Difficulty with machining bismuth, any suggestions/tips/other ideas?

[mesa]

Hello all, we are working with some bismuth for a series of upcoming experiments and are trying to initially keep the purity high for better neutron activation measurements. Unfortunately there has been some difficulty with machining steps on the final pieces with chipping due to the material being so brittle.

Because of this we have been developing casting techniques to minimize machining steps via lost PLA methods from 3D printed specimens (which is also good for keeping contaminants low). This has given very good results with near perfect form as cast, but there are still machining steps to eliminate the sprue, vents, etc.

This particular process has extended cooling times which may be part of the problem from large grain growth.
Would a heat treat and quench potentially help? Normally the opposite would be true, but these crystalls are overwhelmingly large (when they break these 'chunks' are, well.. very chunky).

We can look into alloying for future specimins, but the first needs to maintain high purity. Any other ideas/input are welcome as well!

 

[jrmichler]

I once attended a seminar on ductile machining of brittle materials (search those terms). They were able make ductile chips from machining glass. The process results in a glass smooth surface. Also search diamond turning.

The challenge is to make extremely precise, extremely shallow cuts with an extremely sharp cutting tool. The depth of cut is in microns or even nanometers, and the cutting tool edge radius is a small fraction of the depth of cut. I believe that the cutting tool is usually a diamond.

Grinding with a sharp diamond wheel is a possibility. A flood of coolant might help.

 

[anorlunda]

I once attended a seminar

Your broad knowledge of M.E. subjects is amazing. We're lucky to have you.

 

[Lnewqban]

What type of machining and cuting tools are you using?

 

[DaveE]

I was thinking EDM. But, really, this is way above my pay grade. Maybe you should call these guys and ask them: https://www.thomasnet.com/profile/0...ismuth&heading=5291356&cid=453035&searchpos=7

 

[mesa]

I once attended a seminar on ductile machining of brittle materials (search those terms). They were able make ductile chips from machining glass. The process results in a glass smooth surface. Also search diamond turning.

The challenge is to make extremely precise, extremely shallow cuts with an extremely sharp cutting tool. The depth of cut is in microns or even nanometers, and the cutting tool edge radius is a small fraction of the depth of cut. I believe that the cutting tool is usually a diamond.

Grinding with a sharp diamond wheel is a possibility. A flood of coolant might help.

We haven't tried the diamond, but it seems the machinist managed to get a decent(ish) finish playing with speeds/feeds/cutting tools. Here is a before:



The 'fluffy' areas are where the crystal orientation is 'off', but after some testing he got it to this:



The issue now is with tapping the hole. The material was cast taking into consideration the expansion of the material as it solidifies (bismuth is one of those weird elements with this property), and even though we cast the theards, they need chasing afetr the fact. Unfortunately chipping is proving to be difficult to avoid yet again:



Good suggestions none the less, thanks for the tips!

Any other ideas on how to proceed from here?

 

[mesa]

I was thinking EDM. But, really, this is way above my pay grade. Maybe you should call these guys and ask them: https://www.thomasnet.com/profile/0...ismuth&heading=5291356&cid=453035&searchpos=7

EDM could do it for the straight cuts, so not a bad idea. We don't have one at our shop, but could farm that out, although our guy is getting pretty good at machining this part now (except for chasing the threads that is).

 

[Tom.G]

It sounds obvious and impossible but, how about asking your magician machinist to use the same approach on the threads as he used when facing the part?

Please keep us updated, we're curious too?

 

[mesa]

It sounds obvious and impossible but, how about asking your magician machinist to use the same approach on the threads as he used when facing the part?

Please keep us updated, we're curious too?

Unfortunately the machining steps used to face the pieces are different than that used to tap threads. The hole is only a 3/8" 16tpi, so you have to use a tap.

Regardless, I am sure our machinist will appreciate being called a 'magician'.

Today we are attempting a heat treat and quench to reduce the crystal grain size, should know by the end of the day if this is useful/helps.

 

[Lnewqban]

Could the design be modified to avoid that thread and any other surface requiring slow machining?
The way it is now could compromise the calculated load capability of that thread.

 

[jrmichler]

Some possibilities, none of which I have personal experience with:

1) Thread milling using a carbide tool and light cuts.
2) Thread grinding. If the surface wants to spall, use a diamond wheel and flood with coolant.
3) Electrochemical machining. From the Wikipedia article: High metal removal rates are possible with ECM, with no thermal or mechanical stresses being transferred to the part, and mirror surface finishes can be achieved.
4) EDM can chase threads given the right tool.

All four of the above suggestions require a CNC machine tool with the appropriate axes.

