another candidate for the Wile E Coyote Super Genius Award

[unclejed613]

fortunately not on this board... https://www.reddit.com/r/AskElectronics/comments/8581n5/how_would_this_circuit_behave/

dude wants to build a DC-DC converter for a 10kW arc furnace, and thinks he can use a 555 and a bunch of MOSFETs to do it. he thinks a transformer is too expensive... doesn't look like anybody is going to convince him that 10kW is not a beast for a newbie to mess with.

 

[OBW0549]

I sense a possible future Darwin Award candidate there. Some people simply should NOT be playing with electricity.

 

[alec_t]

At least he's not aiming to run it from a PP3 battery .

 

[OBW0549]

At least he's not aiming to run it from a PP3 battery .

Give him time; I'm sure he'll get there eventually. He might even convince himself that by adding a secondary winding to his inductor plus a rectifier diode, he can feed some of the inductor's energy back into the battery to keep it perpetually charged.

 

[tcmtech]

Give him time; I'm sure he'll get there eventually. He might even convince himself that by adding a secondary winding to his inductor plus a rectifier diode, he can feed some of the inductor's energy back into the battery to keep it perpetually charged.

But I don't need a lot of overunity. I just need a little to make it work therefore it should be possible.

As for powering a 10 KW arc furnace most any 300+ amp constant current welder will do that just fine.

 

[OBW0549]

But I don't need a lot of overunity. I just need a little to make it work therefore it should be possible.

Good point. Wouldn't surprise me to see a supercapacitor appear in that thing soon, either...

 

[tcmtech]

Something I have noticed over the years with people wanting to do such higher level projects is that they rarely have a realistic grasp on reality regarding both what they actually are after in terms of their intended end goal or function and how they perceive things to be the most practical way to get there.

I cant count how many times I have seen someone wanting to do a large complex build or wanting to use a overly simplified and grossly inadequate process or method to do a large and detailed function because the grossly oversimplified method looks cheap and easy and nothing else.

That or wanting to use a way over complicated, inefficient or all around unrealistic method to do what is a simple cheap and easy action if any other path but theirs is used to get where they want to be.

 

[JonSea]

The original poster apparently deleted all of his remarks and went off in search of a more "helpful" audience.

 

[tcmtech]

Seems like a growing theme in the world these days. Embarrass yourself then hide all evidence, so that you can deny it (and the emotional blow it created for you) ever happened, rather than own up to having been set straight on anything no matter how small it was.

Strange world where learning things has become so emotionally damaging that people have to run and hide at all costs to deal with it.

 

[OBW0549]

The original poster apparently deleted all of his remarks and went off in search of a more "helpful" audience.

Not surprising. People who engage in magical thinking (my term for it in this context is "voodoo electronics") often get really upset when people try to debunk their precious notions and set them straight. Some get downright hostile.

 

[tcmtech]

Not surprising. People who engage in magical thinking (my term for it in this context is "voodoo electronics") often get really upset when people try to debunk their precious notions and set them straight. Some get downright hostile.

I think it's part of the anti-science movement where everyone who can not understand basic scientific fact or defend their views is to be assumed correct while those who can are to be proclaimed wrong and mocked in every possible way for it.

 

[unclejed613]

i know there are always going to be "gotchas" when you try to scale something up, and often one of the "gotchas" is that the price of components rises exponentially. but i've had some experience too with "simple" power switching devices (motor contactors, line transfer switches, emergency disconnects) for the kind of power he was talking about. that's the power level where you begin to need wire to be BIG, i.e. from #0000 to 1500MCM, and if you make a mistake at that power level you literally can create huge fireballs of copper vapor just from having one high resistance connection (in this context, anything more than a few milliohms). at currents over about 100A, there's a physics problem that you normally don't see with lower current levels, and that's that the wire actually moves because it's pushing against the earth's magnetic field. when i was a calibrator in the Army, i had to calibrate some clamp-on current meters at 300A. there was a high current supply in the shop that went up to 75A. the calibration procedure was to make a loop using 4 turns of #0000 wire, and run the supply up to 75A (which works out to 300 ampere-turns). when we got the current up past about 100 ampere-turns, the loop began to move, and not from thermal effects. at 300 ampere-turns, the loop had moved until it was almost perpendicular to magnetic north. remember, that's #0000 wire which is about the diameter of my thumb, and it moved itself about 18 inches. we also had to limit the amount of time we spent making the test, because that wire also got very warm, very quickly.

yeah, the OP on that thread on reddit seemed oblivious to how things scale. i don't know where this guy was from, but keep an eye on world news for a smelting works somewhere having an explosion or a fire... if i remember correctly, some arc furnaces use huge graphite rods, and thousands of amps of current, and the power supply uses mercury arc rectifiers the size of bathtubs. and, granted there are semiconductor rectifiers for such large amounts of current [for example](**broken link removed**)

 

[JoeJester]

You failed to provide a trigger warning before you shattered their highly attuned scientific knowledge where anything is possible and those scientific impediments are part of the power structure to keep them down, economically.

 

[tcmtech]

You know you damaged their precious egg shell egos when they make an effort to inform you that they didn't bother to read whatever you wrote.

TL;DR (Too Long; Didn't Read) is the new code for 'I read it but then I got the butthurts cause I can't factually refute it.'

