Induction brass annealer redux

[itchyTF]

Specs say minimum thickness .4mm (.0157”) and minimum ID 10mm (.393”). That would leave out most of the cartridges we shoot.

 

[ToddP]

In the video, the guy had to grind down a couple parts on the tester to accommodate a case, but he was able to do so (and, a few pages back, someone in this thread said he was using a Weber tester as well).

 

[ToddP]

I just finished reading all 179 pages. Whew, what a slog. Initially, I was very excited to give this project a go, but after reading the last 20 pages or so it sounds like there is an almost 100% chance that the ZVS and/or the power supply will show up DOA and that the relay (unless spending for an American made product) will crap out quickly due to below spec components used in its construction. With all that out of my system, I do have some questions:

1) If I go with a 12 VDC timer rather than the 110 VAC timer used in the GenErick 2.0 design, would the Packard C230 still work? If not, what would a comparable, reliable contactor be? (am I right in believing a contactor would be more reliable than a SSR without having to break the bank?)
2) Given the poor quality of ZVS boards, has anyone successfully replaced it's FETs (and/or other components)? I saw Gina was considering doing so, but never saw any follow up.
3) I got the feeling that Mean Wells were the power supplies to favor. Has anyone had bad experiences with them?
4) The Fluxeon ferrite coil is able to be purchased from Fluxeon ( https://fluxeon.com/product/annie-flux-concentrator-coil/ ). Is that a simple drop in replacement for the copper tube coil? Or is there tweaking required to mount it to the ZVS board?

 

[rox]

In the video, the guy had to grind down a couple parts on the tester to accommodate a case, but he was able to do so (and, a few pages back, someone in this thread said he was using a Weber tester as well).

Here are the modifications I had to do to mine. The bonded support pin was removed (comes out easily with heat). It needed to be turned-down off-center, otherwise you will exceed the travel limit of the indenter. (Not realising that on the first attempt lead to replacement of the original pin with a suitable gauge/gage pin).

A little of the cylinder is relieved to prevent interference with the case shoulder and allow measuring further down the neck than would otherwise be possible.

 

[itchyTF]

The SSR I used. Not too bad -

Crydom D1D40 3.5-32V Input 100V 40A Output Solid State Relay | eBay

Find many great new & used options and get the best deals for Crydom D1D40 3.5-32V Input 100V 40A Output Solid State Relay at the best online prices at eBay! Free shipping for many products!

www.ebay.com

 

[David101]

ac1) If I go with a 12 VDC timer rather than the 110 VAC timer used in the GenErick 2.0 design, would the Packard C230 still work? If not, what would a comparable, reliable contactor be? (am I right in believing a contactor would be more reliable than a SSR without having to break the bank?)
2) Given the poor quality of ZVS boards, has anyone successfully replaced it's FETs (and/or other components)? I saw Gina was considering doing so, but never saw any follow up.
3) I got the feeling that Mean Wells were the power supplies to favor. Has anyone had bad experiences with them?
4) The Fluxeon ferrite coil is able to be purchased from Fluxeon ( https://fluxeon.com/product/annie-flux-concentrator-coil/ ). Is that a simple drop in replacement for the copper tube coil? Or is there tweaking required to mount it to the ZVS board?

I dont see any benifit in using a 12v timer over the AC version actually the opposite but either way its of no importance so up to you.
I have replaced many components on the boards either with same or upgraded specs. I do think the cont actor is a more reliable buy as in it is more likely to work the cheap SSRs are cheap for a reason there were corners cut in manufacture.
The power supplies I have used are purchased on Aliexpress and all have been good quality and fully functional and meet specs I have built 3 units so a fair bit of experience with the components.
The Fluxeon ferrite is a drop in part other than for the connections. I soldered the ends of the wire which is very hard to solder into copper pipe as if it was the original copper tube it made fitting easier and kept all the small wires together and protected.

 

[ToddP]

The Fluxeon ferrite is a drop in part other than for the connections. I soldered the ends of the wire which is very hard to solder into copper pipe as if it was the original copper tube it made fitting easier and kept all the small wires together and protected.

Can you please post a picture so I can see this connection?

 

[SGK]

Here is a video where a guy found an "inexpensive" way to test brass hardness (near the end of his video). There is a tool on Amazon ( https://www.amazon.com/dp/B07BNCJ7C...KMLMANYK&psc=1&ref_=list_c_wl_lv_ov_lig_dp_it ) that will test metals hardness. The best use is to measure the hardness of unfired brass, then attempt to achieve that same hardness through annealing after having shot the cartridge. He scored the tester for £35, but it normally sells for $300+.

This might create an opportunity for someone's side job: case hardness testing for the rest of us at a reasonable price (as you only have to cover time, shipping, and the $350 of the device).

