Home |
Search |
Today's Posts |
#1
|
|||
|
|||
Building a circuit with no power transformer ?
Please forgive a newbie question...
I've read in a number of places that a circuit should always have transformer isolation from the wall AC, even if the circuit needs only 120V AC input. Then again, a number of people have told me that it really doesn't matter. That the only reason to use a power transformer is to step voltage up or down... if the circuit needs 120V AC, there's no reason to use a transformer. I'd really like to understand this issue, and I haven't thought of any compelling advantages to having a transformer in the power supply path. With audio transformers, I see how removing DC continuity can prevent stray ground current, but I don't understand what isolation is gained by using a 1:1 *power* transformer. I've heard it said that if a power transformer is not used, that the chassis of the device will be "hot." Does that make sense to anyone ? If I'm using a 3-prong AC connection, I'm wiring my chassis to safety ground, so as long as the wall ground connection is properly wired, isn't it impossible for my chassis to have any voltage on it relative to ground ? (And if the wall safety ground is miswired, doesn't any circuit become potentially unsafe, regardless of its design ?) If I use a 1:1 power transformer, and thus have no galvanic connection to the wall hot and neutral, I still have high DC and AC voltages in my circuit, and my chassis is still connected to the wall safety ground... there are plenty of ways a user could get a serious shock by touching things inside the device. And the chassis is only safe to touch if the house AC ground is wired properly (otherwise a circuit fault or AC polarity problem could result in voltage on the chassis) and if you aren't simultaneously touching some other high voltage source (at which point, the chassis just becomes a path to ground). This is true regardless of whether or not there's a power transformer, right ? The only safety advantage I can immediately see from the power transformer is that it makes it impossible for DC currents to flow to the neutral wall connection... is that a significant advantage, or am I missing something here ? Thanks for any explanation! There's probably a good explanation for the convention of using 1:1 power supply transformers, and I'd really like to understand it! James |
#2
|
|||
|
|||
Building a circuit with no power transformer ?
James Nash wrote:
I've read in a number of places that a circuit should always have transformer isolation from the wall AC, even if the circuit needs only 120V AC input. Then again, a number of people have told me that it really doesn't matter. That the only reason to use a power transformer is to step voltage up or down... if the circuit needs 120V AC, there's no reason to use a transformer. Electrical and fire codes, for one. Your life for another. With audio transformers, I see how removing DC continuity can prevent stray ground current, but I don't understand what isolation is gained by using a 1:1 *power* transformer. It isolates the chassis and knob shafts from the AC line. Reversing or miswiring a power cord or pluggin into a miswired outlet or extension cord, or plugging into an outlet with a ground fault is likely to kill someone. Power Solution used to sell a 2 kW class D bass amplifier, and it had a GFI on the back panel to prevent stupidity. If you absolutely insist on building something without a power transformer, at least do this... |
#3
|
|||
|
|||
Building a circuit with no power transformer ?
On Wed, 22 Oct 2003 21:47:40 GMT, James Nash
wrote: Thanks for any explanation! There's probably a good explanation for the convention of using 1:1 power supply transformers, and I'd really like to understand it! James Yes; it prevents a medical condition known as death. Ed |
#4
|
|||
|
|||
Building a circuit with no power transformer ?
