Induction brass annealer redux

[Bora72]

Testing the mechanical quality is a bit of a challenge. If you had a hardness tester you could compare your annealed brass to commercial virgin brass. Without a tester it is somewhat guesswork. If you get a dull red color for your neck and neck/shoulder junction and not exceed 450 from about mid case to the head, you’re probably OK. If you lightly squeeze the neck and it doesn’t spring back you probably went too hot.

Based on the photos and some guessing, it looks like the inductance of your coil is somewhere around 1.7 to 2 uH. It should be closer to 1uH. I think you would be better served using 1/8” tubing.
Use this calculator to calculate your inductance. Select multi-layer, multi-row.

Coil Inductance Calculator

Use this online calculator to determine the inductance of single-layer and multilayer air-core coils. Both metric and English units of measurement are supported.

66pacific.com

Ok for all.
The coil inductance is about 1,25uH measured with my LCR meter.

 

[rox]

If you had a hardness tester you could compare your annealed brass to commercial virgin brass.

Is the final anneal of commercial brass always after all sizing processes are complete? If not, the target hardness might need to be a little softer than commercial brass, since it would have hardened due to any post-anneal sizing. It's easy to imagine the bean-counters putting a stop to any annealing after all the work on the case at the factory is complete.

I think I read that Norma brass *is* annealed at the factory after all sizing processes are finished though. I'll try and find out from Lapua what they do.

 

[rox]

Testing the mechanical quality is a bit of a challenge. If you had a hardness tester you could compare your annealed brass to commercial virgin brass. Without a tester it is somewhat guesswork.

Not long ago there were some Webster hardness testers available at around the $100 mark, and these can work well for case necks with a little modification, but prices seem to have soared. It's worth keeping a lookout for them though.

One thing that often comes-up in flame annealing circles is the annealing 'ladder test'.
Basically you anneal a series of cases with incremental anneal time, then you size the cases (without adjusting the die). As the anneal time increases the head-to-shoulder dimension starts to change (as the brass springback changes), but at a certain point the graph flattens and the dimensions do not change with increasing anneal time any more. The point at which the graph flattens indicates optimum anneal time, goes the theory. Perhaps that point is the cusp of full recrystallization, which some might argue is too soft. I've never tried this test, so use at your own risk, but it would be interesting to see what hardness results at that point. Once I have my system working I'll run such a test and compare with virgin brass hardness.

 

[Bora72]

Is the final anneal of commercial brass always after all sizing processes are complete? If not, the target hardness might need to be a little softer than commercial brass, since it would have hardened due to any post-anneal sizing. It's easy to imagine the bean-counters putting a stop to any annealing after all the work on the case at the factory is complete.

I think I read that Norma brass *is* annealed at the factory after all sizing processes are finished though. I'll try and find out from Lapua.

Hello rox, this info is very interesting.

If you grab info about Lapua annealing procedure please tell us!

 

[Bora72]

Not long ago there were some Webster hardness testers available at around the $100 mark, and these can work well for case necks with a little modification, but prices seem to have soared. It's worth keeping a lookout for them though.

One thing that often comes-up in flame annealing circles is the annealing 'ladder test'.
Basically you anneal a series of cases with incremental anneal time, then you size the cases (without adjusting the die). As the anneal time increases the head-to-shoulder dimension starts to change (as the brass springback changes), but at a certain point the graph flattens and the dimensions do not change with increasing anneal time any more. The point at which the graph flattens indicates optimum anneal time, goes the theory. Perhaps that point is the cusp of full recrystallization, which some might argue is too soft. I've never tried this test, so use at your own risk, but it would be interesting to see what hardness results at that point. Once I have my system working I'll run such a test and compare with virgin brass hardness.

Also if you take this test let us know!

 

[SGK]

I have published a link to the Norma info here before. Search my posts.

 

[David101]

I would bet that there are multiple anneals during the manufacturing process.
Nice work but so far removed from the original and way more complicated than needed for just about everyone else. Perhaps it deserves it own thread. I am concerned it would put people off attempting to make one of the basic units.

 

[rox]

Nice work but so far removed from the original and way more complicated than needed for just about everyone else. Perhaps it deserves it own thread. I am concerned it would put people off attempting to make one of the basic units.

So using a flame sensor to 'gauge temperature' and having savable programs to anneal by time or Joules = "Nice build ... new addition for others to try, Well done", but following the ideas suggested here by others to refine temperature control and avoid SSR/Contactor problems is "way more complicated than needed", "would put people off" and should be in it's "own thread"?

I'm confused.

 

[oliverpsmile]

Now is the time to implement automatic voltage control (PID of course) to assure the smoothest annealing process ever created. Switching the ZVC by dedicated SSR or gate control by any methods is still discrete control. The smoothness is somehow limited by the semiconductor required time to switch modes (in addition - timing for on or off state are not equal).

 

[Chris82]

Has anyone made a step-by-step video on how to build one of these annealers?? I'm more mechanically minded and have never built anything like this before.

 

[SGK]

Now is the time to implement automatic voltage control (PID of course) to assure the smoothest annealing process ever created. Switching the ZVC by dedicated SSR or gate control by any methods is still discrete control. The smoothness is somehow limited by the semiconductor required time to switch modes (in addition - timing for on or off state are not equal).

Given how fast the IRFP260N can switch* a discrete on/off response to fine temperature boundary will be well within the realms of any sensible level of 'accuracy'. One's ability to measure the hardness of the brass and so calibrate for time (for a given temp) will present far more error/variance (let alone case to case variance). (The ability of the power supply to respond will be a greater limiting factor than the response of the HEXFETs. And of course there's the response time of the pyrometer and all other components in the chain.) You guys are arguing about completely spurious 'accuracy.'

*td(on) 17 ns; tr 60 ns; td(off) 55 ns; tf 48 ns

 

[oliverpsmile]

You guys are arguing about completely spurious 'accuracy.'

But of course. You are absolutely correct. The problem is how fast can you drive the HXFETs.

 

[Foreigner]

First i want to thank everyone that helped me in the past. My device is almost ready… , just the basic version ( more then difficult enough for me) Everything works only de induction itself. When the induction starts the 48 volt drops down and the coil doesn’t heat up. I’ve bought eveything from the list so this should be right. Any idea why the voltage drops down?

 

[itchyTF]

How much does it drop? What is the current rating for the supply?
Are you monitoring current? If so, how much is that?

Could be a shorted coil or something on the ZVS board.

 

[Foreigner]

How much does it drop?
Are you monitoring current? If so, how much is that?

Could be a shorted coil or something on the ZVS board.

It drops to 2V 0,8 Amp

 

[itchyTF]

It drops to 2V 0,8 Amp

From 48v to 2v and only 800 ma?

 

[Foreigner]

Yes and when the seastos jump to B it goes back to 48V but 0 amp

 

[itchyTF]

Wonder if the supply is folding back. Can you load the supply to its maximum current to make sure it functions correctly? You would need around 2.5 ohms and 1000 watts. You can get by with less wattage as long as the connection is short. You can also use a bunch of incandescent bulbs in parallel.

 

[Chris82]

Hey Everyone,
I really need a bit of guidance in building one of these induction annealers. I have never build anything electronic before and am worried that I haven't ordered the right components:

- Power Supply 110V-240V to DC 24V 25A 600W
- 24V-36V ZVS Board (not sure on wattage, it gives "found" results of melting copper, gold at 36V needing 700-900W)
- DH48S-S Time relay

I just want a simple design that the induction coil is timed on and off with a manual trap door to drop the cases. I know it's a dumb question but what size toggle switch will I need between the 240V supply from the wall and the Time Relay?? I can only find 250V 16A rocker switches but wouldn't I need a 25A switch because of the power supply?? I feel so stupid asking. I'd rather be pulling down an engine and rebuilding it, I can do that with my eyes shut.

 

[txtaxman]

Hey Everyone,
I really need a bit of guidance in building one of these induction annealers. I have never build anything electronic before and am worried that I haven't ordered the right components:

- Power Supply 110V-240V to DC 24V 25A 600W
- 24V-36V ZVS Board (not sure on wattage, it gives "found" results of melting copper, gold at 36V needing 700-900W)
- DH48S-S Time relay

I just want a simple design that the induction coil is timed on and off with a manual trap door to drop the cases. I know it's a dumb question but what size toggle switch will I need between the 240V supply from the wall and the Time Relay?? I can only find 250V 16A rocker switches but wouldn't I need a 25A switch because of the power supply?? I feel so stupid asking. I'd rather be pulling down an engine and rebuilding it, I can do that with my eyes shut.

No, you do not need a 25A switch for the 240 Volt input to a 600 Watt power supply.
The formula for power is: P (watts) = I (current) x E (volts), so the switch you referenced would handle: 250 Volts x 16 Amps = 4000 watts.