Author Topic: AC/DC Desktop adapter, non floating ground?  (Read 368 times)

0 Members and 1 Guest are viewing this topic.

Offline ninjanation217

  • Newbie
  • Posts: 4
  • Country: us
AC/DC Desktop adapter, non floating ground?
« on: September 11, 2019, 01:58:00 pm »
Hello everyone, I have what I hope is a simple question.

I need a desktop power supply (wall wart) like this:
https://www.digikey.com/product-detail/en/mean-well-usa-inc/GST60A24-P1J/1866-2152-ND/7703715

but, I need the ground to be connected to earth ground. What spec do I need to be looking for to find this?

Thanks,

- James
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 12888
  • Country: gb
  • Hero999
Re: AC/DC Desktop adapter, non floating ground?
« Reply #1 on: September 11, 2019, 03:28:18 pm »
I think the only way is to read the data sheet or ask the manufacturer.

In the case of the PSU you've linked to. It clearly has a floating output. Refer to page 3 of the data sheet:
Quote
-V not connected to AC FG
https://www.meanwell.com/Upload/PDF/GST60A/GST60A-SPEC.PDF

Why do you need the output to be grounded? Floating is generally better as it offers greater flexibility.
 

Offline ninjanation217

  • Newbie
  • Posts: 4
  • Country: us
Re: AC/DC Desktop adapter, non floating ground?
« Reply #2 on: September 11, 2019, 04:06:57 pm »
Thanks for your reply!

Yeah it looks like I'm just going to contact Mean Well or someone, as not all data sheets list the earth connection, turns out I'm looking for a grounding configuration of E1. I have found most of my laptop power supplies are grounded in this way.

I know its normally better, however in my particular case it helps.

I am making a photometer using TI DDC112. If I am not using an integration time of multiples of 16.66 ms (because of the 60 Hz mains, long story) I see a sine wave in my measurement at extremely low levels. however, I have found attaching earth ground minimizes this. I have tried filtering to no avail, so, I'm just attaching earth ground as its the easiest fix.

- James
 

Offline ejeffrey

  • Super Contributor
  • ***
  • Posts: 1907
  • Country: us
Re: AC/DC Desktop adapter, non floating ground?
« Reply #3 on: September 11, 2019, 04:25:22 pm »
This is probably caused by common mode leakage from the class-Y EMI capacitors in the power supply.  Connecting to ground gives an easier return path for the leakage current that through the rest of your circuit.

Does it have to be a barrel connector? Some power supplies use DIN connectors and connect the DIN shield to the input ground so that you can choose how to connect that with your circuit.  That gives the best of both worlds.

A medical rated power supply will have less ground leakage so will likely be considerably better than standard industrial/telecom power supplies even when floating.
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 12888
  • Country: gb
  • Hero999
Re: AC/DC Desktop adapter, non floating ground?
« Reply #4 on: September 11, 2019, 04:26:37 pm »
If the power supply has an earth connection, it's less likely 60Hz hum will be an issue. The secondary side will most likely be connected to earth via a capacitor, so it's grounded at AC, even if it's floating at DC.
 

Offline ninjanation217

  • Newbie
  • Posts: 4
  • Country: us
Re: AC/DC Desktop adapter, non floating ground?
« Reply #5 on: September 11, 2019, 05:45:47 pm »
Yeah unfortunately it does have to be a barrel connector, its for a product that already exist, trying to fix it without adding a lot of cost / recall :(

I have a couple of supplies that fix the issue in house but they are no longer made. When a company gets back to me I will post a link to the supply here in case someone else stumbles upon this in the future.
 

Offline ninjanation217

  • Newbie
  • Posts: 4
  • Country: us
Re: AC/DC Desktop adapter, non floating ground?
« Reply #6 on: September 11, 2019, 05:52:21 pm »
Zero999, thanks again for the input. Yeah I was hoping my Mean Well GST40A24-P1J (earthed supply) would cut out the noise but its still there. The only supplies that I have found fix the issue have around 1-2k Ohm connecton between the -v and AC Ground on the connector. Unfortuntly the GST40A24-P1J even though it has an earth pin, has no connection between the -v and earth pin on the supply. I have contacted some suppliers and will wait for them to get back to me.

