PDA

View Full Version : Anything special about machining copper???



brian Rupnow
07-22-2008, 04:36 PM
I have a big peice of copper plate, about 3/4" thick that I salvaged from a machine I dismantled a few years ago. The original plan was to cut it up and use it as backing plates for mig welding on some of the antique automobile sheet metal that I play with.--However the plan changed, and now I am thinking of using it as part of a rotary base for my milling vice. How is it to machine---similar to aluminum and brass---or are there too many kinds of copper plate to generalise?

pcarpenter
07-22-2008, 04:46 PM
Most CU alloys are too soft for a vise base and you really don't know for sure what you have....Do what you said first. I am trying to find a similar piece for weld backing. Copper scrap is priced so high nowdays that anyone who has it is scrapping it out so its hard to come by a piece of flat like that. Hang onto it.

Paul

Evan
07-22-2008, 05:32 PM
It isn't at all like brass or aluminum. It work hardens so easily that it is tough to machine cleanly. It often leaves a hard to remove burr on cut edges and is just plain a pain in the asp. If you insist on machining it use very sharp tooling and a very thin coolant such as WD-40 or alcohol. I've heard of milk being used which makes sense since it is an emulsion of oil and water but I wouldn't want my machine smelling like sour milk for the next year or two. A lot depends on exactly what alloy it is (or none) and what trace elements it has as well as the work condition. Copper is very sensitive to trace element alloying and the properties change dramatically depending on what it contains.

I'm curious what sort of machine it was from. I would assume it was used to prevent local distortion caused by local heat sources of some kind. Being the second best metallic conductor of heat it would rapidly equalize the heat distribution in a machine base.

Doc Nickel
07-22-2008, 05:45 PM
I agree with PC- copper is way too soft for something like a vise base, and would be a waste of some very expensive stock.

Doc.

brian Rupnow
07-22-2008, 06:23 PM
It isn't at all like brass or aluminum. It work hardens so easily that it is tough to machine cleanly. It often leaves a hard to remove burr on cut edges and is just plain a pain in the asp. If you insist on machining it use very sharp tooling and a very thin coolant such as WD-40 or alcohol. I've heard of milk being used which makes sense since it is an emulsion of oil and water but I wouldn't want my machine smelling like sour milk for the next year or two. A lot depends on exactly what alloy it is (or none) and what trace elements it has as well as the work condition. Copper is very sensitive to trace element alloying and the properties change dramatically depending on what it contains.

I'm curious what sort of machine it was from. I would assume it was used to prevent local distortion caused by local heat sources of some kind. Being the second best metallic conductor of heat it would rapidly equalize the heat distribution in a machine base.

The machine was an automated welding cell, with fixtures mounted to a rotary table. The copper was a fixture base, and it was made from copper to have good conductivity for the electric welding arc. I bought the machine and reconfigured it to weld seat adjuster assemblies for Chrysler, so of course all the fixtures and there bases changed.

lane
07-22-2008, 08:00 PM
Dont ask take a cut you will find out real fast . Its a bi--h to machine .Worth more a scrap than to try to make something from.

jacampb2
08-19-2008, 11:39 PM
Digging a recently dead post up because I have the same question. I have some .25x4 C110 flat bar that I want to make some buss bar/heat sinks out of. It will require splitting the length of the flat bar, for finished dimensions of about 1.75x5x.25. It does not require any surfacing. I do however need to punch holes in it for TO-3 transistor mounting.

Should I try to cut the bars lengthwise w/ a EM, or should I use some form of saw. As for drilling, is there any thing to look out for? Is water soluble coolant going to be okay, or is kerosene or WD the way to go? I would like to make several of these plates, and it would be great if I could knock a pair at a time out on the CNC w/o having to hand cut the bar stock. Doable? Most of my tooling is high end carbide. I have a hand full of import HSS EMs though as well. My drills are all pretty much HSS.

Any info would be great. The project has to be copper though, these are audio power amp isolation heat sinks. They are going to carry the rail voltage, as well as be part of the heat sink.

Thanks,
Jason

Evan
08-19-2008, 11:52 PM
Why not use a 30% heavier/thicker aluminum? More surface area and the same conductivity.

jacampb2
08-20-2008, 12:32 AM
The main reason is I intend to liquid cool these. With copper, I can easily solder tubing to the plates to circulate coolant. To pull it off with aluminum, the only simple scenario I can think of is machining a coolant passage into a top and bottom plate, and gasketing it to seal it up. Or, drill thicker material and plug the through holes at the corners. I have very limited space to work with, and I have the copper stock, so it seems the best solution. There is not that much machine work that needs done, but this thread has kind of scared me off to it :)

Later,
Jason

Carld
08-20-2008, 12:42 AM
In the late '60's I made a straight line cap and ball muzzle loader reciever out of Copper. It was hell to make it and if I did it again it would be for mucho money. I made the trigger mechanism, plunger hammer, breach plug, etc. out of steel. I assembled it all and the owner made a stock. I guess he did, as I never saw him again. I took photos of it because I didn't think I would see it again and no one would believe me. I did all this on a 10" flatbelt Logan with a milling attachment and a drill press. Can you imagine how many times the belt kicked off, the drill got stuck, the boring bar hung up?

