-
Posted for some information on a variac transformer and the topic came up that it's better to purchase one with a isolation transformer. This has got be thinking why would it not be a best practise to ALWAYS use a isolation transformer when working/repairing AC mains powered equipment. I know that it is never completely safe when using isolation, but could certainly save on buying a new scope if they probing in the wrong place and BANG. Overall it just seems a safer practise, but is there any reason why one would not want to use isolation when working on mains lines equipment.
-
See this Tektronix technical brief.
http://www.cbtricks.com/miscellaneous/tech_publications/scope/floating.pdf
-
The controls of most scopes are not designed to isolate the internal ground from the user, if not and it floats at high voltage you can't touch the scope ... making it mostly useless. You are effectively also plugging a lot of capacitance into your circuit but a common mode choke on the probe cable can help a bit there.
If you use the isolation transformer on the DUT you might put high voltage on it's exterior instead, so also not necessarily safe. -
I got the feeling the OP was referring to always putting the device under test on an isolation transformer, rather than the scope.
If you hook a properly grounded scope to a circuit on an isolation transformer then the circuit becomes ground referenced, and you can still get a shock from the circuit to ground.
Also, a circuit on an isolation transformer will not trip a GFCI, which you probably want to have on every outlet in your lab.
So the isolation transformer protects you from getting a shock if you touch one live wire and an external ground (until your scope is connected). However, the GFCI may be better protection in terms of preventing harm to the technician.
Hopefully some experts will chime in an correct/augment my posting, as I am a hobbyist. -
Just to clarify a few things.
It is never a good idea to float a scope by using an isolation transformer or cheater plug. This can induce a lot of additional stress on the equipment. Also, if an electrical occurs failure within the scope, the chassis may be elevated to line potential. This introduces a serious shock hazard.
I see nothing wrong with using an isolation transformer to float the circuit under test. There is however, a right and a wrong why do do this. If you simply connect a standard isolation transformer to your mains powered equipment (DUT) while keeping the ground connection intact, you aren't really achieving any isolation at all. Using an isolation transformer in this way only serves to filter out EMI and other undesirable artifacts present on the mains. This is why they are used so extensively in hospitals to protect sensitive medical equipment from interference.
If you truly want to "isolate" the mains powered circuit you are analyzing, you need to remove the ground connection on the output of the isolation transformer while keeping the ground connection on your scope intact. At this point, if you measure from either end of the output winding to ground, you will see 0 VAC. The only way to receive a shock using this method is to simultaneously contact both output wires of the transformer. This is was you might call an isolation transformer for "technical purposes". The danger of smoking your scope by connecting the earth lead to an energized part of the equipment has been eliminated. The risk of shock has also been greatly reduced.
If you don't believe me take a few measurements with the ground connection intact and then lifted on your isolation transformer. With the ground intact, you will measure full line voltage from either end of the secondary winding to ground. Now lift the ground on the transformer. Measure from each end of the secondary winding to ground. That mains voltage that you just measured is no longer present. Probing either end of the winding to ground will not show any voltage because the winding is no longer referenced to earth ground. The only way to measure mains voltage in this scenario is to probe directly across the secondary winding. Presto, instant isolation. -
This video does a decent job of explaining how an isolation transformer can add benefit to the circuit under test. It also shows you what is mentioned above by 3DG4 about how you must also isolate the ground on a typical consumer tranformer (the gentleman in the video refers to these as public transformers).
http://youtu.be/11Yve2ijWyk -
No one has the guts to say just pull out the ground plug. You could get a 3 foot outlet strip and take pliers and rip out the ground prong then plug the transformer into the strip. NO GROUND. Why open the transformer and monkey around with wires?
But this is like saying puppies are bad.
-
The table posted by fivefish is the industy standard recommended solution.
Particularly Differential probes that remove any link to mains GND by the scope reference (GND) lead.
But wide BW products like these are expensive, often much more than a hobbyists scope, however this is the case with most advanced measurent techniques.
One should ask: how much BW do you really need along with safe isolation?
Siglent offer a well priced solution for basic safe 2 channel isolation, however at a low BW of 1 MHz.
ISFE = Isolated Front End.
My listings show RRP US$135
http://www.siglentamerica.com/prodcut-fjxx.aspx?fjid=404&id=25&tid=1&T=2
-
I got the feeling the OP was referring to always putting the device under test on an isolation transformer, rather than the scope.
Yes I meant STRICTLY the DUT, never the test equipment. -
No one has the guts to say just pull out the ground plug. You could get a 3 foot outlet strip and take pliers and rip out the ground prong then plug the transformer into the strip. NO GROUND. Why open the transformer and monkey around with wires?
