Posted: 10/27/2008 10:22:10 AM EDT
|
As a kid, I played with transformers. My dad actually built me a 6V AC power supply to play with. I knew then that a transformer would bump down or bump up AC voltages. I knew for instance that If I were to connect a transformer to the wall to bump voltages up (reversing a 12V output transformer) I should get high voltages out the other end––-I wanted to make some sparks. But, upon connecting such transformers I would get one of two responses: blow the breaker (making sparks at the wrong end) or let the smoke out of the transformer.
So, my question is: how does one design a transformer to bump voltages up without blowing breakers or letting the smoke out. My guess is that the impedance of the primary winding needs to be greater than say 12 ohms for a 120V 10A household plug. But if this is the case, how does one choose the size of the core, the number of windings, and the size of the primary wire? Also, say I want to have the output current limited to some value like 1mA at 2400V. Do I essentially assume no power loss so the input current needs to be about 0.02A on the 120V end. If so, then again how do I choose the core, wire and windings? Thanks. |
|
The output voltage of the transformer will be a ratio of the input voltage determined by the ratio between the primary and secondary windings.
This link explains transformers and their principals fairly well. |
|
It all relates to the impedance of the coils and the amperage capacity of the coils.
Plugging a 12 volt transformer in backwards didn't work because of these two factors. You have to design the primary coil to (A) exhibit the desired impedance and (B) be made of sufficiently heavy gauge wire to carry the current required at the operating voltage. The secondary coil must have those two features plus also be of the correct turns ratio relative to the primary coil that is required to deliver the intended output voltage and current capacity. CJ |
| If you want to run a transformer in reverse, that is using the primary winding as the output, then you need to figure out what (in VA) the transformer is designed for. Once this number is known, then you can work back using Ohm's law and figure out how many amps the transformer is capable of handling. Once that is known, then just limit it with a fuse, on the secondary, to less than those amps to protect the transformer. |
| Why don't you just find an old transformer that is capable of producing the voltages you want? IE, an old neon sign tranformer, oil burning furnance ignition transformer, etc. Sometimes these things can be had for little to nothing in a junk yard. I made a jacobs ladder from an old ignition transformer that I salvaged off of an old boiler they were replacing at work. It's secondary output is 12kv. |
|
Quoted:
...Also, say I want to have the output current limited to some value like 1mA at 2400V. Do I essentially assume no power loss so the input current needs to be about 0.02A on the 120V end. If so, then again how do I choose the core, wire and windings? Thanks. If you have a set input voltage, you can't really choose both the output voltage and output current. You keep raising the secondary windings (which raises the output voltage) until the current drops to where it won't kill you. Quoted:
...Plugging a 12 volt transformer in backwards didn't work because of these two factors. You have to design the primary coil to (A) exhibit the desired impedance and (B) be made of sufficiently heavy gauge wire to carry the current required at the operating voltage. ... True. The inductive load of the primary has to be enough that it won't short the outlet, even when the secondary is left open (unloaded). ~ |
|
Quoted:
The output voltage of the transformer will be a ratio of the input voltage determined by the ratio between the primary and secondary windings. This link explains transformers and their principals fairly well. I am well aware of this, It is the power output that I wish to control, and not burn or blow breakers because of current draw in the primary windings. If you take a wall wart apart, remove the bridge rectifier and connect a power cord to what was the output end, you will either blow your circut breaker or burn those windings up. I don't want either of these to happen. BTW, I know how to connect leads so that I won't electorcute myself, and how to prevent serious fires and avoid fumes. I'll keep my guns and stuff, thank you. |
|
Quoted:
It all relates to the impedance of the coils and the amperage capacity of the coils. Plugging a 12 volt transformer in backwards didn't work because of these two factors. You have to design the primary coil to (A) exhibit the desired impedance and (B) be made of sufficiently heavy gauge wire to carry the current required at the operating voltage. The secondary coil must have those two features plus also be of the correct turns ratio relative to the primary coil that is required to deliver the intended output voltage and current capacity. CJ Precisely. Now how do I figure this out? How do I detrmine the impedance beforehand? |
|
Quoted:
Why don't you just find an old transformer that is capable of producing the voltages you want? IE, an old neon sign tranformer, oil burning furnance ignition transformer, etc. Sometimes these things can be had for little to nothing in a junk yard. I made a jacobs ladder from an old ignition transformer that I salvaged off of an old boiler they were replacing at work. It's secondary output is 12kv. This is more of an academic thing than practical. I just want to know how it is done. |
|
As others have said, for the primary (the one you are going to apply 110v to) its the impedance that matters.
