7

If I have two identical AC/DC adapters(Such as laptop adapters) can I achieve this as in the following illustration? Would it be safe?: enter image description here

user16307
  • 11,802
  • 51
  • 173
  • 312
  • 2
    It is not duplicate! His question is about two "non-identical" wallwarts. My question is about two "identical" adapters. Please re-check! – user16307 Mar 30 '15 at 13:22
  • He used two identical 9V, 300mA adapters. This looks the same as your question. – Greg d'Eon Mar 30 '15 at 13:22
  • 2
    Re read what he wote: "I have two 9v, 300mA wallarts and would like to connect them in series to produce +9v, 0v and -9v. This can be done with 9v batteries (as seen below), but can I do this with two non-identical wallwarts?" The question involves non-identical so it is not duplicate, it is similar. – user16307 Mar 30 '15 at 13:23
  • @user16307: It applies here too, you can never find two identical ones anyways, and if it works with two unspecifie non identical ones... – PlasmaHH Mar 30 '15 at 13:24
  • 1
    Oops - I misread your comment. The answer does not change for non-identical wall-warts. My answer on the linked question does not depend on the fact that they are both 9V. – Greg d'Eon Mar 30 '15 at 13:24
  • 1
    Of course nothing is identical in that sense. Identical means the current voltage ratings are the same from the same brand ect in this context. – user16307 Mar 30 '15 at 13:25
  • 4
    If two non-identical wall warts can be used for this purpose, why would you think two identical ones could not? If anything this should have been a comment on Greg's answer to the duplicate question. – I. Wolfe Mar 30 '15 at 13:28
  • 1
    this is not a duplicate question! the thing that answer is the same, does NOT MAKE IT DUPLICATE. – user16307 Mar 30 '15 at 13:29
  • Having the same answer is a very strong indicator that the question is a dup (See some discussion on meta.SE). Thinking that there is some difference while in fact the information is irrelevant is no reason too. A big point oft marking things AS dup is that we keep the alternative formulations of a problem for people to find – PlasmaHH Mar 30 '15 at 22:42
  • 1
    no i don't have to agree. to me it is not logical. go and ask to philosophy department. those questions are NOT duplicate. – user16307 Mar 30 '15 at 23:10

2 Answers2

2

There is nothing fundamentally wrong with what you propose. Any purpose built dual output power supply is essentially the same thing. So using this approach can be a workable scheme to achieve a dual output + and - rail output.

One thing that must be considered here is that the two power supplies need to have fully floating outputs.

There are a number of serious disadvantages however. Here are some things to consider:

  1. Can the target device powered by this scheme survive safely if one or the other output becomes 0V because one of the wall warts gets unplugged.

  2. A purpose built dual output power supply will have certain fault detection features that will apply equally to both the + and - outputs which will safely shut things down in case of problems. One example would be a primary side safety fuse that blows and disconnects both the outputs from the mains.

  3. Invariably a dual output supply will share some common components and circuits for supporting both outputs. In particular this would be in the mains conversion and isolation circuit. This can lead to a smaller overall design than using two completely separate power supplies.

