Induction brass annealer redux

[ST42]

Oh, 9.2-9.8 amps

 

[itchyTF]

9.2 at 45v and 9.8 at 48v?

 

[ST42]

yes

 

[Ron Blain]

I dropped the PS voltage to minimum.
1. PS at idle, 45.4 V
2. ZVS, annealing, no case, 44.2 V
3. ZVS at idle, 0 V
4. ZVS annealing with case in the coil, 3.5 V and then failure. Again it was just a flash on the voltmeter.

Yes and what that tells me is the power supply is folding over. If you were measuring current to your ZVS module it would show you that when a case is inserted your current exceeds what your supply can deliver. This is when the supply folds over as a means of survival. The likely reason is the module ceases to oscillate and one or the other MOSFET on the ZVS module is in full conduction and the other is off. If things did not shut down one of the MOSFETS would fail as a Source Drain short.

The fact that it works fine till a case is inserted leads me to believe the inductance of the coil changed enough to cause oscillation to change enough that it stopped. The Coil and the Capacitors on the board form a LC network, commonly called a tank circuit or resonant circuit. What bothers me is the change of inductive value placing a cartridge in the coil should not effect the frequency of the tank circuit to make it stop oscillating. Then too, it's the only change happening. Typical frequency on these things is about (a big about) 18 KHz to 24 KHz.

Ron

 

[ST42]

Thank you Ron. You obviously have a very well versed understanding of this science. I unfortunately do not.
What is your suggestion to get this running? New coil? or new PS?

Thank you very much,
ST42

 

[ST42]

If I work up a new coil, is there another copper tubbing diameter that would be easier to obtain the desired frequency? It seems odd that it is not working with the present coil. I know many other builders have used the same recipe with success.

 

[itchyTF]

My understanding is that the frequency runs around 100KHz.

I made multiple coils based on the case size. I tried to keep the ID a little more than the case OD. Also, I tried to keep the case mouth just below the top turn and the bottom turn around 1/4” below the shoulder/body junction. Also tried to keep the inductance to around 1uH.

I would suggest a power supply with more wattage. I built one for a friend using a 36v, 1,000W supply. Some cases were drawing in excess of 20 amps.

 

[itchyTF]

There are numerous inductance calculators but you could use this one to monitor (or predict) your coil inductance.

Inductance calculator

 

[Ron Blain]

If I work up a new coil, is there another copper tubbing diameter that would be easier to obtain the desired frequency? It seems odd that it is not working with the present coil. I know many other builders have used the same recipe with success.

This is what bothers me. It works right till a case goes into the work coil. OK, as mentioned there are many of these things working just fine. Aside from the coil inductance changing the only other change is there will be an increase in current draw from the supply. Now thinking further about this I have to wonder if the power supply is simple wimping out? I played around with a ZVS module years ago but running 24 volts as I recall. My power supply was an old linear type weighing about 1/2 of what I do.

My earlier guess was the module quit oscillating because of a shift in inductance of the work coil. Maybe I need to reconsider that. Maybe when a case goes in and the demand for more power happens the supply just folds over because it can't deliver the current and maintain the voltage out.

I would suggest a power supply with more wattage. I built one for a friend using a 36v, 1,000W supply. Some cases were drawing in excess of 20 amps.

I am thinking this is what is going on.

Ron

 

[ST42]

Thanks Ron. I have two new power supplies inbound. Hopefully one of them will work out. I will keep you all updated. Once again, Thank You all for your help.

ST42

 

[Ron Blain]

Please do keep things updated. I really would like to know. It's a shame you have to go through all of this. As I mentioned I played around with the concept several years ago. Never overly pursued it. Never really took notes on my current draw and today haven't a clue what the numbers were. I used an ole linear power supply which is a brute and ran things at 24 volts. Anyway I will be curious as to how things work out for you so please let us know.

Ron

 

[VenatusDominus]

I had a similar problem with my first ZVS board. I inadvertently energized it without the coil attached and the board produced similar symptoms to what you are describing.

 

[hdmunger]

Forgot the coil,

I don't know if you took the pic with the case so high in the coil just for demonstration purposes, but if you can heat up the tip of a screwdriver, then maybe if you lowered the case until the shoulder was just above the bottom coil then there might be enough current in the PS to heat up the lesser amount of brass in the coil...in this case, just the shoulder and neck...

 

[ST42]

VenatusDominus, I was very careful to not do that. I noticed a few builders made that mistake, and posted about doing so. Very grateful for those builders doing so too.

Hdmunger, the photo was to provide a size reference of the coil. When I do lower it so that just the case and shoulder are entering the coil from below, I can get it to anneal. However it is a Very small window, and it take almost 5 secs to get the case mouth to begin to glow. Is the annealing window of the coil usually so tight? Maybe I am expecting more from the build than I should.

 

[VenatusDominus]

VenatusDominus, I was very careful to not do that. I noticed a few builders made that mistake, and posted about doing so. Very grateful for those builders doing so too.

Hdmunger, the photo was to provide a size reference of the coil. When I do lower it so that just the case and shoulder are entering the coil from below, I can get it to anneal. However it is a Very small window, and it take almost 5 secs to get the case mouth to begin to glow. Is the annealing window of the coil usually so tight? Maybe I am expecting more from the build than I should.

I made my coil inside diameter much smaller. The closer the coil is to the case, the less energy (amperes) is required. The annealing is taking place by the magnetic field produced by the coil and this field is subject to the inverse cube law with regards to distance. So even a small change in the distance from the coil to the case, will produce a significant change in the current draw. I would try winding a tighter coil and see if that reduces you current draw. My coil is a double layer, ID=0.645", 6.5 turns, with 1/8" copper tubing.

 

[hdmunger]

5 seconds doesn't sound all that long to me, considering the lower power output of your power supply. And, frankly, a little longer is not necessarily a bad thing...if your cycle is too fast, it makes the timing target window that much harder to determine.
Also, having too much of the body in the coil is less desirable, because it means your not just heating up the neck/shoulder, which is all you want annealed anyways.

Edit:
Also, I agree with VD, your coil is bigger than it needs to be, and I also use the double layer design (much the same as VD's, except with 1/4" tubing). I believe that the double layer design focuses the coil magnetic field in a smaller vertical area. (I discuss making my coil here: https://forum.accurateshooter.com/t...annealer-redux.3908353/page-153#post-38361561 )

 

[ST42]

Gents,
She’s running like a Swiss watch!!!! I upgraded the power supply to this,

https://www.amazon.com/dp/B07FBQJTWZ?psc=1&ref=ppx_yo2ov_dt_b_product_details

My annealing, 5.56 cases, is running at 4.3 seconds, and I am drawing 8.9 amps.
I’ll get a video up once I get my bench cleaned up and I establish a YouTube channel.
VD, and hdmunger. I am going to try another coil based off of your suggestions. I really like the idea of 1/4” line to facilitate cooling.
itchyTF, and Ron. Thank you for all of your help gentlemen. The power supply upgrade was the trick for me.
Lastly, Thank you Gina and Erick! Had I not found your your thread, I never would have started my investigation into building this annealer. It has been Very rewarding!!!

 

[itchyTF]

Looks like the same supply I used for my friend’s build.
1/8” tubing cools fine and I think it provides more flexibility in coil design.
As I previously mentioned, the ID of my coils are a little more than the case OD for better magnetic coupling. I believe a 5.56/.223 case annealed in around 1.4 seconds. I know it was less than 2.

 

[WildBill3/75]

Over 3,000 posts in this thread.
This thread should really be a sticky by now.

Love all the open source information and how this project has evolved.

After extensive research I was inspired to buy the mgnz Arduino + mgnz shield as well as a 3d printer and go that build route with my own ideas thrown in there.


Word to the wise, Amazon is selling the exact same components you can buy on AliExpress, with a large markup... That was a no brainier to just buy direct.

I was at least able to source Mean Well Taiwanese power supplies... Damn near everything else is only produced in China...

Should be a fun project.

Thanks to all that contributed to this thread and on other sites.

 

[ASD9055]

Over 3,000 posts in this thread.
This thread should really be a sticky by now.

Love all the open source information and how this project has evolved.

After extensive research I was inspired to buy the mgnz Arduino + mgnz shield as well as a 3d printer and go that build route with my own ideas thrown in there.


Word to the wise, Amazon is selling the exact same components you can buy on AliExpress, with a large markup... That was a no brainier to just buy direct.

I was at least able to source Mean Well Taiwanese power supplies... Damn near everything else is only produced in China...

Should be a fun project.

Thanks to all that contributed to this thread and on other sites.

Looking forward to hear about your progress. Maybe you can post a list of the materials and design to help others.
Cheers