Thanks again everyone for the input, I have always liked Dave's channel and have been meaning to make my way to the forums, you guys are very helpful!

- James
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 12888
  • Country: gb
  • Hero999
Re: AC/DC Desktop adapter, non floating ground?
« Reply #7 on: September 12, 2019, 07:46:04 am »
You've mentioned filtering but have you tried screening? A screened enclosure, connected to 0V, surrounding the low level signal parts will probably help to reduce mains pick-up substantially.
 

Online magic

  • Frequent Contributor
  • **
  • Posts: 855
  • Country: pl
Re: AC/DC Desktop adapter, non floating ground?
« Reply #8 on: September 12, 2019, 08:43:10 am »
Depends. It may only make it worse if the problem happens when ground bounces around due to Y capacitors while some high impedance signal nodes bounce at a lower amplitude due to capacitive coupling to ambient.
Electrostatic shields should be referenced to the ground of the circuit they protect.

A medical rated power supply will have less ground leakage so will likely be considerably better than standard industrial/telecom power supplies even when floating.
These are treacherous. They still have some capacitive injection of primary-side junk, their outputs are not grounded due to regulations, they look like they might be grounded because they have three-prong plugs, but that's actually because they are shielded to reduce EMI.
« Last Edit: September 12, 2019, 08:45:22 am by magic »
 

Offline EEEnthusiast

  • Regular Contributor
  • *
  • Posts: 66
  • Country: in
  • Expert on RF boards
Re: AC/DC Desktop adapter, non floating ground?
« Reply #9 on: September 12, 2019, 11:30:54 am »
I have seen an old Dell adapter which had three terminals on the barrel jack. one of them was connected to the shield on the output DC cable and was connected to the main earth ground. So these are for sure available, mostly for laptop chargers.
 

Offline Zero999

  • Super Contributor
  • ***
  • Posts: 12888
  • Country: gb
  • Hero999
Re: AC/DC Desktop adapter, non floating ground?
« Reply #10 on: September 13, 2019, 10:26:44 am »
Depends. It may only make it worse if the problem happens when ground bounces around due to Y capacitors while some high impedance signal nodes bounce at a lower amplitude due to capacitive coupling to ambient.
Electrostatic shields should be referenced to the ground of the circuit they protect.
I don't see how a shield can make it worse.

As long as the shield is connected to the supply rails, via a low impedance AC path and surrounds the low level signals, it will only improve things. The shield will form a Faraday cage, which can bounce up and down due to the Y capacitors, but as long as the circuit in side is referenced to it, there won't be any interference.
 

Offline Someone

  • Super Contributor
  • ***
  • Posts: 2063
  • Country: au
Re: AC/DC Desktop adapter, non floating ground?
« Reply #11 on: September 13, 2019, 11:59:18 am »
A medical rated power supply will have less ground leakage so will likely be considerably better than standard industrial/telecom power supplies even when floating.
These are treacherous. They still have some capacitive injection of primary-side junk, their outputs are not grounded due to regulations, they look like they might be grounded because they have three-prong plugs, but that's actually because they are shielded to reduce EMI.
Not all medical standard mains supplies are floating, you can get them with the DC output connected through to the input earth.
 

Offline TimNJ

  • Frequent Contributor
  • **
  • Posts: 766
  • Country: us
Re: AC/DC Desktop adapter, non floating ground?
« Reply #12 on: September 14, 2019, 02:31:36 am »
Hi James,

This power supply has a series impedance in the earth return path as you described above (Class II with functional earth):
https://www.digikey.com/product-detail/en/globtek-inc/TR9CI1500CCPIMR6B/1939-1768-ND/9837833
 
The following users thanked this post: ninjanation217


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf