Like This?
The point of the whole exercise was to do it by hand, with a file.
These days, you just upload the program, throw a piece of metal into the machine, and the finished part comes out the other end. ;-)
Cheers! Rich
Exercise? I thought he wanted to make a heatsink?
For a one-off of a simple operation like milling a surface flat, it's a lot easier to just stick it in a manual mill. If you want 100 of them, or you want a complex part, then CNC is the way to go.
Yup, like this:
ftp://jjlarkin.lmi.net/Amp.jpg
The shiny things are copper heat spreaders, nickel plated. I did one DSRD-based [1] HV pulser that was built on a gold-plated copper block, water cooled; it looked great. You can get a peek of it here:
Some, perhaps even most, aluminum alloys machine nicely. Pure aluminum, and just about anything designed to be easy to bend up, is 'gummy', and hard to get a good finish on.
But 6061 is a treat, and many extrusions seem to be nicely machinable.
I would've thought that any type would be fine for this purpose.
Surface quality, as well as flatness, are the two factors here.
Don Klipstein wrote: [...]
Indeed. The most important factor would be that the spacer be as flat as possible on both sides, for best heat transfer.
What's the heat sink made out of?
Can it be machined to include a suitable mounting pad? This will eliminate one junction.
Yes, that's the right AN. Great reading. Thanks.
Yes, very careful lapping on a piece of glass with finer and finer grinding paste should produce a suitable finish.
Of course you could try scraping. A friend of mine used to hand scrape surface plates.
Hand-scraped TO-220 packages. The mind boggles.
Right idea, but to actually work It's three metal blocks, a tube of bluing, and a scraper.
You need two to be able to mark the high spots on each other, and you need three to avoid developing complementary curves.
An old optical technique is to rub three surfaces against one another. Eventually all three become flat. There are automatic rotating grinder things that do this.
John
Is scraping really better than lapping? If anyone here can speak definitively about the difference, I'd be very interested to hear about it.
I've mastered scraping flat--how does one ensure "square" ?
TIA, James Arthur
When I was given that task, (not quite a century ago!), we were given a vernier caliper, whose jaws were "square". The metal was a chunk of cast iron, and the task was to file a one inch cube, accurate to +/-
0.001", "square" on all sides.
Have something to add? Share your thoughts — no account required.
Ask the community — no account required