What they think their butthurt is bringing to the world. (12 string bass guitar.)

What they actually have going for them.

 

[large_ghostman]

Not sticking up for the guy, but i am careful these days when dismissing things that at first glance wont work. Not electrical related but within the last year, i have seen two impossible (as far as was known) chemical related procedures become common, one was taken up by a university in the end and while they didnt credit the original idea (poor show that), They wrote a paper and got an award for it.

Not saying that applies here, but currently i am sailing really close to what at first glance looks like over unity, but in reality the truth is the physics work out ok, its a simple case of using something in a way that it hasnt been used before. In my own case it does help that houses now require less energy to run if they have been modernized.

If you go back just 7-8 years when i joined here, LED lights or even CFC bulbs were all but unusable, now you can get decent ones. So once upon a time you used 100W bulb to light a room, now you can light most the house to an acceptable level foe a little over that.

 

[unclejed613]

it's difficult enough to design a DC-DC converter for a 100W load, or even a 100W amplifier let alone trying to scale it up 100 times bigger. the guy is definitely not going to be able to build it in his garage. most residential service is between 5 and 10kW total, for the whole house.

looking up arc furnaces provided some interesting results, and a whole new level of "ming boggling"...

A mid-sized modern steelmaking furnace would have a transformer rated about 60,000,000 volt-amperes (60 MVA), with a secondary voltage between 400 and 900 volts and a secondary current in excess of 44,000 amperes.

 

[tcmtech]

looking up arc furnaces provided some interesting results, and a whole new level of "ming boggling"...

It's like everything else. There is the most common commercial huge scale devices and then there are the much smaller private and even DIY mini versions. I have had a fascination with the commercial end of electrical and electronics all my life and thusly have worked first hand with systems and devices that are well above what the average private consumer would ever play with.

As impressive as the huge industrial process units are I do have to say I find the miniaturized private versions as equally impressive in how they can do the same jobs but at equivalently miniscule scales. The biggest arc welder systems I have ever been around was a huge industrial process submerged arc rig that could run ~1/8" solid core welding wire at something like 2000 amps (@ 50 - 60 volts? )continuous duty! Every current carrying component was either heavy copper bus bar or 10 - 12 4/0 fine strand welding cables in parallel! Something that no realistic private home owner would ever build let alone power given it needed a 3 phase 480 VAC 150 amp input.

Now that said I do have to say that the tiny 120 VAC powered portable MIG and mini stick welders are just as much of a marvel to me for what they can do with such a equivalent limited power capacity. That and at their relevant power levels I think that most electronically proficient people could conceivably build one or even a slightly larger version if they really set their mind to it.

Where I am at most every modern home comes with a at minimum 120/240 VAC 200 amp service that typically fed from a 15 - 25 KVA high duty cycle (200+% overload for 2 - 4 hours rated) utility transformer or a 100+ KVA if it part of a local residential system. Given that for most of us with such systems adding our own personal 100 amp 240 VAC power circuit is a somewhat trivial issue that also gives us a honest ~ 20 KW capable service supply point using nothing but off the shelf electrical components anyone can buy at any local home building supply chain store. The rest of what we do with it is up to the individual.

To me 10 KW is not a major undertaking given how much off the shelf components to make such a setup is already available enmasse to everyone. Sure, by the hobiest view 10 KW is huge but by common residential and utility scale level power consumption 10 KW is just something that uses a bigger than normal extension cord. 6 ga SO cord fed from a common 50 amp welder or electric stove output coming from a common 50 amp breaker set would easily power it at 240 VAC and its all 100% off the shelf private level electrical components anyone can buy and install if they see fit to try.

 

[large_ghostman]

No I agree with all thats been said, but I am learning to be cautious with some things.As for Darwin..... There are a couple of guys on a science forum, they are playing with solutions to make pretty patterns in wood by arching with high voltage etc. They are great at chemistry but really thought i was being unreasonable when i suggested what precautions to take when using TWO MOT transformers to burn the patterns, they are sure a bit of a stick and a pair of gloves are all thats needed to work safely......

 

[unclejed613]

i had a friend who bought a welding rig rated at 1800W. he had tested it when he bought it, and it worked ok, but when he got it home it would work for about a second each time he touched the rod to the workpiece. the unit was on a 20A circuit, but the garage was the furthest part of the house from the panel. i stuck a voltmeter in the socket next to where the welder was plugged in. when he touched the rod to the workpiece the voltage would drop to 90. turns out the "20A" circuit was wired with #14 not with #12.

 

[gophert]

i had a friend who bought a welding rig rated at 1800W. he had tested it when he bought it, and it worked ok, but when he got it home it would work for about a second each time he touched the rod to the workpiece. the unit was on a 20A circuit, but the garage was the furthest part of the house from the panel. i stuck a voltmeter in the socket next to where the welder was plugged in. when he touched the rod to the workpiece the voltage would drop to 90. turns out the "20A" circuit was wired with #14 not with #12.

1800W is 15A on a 120VAC circuit. Or an 8 ohm load.
To drop the 120 VAC supply down to 90VAC means there is a series resistance of 2.66 ohms (wire resistance) from the panel.

That is 1000 feet of 14 gauge wire! The garage was 500 feet from the panel?

Either that or a loose wire nut somewhere.