| Author |
Message |
|
|
 Posted: Aug 01, 2009 - 01:42 AM |
|
Joined: Dec 11, 2007
Posts: 6980
Location: Cleveland, OH
|
|
I searched a few words and threads but did not seem to fine what I was looking for.
In "the old days" wall wart power supplies had a transformer in them, and one could power separate portions of a circuit with two or three wall warts, and they were all isolated, floating, and independant of each other. The grounds were not tied to each other back through the 120 VAC building wiring.
With the newer, physically smaller and more effieicent switching wall warts, are the V+ and Gnd leads isolated from the 120 VAC terminals, and hence floating with respect to earth ground, and building neutral?
A simple test with an ohm meter indicated that they were, but I'm not familiar with the circuitry used inside them.
For a quick testbed project it would be easier if they truely floated with respect to each other.
Any thoughts?
JC |
|
|
| |
|
|
|
|
|
|
Posted: Aug 01, 2009 - 03:04 AM |
|
Joined: Mar 28, 2001
Posts: 20624
Location: Sydney, Australia (Gum trees, Koalas and Kangaroos, No Edelweiss)
|
|
|
Quote:
are the V+ and Gnd leads isolated from the 120 VAC terminals, and hence floating with respect to earth ground, and building neutral?
I would sincerely hope so or you would have many more fried patients in the ER ward. 
Usually there is a bridge with a big cap smoothing/storing the rectified mains, then there is the switch mode regulator feeding a transformer which supplies the DC to your circuit. Again usually there will be an optocoupler for feedback to the regulator to keep the voltage stable.
You may want to have a look at a site like "power integration" and you should have many examples in the data sheets. www.powerint.com |
_________________
John Samperi
Ampertronics Pty. Ltd.
www.ampertronics.com.au
* Electronic Design * Custom Products * Contract Assembly
|
| |
|
|
|
|
|
|
Posted: Aug 01, 2009 - 03:40 AM |
|
Joined: Dec 11, 2007
Posts: 6980
Location: Cleveland, OH
|
|
John,
Got it!
Thank you for the link, also.
Your description makes perfect sense, but I wanted some guidance before I fried something, (like myself  ).
JC |
|
|
| |
|
|
|
|
|
|
Posted: Aug 01, 2009 - 04:06 AM |
|
Joined: Mar 28, 2001
Posts: 20624
Location: Sydney, Australia (Gum trees, Koalas and Kangaroos, No Edelweiss)
|
|
All care taken but no responsibility accepted. 
When I lived in Italy we had 2 pins power plugs. To save costs things like TVs and Radio had autotransformers which meant that 1 leg of the mains went to chassis (metal). It was meant that the neutral went to ground but the plugs are reversible, so one learned very quickly to either have a neon light screwdriver that would indicate a live chassis or stand on your heels and quickly brush the chassis with your hand.
A tingle would mean "reverse the plug"!! |
_________________
John Samperi
Ampertronics Pty. Ltd.
www.ampertronics.com.au
* Electronic Design * Custom Products * Contract Assembly
|
| |
|
|
|
|
|
|
Posted: Aug 01, 2009 - 05:02 AM |
|
Joined: Dec 11, 2007
Posts: 6980
Location: Cleveland, OH
|
|
|
Quote:
All care taken but no responsibility accepted.
I think you are safe... If I die I won't be around to complain!
I remember working on some of those old AC/DC tube radios with live chassis long ago. In retrospect I'm lucky I didn't kill myself a time or two.
It was some of those memories, and others, that elicited the original question above. I saw, on the link you provided, several data sheets with exactly the circuit you described, (a transformer included which provides isolation), and a few others without transformers, (scary thought).
I appreciate your insight!
JC |
|
|
| |
|
|
|
|
|
|
Posted: Aug 01, 2009 - 07:22 AM |
|
Joined: Nov 29, 2007
Posts: 3219
|
|
|
DocJC wrote:
With the newer, physically smaller and more effieicent switching wall warts, are the V+ and Gnd leads isolated from the 120 VAC terminals, and hence floating with respect to earth ground, and building neutral?
Not always.
The low-voltage and the high-voltage parts are isolated, but
a) At least in the EU authorities regularly find and withdraw wall warts from the market where the required isolation, creepage, etc. was not observed.
For a thrill, you can go to http://ec.europa.eu/consumers/dyna/rape ... ves_en.cfm and type "charger" or "electric" into the search field. Note the country of origin of most violators.
Because of the country of origin I assume that similar wall warts could be on the market in the US.
b) Even when within regulations, the high-voltage and low-voltage side need not the be "100%" isolated. The use of capacitors to connect the low-voltage and high-voltage ground (there is a high-voltage ground, because the AC is rectified into DC on the "primary" side) is permitted under certain circumstances (usage of "Y" class capacitors etc. required)
Generally speaking, medical equipment must not be powered from consumer-graded wall warts because of these possibilities. Medical-equipment grade power supplies require more isolation (double? tripple? sure depends on local regulations) for a good reason. Consumer.grade wall warts, especially the budget ones, don't provide the isolation.