James Nash wrote:
I've read in a number of places that a circuit should always have transformer isolation from the wall AC, even if the circuit needs only 120V AC input. Then again, a number of people have told me that it really doesn't matter. That the only reason to use a power transformer is to step voltage up or down... if the circuit needs 120V AC, there's no reason to use a transformer. If you don't care about noise or isolation from the line, and you can sheild everything so that there is no way anybody can ever touch the chassis, you can do hot chassis work. A lot of TV and radio gear was this way. I'd really like to understand this issue, and I haven't thought of any compelling advantages to having a transformer in the power supply path. With audio transformers, I see how removing DC continuity can prevent stray ground current, but I don't understand what isolation is gained by using a 1:1 *power* transformer. MASSIVE isolation from line noise. Also, your secondary is now isolated from the power line which prevents the chassis from floating at high voltages. I've heard it said that if a power transformer is not used, that the chassis of the device will be "hot." Does that make sense to anyone ? If I'm using a 3-prong AC connection, I'm wiring my chassis to safety ground, so as long as the wall ground connection is properly wired, isn't it impossible for my chassis to have any voltage on it relative to ground ? (And if the wall safety ground is miswired, doesn't any circuit become potentially unsafe, regardless of its design ?) There is a difference between being potentially unsafe (if both the plug is miswired AND the transformer fails AND the ground is open, you get hurt), and just plain unsafe (if any one of those three takes place, you get hurt). The only safety advantage I can immediately see from the power transformer is that it makes it impossible for DC currents to flow to the neutral wall connection... is that a significant advantage, or am I missing something here ? That's another issue, but frankly just the RFI rejection from the transformer acting as a line filter is worth the price of admission. Thanks for any explanation! There's probably a good explanation for the convention of using 1:1 power supply transformers, and I'd really like to understand it! Talk to any of the number of musicians who have been killed due to lifting grounds on hot-chassis guitar amps. --scott -- "C'est un Nagra. C'est suisse, et tres, tres precis." |
#5
|
|||
|
|||
Building a circuit with no power transformer ?
James Nash wrote:
Please forgive a newbie question... I've read in a number of places that a circuit should always have transformer isolation from the wall AC, even if the circuit needs only 120V AC input. Then again, a number of people have told me that it really doesn't matter. That the only reason to use a power transformer is to step voltage up or down... if the circuit needs 120V AC, there's no reason to use a transformer. ATTENTION! Using an isolation transformer is a serious safety matter -- a matter of life and death! If you are a hobbyist building an electronic circuit using wall AC, you could kill yourself or someone else by not using an isolation transformer. DON'T TRY THIS AT HOME! Pros can save money by designing circuits that can be manufactured without an isolation transformer, but it would be very dangerous for an amateur to try it, and even pros would be wise to plug such equipment into an external isolation transformer before working on it. Here's the secret of why an isolation transformer gives you some protection: because electrical current only flows if a circuit is completed. Consider the paths that the electricity follows with and without an isolation transformer. If you just used the wall AC without an isolation transformer, one lead is "neutral" and is connected to a ground near the service entrance, such as a cold water pipe. Another lead is "hot" and has 120 VAC relative to the neutral lead. (There may also be a safety ground wire.) Since the neutral wire is grounded somewhere, anything (such as you) connecting the hot lead to a ground will complete the circuit, and current will flow through it. If the current passing through you goes through your heart, you could die. A 1:1 isolation transformer still gives you two leads with 120 VAC relative to each other. The difference is that neither of these is leads is electrically grounded. As long as your circuitry doesn't ground a lead, neither lead is "hot" relative to ground; they are only "hot" relative to each other. So, accidentally touching either lead while some other part of your body is grounded will not complete the circuit, current doesn't flow through your heart, and you remain alive. If you touch both leads, however, you'll still get shocked. A good safety practice attributed to Tesla is to always keep one hand in a pocket while working around dangerous voltages, to reduce the risk of a fatal shock. |
#6
|
|||
|
|||
Building a circuit with no power transformer ?
James Nash wrote: If I use a 1:1 power transformer, and thus have no galvanic connection to the wall hot and neutral, I still have high DC and AC voltages in my circuit, and my chassis is still connected to the wall safety ground... there are plenty of ways a user could get a serious shock by touching things inside the device. And the chassis is only safe to touch if the house AC ground is wired properly (otherwise a circuit fault or AC polarity problem could result in voltage on the chassis) and if you aren't simultaneously touching some other high voltage source (at which point, the chassis just becomes a path to ground). This is true regardless of whether or not there's a power transformer, right ? If you have no power transformer, and everything is wired correctly (that's putting a LOT of faith in electricians, from the wall outlet to the incoming service panel over how many years?) you get what you're asking for. But if the hot side of the AC ever leaks, in any way (that includes stray capacitance, corrosion, who knows), to the outside world, the return path of the circuit is ANY grounded point nearby. This is because the mains return is ground. If you have a transformer, no current can flow that isn't going back inside the unit to the transformer. |
#8
|
|||
|
|||
Building a circuit with no power transformer ?
|
#9
|
|||
|
|||
Building a circuit with no power transformer ?