Yeah there are plenty of these tools available, albeit at prices much higher than what he paid. One challenge however is that they require a minimum material thickness. If you have turned your brass thinner than that they're of no use.

You can modify the tool to allow better fit with a cartridge case but are still limited to a minimum neck thickness as the tool needs a minimum amount to indent the brass.

 

[tstanek]

Hey guys,
I have a question about the induction annealer I made which is based on the original GinaErik design. I use MGNZ arduino board and basically the whole setup. I was thinking if it's feasible, and most importantly meaningful to use 2000 W ZVS, instead of 1000 W one. Would I be able to run it safely with my 600 W Power supply? I kinda feel the answer is no, from the logical reasons. I'm willing to buy more powerful power supply if my idea makes any sense. And last but not least, if this could work, would i need some more adjustments to the adruino DW. / different hardware. Thanks a lot for your help.
Best

Teodor

 

[itchyTF]

You could use the 2000 watt board. The wattage of the board is what it can handle not what it will try to draw from the supply. Can't answer the Arduino questions, no experience with the MGNZ.

 

[Ron Blain]

OK, my read is that the MGNZ board is just an Arduino Uno Rev 3 board clone. More on that later.

Hey guys,
I have a question about the induction annealer I made which is based on the original GinaErik design. I use MGNZ arduino board and basically the whole setup. I was thinking if it's feasible, and most importantly meaningful to use 2000 W ZVS, instead of 1000 W one. Would I be able to run it safely with my 600 W Power supply? I kinda feel the answer is no, from the logical reasons. I'm willing to buy more powerful power supply if my idea makes any sense. And last but not least, if this could work, would i need some more adjustments to the adruino DW. / different hardware. Thanks a lot for your help.
Best

Teodor

Just a few things to keep in mind without getting into Electronics 101. Been a few years since I played around with any of the induction heaters and my last was a 48 volt 1,000 watt unit. The applied voltage is important here. Discounting any loss in the conversion process a 2,000 Watt ZVS we will use this unit merely as an example.

12‑48V 2000W Induction Heating Module ZVS High Frequency Low Voltage Induction Heating Board Module DIY Induction Heater Kit​

Here is how things play out. The module reflects 12 to 48 volts. The reality is you only have a 2,000 watt unit when 48 volts is applied and at 48 volts a 2.0 KW heater (ZVS Module) will draw about 2,000 Watts / 48 Volts = 41.6 amps and since you need a buffer you need a 48 VDC 50 to 60 amp supply. Normally in design you want a 20% overhead for current. That is a pretty beefy 48 volt supply. Keep in mind your supply to module wire gauge should be a minimum of AWG 8 wire gauge. Power the module with less voltage and you will never get 2.0 KW power. 12 volts would be 500 watts, 24 volts would be 1,000 watts.

Next never ramp up the ZVS voltage. Power to these modules needs to be applied instantaneous. This is why some units just lock up and go max current and burn out the MOSFET drivers. Not that you plan on ramping the voltage up but for general knowledge.

Finally the MGNZ is actually a clone of what I believe to be an Arduino UNO Rev 3 board. Assuming a quality clone it should behave exactly like an Arduino manufacture board. Rather that relay logic to power the ZVS On/Off I would find a suitable Logic Level MOSFET rated for the needed voltage and current. I suggest a Logic Level MOSFET so the Gate voltage can be 5.0 volts to turn it On. This makes it compatible with a Digital Out from a 5.0 volt uC like the Arduino UNO boards.

No your 600 Watt power supply will not drive a 2.0 KW ZVS module. The 2,000 watts is a direct function of the Voltage * the Current. That is how power is derived. Actually yes, it would power the board but you won't see anything close to 2.0 KW and after inefficiencies you would be lucky to see 500 watts actually converted to coil heat. Also if you were to run at a full 2.0 KW I would be water cooling the copper coil.

I will be happy to explain anything and maybe provide you with a few schematics and design ideas.

Ron

 

[tstanek]

Thank you for a quick response I have one more question. I would like to control voltage output of power supply to ZVS in some more practical way than that tiny trimmer on the power supply itself. Is there some easy way to do that? I found some stuff on Aliexpress, but those switches are usually with too low power output, which would limit the power going to ZVS quite a lot. Totally perfect scenario would be to have something able to measure and display the actual voltage. Any ideas? Thanks in advance!

 

[tstanek]

OK, my read is that the MGNZ board is just an Arduino Uno Rev 3 board clone. More on that later.

Just a few things to keep in mind without getting into Electronics 101. Been a few years since I played around with any of the induction heaters and my last was a 48 volt 1,000 watt unit. The applied voltage is important here. Discounting any loss in the conversion process a 2,000 Watt ZVS we will use this unit merely as an example.

12‑48V 2000W Induction Heating Module ZVS High Frequency Low Voltage Induction Heating Board Module DIY Induction Heater Kit​

Here is how things play out. The module reflects 12 to 48 volts. The reality is you only have a 2,000 watt unit when 48 volts is applied and at 48 volts a 2.0 KW heater (ZVS Module) will draw about 2,000 Watts / 48 Volts = 41.6 amps and since you need a buffer you need a 48 VDC 50 to 60 amp supply. Normally in design you want a 20% overhead for current. That is a pretty beefy 48 volt supply. Keep in mind your supply to module wire gauge should be a minimum of AWG 8 wire gauge. Power the module with less voltage and you will never get 2.0 KW power. 12 volts would be 500 watts, 24 volts would be 1,000 watts.

Next never ramp up the ZVS voltage. Power to these modules needs to be applied instantaneous. This is why some units just lock up and go max current and burn out the MOSFET drivers. Not that you plan on ramping the voltage up but for general knowledge.

Finally the MGNZ is actually a clone of what I believe to be an Arduino UNO Rev 3 board. Assuming a quality clone it should behave exactly like an Arduino manufacture board. Rather that relay logic to power the ZVS On/Off I would find a suitable Logic Level MOSFET rated for the needed voltage and current. I suggest a Logic Level MOSFET so the Gate voltage can be 5.0 volts to turn it On. This makes it compatible with a Digital Out from a 5.0 volt uC like the Arduino UNO boards.

No your 600 Watt power supply will not drive a 2.0 KW ZVS module. The 2,000 watts is a direct function of the Voltage * the Current. That is how power is derived.

I will be happy to explain anything and maybe provide you with a few schematics and design ideas.

Ron

Thank you very much Ron. It brought a light to this problematics really, cause I am far from being educated in this field. As you correctly suggested the MGNZ board is Arduino rev 3 c board clone. I mentioned that just to outline my setup. So I guess I won't go any further than classic 1000 W ZVS circuit with 600 W supply which I'm running now. As my previous reply asks, I'm also interested in voltage regulation. Any thoughts on that?

 

[Ron Blain]

The ability to vary the voltage to the load is a function of the power supply. Many fixed voltage supplies allow some small adjustment range but they aren't adjustable over a wide range. This all goes back to the supply design and intended application. If the 1.0 KW ZVS works for you with the 600 watt supply and you can anneal I would just run with it. Power supplies allowing a wide range of voltage adjustment are typically called "programmable" and allow adjustment via an internal pot or an external voltage applied. Trying to keep this simple.

Ron

 

[itchyTF]

Decent ones are also more expensive.

 

[Phys86]

How critical is placement of the case on the coil for consistency?

I just place mine in the coil on a wood block and maneuver the block the get the case centered. I don't think it's perfect every time but as close as I can quickly eyeball it.

 

[Bora72]

I just finished reading all 179 pages. Whew, what a slog. Initially, I was very excited to give this project a go, but after reading the last 20 pages or so it sounds like there is an almost 100% chance that the ZVS and/or the power supply will show up DOA and that the relay (unless spending for an American made product) will crap out quickly due to below spec components used in its construction. With all that out of my system, I do have some questions:

1) If I go with a 12 VDC timer rather than the 110 VAC timer used in the GenErick 2.0 design, would the Packard C230 still work? If not, what would a comparable, reliable contactor be? (am I right in believing a contactor would be more reliable than a SSR without having to break the bank?)
2) Given the poor quality of ZVS boards, has anyone successfully replaced it's FETs (and/or other components)? I saw Gina was considering doing so, but never saw any follow up.
3) I got the feeling that Mean Wells were the power supplies to favor. Has anyone had bad experiences with them?
4) The Fluxeon ferrite coil is able to be purchased from Fluxeon ( https://fluxeon.com/product/annie-flux-concentrator-coil/ ). Is that a simple drop in replacement for the copper tube coil? Or is there tweaking required to mount it to the ZVS board?

Hello, I've replaced the FETs, china market, the zvs work.

Can anyone explain me how I can test the FETs for find the one damaged.
I replaced all the two....

I've a multimeter and lcmeter.

Thanks

 

[itchyTF]

How critical is placement of the case on the coil for consistency?

I just place mine in the coil on a wood block and maneuver the block the get the case centered. I don't think it's perfect every time but as close as I can quickly eyeball it.

I’ve noticed the magnetic field tends to center the case.

 

[itchyTF]

Hello, I've replaced the FETs, china market, the zvs work.

Can anyone explain me how I can test the FETs for find the one damaged.
I replaced all the two....

I've a multimeter and lcmeter.

Thanks

Use the ohmmeter and compare with a good one. The bad one should indicate a significantly lower resistance in at least one of the readings.

 

[Bora72]

Use the ohmmeter and compare with a good one. The bad one should indicate a significantly lower resistance in at least one of the readings.

Hello I've found this on YouTube and the datasheet also find on the net...
The video are on italian language but I think the explanation are comprensible..