Consider using batteries. |
|
|
| |
|
|
|
|
|
|
Posted: Aug 01, 2009 - 07:47 AM |
|
Joined: Apr 25, 2004
Posts: 3817
Location: Denmark
|
|
|
ArnoldB wrote:
For a thrill, you can go to http://ec.europa.eu/consumers/dyna/rape ... ves_en.cfm and type "charger" or "electric" into the search field. Note the country of origin of most violators.
With 1 Bn. ppl , they might think they can afford to loose a few 
/Bingo |
|
|
| |
|
|
|
|
|
|
Posted: Aug 02, 2009 - 12:09 AM |
|
Joined: Dec 11, 2007
Posts: 6980
Location: Cleveland, OH
|
|
Hi Arnold,
Thank you for the info.
This is just for a quickie weekend project, not for a "real application", or for anything medically oriented. (An AVR ammeter to watch another AVR PCB...).
Hi Bingo,
I just wanted to make sure I was not going to be one of the individuals zapped!
JC |
|
|
| |
|
|
|
|
|
|
Posted: Aug 02, 2009 - 01:20 AM |
|
Joined: May 26, 2004
Posts: 2538
Location: Las Vegas, Nevada
|
|
|
js wrote:
A tingle would mean "reverse the plug"!!
American guitar amplifiers from the tube era (showing my age here) often had a 2 pin plug and a toggle switch on the back that would connect the chassis to one side or the other of the 110V supply, for hum cancellation. Allowing one's lips to touch the microphone would generally reveal whether the switch needed reversing, if you survived. Mortality in the Hamburg clubs was further enhanced by the German practice of using red for the ground wire, whereas the British used red for live, before the brown-blue-green/yellow convention was adopted. I still have a Schaller reverb unit with a red ground.
Regarding switch mode supplies, I have known equipment with multiple PSUs in it to fail compliance testing because the combined ground leakage current was too high. Changing to medical grade supplies fixed it, though at substantial extra cost. The leakage current has to go somewhere, presumably into the low voltage equipment in the case of a wall wart, and thence to ground through someone's body, but the current won't be hazardous unless a fault develops. |
|
|
| |
|
|
|
|
|
|
Posted: Aug 02, 2009 - 01:49 AM |
|
Joined: Sep 07, 2004
Posts: 2563
Location: New York State
|
|
I had the same question as you. I decided to use a wall wart with transformer.
Using batteries is a good idea when practical. But my handheld DMM that I converted to use a wallwart is one of my best inventions. It's been on for a year now and may last a lifetime. It's generally monitoring the current draw of AVRs. What a relief not having to change the bleepin' 9 volt battery.
You do bring up an interesting question. I have a dozen or so of the transformer type wall warts, but if I ever have to buy one, how can I be sure it has a transformer?
I guess I would avoid regulated ones. Maybe the AC output ones would be most likely to have a transformer. |
|
|
| |
|
|
|
|
|
|
Posted: Aug 02, 2009 - 01:51 AM |
|
Joined: Oct 30, 2002
Posts: 5727
Location: The Netherlands
|
|
| Weight is usually a good indicator |
|
|
| |
|
|
|
|
|
|
Posted: Aug 02, 2009 - 03:34 PM |
|
Joined: Jun 17, 2002
Posts: 300
Location: Melbourne, Australia
|
|
ArnoldB wrote:
Quote:
b) Even when within regulations, the high-voltage and low-voltage side need not the be "100%" isolated. The use of capacitors to connect the low-voltage and high-voltage ground (there is a high-voltage ground, because the AC is rectified into DC on the "primary" side) is permitted under certain circumstances (usage of "Y" class capacitors etc. required)
This is a point I have made many times regarding the 2 pin variety of Plug Packs / Wall Warts.
The capacitor if fitted will capacitively couple the mains to the output.
Please take the time to measure either output terminal to Earth with an AC meter. Then with a CRO.
Then You decide if the supply is suitable for your application.
Most dammage occurs when connecting power supplies to grounded circuits.
SMPS power supplies that are grounded, have these capacitors between the output and earth, Not output to rectified mains, so usually dont pose any problems.
Ron. |
|
|
| |
|
|
|
|
|
|
Posted: Aug 03, 2009 - 09:08 AM |
|
Joined: Jun 11, 2004
Posts: 2422
Location: Vilanova (Barcelona)
|
|
I had developed one SMPS from mains to 5V and 12V once. In fact, many regular (except some low power) SMPS's have a little cap (Y class safety type) that connect a 'low noise point' on the HV side to a 'low noise point' on the LV side for conducted noise supression. A poor design can lead to some little AC, 'HF' (low KHz) to pass to the LV side. Check PowerInt app notes (they have a lot!!) about that. Sizing this cap is usually done by "succesive approximation".
Isolation transformers also have some low capacitive coupling between both sides that can affect it.
Anyway, for many hobbyist applications at my home, I used these kind of supplies, but my preferred for +5V supply is the USB bus from PC's. |
_________________
Guillem.
