Isolation Transformer and GFI

[W8LV]

My question has to do with Isolation Transformers and Ground Fault Interrupters. (My first post, so if I messed it up, missed something completely, okie dokie, just let me know.) So I used to use an Eico Oscilloscope back in High School in the 1970's. I actually don't recall using the ground lead on that scope when checking out the televisions and "All American 5" (tube) radios that we fixed back in the day.. and I don't recall using an Isolation Transformer. Ever. Maybe our common bench power strips were isolated..I am not sure. But all of us seemed to graduate from High School Vocational Electronics unscathed.
Now setting up a lab thirty plus years later, I purchased a B&K Precision Model 2160 Oscilloscope (Whoo Hoo!) and I have a GFI protected outlet. So I follow Dave's video, read and read, and I am told I plug in the scope (three prong) and "float" whatever I am testing through an Isolation Transformer. So here I see a B&K Precision Model 1604A Isolation Transformer (http://www.bkprecision.com/products/power-supplies/1604A-single-output-isolation-transformer.html) but IT has a GROUNDED OUTLET on the front of it!

So my questions are: Doesn't that GROUNDED outlet sort of defeat the purpose of the Isolation Transformer pretty much COMPLELELY? And... if I use this setup, by plugging the isolation transformer into the Ground Fault Interrupter, am I just defeating IT altogether and allowing my wife to collect on my Life Insurance Policy a bit too early, at least for me?

[andtfoot]

I can't find anything in a quick look at the doco, but does the pin on the front actually connect through to ground?

Maybe it could be there for compatibility with plugs that would normally be grounded?

[Rerouter]

i would agree with the line of reasoning that its likely to allow plugging in earthed chassis devices,

while this may spill out into a debate its not too uncommon for an isolation transformer to terminate its isolated ground to one side of the transformer output to provide a return path, which mirrors strapping ground and neutral for a home,

as for plugging the transformer into a GFCI, it does render it invalid as the full power will enter the transformer but it may exit on the output elsewhere,

to alleviate this you would need to install one on the output of the isolation transformer as a gfci only cares about the hot and neutral current, how you choose to install this can also open into a debate, as before the isolated ground point will pick up any isolation leakage, e.g. your device is raised 300V above ground and you touch it and mains ground, or after the tie in point and it will pick up a ground fault current, e.g. you touching it and its isolated ground,  (all the fun ways to claim that life insurance policy eh?)

[nctnico]

In short: never ever use an isolation transformer. Once you ground a part of the circuit you'll be creating a dangerous situation and the GFI won't protect you. Do yourself a favour and buy a proper differential probe. There are plenty of those (new!) on Ebay for reasonable prices.

[madires]

The problem with using a scope and DUT powered by an isolation transformer is that if you connect a probes ground clip to the DUT that point/potential will be grounded (connected to earth/PE via the scope). Either get a differential probe or setup two probes as differential probe (remove ground clips, invert one channel, add both channels). And keep the isolation transformer, because it will protect you in case you touch a hot wire by accident (keep the other hand in a safe place).

The grounded outlet of the isolation transformer shouldn't be a problem, as long as it's just connected to the PE of the wall socket. Also check that no tap of the secondary winding is grounded.

[vk6zgo]

In short: never ever use an isolation transformer. Once you ground a part of the circuit you'll be creating a dangerous situation and the GFI won't protect you. Do yourself a favour and buy a proper differential probe. There are plenty of those (new!) on Ebay for reasonable prices.

An isolation transformer,by definition,does not have any part of  its secondary winding connected to earth in the manner used for normal house power wiring.
If there is a connected "Earth" pin,it just connects to the Mains Earth & nothing else.

What were normally "Active" & "Neutral" inputs to the DUT power supply are now both floating with respect to mains Earth.

Thus you are able to take either of these to Mains Earth potential .
In other words,as they normally possess no potential to Mains Earth,they can take up whatever potential you like,so you can put the earth clip on either side.

OK,if you are seriously stupid,you could then get across the now earthed side & the other,but how much do we have to dumb things down?

Without an isolation transformer,get it wrong & you will blow the earth clip apart,& maybe damage your 'scope.

If the Mains Earth connection is made from the transformer input plug to the socket on its output,all it will do is take the common side of the output of the SMPS in the DUT to Mains Earth.
This output is isolated from the "mains" side of the SMPS by the inbuilt galvanic isolation  of the supply.
After all,this is the main reason why we use SMPS in the first place!

If we didn't care about isolation,we could use the old AC/DC type circuits used in many post  WW2 US radios.
Why they went to them is beyond me--pre WW2 US radios used power transformers.

[ddavidebor]

It's grounded when you ground it with the ground clip of the oscilloscope

[W8LV]

Thanks So Much for the replies... It's a bit like I am in the position of Rip Van Winkle returning to the Test Bench...
;-)
So let's say I set it up this way: the three-pronged Oscilloscope is plugged into the GFI... In other words, the scope is not floating. The isolation transformer is plugged into another outlet, and the DUT is then plugged into the isolation transformer... In other words the scope is grounded, and the DUT is floating. So, now there is no path from the scope probe to ground. Is that correct and the best practice, short of using a differential probe?

I don't see any "cheap" differential probes on the market, by the way. At least not in my budget...maybe I missed seeing goodly priced ones.

Why did the "All American Five Tuber" dispense with the power transformer? I think it was to shrink the size, weight, and cost of radios. Well, I remember my Electronics teacher telling us that "all" engineering is about just three things: Time, Space, and.... Money! ;-)

Sent from my VS920 4G using Tapatalk 2

[IanB]

I think the information missing from this thread is what you intend to measure and test with your oscilloscope?

For example, in many devices the low voltage side of the power supply is already isolated from the mains through a transformer. So typically it will not be a problem to attach the ground probe to some part of this circuit for testing.

If you want to measure live mains equipment, you could do what you did in the past and leave the ground clip unconnected. You will measure voltages referenced to mains ground. It may not be cleanest signal, but it will not cause any short circuits.

[G7PSK]

From your call sign you are in the US I guess, as I understand it from what I have read elsewhere on this forum the practice is in the US to bond the earth/ground and neutral wires in the appliance (shoot me down if this is wrong) therefore if the earth wire goes through that transformer your isolation transformer has just become redundant unless you lift the earth in the DUT. Can't you just use two probes in differential mode when required.

[W8LV]

What I am measuring is just an Arduino but it's directly connected to a USB Wall Wart.
I am using the Arduino to control/sequence some electrodes/ magnets of an experimental Magnetohydrodyamic Drive. 

Arduinos, USBs and wall warts(!) didn't exist in my era, back when I was working as an assistant in Nikola Tesla's lab during the Colorado Spring Days... Just kidding, but I HAVE seen many a sun rise, and asking for the safety advice in hopes of not missing any future sunrises unnecessarily (hopefully, I have a few left), and appreciate your advice!

Best Regards from W8LV in Pickaway County, Ohio, USA.

[W8LV]

Yes, I am in the US/Canada, and that's how they do it here... So you are okay if it's wired correctly. Once in a while, you also come across a polarised plug here on this side of the pond. But the unpolarised hot chassis are the nasties.. It just gets confusing you toss in a gfi AND a grounded oscope, PLUS the "unit under test" connected TO an isolation transformer, AND a wall-warted USB. Better yet... A three pronged cord that on the other side of THAT wallwart ends up with a NON-polarised two wire cord that plugs into your computer... Too many chances in the equation for a nasty ground loop to get in there somewhere... The differential probe sounds like good advice, hope I can find one for a good price to have and to hold, so to speak.. ;-)

I HAVE learned that it is a "good thing" to have an emergency power outage light in my basement lab, so at least when I do short something out, I am not holding something delicate (or a hot soldering iron) in complete darkness...

Best Regards from W8LV in Pickaway County, Ohio, USA.

[W8LV]

I am not sure, either. I downloaded the PDF Poop Sheet from B&K  (gotta Love B&K for always having the Poop Sheets for everything they have ever manufactured up on their website!) but I couldn't tell how that ground pin was hooked up, either. I just may have to purchase one and check out that mystery personally...

Best Regards from W8LV in Pickaway County, Ohio, USA.

[nctnico]

Thanks So Much for the replies... It's a bit like I am in the position of Rip Van Winkle returning to the Test Bench...
;-)
So let's say I set it up this way: the three-pronged Oscilloscope is plugged into the GFI... In other words, the scope is not floating. The isolation transformer is plugged into another outlet, and the DUT is then plugged into the isolation transformer... In other words the scope is grounded, and the DUT is floating. So, now there is no path from the scope probe to ground. Is that correct and the best practice, short of using a differential probe?

No. Once you clip the probe's ground lead to the DUT the DUT is grounded and the GFI won't protect you if you touch a live part of the circuit. So either use a differential probe or use a scope with isolated inputs.

I don't see any "cheap" differential probes on the market, by the way. At least not in my budget...maybe I missed seeing goodly priced ones.

I have this one:
http://www.ebay.com/itm/New-Professional-Differential-probes-DC-25Mhz-ADP25-Max-Voltage-1300V-1000V-RMS-/380411931457

[madires]

There are some nice videos about this topic:





and

[vk6zgo]

From your call sign you are in the US I guess, as I understand it from what I have read elsewhere on this forum the practice is in the US to bond the earth/ground and neutral wires in the appliance (shoot me down if this is wrong) therefore if the earth wire goes through that transformer your isolation transformer has just become redundant unless you lift the earth in the DUT. Can't you just use two probes in differential mode when required.