But this is like saying puppies are bad.
That could certainly work, if you clearly label the power strip, and the isolation transformer that they should used together to float the ground. You do lose any protection from having the transformer itself grounded. I other words, if the transformer had a wiring fault it would be better if the case and the primary side were grounded.
Your suggestion would have the added benefit of not having the case grounded should somebody decide to use it as a foot rest.... Not that anybody would ever do that with equipment under their desk. -
Use a HV Differential probe instead or a handheld oscilloscope , generally just not a good idea to use a isolation transformer with a scope, as no ground (it will be "floated"). For example if oscilloscope is plugged into a isolation transformer or a cheater plug, then you connect your scope probe ground to ground in your AC circuit (perhaps at the ground point of a bridge rectifier), you now have a ground reference that is actually at a very high voltage. This means that the metal case of your oscilloscope is also at this same high voltage. If you are touching the case of the oscilloscope and at the same time touch the metal case of another instrument that has a proper earth ground, you may get a serious, even lethal, shock. Plus then you cumulative stresses on the oscilloscope? power transformer insulation and other components.
One use for a Isolation transformer, when working on older old vacuum-tube equipment with live chassis, or transformerless designs allowing the chassis to be at 110V. Then some Isolation transformers sold, also ties secondary coil to ground for common mode noise and transients, these are mainly design for medical for sensitive equipment they use. You have to remove that ground point on secondary of transformer, if your going to use these. -
Just to clarify a few things.
It is never a good idea to float a scope by using an isolation transformer or cheater plug. This can induce a lot of additional stress on the equipment. Also, if an electrical occurs failure within the scope, the chassis may be elevated to line potential. This introduces a serious shock hazard.
I see nothing wrong with using an isolation transformer to float the circuit under test. There is however, a right and a wrong why do do this. If you simply connect a standard isolation transformer to your mains powered equipment (DUT) while keeping the ground connection intact, you aren't really achieving any isolation at all. Using an isolation transformer in this way only serves to filter out EMI and other undesirable artifacts present on the mains. This is why they are used so extensively in hospitals to protect sensitive medical equipment from interference.
If you truly want to "isolate" the mains powered circuit you are analyzing, you need to remove the ground connection on the output of the isolation transformer while keeping the ground connection on your scope intact. At this point, if you measure from either end of the output winding to ground, you will see 0 VAC. The only way to receive a shock using this method is to simultaneously contact both output wires of the transformer. This is was you might call an isolation transformer for "technical purposes". The danger of smoking your scope by connecting the earth lead to an energized part of the equipment has been eliminated. The risk of shock has also been greatly reduced.
If you don't believe me take a few measurements with the ground connection intact and then lifted on your isolation transformer. With the ground intact, you will measure full line voltage from either end of the secondary winding to ground. Now lift the ground on the transformer. Measure from each end of the secondary winding to ground. That mains voltage that you just measured is no longer present. Probing either end of the winding to ground will not show any voltage because the winding is no longer referenced to earth ground. The only way to measure mains voltage in this scenario is to probe directly across the secondary winding. Presto, instant isolation.
OK,I'll bite!
If neither end of the transformer secondary is connected to Mains Earth,how do you "measure full Mains voltage from either end of the secondary winding to ground"?
This is what normal transformers do---they isolate the secondary from the primary.
An Earth connection which is just basically a wire to the frame of the transformer,hence to the "Earth" pin of the 3 pin socket which carries the normal secondary voltage,has no connection whatever to the secondary itself,& cannot be at any voltage w.r.t that secondary.
Its only real purpose is to provide protection in the case of a primary winding-to-core breakdown of the transformer,so it really doesn't do anything for the DUT,in any case,whether connected or disconnected.
If,for some strange reason,it is connected to one side of the secondary,it is no longer an "isolation" transformer.
Even in this case,you will only see the same voltage as the Mains (not really the "Mains" voltage) between one side of the secondary & Earth.
In the SMPS on which I have worked,the DUT chassis which is earthed via the 3 pin power lead is effectively isolated from both sides of the incoming Mains,as well as from the common of the high voltage rectifier & the switching devices.
If neither side of the incoming Mains is connected to Earth,the "earth clip" of the Oscilloscope may be connected to the "common" of the SMPS switching circuitry without risk.
PS: The gentleman in the video covered the whole thing very well!
"Isolation transformers" with one side of the secondary tied to Earth are not common in Australia---at least in my experience.
I've only ever seen one standalone transformer wired that way over many years,& that was used for another purpose altogether.