The impedance has a small resistive component, this really needs to be kept small, because it will absorb power and convert it to heat. The main component of the impedance will be inductive reactance. The inductive reactance is a product of the number of turns and the permittivity of the core material. Higer permittivity cores (such as iron) require less turns compared to the same inductance using (say) air as the core. Transformer cores often give a guideline to the number of turns required for the primary by a guide-line of turns per volt - this is the turns per volt rating of a given core is (say) 2.5, then a winding to accomodate 100v would require 2.5 * 110 = 275 turns. One issue to watch is that as current is drawn from the secondary, the current in the primary will increase - which increases the magnetic flux in the core material. For core materials other than free space (air) there is a point at which the core material saturates, and can't carry any more magnetic flux. The output voltage is simply the ratio of the number of turns on the primary to the number of turns on the secondary. |
|
Quoted:
As others have said, for the primary (the one you are going to apply 110v to) its the impedance that matters. The impedance has a small resistive component, this really needs to be kept small, because it will absorb power and convert it to heat. The main component of the impedance will be inductive reactance. The inductive reactance is a product of the number of turns and the permittivity of the core material. Higer permittivity cores (such as iron) require less turns compared to the same inductance using (say) air as the core. Transformer cores often give a guideline to the number of turns required for the primary by a guide-line of turns per volt - this is the turns per volt rating of a given core is (say) 2.5, then a winding to accomodate 100v would require 2.5 * 110 = 275 turns. One issue to watch is that as current is drawn from the secondary, the current in the primary will increase - which increases the magnetic flux in the core material. For core materials other than free space (air) there is a point at which the core material saturates, and can't carry any more magnetic flux. The output voltage is simply the ratio of the number of turns on the primary to the number of turns on the secondary. Precisely. The best bet to limit output current and thereby protecting the higher impedence side, is to use a current limiting resistor in series with the secondary side. I had an old weather station kit that would output wind direction with a series of neon lamps. It plugged directly into 120 VAC and used a pair of resistors in series, one on each lead from the 120, to reduce the current to a few milliamps. No transformer, no GFCI. The only thing keeping dangerous currents from zapping humans were these two resistors! V=I*R. Make R high enough and you can withstand any V because I will be tiny. |
|
Quoted:
... Precisely. Now how do I figure this out? How do I detrmine the impedance beforehand? You calculate the inductance of the primary alone. Look up info on making DIY inductors. Designing transformers properly is kind of like designing radio antennas,,,,, it looks simple but there's a LOT of math involved to do it the most-efficient way. –––––– Another more practical possibility would be for you to open the outer case a 12V transformer (just the case) and measure the core size (the area of the core that the wire coils are wrapped around), the core window (the holes in the transformer core that the wires run through) and then measure the primary and secondary wire sizes. Then find the cross-areas of the two wire sizes (you can get these online, from Tesla-coil building sites) and using that figure and the transformer's known input:output ratio, you can estimate the number of turns on each of the windings. If you use a core of about the same cross-section, this will be close enough to work. Strictly speaking it does not "scale" up and down though, because the flux density varies with the physical size of the windings. 100 turns of #20 wire on a 1-inch diameter bobbin will not have the same impedance as 100 turns of the same wire on a 10-inch diameter bobbin. ...AC-freq transformers are just steel core, so any piece of steel will be close enough. When you get into RF transformers, you get into issues with flux density and saturation and they may not work at all without the proper core material. –––––– Also––if you eventually build a transformer and want it to last "permanently", then it has to be impregnated. You submerge it in a pot of sealant inside a vacuum. There are industrial chemicals just for this, but varnish works too. Audiophiles and vintage equipment aficionados use beeswax so that the coils can be easily unwound from the bobbins if it becomes necessary. ....If you don't do this, the wires of the coils will stretch and become slightly loose and vibrate against each other, and eventually the enamel will wear through and result in short turns. ~ |