NO! NO! NO!----That is not done anywhere!

The only place that the Neutral & Earth are bonded together is at the  entry point to your house,never in the appliance.
It is against the wiring rules in Australia,the UK,the USA,& as far as I know,every other country in the world to make any other such connection.

If you were absolutely certain which wire was the Neutral,& that your house wiring was perfect,it would probably be safe to place the earth clip on it BUT,it would then also contravene the above wiring rule & be illegal.

[vk6zgo]

Yes, I am in the US/Canada, and that's how they do it here... So you are okay if it's wired correctly. Once in a while, you also come across a polarised plug here on this side of the pond. But the unpolarised hot chassis are the nasties.. It just gets confusing you toss in a gfi AND a grounded oscope, PLUS the "unit under test" connected TO an isolation transformer, AND a wall-warted USB. Better yet... A three pronged cord that on the other side of THAT wallwart ends up with a NON-polarised two wire cord that plugs into your computer... Too many chances in the equation for a nasty ground loop to get in there somewhere... The differential probe sounds like good advice, hope I can find one for a good price to have and to hold, so to speak.. ;-)

I HAVE learned that it is a "good thing" to have an emergency power outage light in my basement lab, so at least when I do short something out, I am not holding something delicate (or a hot soldering iron) in complete darkness...

Best Regards from W8LV in Pickaway County, Ohio, USA.

The old "hot chassis" radios & TVs relied on insulation to  protect the consumer from the possibility of the Active & Neutral being swapped over.
The Chassis "common" was (hopefully) returned to the Neutral line,but this is totally different to connecting the Neutral & Earth in the Appliance!
They normally had no earth connection.

Australian radios all used power transformers & were of similar size & weight to the "All American Five".
They didn't have to use series strings of weird filament voltages,or have elaborate insulation.
They could use fullwave rectifiers too!
I personally think Americans got "dudded" with those radios.

Modern switchmode power supplies only have a small part of their circuitry directly Mains connected,as distinct from the whole device as in the old style AC/DC supplies.

The output voltages are isolated from the Mains because of the High frequency transformer included as part of their operation.

IanB's suggestion of not putting your earth clip anywhere,& just reading the Mains Active w.r.t the 'scopes earth will work in most of the cases where you might want to check for the presence of Mains.

[W8LV]

Well, I think I lost a message somehow.. Anyway, I want to THANK YOU ALL for keeping me out of SERIOUS TROUBLE on the isolation/grounding issue. I ended up obtaining a B&K Precision 1604A Isolation Transformer, and it checks as "open ground"... So I do not think they tied the secondary to ground on that unit. I also picked up via EBay a Tenma Model 72-6670 Isolation Transformer, curiously, it has both "direct" outlets (which test as  "normal" three prong grounded outlets) and "isolated" outlets which (surprise !) test as "open ground". Not sure why they placed isolated AND direct outlets on the front, but I do recall seeing this setup on similar units back in the Twentieth Century... So I placed some plastic "baby protectors" on the direct ones... Maybe the "direct" ones are there for ease of plugging in your scope and signal generator, etc... But the scope is plugged into my regular three pronged grounded GFI unit on the wall.
With the whole safety issue I am reminded of a Claymore Mine which was labeled on the "business" side: "Face Towards Enemy"...;-)  I hope to NOT receive the "Electronics Engineering Workbench Darwin Award" this year, and thanks AGAIN for helping  me to merely be a "runner-up" to the Award.
 Now I can finally probe around the Arduino and motor shield safely I think...but let me know if I have strayed in my assumption!

Best Regards from W8LV in Pickaway County, Ohio, USA.

[Nerull]

It's fairly likely your arduino already has an isolated supply, and in any case should never have more than 12V on it, so you're fairly safe in any case, but just to restate: All the isolation in the world won't do anything at all to protect you - the user - once the scope ground is clipped into the circuit. All the potentials will be just as high as they would be without the isolation. You've bypassed the ground isolation and provided a ground path through the scope. If you touch a line that is at a high potential relative to whatever you've grounded, you'll get shocked, isolation or no. And to make matters worse, the GFI won't trip.

An isolation transformer allows you to ground any part of the circuit without blowing up a ground lead, but it doesn't make a scope-grounded circuit safe.

[rstor22]

I realize this is an old post however I thought the following should be added for the benefit of those coming across it at a later time. The opening post was regarding the BK 1604A and the outlet at the front having a three prong connector.  I have a BK1604A and I can confirm on my unit, the front output connector is connected to earth ground (see attached picture).