 

[Matcon]

Hello all, we are working with some bismuth for a series of upcoming experiments and are trying to initially keep the purity high for better neutron activation measurements. Unfortunately there has been some difficulty with machining steps on the final pieces with chipping due to the material being so brittle.

Because of this we have been developing casting techniques to minimize machining steps via lost PLA methods from 3D printed specimens (which is also good for keeping contaminants low). This has given very good results with near perfect form as cast, but there are still machining steps to eliminate the sprue, vents, etc.

This particular process has extended cooling times which may be part of the problem from large grain growth.
Would a heat treat and quench potentially help? Normally the opposite would be true, but these crystalls are overwhelmingly large (when they break these 'chunks' are, well.. very chunky).

We can look into alloying for future specimins, but the first needs to maintain high purity. Any other ideas/input are welcome as well!

For removing sprues and vents, surely a wire diamond saw, such as those used for slicing semiconductors, is worth considering?

They are designed to avoid surface damage on brittle materials

 

[mesa]

Well, here are the results of the heat treat and quench:



The piece on the right was heat treated and quenched while the piece on the left was faced and threaded as it came from the mold. As we can see there is a marked improvement on chipping with the quenched part (both are from the same melt/cast/mold).

We brought the piece to within 20C of melting and held it for over 3 hours followed up by an immediate quench in water.

Although this is a big improvement, the threads are still experiencing an unacceptable level of chipping (may be a little difficult to see in this pcture, but there are a few 'chunks' missing on some portions of the threads down the borehole):



We plan on trying an oil quench since the specific heat is high, and boiling is above our material temp so less gas pockets from boiling so a better quench. We only need recrystallization of near surface for machining as the part is very close to its final form factor.

For kicks we figured a liquid nitrogen quench would be fun (although it only has being super cold going for it so probably won't work as well as the oil), but we will try it anyway.

 

[mesa]

Could the design be modified to avoid that thread and any other surface requiring slow machining?
The way it is now could compromise the calculated load capability of that thread.

We have considered that, although we seem to be making decent progress; we will see how deep down the rabbit hole we need to go to see if this is a fruitful path.

The total load capacity is fairly low and the threads, as is, are well within our factor of safety for use, but we don't like inserting pieces with rough finishes in these reactors (well machined parts also lend the ROs and SROs more confidence in our pieces/processes).

Either way, good suggestion and certainly up for consideration if the threads prove to be a no-go!

 

[mesa]

Some possibilities, none of which I have personal experience with:

1) Thread milling using a carbide tool and light cuts.
2) Thread grinding. If the surface wants to spall, use a diamond wheel and flood with coolant.
3) Electrochemical machining. From the Wikipedia article: High metal removal rates are possible with ECM, with no thermal or mechanical stresses being transferred to the part, and mirror surface finishes can be achieved.
4) EDM can chase threads given the right tool.

All four of the above suggestions require a CNC machine tool with the appropriate axes.

We have been threading on the mill with indicating for precision placement of the tap (which helped vs lathe cutting since this gives us a better 'feel' of the cutting).

The machinist is thinking of trying a roller tap (minimal fluting) as he says the material has a bit of 'stickiness' (like lead) when running the tap (surprising considering the brittleness that makes this so hard to thread).

 

[mesa]

For removing sprues and vents, surely a wire diamond saw, such as those used for slicing semiconductors, is worth considering?

They are designed to avoid surface damage on brittle materials

That could certainly work, although (especially with the heat treatment) we seem to be able to get a really nice surface with standard machining tools now once the speeds ad feeds are set up correctly.

Our biggest issue now is getting the holes tapped for good thread engagement/looks. We are thinking of digging a little deeper on the heat treat/quench to see if we can make this work along with a roller tap (minimal fluting).

Fingers crossed!

 

[synch]

just a thought - bismuth has a low melting point, maybe heat the cutting tool to above the mp and melt-cut it ? Like a hot wire through polystyrene foam etc. A hollow bolt might melt-form the thread in-situ as it screws in, hopefully molten bismuth does not weld to steel .
ed- Ultrasonic welding uses local heating effects, maybe an ultrasonic cutter would help

 

[mesa]

just a thought - bismuth has a low melting point, maybe heat the cutting tool to above the mp and melt-cut it ? Like a hot wire through polystyrene foam etc. A hollow bolt might melt-form the thread in-situ as it screws in, hopefully molten bismuth does not weld to steel .
ed- Ultrasonic welding uses local heating effects, maybe an ultrasonic cutter would help

That is an interesting idea.

We have gotten pretty good at machining this stuff post heat treat, but it is still a pain. Certainly is worth a shot on the next run and is something we haven't thought about; I'll let you know if we have any success.