DIY Electrical Annealing Of Rifle Cases
- Thread starter JeepHammer
- Start date
What you are seeing is a reaction to the magnetic field due to eddy currents induced in the brass. Any conductor will do this. It's not a ferromagnetic phenomenon; just a magnetic one. JH is using ceramics (non-conductor) and/or fiberglass to keep the case in place. It doesn't stop the field from being there or from pushing.
An old fashioned demo used to be to make up a copper or aluminum disc and put radial slots in half of it, then mount it on a hand-cranked shaft oriented so the outer edge of the disc passed between the poles of a strong magnet. When you rotate the disc with the crank, it gets very hard to turn when the solid half of the disc is going between the PM poles, but suddenly gets much easier when the slotted part goes through. What is happening is the copper moving through the magnetic field induces eddy currents that have their own magnetic field that interacts with the permanent magnet to produce drag. Basically, it's a generator with a shorted output. But when the slots pass through they interrupt the eddy currents so they are much smaller and weaker and you don't feel much. But note that this works with aluminum or any other conductor, too, and is the principle the little magnetic damper on beam scales depends on to work.
In the case of the induction heater, you can look at the freqency of operation as the rotation rate of a virtual "disc" and the case as an unslotted disc. In this instance the eddy currents are of such magnitude that they make the brass hot. But they also have that induced magnetic field reacting against the field in the coil. BTW, you can put multiple ac magnets of opposing polarity, as you see with the poles of an induction motor stator, and tweak the phase (different start capacitances will do this in the motor stator) to cause the field reactions to either attract or repel a conductor. You could, in fact, make a brass magnet this way to pick up your brass. The problem is the power requirement is large for the amount of work it actually does.
Jeephammer,
The letter Q actually stands for "Quality". The electrical quality of an inductor has to do with how quickly it dissipates ac energy that is cycling through it at a given frequency. A coil that dissipates a lot of energy is said to be lossy or low quality for the application, and vice versa. Both inductive reactance and resistance are measured in ohms, but only the resistance dissipates any energy, so the Q of a coil is defined as its inductive reactance at the frequency you are finding Q for, divided by the resistance of the coil. The ohms cancel and you are left with a unitless ratio called Q or the Q Factor of the coil. As long as this is a high number, you can can take the ac volts times the ac current through the coil and divide by the Q to get close to how many watts are being dissipated in the coil itself. As Q gets lower, the phase shift between the reactance and the resistance, aka, the Power Factor, has to be factored into the calculation by multiplying voltage and current by the cosine of the phase angle between voltage and current. But we don't have to go there in this case. The bottom line is that you would ideally have a high Q coil for the induction heater so that it lost very little of the energy you put into operating the system. You would like the losses all to be in the brass, where the heat is needed. Physical reality just makes it hard to get there. Litz wire gives you high Q, but costs you the ability to easily get rid of whatever heat is still made.
An old fashioned demo used to be to make up a copper or aluminum disc and put radial slots in half of it, then mount it on a hand-cranked shaft oriented so the outer edge of the disc passed between the poles of a strong magnet. When you rotate the disc with the crank, it gets very hard to turn when the solid half of the disc is going between the PM poles, but suddenly gets much easier when the slotted part goes through. What is happening is the copper moving through the magnetic field induces eddy currents that have their own magnetic field that interacts with the permanent magnet to produce drag. Basically, it's a generator with a shorted output. But when the slots pass through they interrupt the eddy currents so they are much smaller and weaker and you don't feel much. But note that this works with aluminum or any other conductor, too, and is the principle the little magnetic damper on beam scales depends on to work.
In the case of the induction heater, you can look at the freqency of operation as the rotation rate of a virtual "disc" and the case as an unslotted disc. In this instance the eddy currents are of such magnitude that they make the brass hot. But they also have that induced magnetic field reacting against the field in the coil. BTW, you can put multiple ac magnets of opposing polarity, as you see with the poles of an induction motor stator, and tweak the phase (different start capacitances will do this in the motor stator) to cause the field reactions to either attract or repel a conductor. You could, in fact, make a brass magnet this way to pick up your brass. The problem is the power requirement is large for the amount of work it actually does.
Jeephammer,
The letter Q actually stands for "Quality". The electrical quality of an inductor has to do with how quickly it dissipates ac energy that is cycling through it at a given frequency. A coil that dissipates a lot of energy is said to be lossy or low quality for the application, and vice versa. Both inductive reactance and resistance are measured in ohms, but only the resistance dissipates any energy, so the Q of a coil is defined as its inductive reactance at the frequency you are finding Q for, divided by the resistance of the coil. The ohms cancel and you are left with a unitless ratio called Q or the Q Factor of the coil. As long as this is a high number, you can can take the ac volts times the ac current through the coil and divide by the Q to get close to how many watts are being dissipated in the coil itself. As Q gets lower, the phase shift between the reactance and the resistance, aka, the Power Factor, has to be factored into the calculation by multiplying voltage and current by the cosine of the phase angle between voltage and current. But we don't have to go there in this case. The bottom line is that you would ideally have a high Q coil for the induction heater so that it lost very little of the energy you put into operating the system. You would like the losses all to be in the brass, where the heat is needed. Physical reality just makes it hard to get there. Litz wire gives you high Q, but costs you the ability to easily get rid of whatever heat is still made.
JeepHammer
Moderator
Unclenick is correct, I use ceramic to physically restrain the brass from moving,
While the fiberglass is heat resistant insulation to keep brass moving around from shorting out or changing resistance of the coil.
Fiberglass strips (pre made) guide brass into the ferrite on production feeder, keeping the brass from moving around too much and gets the necks into the ferrite without snagging.
On a side note, related but not directly applied to anealing,
What you are seeing is kind of the same principle a 'Rail Gun' works on.
Non-magnetic but electrically conductive between capacitively charged conductors.
A MUCH weaker version of the eddie current effect used on rail guns to get a non-magnetic projectile moving.
While the fiberglass is heat resistant insulation to keep brass moving around from shorting out or changing resistance of the coil.
Fiberglass strips (pre made) guide brass into the ferrite on production feeder, keeping the brass from moving around too much and gets the necks into the ferrite without snagging.
On a side note, related but not directly applied to anealing,
What you are seeing is kind of the same principle a 'Rail Gun' works on.
Non-magnetic but electrically conductive between capacitively charged conductors.
A MUCH weaker version of the eddie current effect used on rail guns to get a non-magnetic projectile moving.
DIY innealer (you build)
Hi guys...
Thought my may be interested in this youtube video. The link from the video will take you to www.accurateshooter.com and a thread on building this induction annealer. Parts and where to get them.
Works well (almost easy) to build. So far I know of, 10 that have been built and everyone likes theirs.. Cost is about $350-400.
Check it out
https://www.youtube.com/watch?v=57RL5v54FhM
Gina
Hi guys...
Thought my may be interested in this youtube video. The link from the video will take you to www.accurateshooter.com and a thread on building this induction annealer. Parts and where to get them.
Works well (almost easy) to build. So far I know of, 10 that have been built and everyone likes theirs.. Cost is about $350-400.
Check it out
https://www.youtube.com/watch?v=57RL5v54FhM
Gina
JeepHammer
Moderator
What I don't like about the video,
The case is being dropped into position with the timer running, so hesitation getting the case into place wastes time and case doesn't reach target temp.
Timer NEEDS to start when case is in place for consistancy so each case gets full annealing.
This is pretty easy to do from mechanical switch to proximity sensor.
This *Looks* like a continuously 'ON' coil unit, but I just can't tell from the video. It could be switched on/off by the timer, but that means the 'Off' time is tiny, it would have to be a pretty much 100% duty cycle induction unit.
Switched (On/Off) units NEED superior switching semi-conductors, which most of the 'China' units I deal with don't have. The switching gives up...
So normally, I cool the coil with water/alcohol mix & run continuous, doing like this guy and controlling the time the case stays in the coil.
One word of caution here, up around 1,000 or 1,200 watts & ferrite, particularly using Litz wire around the ferrite, you CAN seriously overheat the case neck, going well beyond 800*F pretty easily.
A slower 'Cook' will allow heat to creep further down the case, but will allow you to control the maximum heat applied to the neck/shoulder.
I'm still not knocking into the 450*F mark, just annealing creeping a little further down the case.
The open tubing coil helps with control, but again, it takes a little longer per case.
Ferrite REALLY focuses the magnetic field so you get a much more concentrated cook, and that is faster (time wise), but it's much easier to over cook the brass.
The smaller the coil, the closer the case is to the coil, and the faster the case reaches target temp. Rewinding the coil so it's tight to the case is more difficult with smaller tubing, but worth the effort...
Use fiberglass wrap to keep Case from contacting coils directly.
Fiberglass is electrical insulator, heat proof at these temps, and abrasion resistant so it lasts a good long time.
The case is being dropped into position with the timer running, so hesitation getting the case into place wastes time and case doesn't reach target temp.
Timer NEEDS to start when case is in place for consistancy so each case gets full annealing.
This is pretty easy to do from mechanical switch to proximity sensor.
This *Looks* like a continuously 'ON' coil unit, but I just can't tell from the video. It could be switched on/off by the timer, but that means the 'Off' time is tiny, it would have to be a pretty much 100% duty cycle induction unit.
Switched (On/Off) units NEED superior switching semi-conductors, which most of the 'China' units I deal with don't have. The switching gives up...
So normally, I cool the coil with water/alcohol mix & run continuous, doing like this guy and controlling the time the case stays in the coil.
One word of caution here, up around 1,000 or 1,200 watts & ferrite, particularly using Litz wire around the ferrite, you CAN seriously overheat the case neck, going well beyond 800*F pretty easily.
A slower 'Cook' will allow heat to creep further down the case, but will allow you to control the maximum heat applied to the neck/shoulder.
I'm still not knocking into the 450*F mark, just annealing creeping a little further down the case.
The open tubing coil helps with control, but again, it takes a little longer per case.
Ferrite REALLY focuses the magnetic field so you get a much more concentrated cook, and that is faster (time wise), but it's much easier to over cook the brass.
The smaller the coil, the closer the case is to the coil, and the faster the case reaches target temp. Rewinding the coil so it's tight to the case is more difficult with smaller tubing, but worth the effort...
Use fiberglass wrap to keep Case from contacting coils directly.
Fiberglass is electrical insulator, heat proof at these temps, and abrasion resistant so it lasts a good long time.
I don't think so. The timer is on the left display and is behind the operator's arm most of the time, so it's a little hard to tell, but the coil is on only when the green LED above the work "annealing" turns on. That doesn't happen until the case is in place and on a second viewing you can see there is a pause during which the timer appears to count down briefly before turning on the power and counting up. So I think that countdown is the switch debounce and case settling time. In one run it doesn't trigger properly and the operator has to trigger it again. In the schematic you can see the green LED gets no power until the induction heater does.
JeepHammer
Moderator
I believe you are right. I watched it again.
The timer runs, seems to be tracking down time before the case is dropped in.
The timer seems to reset when the brass is dropped into the socket.
The timer runs, seems to be tracking down time before the case is dropped in.
The timer seems to reset when the brass is dropped into the socket.
Last edited:
Took me awhile being slow on the uptake but I see what they are doing and why they labeled it "Auto Cycle". The annealing is a 5.4 second process and between annealing you have 3 seconds to load the next cartridge in. The counter increments to 3.0 seconds and the process starts. The process runs for 5.4 seconds, the annealed case drops and 3.0 seconds later the process repeats. Also noticed that they just turn the entire "Inductor PCB" On and OFF. That apparently works well as initially I was thinking about just gating the MOSFETS. Not really required, just turn the board On and Off using the contactor.
The 48 VDC supply powering the inductor pcb is showing about 43.1 VDC @ 12.2 Amps so they get about 526 watts. THe PCB Inductor Fan and the Radiator fan both run apparently on 120 VAC. I would likely use 12 VDC fans since the pump is 12 VDC. I don't know what the thinking was with the 120 VAC fans when 12 VDC home computer fans are inexpensive. Find a low cost CPU cooling radiator, pump and fan(s) and that should work. The 48 VDC 600 watt PSU won't be cheap, about $85 to $100 on Amazon. Using a cheap one shot timer would keep cost down. You just have to push a button for each cycle.
Ron
The 48 VDC supply powering the inductor pcb is showing about 43.1 VDC @ 12.2 Amps so they get about 526 watts. THe PCB Inductor Fan and the Radiator fan both run apparently on 120 VAC. I would likely use 12 VDC fans since the pump is 12 VDC. I don't know what the thinking was with the 120 VAC fans when 12 VDC home computer fans are inexpensive. Find a low cost CPU cooling radiator, pump and fan(s) and that should work. The 48 VDC 600 watt PSU won't be cheap, about $85 to $100 on Amazon. Using a cheap one shot timer would keep cost down. You just have to push a button for each cycle.
Ron
Maybe this description of the timer may help...
The timing heart of this annealer is the Seatos digital quartic timer relay switch.
It is in fact 4 timers in one unit Timers A, B, C, and D. Timers A and C have SPDT relays controlled by the timers. In this annealer timer "A" is used to set up the amount of time a case is being annealed. it controls the 30 amp contactor/relay and the power (48 volts) going to the inducter annealing PCB.
Timer B, is not used and is set to "zero" time.
Timer "C" is usually set to .5 seconds and controls the trap door solenoid. Dropping the newly annealed case in the collector pan.
Timer "D" is usually set to 2.5 to 3.0 seconds. and is used for a delay to allow the operator time to insert another case into the annealing coil.
In recycle mode, this A, B, C, D runs continually. Annealing one case right after another. Remember timer D is a "delay" to allow the operator to insert another case on autocycle. The operator sets it up.
As far as 110 volt AC fans.. Yes 12 volt DC fans are cheaper.. But that is what I had on hand already, so I used them. Did not have to spend any more $$. And as I said in the tread.. GO for 12 volt fans !!
The whole idea of this annealer is a relativity low cost high volume induction annealer. There have been variatrions to this design since it was first published. Read the entire thread to see them
The timing heart of this annealer is the Seatos digital quartic timer relay switch.
It is in fact 4 timers in one unit Timers A, B, C, and D. Timers A and C have SPDT relays controlled by the timers. In this annealer timer "A" is used to set up the amount of time a case is being annealed. it controls the 30 amp contactor/relay and the power (48 volts) going to the inducter annealing PCB.
Timer B, is not used and is set to "zero" time.
Timer "C" is usually set to .5 seconds and controls the trap door solenoid. Dropping the newly annealed case in the collector pan.
Timer "D" is usually set to 2.5 to 3.0 seconds. and is used for a delay to allow the operator time to insert another case into the annealing coil.
In recycle mode, this A, B, C, D runs continually. Annealing one case right after another. Remember timer D is a "delay" to allow the operator to insert another case on autocycle. The operator sets it up.
As far as 110 volt AC fans.. Yes 12 volt DC fans are cheaper.. But that is what I had on hand already, so I used them. Did not have to spend any more $$. And as I said in the tread.. GO for 12 volt fans !!
The whole idea of this annealer is a relativity low cost high volume induction annealer. There have been variatrions to this design since it was first published. Read the entire thread to see them
JeepHammer
Moderator
For a guy wanting to anneal 100 at a time it's not too bad, pretty functional and fairly easy to operate.
Get beyond about a 100 cases at a time and auto feed comes in REAL handy!
Pyrex tube or ceramic end on the Dillon tube, and putting a Dillon case feeder on it wouldn't be difficult and would sure free some time up for other things...
Get beyond about a 100 cases at a time and auto feed comes in REAL handy!
Pyrex tube or ceramic end on the Dillon tube, and putting a Dillon case feeder on it wouldn't be difficult and would sure free some time up for other things...
JeepHammer
Moderator
Everything is 'Rube Goldburg' until someone works the kinks out and puts it into a slick case...
And marks the price up 1,000% and retails it!
'Open Source' just keeps the manufacturers honest & prices reasonable!
Drop through or belt feeding is as old as the industrial revolution,
Electrical Magnetic annealing is as old as semi-conductors that made it practical,
Nothing new or revloutional here, it's sorely needed for guys that anneal.
And marks the price up 1,000% and retails it!
'Open Source' just keeps the manufacturers honest & prices reasonable!
Drop through or belt feeding is as old as the industrial revolution,
Electrical Magnetic annealing is as old as semi-conductors that made it practical,
Nothing new or revloutional here, it's sorely needed for guys that anneal.
fascinating thread. Love DIY projects, but I think I will pass on this one.
