PSU question...

133 views
Skip to first unread message

Nicholas Stock

unread,
Sep 18, 2016, 3:06:12 PM9/18/16
to neoni...@googlegroups.com
Hi Fellow lovers of neon...

I have a question from a bit of a novice here (me..). I'm working on a project that involves a dekatron or two and want to make sure that there's some degree of safety when placed inside a metal case and HV is involved. Instead of an IEC inlet and transformer inside the device, I want to use a 12V DC supply with a ground connection or -ve side tied to Earth. I found the following picture in my PSU stash and checked the continuity of the barrel side of the supply and Earth and there's continuity. Can anyone tell me which symbol on the PSU denotes this (instead of it being a truly isolated supply) or if anyone has any other ideas about how to go about a 'safe' 12V DC supply (I looked at the medical grade PSU's with 4 pin DIN and earthed shell, but they're quite expensive).

Apologies if this seems like a stupid question...

Thanks,

Nick

image1.JPG

David Forbes

unread,
Sep 18, 2016, 3:25:39 PM9/18/16
to neoni...@googlegroups.com
Nick,

The mark that indicates that the input and output are fully isolated
from each other is a square within another square, called "double
insulated". If you don't see that mark, and you see a UL or ETL mark,
then it's reasonable to assume that the barrel is grounded, because
otherwise the unit wouldn't pass UL or ETL testing.
> image1.JPG
>
> --
> You received this message because you are subscribed to the Google
> Groups "neonixie-l" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to neonixie-l+...@googlegroups.com
> <mailto:neonixie-l+...@googlegroups.com>.
> To post to this group, send email to neoni...@googlegroups.com
> <mailto:neoni...@googlegroups.com>.
> To view this discussion on the web, visit
> https://groups.google.com/d/msgid/neonixie-l/CAOX%2BRHJt%2Biv3jUBFWe7PYzfX5CotfRPUf7ZYMBUzdeSk6p1TMQ%40mail.gmail.com
> <https://groups.google.com/d/msgid/neonixie-l/CAOX%2BRHJt%2Biv3jUBFWe7PYzfX5CotfRPUf7ZYMBUzdeSk6p1TMQ%40mail.gmail.com?utm_medium=email&utm_source=footer>.
> For more options, visit https://groups.google.com/d/optout.


--
David Forbes, Tucson AZ

Nicholas Stock

unread,
Sep 18, 2016, 3:27:01 PM9/18/16
to neoni...@googlegroups.com
Much appreciated David!

Nick


To post to this group, send email to neoni...@googlegroups.com


--
David Forbes, Tucson AZ

--
You received this message because you are subscribed to the Google Groups "neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email to neonixie-l+unsubscribe@googlegroups.com.
To post to this group, send an email to neoni...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/a3b38c64-a24f-9ddd-4a9b-a4a3f97c50c7%40dakotacom.net.

Charles MacDonald

unread,
Sep 18, 2016, 9:16:32 PM9/18/16
to neoni...@googlegroups.com
On 16-09-18 03:25 PM, David Forbes wrote:
> Nick,
>
> The mark that indicates that the input and output are fully isolated
> from each other is a square within another square, called "double
> insulated". If you don't see that mark, and you see a UL or ETL mark,
> then it's reasonable to assume that the barrel is grounded, because
> otherwise the unit wouldn't pass UL or ETL testing.

Assuming of course that the input has a ground pin. if it is a two
prong plug it would be reckless to pass along one side of the line.
Although I suppose that they might trust that the Neutral prong should
be safe to pass on.



--
Charles MacDonald Stittsville Ontario
cm...@zeusprune.ca Just Beyond the Fringe
No Microsoft Products were used in sending this e-mail.

gregebert

unread,
Sep 19, 2016, 1:49:35 AM9/19/16
to neonixie-l
If you do tie your circuit ground to the case, make sure it is only at a single point, otherwise you can get ground loops and in some cases that can cause noise problems even in digital circuits. Many years ago my home-built computer had such a horrible mess of ground loops that I could hear radio interference on a pocket radio several houses away. Once I removed the ground loops, there was no detectable interference outside the metal cabinet.

If my project has a metal (conductive) case, I connect it directly to the ground terminal of the line cord with high-current wiring (no connectors, splices, etc). As far as I know, the reason consumer and industrial equipment do the same is to provide user protection in case the device gets damaged and the AC line contacts the exposed conductive areas. If that happens, the ground protects you until the circuit-breaker/GFCI trips. But, I dont tie the circuit gnd directly to the case; instead I use about 1 meg of resistance to bleed-off ESD and provide a DC path to GND. Circuit-wise, if you are using an isolated supply there's really no need for your circuit GND to be tied to the earth.

Now for high-voltage circuits that have an isolated supply, if your circuit GND is tied directly to the earth, you probably will get shocked if you touch the high-voltage supply, depending upon your body resistance (varies wildly from 500 ohms to about 100K), and possibly additional protection from clothing. However, if your circuit GND has a 1Meg path to earth ground, you may not get shocked at all; even if you touch a 500V supply, the current will be less than 500uA which I think is below the threshold of sensitivity.

Where things get tricky is if you are going to connect the device you built to another device, such as a USB port on a computer. If both devices have a low-impedance path from their respective circuit-gnd to the AC power-cord (earth) ground, you will get a ground loop.

Alex

unread,
Sep 19, 2016, 6:05:23 PM9/19/16
to neonixie-l
I am not sure I really agree with this, the double insulation mark indicates that there is two levels of insulator (case materials, fiberglass / mylar sheeting, cable insulation etc) between the user and anything live. A double insulated appliance (such as most consumer electronics / appliances) should not need or indeed have an earth connection. For more info on that, see here : http://www.double-insulated.com/

Now, if you want an electrically isolated output the only true way of doing this is galvanomically - either an old style 50Hz transformer / rectifier (heavy power brick kind of thing) or a good quality switching supply with a guaranteed isolation level. You can then pass through the true earth to your chassis and, since it is isolated, peg the neutral output to earth as well - though I would argue this actually reduces the safety somewhat, I would always prefer to leave the isolated supply floating, that way you can contact either conductor and be safe.

The symbol you really want to see, is two overlapping circles like those seen on this page : http://sc3c.iec.ch/ap60417/TC96gif.html

Hopefully I have not spread any inaccuracies here, I recently did a large commercial project involving many 3kva isolating transformers so had to read up a bit on all this, but that was a few months ago now... 

- Alex

gregebert

unread,
Sep 20, 2016, 2:11:14 AM9/20/16
to neonixie-l
When I was a computer tech in 1980, my employer started changing from traditional linear (ie, transformer) supplies, to switchers, and it was pretty scary powering these up the first time.

Most systems were shipped within the US/Canada, so they ran on 120VAC. There was no shield around the switcher. The way the desktop system was built, your face was a few inches from the power supply in order to reach the power switch.  The first time someone (not me, fortunately...) powered-up a 220V/50Hz system, it went KABOOM! Back then, the power supplies were not universal, so a jumper had to be changed to operate at 220V. Needless to say, any time I had to power-up a 220V system the first time I was really on edge. And despite the UL listing, I was always worried about getting shocked; switchers just didn't seem to have the same degree of isolation.

Maybe I'm just too old; I've always felt "safe" around transformer (linear) supplies and leery of switchers, especially the open-frame style.


Reply all
Reply to author
Forward
0 new messages