Michael Karas
  • 56,889
  • 3
  • 70
  • 138
  • How can I be sure that the Adapters are galvanically isolated. Is there a technic to check that. im afraid of a possible short circuit. – user16307 Mar 30 '15 at 13:38
  • The check for galvanic isolation is simple. Use an ohm meter and make sure that there is no connection between either the output wires and any of the input terminals. – Michael Karas Mar 30 '15 at 13:41
  • If there is galvanic isolation, is that because there is transformer in between as an interface? – user16307 Mar 30 '15 at 13:44
  • 2
    Be aware that if you are using "laptop" adapters for this scheme that the the primary side of the adapter will likely have a three contact plug which includes a safety ground connection. Often the low side of the output connects through the supply to this safety ground. In this case you cannot use the scheme pictured. – Michael Karas Mar 30 '15 at 13:45
  • will i short and burn it in that case? – user16307 Mar 30 '15 at 13:46
  • i should remove earth and it is safe? – user16307 Mar 30 '15 at 13:47
  • Yes there is almost for sure a transformer providing mains voltage isolation from the outputs. But the adapters could still have a safety ground through connection which would mean that the output is not fully isolated from the inputs. – Michael Karas Mar 30 '15 at 13:47
  • hmmm.. can we say that if in the primary end we do not have earth we wont face problem? – user16307 Mar 30 '15 at 13:49
  • You need to have 100% isolation for the two supplies to work together safely. I can never recommend that you remove safety ground connections from any equipment designed to be operated with such connection. – Michael Karas Mar 30 '15 at 13:50
  • if we have 2 prong adapters can we say that we are in safe heaven? – user16307 Mar 30 '15 at 13:52
  • Make the ohm meter check whether you have a two prong AC side or not. Make sure the adapter is 100% isolated period. Also, as I said before, I cannot recommend that you operate ANY line operated device without its safety ground connection if it is designed and equipped with such connection. – Michael Karas Mar 30 '15 at 13:57
  • Given that both power supplies are isolated (or this wouldn't work at all), how would you expect 1. to work - wouldn't a single power supply being unplugged create an open in the circuit? – user2813274 Mar 30 '15 at 15:55
  • @user2813274 good question. i wonder if anybody really tried this? im scared to experiment.. – user16307 Mar 30 '15 at 20:53
2
  • This architecture would work if the wall adapters are isolated (some wall adapters aren't on purpose). No power applied (offline), connect the - terminal of the primary to the - terminal of the secondary, and test the resistance between the remaining + terminals.
  • HOWEVER when the Earth is isolated/not present (2 prongs or 3 prongs with the earth prong being made out of plastic for example), you are giving up on a very important safety feature: residual current devices. They cut off the supply when they detect a leak to Earth (a person or a short to case) which can be lethal. You NEED to make sure the secondary voltages are inherently safe for you to accidentally touch, given the environment. The voltage between V+ and V- will be twice V+, in particular. Normally, the isolation would prevent any person from closing the circuit rendering it safe, but if not tied to Earth the voltage can float quite high and a fault would be dangerous.

Edit/note: The safest is 2 terminals plus Earth connected to case, and Line-Neutral of primary isolated from secondary. This way even if there is a fault, the residual current devices of your house will trigger. 2 terminals which are isolated from the primary, Earth absent, is safe in the absence of fault but dangerous if there is one. 2 terminals which are not isolated from the primary, Earth absent, is generally very risky: consider a very simple voltage divider as an example, and imagine the bottom one breaks - the entire line voltage is present at the output...

  • AND protect the circuit against overcurrents/short circuits if the wall adapters do not include any, to prevent fire.
  • The following is circuit specific: some circuits will behave dangerously if one of the supply rails is missing. Since it's very likely that one adapter may be plugged without the other, this should be taken care of. Say, with some logic that inhibit both if one of them is missing.
Community
  • 1
Mister Mystère
  • 9,477
  • 6
  • 51
  • 82
  • can we say that in 3 prong case GND of secondary ends might be connected to earth and we are in danger; but in case of 2 pong we do not have earth and it is safe?? – user16307 Mar 30 '15 at 13:50
  • I have clarified my answer, I think. – Mister Mystère Mar 30 '15 at 14:01
  • There are some failure modes where having a safety ground connection can create dangers that would not exist in its absence. For example, it is legal in the U.S. to have three-prong outlets guarded by a GFCI (U.S. equivalent to an RCD) without the earth terminal connected, if such outlets are marked "No equipment ground". If two devices are plugged into outlets guarded by the same GFCI, and a fault shorts the ground of the first to "hot", the case of the second may be electrified until such time as a current path exists from the ground wire to something else, whereupon the GFCI would trip. – supercat Mar 30 '15 at 19:39
  • The fact that the GFCI would trip would likely render any shock issued as a consequence of such a fault non-lethal, but the current could still easily be sufficient to cause severe equipment damage or harm unusually-sensitive individuals. – supercat Mar 30 '15 at 19:40
  • Isn't the short to case a leak through Earth which triggers the RCD/GFCI? The current ought to be high enough, since everything is metallic on its path. – Mister Mystère Mar 31 '15 at 10:02