Inductor for magnet power supply

[dwlee6vn]

New inductor - works great!

Made up a new inductor with 36 pieces of 6" long 1/16" dia mild steel tig filler rod. Sprayed with clear insulating varnish and dried. Bundled together and wound 500 turns of 19 gauge magnet wire onto it in 4 layers. Put it in place of the former solid rod inductor and tried it out. Now OCV is 168 volts and voltage under load of the magnet is 147. Let it run for several minutes and it only got very slightly warm. Also notably less "buzzing" of the inductor assembly - just a low hum.

Also while I was in there I noticed the capacitor was 500 mfd not 200 as I previously recalled. I did some searching on the web and it appears that to have a substantial smoothing effect it would need more like 5000 mfd. Seems like a lot and at this point, the thing works fine so I am going to apply the "if it ain't broke, don't fix it" rule.

Thanks to all. I'll add the flyback diode when it the Amazon order gets in.

 

[metalsculptor]

Thanks for complement. Kind of amaze myself when I remember I built the basic structure when I was a teenager using a torch, welder, grinder, and hand held 3/8" drill. And a hammer of course. Reminds me of the saying cut with an ax, beat to fit, and paint to match.

Woke up this morning with a few questions:

• Is the length of my new inductor critical? Making it a couple inches shorter would package better in the power supply box.
• Regarding suggestion to just eliminate all filtration, doesn't the capacitor help protect the system (esp. the diodes) from the voltage spike when I shut it off and the magnetic field collapses?

That's how I used to make things when I was young and tools were scarce.

The length of the inductor is not critical, the area of the core has the biggest effect. The permeability of the air gap in the magnetic circuit is so much greater than the iron any slight decrease in magnetic path on the iron side will be swamped by the effects of the air gap.

The inductor will freewheel through 2 diodes of the bridge rectifier, this happens 120 x per second when the power is on. If you measure the resistance across a bridge rectifier on the DC side it will measure like a diode, that is what becomes the freewheeling diode at times when the mains cannot supply current.

Adding a freewheeling diode across the inductor can't hurt.

 

[jneutron]

Made up a new inductor with 36 pieces of 6" long 1/16" dia mild steel tig filler rod. Sprayed with clear insulating varnish and dried. Bundled together and wound 500 turns of 19 gauge magnet wire onto it in 4 layers. Put it in place of the former solid rod inductor and tried it out. Now OCV is 168 volts and voltage under load of the magnet is 147. Let it run for several minutes and it only got very slightly warm. Also notably less "buzzing" of the inductor assembly - just a low hum.

Also while I was in there I noticed the capacitor was 500 mfd not 200 as I previously recalled. I did some searching on the web and it appears that to have a substantial smoothing effect it would need more like 5000 mfd. Seems like a lot and at this point, the thing works fine so I am going to apply the "if it ain't broke, don't fix it" rule.

Thanks to all. I'll add the flyback diode when it the Amazon order gets in.

Nice work. Thanks for updating us.

John

 

[powerbob]

If you have 80 ohms resistance on your magnet and 160vdc you would draw 2A, witch is what you measured. In other words your inductor capacitor is not doing anything.

This is what i would do. A transformer wound with 22awg might only have a 1.25 amp rating because of the enclosed space and the temperature rise. You have used a full 10 lbs enclosed in a metal cylinder. The heat buildup in the core is 2A x 160vdc = 320 Watts! I would limit the current to no more than 1 amp and see if you had enough holding power.

It will be very simple to try for 1 amp. Just use half wave rectification. All you need is a line cord and one diode. Go to radio shack and get a diode rated 3A or more at 400 volts. Put this diode in series with the hot line, the direction does not matter. The neutral end of the line cord goes directly to one wire of the magnet. The hot of the line cord goes to the diode and the other end of the diode goes to the magnet. You will now use 1/2 the amperage you were using before.

 

[jneutron]

If you have 80 ohms resistance on your magnet and 160vdc you would draw 2A, witch is what you measured. In other words your inductor capacitor is not doing anything.

This is what i would do. A transformer wound with 22awg might only have a 1.25 amp rating because of the enclosed space and the temperature rise. You have used a full 10 lbs enclosed in a metal cylinder. The heat buildup in the core is 2A x 160vdc = 320 Watts! I would limit the current to no more than 1 amp and see if you had enough holding power.

It will be very simple to try for 1 amp. Just use half wave rectification. All you need is a line cord and one diode. Go to radio shack and get a diode rated 3A or more at 400 volts. Put this diode in series with the hot line, the direction does not matter. The neutral end of the line cord goes directly to one wire of the magnet. The hot of the line cord goes to the diode and the other end of the diode goes to the magnet. You will now use 1/2 the amperage you were using before.

While your suggestion seems reasonable, lets look at a few details.

1. If he uses one diode, there will be half sine excitation to a magnet which is not made of transformer laminations. It will dissipate more power as a result of the high ac excitation. In addition, the magnetic material he attaches it to will also dissipate heat as a result of eddy current dissipation.

The present setup has some smoothing, this reduces load eddy dissipation.

2. I do not recommend that source for diodes. They used to buy their diodes by the pound from a manufacturer... That pound of product being comprised of material which was returned to the factory due to failure to meet customer incoming inspection sampling test. Sometimes those failures were HTRB failure, sometimes reverse leakage, sometimes VF, and sometimes....THE CATHODE BAND WAS ON THE WRONG END OF THE PACKAGE!!.

Standard protocol was to send rejected material to less discerning customers a total of 3 times. Once rejected the third time, it was put into a 50 gallon drum, and sold by the pound. To guess who?

ps..I was the grunt who verified failures.

jn

 

[FoMoCo]

2. I do not recommend that source for diodes. They used to buy their diodes by the pound from a manufacturer... That pound of product being comprised of material which was returned to the factory due to failure to meet customer incoming inspection sampling test. Sometimes those failures were HTRB failure, sometimes reverse leakage, sometimes VF, and sometimes....THE CATHODE BAND WAS ON THE WRONG END OF THE PACKAGE!!.

Standard protocol was to send rejected material to less discerning customers a total of 3 times. Once rejected the third time, it was put into a 50 gallon drum, and sold by the pound. To guess who?

Thanks for the info. I had always thought that they sold floor sweepings and rejects but had no real proof. Other than a lot of failures!

 

[jneutron]

Thanks for the info. I had always thought that they sold floor sweepings and rejects but had no real proof. Other than a lot of failures!

It shocked me to find it out. One of the sales guys told me a story about how a Zenith (television fame) accosted him because they opened a competitor's product only to find "zenith" marked parts inside. They were very upset that they were paying more for their parts but others were getting them cheaper. When the date code on the parts was pointed out to them and it was shown that they were from a lot rejected by Zenith for HTRB failures, they laughed. (High Temperature Reverse Bias was the most stringent and difficult test to pass, it really stresses the passivation integrity.)

I used to buy my parts there and other cheap places back in the mid to late 70's. I attributed my constant circuit failures to my stupidity and my youth.

But that was far in the past.. Now, I am no longer young..

jn