You would be better to machine cooling passages in an aluminum plate, attach a cover plate and mount your parts. It will run cooler than the copper with soldered on tubes and you won't use any new cuss words.

Evan
08-20-2008, 01:09 AM
Incidentally, conductivity of metals is normally ordered according to conductivity per unit volume. However, if you order the list by unit mass instead then aluminum comes out well ahead of the other metals.

aostling
08-20-2008, 02:01 AM
Incidentally, conductivity of metals is normally ordered according to conductivity per unit volume. However, if you order the list by unit mass instead then aluminum comes out well ahead of the other metals.

I've never seen an ordering of thermal conductivity per unit volume -- that makes no sense. However, thermal diffusivity brings in the effect of material density (and specific heat).

Evan
08-20-2008, 04:33 AM
I was speaking of electrical conductivity since the OP is using the heat sinks as buss bars. But, thermal conductivity and electrical conductivity in metals are a function of the same atomic properties of free valence electrons. The ratio of thermal to electrical conductivity times the temperature is constant in elemental metals.

Here is a table of thermal conductivity ordered by conductivity per unit volume, specifically (cal/sec)/(cm2 C/cm).

http://hyperphysics.phy-astr.gsu.edu/Hbase/tables/thrcn.html#c1

aostling
08-20-2008, 05:23 AM
Here is a table of thermal conductivity ordered by conductivity per unit volume, specifically (cal/sec)/(cm2 C/cm).

http://hyperphysics.phy-astr.gsu.edu/Hbase/tables/thrcn.html#c1

That's a table of thermal conductivity k, in two equivalent sets of units. Another common unit is (Btu/hr)/ft2/(˚F/ft), or Btu/hr-ft-˚F. You may have thought this was k/unit volume -- it superficially looks like it might be -- but http://en.wikipedia.org/wiki/Thermal_conductivity describes how it is heat flow (Btu/hr) per unit area (ft2) per unit of linear temperature gradient (˚F/ft). Note how the length terms cancel to give ft in the denominator, not ft3.

As a spacecraft thermal engineer I always needed to beware of thermal conductivity units. Sometimes you'll see the area term given in ft2 and the gradient term in (˚F/in), resulting in the odd-looking unit for k of Btu-in/hr-ft2-˚F.

The SI unit of thermal conductivity is W/m-K.

Evan
08-20-2008, 05:36 AM
According to the link you gave it says:

"In other words, it is defined as the quantity of heat, ΔQ, transmitted during time Δt through a thickness x, in a direction normal to a surface of area A, due to a temperature difference ΔT, under steady state conditions and when the heat transfer is dependent only on the temperature gradient."

A surface of area A through a thickness x is a unit volume with t as the variable.

derekm
08-20-2008, 06:06 AM
Incidentally, conductivity of metals is normally ordered according to conductivity per unit volume. However, if you order the list by unit mass instead then aluminum comes out well ahead of the other metals.
since this is liquid cooled its going to be limited by the wetted area and flow rate. and how close you can get the liquid to the heat source. machining a labyrinth that runs in close to the part is going to be much better - Aluminium wins I think as there is much more scope than soldered tube.

derekm
08-20-2008, 07:21 AM
According to the link you gave it says:

"In other words, it is defined as the quantity of heat, ΔQ, transmitted during time Δt through a thickness x, in a direction normal to a surface of area A, due to a temperature difference ΔT, under steady state conditions and when the heat transfer is dependent only on the temperature gradient."

A surface of area A through a thickness x is a unit volume with t as the variable.

The quantity of heat per unit time per unit of temperature difference is proportional to the area but inversely proportional to the thickness all nice and intuitive. But here's the rub because one is proportional and one inversely proportional per unit volume it isnt - so its per unit length Not so intuitive

Evan
08-20-2008, 07:34 AM
OK, thanks for the correction Allan and Derekm.


[added]
Conductivity is however actually characterized by unit volume although not ordered so in the tables (my mistake).


The nature of conductivity in metals may be studied in greater depth by considering the electrons within the bulk metal. This approach is termed microscopic, in contrast to the macroscopic properties of a metal specimen. Under the influence of an internal electric field in the material, the electron cloud will undergo a net drift toward the battery anode. This drift is very slow in comparison with the random thermal motions of the individual electrons. The cloud may be characterized by the concentration of electrons, defined as total number per unit volume: where n is the concentration, N the total number, and U the volume of metal (U is used here for volume instead of V, which as an algebraic symbol is reserved for voltage).

Unfortunately the formulae are missing from the page: http://science.jrank.org/pages/2322/Electrical-Conductivity-Metals.html

Swarf&Sparks
08-20-2008, 08:26 AM
I've turned a bit of copper for thermal/electrical use (spot welders)
Soluble oil/water emulsion works fine, same as stainless.
They are similar to machine (gummy, and work harden), though a gulf apart in elctro-thermal.