"James Nash" wrote in message ... I hate to belabor this point, but if someone wouldn't mind explaining an example situation in which electrocution could be prevented by the insertion of a power isolation transformer, I'd really appreciate it. James, if you make yourself knowledgeable about the power system in the USA it might help you understand the value of isolating power from the phase source. You obviously have an understanding of phase current, neutral and ground. If you grab a 120v phase current source in one hand and neutral in the other, you complete the circuit and kill yourself. If you grab 120v phase current and ground, you complete the circuit and kill yourself. If you go back in your post and read the number of times you used "but" and "if" you can count up the number of times you may accidentally grab the phase current and either neutral or ground (according to code, they should return to the same spot in your electrical service panel) and kill yourself. The use of an isolation transformer decouples the phase current from earth grounding so that you cannot as easily complete the circuit accidentally and kill yourself. The second time I lifted the ground on a guitar amplifier was the last time for me. (I'm a slow learner. Fortunately I had a second chance to make the same stupid mistake.) (And to clarify why I'm asking these questions, two reasons. It's not that I'm too stubborn or cheap to buy a power transformer :-) First, it's because I've been unsuccessful in finding a 110V transformer that will fit in the chassis of the power supply I'm working on. And second, I like to understand the underlying concepts as best as I can, instead of just blindly following the advice of people who clearly DO know more than I do about electronics.) Others may not agree, but I'll separate electricity from electronics for the simple reason that electricity is basically understood as votage potential, load, and current flow. Electronics, while far more interesting to me, is less likely to kill me than a poorly wired piece of gear plugged into a poorly wired outlet. I'm sure if you Googled the subject, you could spend the next month on it. Or head to your local library and read. Of head to a Lowes of Home Depot and buy a book on basic electricity. It's not a difficult subject to understand, but if you're going to muck around with it, you're better off in the know. Thanks again for taking the time to explain this stuff! Don't know if I helped, but I know *I* feel better. John |
#10
|
|||
|
|||
Building a circuit with no power transformer ?
|
#11
|
|||
|
|||
Building a circuit with no power transformer ?
In article ,
"John LeBlanc" wrote: If you grab a 120v phase current source in one hand and neutral in the other, you complete the circuit and kill yourself. If you grab 120v phase current and ground, you complete the circuit and kill yourself. Makes sense to me. The use of an isolation transformer decouples the phase current from earth grounding so that you cannot as easily complete the circuit accidentally and kill yourself. I guess I don't fully understand the implications of decoupling here. Are you saying that if I connect 120v phase current to one side of a 1:1 transformer primary and neutral or ground to the other side of the primary, that the two sides of the secondary are somehow less lethal than the original wall current that's connected to the primary ? There's still 120v between the secondaries, so I guess I don't understand why I wouldn't still get the same shock if I grabbed the two sides of the secondary. And wouldn't there still be roughly 120v between the higher voltage side of the secondary and either the wall's neutral or earth ground ? Logically, as a general principle, wouldn't there *have* to be at least 60v minimum potential difference between one side of the secondary and *any* neutral or ground reference ? I understand that there's no physical connection between the secondary and the house current, but AC current is still induced through the transformer, right ? There *must* be something here that I fundamentally don't understand... if there's a 120v swing inside the circuit, doesn't there exist a lethal potential between *some* places in the circuit and *any* earth reference you choose ? Does the safety lie in the fact that the transformer limits the amount of current that can flow from the wall to the secondary ? James |
#12
|
|||
|
|||
Building a circuit with no power transformer ?
In article znr1066869692k@trad, (Mike Rivers)
wrote: If I use a 1:1 power transformer, and thus have no galvanic connection to the wall hot and neutral, I still have high DC and AC voltages in my circuit, and my chassis is still connected to the wall safety ground... there are plenty of ways a user could get a serious shock by touching things inside the device. But the "hot" side of your transformer would no longer be connected to the wire that's hot relative to the chassis ground. I guess this is what I fundamentally don't understand... what's the difference ? If there's a 120v difference between the two sides of the secondary, doesn't the chassis ground have some quantifiable voltage potential in relation to the two sides of that secondary in spite of the fact that there's no galvanic connection (at least, until someone starts touching things inside the circuit unwisely, of course) ? And doesn't at least one side of the secondary *have* to be at least 60v different from chassis ground, no matter *what* the chassis is connected to ? AC current flows through the transformer even though there's no galvanic connection, so I guess I don't understand how a voltage on one side of the transformer is any safer than a voltage on the other side. If there's a potential difference between a tap on the secondary and earth ground, and you touch both spots, won't current flow from the wall right through the transformer, then through you to ground ? Thanks again for taking the time to tutor me in electrical safety! James |
#13
|
|||
|
|||
Building a circuit with no power transformer ?
James Nash wrote:
I guess I don't fully understand the implications of decoupling here. Are you saying that if I connect 120v phase current to one side of a 1:1 transformer primary and neutral or ground to the other side of the primary, that the two sides of the secondary are somehow less lethal than the original wall current that's connected to the primary ? There's still 120v between the secondaries, so I guess I don't understand why I wouldn't still get the same shock if I grabbed the two sides of the secondary. And wouldn't there still be roughly 120v between the higher voltage side of the secondary and either the wall's neutral or earth ground ? Logically, as a general principle, wouldn't there *have* to be at least 60v minimum potential difference between one side of the secondary and *any* neutral or ground reference ? I understand that there's no physical connection between the secondary and the house current, but AC current is still induced through the transformer, right ? There *must* be something here that I fundamentally don't understand... if there's a 120v swing inside the circuit, doesn't there exist a lethal potential between *some* places in the circuit and *any* earth reference you choose ? Does the safety lie in the fact that the transformer limits the amount of current that can flow from the wall to the secondary ? I think I understand what you're asking, and I think I know the answer. You hook up the 120VAC mains to the primary of a 1:1 power transformer. The secondary has 120V across it. But no, there is no 120V circuit that you can complete with your body between a secondary lead and a primary lead. Well, there is, but it's only due to the inter-winding capacitance of the transformer, which is small, and will not let very much 60Hz current through at all. But the 120VAC *secondary* circuit is ISOLATED from the primary circuit. They're not connected. So, suppose your body is grounded (as it usually is). Suppose you touch one of the leads on the transformer secondary by accident. Will you get a shock? Probably. Will it kill you? Probably not. Now, suppose you touch the hot lead on the primary. Much bigger shock. Because the 120VAC mains supply is referenced to "earth" and that's where you are. But the transformer secondary is referenced to the other end of the transformer secondary, which you are hopefully not connected to. Of course, if the circuit ground is connected to the chassis also, then you ARE connected to it. On a fairly separate point from one of your other posts that I think you had a grasp of but that warrants further discussion... You intuitively suppose that with the chassis bonded to safety ground, you're pretty much safe with or without a power transformer, and if the safety ground is disconnected then you're pretty unsafe either way. But the thing you need to remember is that the safety ground is NEVER really ground. It's sitting on the other end of a long resistor (a piece of copper wire) from "real" ground. So there's more than simple "connected" and "disconnected" safety ground. It's always somewhere in between. Hopefully it's sitting at a few tens of millivolts or less, but it's not zero. In very many cases, the safety ground is not carried by copper wire, but by glavanized steel conduit that goes through perhaps dozens of junctions between the panel and the outlet. The impedance of that ground connection is determined in large part by how clean those junctions are and how tightly fastened they are. Some appliances don't connect to the safety ground at all. But if you do, you have to make sure you're protected from the possibility of that ground having voltage on it. ulysses |
#14
|
|||
|
|||
Building a circuit with no power transformer ?
James Nash wrote:
AC current flows through the transformer even though there's no galvanic connection, so I guess I don't understand how a voltage on one side of the transformer is any safer than a voltage on the other side. If there's a potential difference between a tap on the secondary and earth ground, and you touch both spots, won't current flow from the wall right through the transformer, then through you to ground ? No. AC current does not flow from the primary to the secondary, except through parasitic capacitance. It's only a magnetic field that connects the two sides. Magnetic fields don't electrocute you. Try this: Take an ungrounded 12VAC wall wart and plug it into an outlet. Now take an AC voltmeter and measure the voltage between one of the wallwart output terminals and the powerline safety ground. How many volts would you expect to see? 120 or 12? Or somewhere in between? I just tried it, and I got about 0.8 volts. That small voltage is only there because of interwinding capacitance, and even that would disappear if the voltmeter had a lower impedance. If you measured between the wallwart output and the powerline hot, you'd get about the same thing (I got about 2.3V). Even though one is at 120V and one is at 12V, there is no real voltage between them. ulysses |
#15
|
|||
|
|||
Building a circuit with no power transformer ?
James Nash wrote:
I guess I don't fully understand the implications of decoupling here. Are you saying that if I connect 120v phase current to one side of a 1:1 transformer primary and neutral or ground to the other side of the primary, that the two sides of the secondary are somehow less lethal than the original wall current that's connected to the primary ? Yes, because they are not referenced to the building ground. There's still 120v between the secondaries, so I guess I don't understand why I wouldn't still get the same shock if I grabbed the two sides of the secondary. And wouldn't there still be roughly 120v between the higher voltage side of the secondary and either the wall's neutral or earth ground ? Logically, as a general principle, wouldn't there *have* to be at least 60v minimum potential difference between one side of the secondary and *any* neutral or ground reference ? You are always in some contact with the building ground reference. --scott -- "C'est un Nagra. C'est suisse, et tres, tres precis." |
#16
|
|||
|
|||
Building a circuit with no power transformer ?
|
#17
|
|||
|
|||
Building a circuit with no power transformer ?
|
#18
|
|||
|
|||
Building a circuit with no power transformer ?
One additional point, when discussing "hot chassis" circuits (without a power
transformer) is that no assumptions can be made about which side of the line IS grounded. The code is specific (White wire, wide blade, silver screw, etc), but this relies on everyone (electrician or not) who ever worked on the wiring, from the outlet to the power pole, to do meticulous work. It's probably not a good idea to bet your life on this. Mike Rivers wrote: In article writes: I guess this is what I fundamentally don't understand... what's the difference ? If there's a 120v difference between the two sides of the secondary, doesn't the chassis ground have some quantifiable voltage potential in relation to the two sides of that secondary in spite of the fact that there's no galvanic connection (at least, until someone starts touching things inside the circuit unwisely, of course) ? And doesn't at least one side of the secondary *have* to be at least 60v different from chassis ground, no matter *what* the chassis is connected to ? Grounding the chassis can't protect idiots and those really careless from getting shocked. You're correct that with one side of the power supply going to earth through the safety ground, you can indeed get shocked if you stand in your bare feet on a reasonably well grounded surface and touch the hot lead of the power supply. What grounding the chassis does is assures that if something goes wrong inside the box, like a component shorts out and connects the HOT side of the power supply to the chassis, this will short out the power supply and hopefully blow a fuse before doing any other damage. If it was possible for the hot side to get connected to the chassis inadvertently, by touching the chassis, you could become the wire that completes the circuit from the hot side (which is now connected to the chasis) to earth. This is why it's not a good idea to lift a safety ground. -- I'm really Mike Rivers - ) |
Reply |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Forum | |||
System warm-up | Audio Opinions | |||
rec.audio.car FAQ (Part 2/5) | Car Audio | |||
rec.audio.car FAQ (Part 1/5) | Car Audio | |||
FS: SOUNDSTREAM CLOSEOUTS AND MORE!! | Car Audio | |||
old solid state circa 70-80's` | Audio Opinions |