With a propane torch and a long 1/4 inch drive socket of the proper size chucked into a cordless drill I can anneal 50 cases in less than 30 minutes. The case spinning in the drill assures even heating and when the case neck glows I just tilt the drill to drop the case into the water
Of course I built a case dryer when the oven was working just fine to dry my cases so who am I to judge
With a propane torch and a long 1/4 inch drive socket of the proper size chucked into a cordless drill I can anneal 50 cases in less than 30 minutes. The case spinning in the drill assures even heating and when the case neck glows I just tilt the drill to drop the case into the water
Of course I built a case dryer when the oven was working just fine to dry my cases so who am I to judge
I'm always surprised, given the need to supply power to these simple oscillators abruptly enough to ring the tank circuit, that folks still use mechanical relays for it. Cheap high current P-channel MOSFETs can be had to do the job and electronic timers with high speed TTL compatible outputs are a dime a dozen. The schematic topography would be something like:
Attachments
Unclenick:
While I agree with you 100% the use of relays and contactors does not really surprise me. My experience with these annealing units is that the guys making them are not for the most part engineering types but rather electronic hobby enthusiast. Logic Level gated MOSFETS are a great solution and have the fast turn on times required. While they can be connected in crude fashion to really incorporate them you need a board and heat sink.
While neither is difficult they make a relay with screw or lug terminals, like an automotive relay lucrative. Anyway, this would be my best guess as to why relays and contactors lead in the popularity roll. Pure speculation on my part. Heck, I would also consider the use of some uC units. Figure you can buy uC chips pretty inexpensive and they are not that difficult to program for timing functions. While I have never really played around with them a Rasberry Pi can be had for less than about $35 and could likely run all the control functions for an annealer. I really don't know.
Ron
While I agree with you 100% the use of relays and contactors does not really surprise me. My experience with these annealing units is that the guys making them are not for the most part engineering types but rather electronic hobby enthusiast. Logic Level gated MOSFETS are a great solution and have the fast turn on times required. While they can be connected in crude fashion to really incorporate them you need a board and heat sink.
While neither is difficult they make a relay with screw or lug terminals, like an automotive relay lucrative. Anyway, this would be my best guess as to why relays and contactors lead in the popularity roll. Pure speculation on my part. Heck, I would also consider the use of some uC units. Figure you can buy uC chips pretty inexpensive and they are not that difficult to program for timing functions. While I have never really played around with them a Rasberry Pi can be had for less than about $35 and could likely run all the control functions for an annealer. I really don't know.
Ron
Last edited:
JeepHammer
Moderator
Not for this particular application, but I use a LOT of common relays in projects,
Common maintiance men can find a bad relay and replace it, no specalized test equipment or training required.
It's just too easy to add a LED to a relay to show when it's stopped working,
That makes it super easy for the guy doing maintiance to find the fault.
I can't imagine trying to keep up with a relay switching an induction circuit, the switching rate is so fast I can't imagine a relay would last long at all.
Common maintiance men can find a bad relay and replace it, no specalized test equipment or training required.
It's just too easy to add a LED to a relay to show when it's stopped working,
That makes it super easy for the guy doing maintiance to find the fault.
I can't imagine trying to keep up with a relay switching an induction circuit, the switching rate is so fast I can't imagine a relay would last long at all.
Well, the relay function here and the circuit I drew are just for switching the power to the oscillator circuit to turn it on and off. I figure if we have an open source design, that will mean the circuit and its components and circuit board layout will all be part of the package that is maintained. Adding the MOSFET on/off switch to the board then becomes a trivial addition. A simple 555 circuit could be the control timer, unless it is felt that digital is a must.
But if the timing goes digital, or we talk about adding a BASIC Stamp or Arduino or Raspberry Pi piggy back board or some other added degree of sophistication, then the MOSFET opens up the possibility of power control by pulse width modulation. That would let us heat fast at high average current, then back the average current off to hold a temperature for some predetermined amount of time. There are some long wave IR camera modules now that could be used to indicate the case temperature and adjust the modulation accordingly. We can make this as simple or as complicated as we want to.
But if the timing goes digital, or we talk about adding a BASIC Stamp or Arduino or Raspberry Pi piggy back board or some other added degree of sophistication, then the MOSFET opens up the possibility of power control by pulse width modulation. That would let us heat fast at high average current, then back the average current off to hold a temperature for some predetermined amount of time. There are some long wave IR camera modules now that could be used to indicate the case temperature and adjust the modulation accordingly. We can make this as simple or as complicated as we want to.