"Common sense is the least common of the senses" Anonymous.
|
| |
|
|
|
|
|
|
Posted: Aug 03, 2009 - 10:29 AM |
|
Joined: Oct 01, 2002
Posts: 304
Location: Redbourn, Herts, UK
|
|
JC,
You can perform the test on a switch mode supply yourself using a DVM. You may be surprised at the results and at what performance is actually legal.
1. Measure the AC volts between DC output and ground. You could find approx. half mains voltage here.
2. Measure the AC leakage current between DC output and ground. On many examples I've tried I find approx 110uA.
If your project involves any ADC work, then, if that leakage current flows in signal lines significant errors can result.
Nigel |
|
|
| |
|
|
|
|
|
|
Posted: Aug 03, 2009 - 12:45 PM |
|
Joined: Dec 11, 2007
Posts: 6980
Location: Cleveland, OH
|
|
Hi everyone,
I appreciate everyone's insight.
Ron and Nigel have both suggested simply measuring the AC potential, and leakage. I haven't done that yet, but will, either this evening or tomorrow.
I sure hope not to find Vmains/2 on the plug. That would suprise me, but then that's why I started the thread. I don't know much about switching supplies, or these wall wart modules.
As it turns out it is for an ADC, (ammeter), circuit. I ended up using two ADC inputs, measuring across the current sense resistor with respect to ground, instead of just one ADC, (single ended), across the sense resistor. This avoided the entire issue of having a secondary supply floating with respect to the device under test.
Either way, it is still a good topic for me to learn about, as I am sure the subject will present itself again.
Hi Guillem. USB supplies are convienient , but not for this particular project. I'll take a look at those app notes, too.
Hi Steve. Yes. I, too, do not wish to keep replacing batteries!
Hi Peret,
I can't imagine using Red for Ground connections...Maybe there is a reason for Standards after all!
Quote:
Weight is usually a good indicator
JayJay, is that a reference to the Wall Wart, or the Designer
JC |
|
|
| |
|
|
|
|
|
|
Posted: Aug 03, 2009 - 12:53 PM |
|
Joined: Oct 30, 2002
Posts: 5727
Location: The Netherlands
|
|
The wall wart  |
|
|
| |
|
|
|
|
|
|
Posted: Aug 03, 2009 - 01:32 PM |
|
Joined: Apr 25, 2006
Posts: 2304
Location: Germany
|
|
|
Quote:
sure hope not to find Vmains/2 on the plug. That would suprise me, but then that's why I started the thread. I don't know much about switching supplies, or these wall wart modules.
If you use a high-impedance Voltmeter (or a high
impedance scope) you may well find Vmains/2.
One explanation is the following: In many
designs the output ground is connected
to the both input lines with a Y capacitor
in the order of a few nF. This
capacitors are a voltage divider with respect
to the mains, and the midpoint is connected
to the secondary. So you have a open loop
voltage of Vmains/2 with a internal impedance
of somehundred kOhms. |
|
|
| |
|
|
|
|
|
|
Posted: Aug 03, 2009 - 01:59 PM |
|
Joined: Dec 11, 2007
Posts: 6980
Location: Cleveland, OH
|
|
Hi Ossi,
ArnoldB mentioned the Y cap above, also, but I had not tracked down the reference, yet.
Your explaination is great.
Thanks,
JC |
|
|
| |
|
|
|
|
|
|
Posted: Aug 03, 2009 - 03:51 PM |
|
Joined: Dec 07, 2008
Posts: 347
Location: germany
|
|
wow, gosh ... now I TOTALLY know why we have to connect ground to ANY metal chassis in germany. ZAPPP there goes the fuse.
Quote:
Anyway, for many hobbyist applications at my home, I used these kind of supplies, but my preferred for +5V supply is the USB bus from PC's.
Is there any over current protection for USB? I know, that USB slaves have to request to use the full 500 mA scale. But some devices do it without asking.
I use a PC power supply for experiments and it saved my circuits at least two times. I produced shorts between Vcc and GND and my supply simply turned off^^ |
_________________
Jan Kießling
74HC Overview
Flowchart CAD
schematic CAD
|
| |
|
|
|
|
|
|
Posted: Aug 03, 2009 - 05:19 PM |
|
Joined: Nov 01, 2005
Posts: 6354
Location: Hilversum - the Netherlands
|
|
Nephazz, then call yourself lucky 
The use of a PC PowerSupply as lab-supply has been discussed several times, and the general conclusion was: don't !
Back to Doc's question: I also wondered what these SMPS's give as a leakage current. So I took a bunch of them and did measurements. Preliminary results can be found here:
http://www.aplomb.nl/SMPS_leakage/Doc_ie.html
I will add more info as I collect it.
Nard |
_________________
Dragon broken ? Or problems with the Parallel Port Programmer ? Scroll down on my projects-page http://www.aplomb.nl/TechStuff/TechStuff.html for tips
|
| |
|
|
|
|
|
FAQ
Search
Register
Log in to check your private messages
Log in
Maximize
RSS
