USB power from mains

Go To Last Post
16 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I started using a 1284 xplained board yesterday to provide a USART test signal to a new project. While getting it going, I discovered that it's powered by USB. Having no PC USB ports handy, I decided to plug the USB cord into a USB port on one of those power strips plugged into mains power. That worked fine, and I set to troubleshooting the new circuit. After a few minutes I noticed strange behavior in the circuit, and I looked at it with my scope. What I saw was 57VAC at 60 HZ on the ground radiating all through the xplained and my circuit under test.

 

I felt like I had been violated. I pulled the USB cord out of the wall strip and the voltage disappeared. My question for you guys is how are those things isolated (or supposed to be anyway)? There's no transformer. Is it just a little regulator? Also, any guesses on how much current that connection could carry?  The AVR on my new circuit is still working and I saw no smoke, but the DSP may be inop now. How will stray AC on the ground plane affect these modern IC's, any idea?

 

I'm not going to do that again.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

jimlake wrote:
There's no transformer.
Should ... therefore it's fraudulent (UL?  IC?  etc)

Am glad there was no harm and the loss was limited.

jimlake wrote:
Also, any guesses on how much current that connection could carry?
Multiples of 100mA

jimlake wrote:
How will stray AC on the ground plane affect these modern IC's, any idea?
An AVR's signals are clamped by AVR's ESD diodes and AVR's VCC shunt.

jimlake wrote:
I'm not going to do that again.
The reverse case can occur where a circuit malfunction, or lack of operator headroom (been there, done that, micro-mushroom cloud), results in a popped USB host port or worse (dead notebook PC motherboard)

Some USB isolators also provide isolated USB VBUS.

 


USB in a NutShell - Chapter 2 - Hardware, Power (VBUS)

Earth Ground

by Dr. Howard Johnson

...

In my opinion the most important point to make with regard to grounding is that the input to every digital logic gate is a DIFFERENTIAL amplifier. That's right--a differential amplifier. This differential amplifier compares the digital input signal to some local reference (often generated inside the chip), and decides which is bigger (more positive).

...

 

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 1

Perhaps this Thread would be worth a read through:  Switching Wall Warts, Isolated Mains.

 

Nard, (Plons), did some rather extensive testing on little transformerless power supplies.

 

JC

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

This device is not a wall wart. It's a power strip with a bunch of 120 receptacles and a couple of USB ports.

I looks like this:

 

https://www.amazon.com/KMC-4-Out...

 

It does not have a transformer, as far as I can tell.

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Are you sure? The transformer will be very small.

#1 This forum helps those that help themselves

#2 All grounds are not created equal

#3 How have you proved that your chip is running at xxMHz?

#4 "If you think you need floating point to solve the problem then you don't understand the problem. If you really do need floating point then you have a problem you do not understand." - Heater's ex-boss

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Just about every switching psu will do this that doesn’t use an earth. Even my macbook pro psu does this.
There is a capacitor from live to the 0V output to shunt the switching energy due to the transformer inter-winding capacitance. The current is less than 1mA. Enough to give you a tingle. Especially at 240V.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Would an isolation transformer solve this problem? By that I mean if I plugged the power strip into an isolation transformer and plugged it into mains.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Yes, the leakage through a true isolation transformer is very low, much less than you are currently experiencing, (although not 0.0...)

 

The circuit is then floating with respect to the Mains.

 

The Xplain board does need a Ground connection to the rest of the circuit, (The target for the USART signal).

This Ground then becomes the reference Ground for the Xplain board.

 

The point of the link was that the USB gizmo's and the wall warts in the link often have a very similar circuit inside the case.

 

JC 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

jimlake wrote:
There's no transformer.

 

Maybe it's one of these?

https://en.wikipedia.org/wiki/Ca...

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Snap!.... Crackle!....Pop!.....Chinese Electronics...  from:https://www.youtube.com/watch?v=...

 

Looks like you got another magic power cord from the OneHungLow factory.   Hey! if +5V is good, then +50V is better!  10 times better!  Why you have problem?

I have another cord from the same factory, the 19V/3A power brick/adapter for my laptop.  I get a half centimeter white arc every time that I plug it into the wall.  I'm told that it's considered normal. I don't know what causes it.  Possibly a giant surge through the switching-mode power circuitry when the internal components are discharged?

 

Ever see Dave Jones' EEB video about the most dangerous device ever sold to the West by a Chinese factory?  It's a "tea maker": it boils a cup of water for a quick cup of tea.  It's two bare pieces of metal that plugs directly into 220VAC mains on one end and inserts into your cup of cold water on the other end, with nothing but wire in between.  Guaranteed to kill anything from a hamster to a rhino!

 

A USB jack on the side of a Chinese power strip...what could possibly go wrong?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Simonetta wrote:
Ever see Dave Jones' EEB video about the most dangerous device ever sold to the West by a Chinese factory?  It's a "tea maker": it boils a cup of water for a quick cup of tea.  It's two bare pieces of metal that plugs directly into 220VAC mains on one end and inserts into your cup of cold water on the other end, with nothing but wire in between.  Guaranteed to kill anything from a hamster to a rhino!

 

Who needs that chinese crap? This is how it's done:

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Who needs safety stuff anyway. As he shows, you just need to be careful! I wonder if he heats his vodka drinks on those cold Russian nights that way. After the fourth brimming cup of hot vodka slam, you reach for your fork for another bite of jellied perch and.............slam, no need for social security.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Somehow we've managed to confuse standard design practice with low quality devices from the Orient. With switching supplies, the inter-winding capacitance becomes significant at the switching frequencies involved - usually 50kHz+. If you don't shunt this, then the switching energy gets coupled into your output which radiates and can also upset your load as the energy wants to find its way back to earth. I learnt this the hard way with isolated (low voltage) dc/dc converters. 300kHz upset my 1 wire comms. To shunt this inter-winding capacitance, we need a capacitor to shunt it back to whence it came. The downside of this is we get leakage. In an earthed system it is shunted back to earth, but with no earth, it goes back to active. So, if you have a switchmode supply that has an earth connection - it needs to be earthed.

 

For those that use laptops and use an oscilloscope you'll find you get significant mains voltage imposed onto your supposed 5V from the laptop USB. Once you ground the circuit via the scope probe, the problem goes away (unless the leakage current is enough to cause issues on its way back to earth). For a 'normal' computer you never see it as the power cord will most likely have an earth.

 

The audio guys get hit with this problem as it causes significant audible hum if you're not careful.

 

 

As an aside, I was asked to investigate why someone got electrocuted (no death) by some equipment that was supplied by the company I worked for. These were some video to ethernet converters powered by a wall wart. The customer had 20 of these mounted on a 19" rack tray with 20 wall warts on plug bars (or whatever they're called). The combined leakage current of the wall warts became significant and posed a danger. So don't do it - use one suitably rated power supply for all.

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Here is the way that I decided to solve this problem. There is a 5v regulator in the black shrink tube. Rechargeable NiMH batteries. Nice clean 5v from a USB Mini B.

Attachment(s): 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

the ubiquitous power banks work a treat. When I was in the 'States earlier this year I got a couple from Fry's at $2.90 each. The batteries were of dubious quality, but I couldn't walk past the price!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

"Never mind the quality, feel the width."

 

Ross McKenzie ValuSoft Melbourne Australia