Your "full Mains voltage between Earth & either end of the secondary" must be a "phantom voltage" as referred to in the video. -
If you truly want to "isolate" the mains powered circuit you are analyzing, you need to remove the ground connection on the output of the isolation transformer while keeping the ground connection on your scope intact. At this point, if you measure from either end of the output winding to ground, you will see 0 VAC. The only way to receive a shock using this method is to simultaneously contact both output wires of the transformer.
The moment you attach the ground clip this stops being true of course. -
Just to clarify a few things.
It is never a good idea to float a scope by using an isolation transformer or cheater plug. This can induce a lot of additional stress on the equipment. Also, if an electrical occurs failure within the scope, the chassis may be elevated to line potential. This introduces a serious shock hazard.
I see nothing wrong with using an isolation transformer to float the circuit under test. There is however, a right and a wrong why do do this. If you simply connect a standard isolation transformer to your mains powered equipment (DUT) while keeping the ground connection intact, you aren't really achieving any isolation at all. Using an isolation transformer in this way only serves to filter out EMI and other undesirable artifacts present on the mains. This is why they are used so extensively in hospitals to protect sensitive medical equipment from interference.
If you truly want to "isolate" the mains powered circuit you are analyzing, you need to remove the ground connection on the output of the isolation transformer while keeping the ground connection on your scope intact. At this point, if you measure from either end of the output winding to ground, you will see 0 VAC. The only way to receive a shock using this method is to simultaneously contact both output wires of the transformer. This is was you might call an isolation transformer for "technical purposes". The danger of smoking your scope by connecting the earth lead to an energized part of the equipment has been eliminated. The risk of shock has also been greatly reduced.
If you don't believe me take a few measurements with the ground connection intact and then lifted on your isolation transformer. With the ground intact, you will measure full line voltage from either end of the secondary winding to ground. Now lift the ground on the transformer. Measure from each end of the secondary winding to ground. That mains voltage that you just measured is no longer present. Probing either end of the winding to ground will not show any voltage because the winding is no longer referenced to earth ground. The only way to measure mains voltage in this scenario is to probe directly across the secondary winding. Presto, instant isolation.
Incorrect.
I don't know what you're measuring mate, but an isolation transformer it is not!
-
No one has the guts to say just pull out the ground plug. You could get a 3 foot outlet strip and take pliers and rip out the ground prong then plug the transformer into the strip. NO GROUND. Why open the transformer and monkey around with wires?
But this is like saying puppies are bad.
here most things don't have ground and in older house there are no ground in the outlets
The problem is that anything with a switch mode supply the gnd/chassis will usually float at 110V via the input filter capacitors
so you have to be very careful if anything is grounded because it will zap most inputs
-
Quote
here most things don't have ground and in older house there are no ground in the outlets
The problem is that anything with a switch mode supply the gnd/chassis will usually float at 110V via the input filter capacitors
so you have to be very careful if anything is grounded because it will zap most inputs[/quoteb
Where is here? You do not have your country flag posted.
I also live in an old house with no grounds and some members cannot believe this
Their response is I have to re-wire the house. -
Quote
here most things don't have ground and in older house there are no ground in the outlets
No grounds here ether. The place is nearly 65 years old.
The problem is that anything with a switch mode supply the gnd/chassis will usually float at 110V via the input filter capacitors
so you have to be very careful if anything is grounded because it will zap most inputs[/quoteb
Where is here? You do not have your country flag posted.
I also live in an old house with no grounds and some members cannot believe this
Their response is I have to re-wire the house. -
Quote
The place is nearly 65 years old.
I am 67 and my house is 102 years old. We both are showing our age. The front of the house still has the original wiring and the back half was re-wired in the 70's (still no grounds). The original wiring is thick (maybe 10 ga) covered with heavy cloth. It is very hard to work with. In the 90s I connected some ceiling fans and really hate to touch this wiring. So I will do anything to avoid touching an outlet (like running extension cords all over the place)
If I get an isolation transformer, it will also be an old one (no ground) so there is no question about a ground.
-
Quote
here most things don't have ground and in older house there are no ground in the outlets
The problem is that anything with a switch mode supply the gnd/chassis will usually float at 110V via the input filter capacitors
so you have to be very careful if anything is grounded because it will zap most inputs[/quoteb
Where is here? You do not have your country flag posted.
I also live in an old house with no grounds and some members cannot believe this
Their response is I have to re-wire the house.
Denmark, I think it wasn't until after mid 1975 that outlets had to have ground, but since almost everything comes with a schuko plug that doesn't connect ground
in a danish outlet almost nothing is grounded -
Sounds like the farm house my uncle was renting for a couple of years. in 1984 it was 83 years old. Plumbing electrical all added in the thirties. The oud Outhouse was still on the property, and functional.QuoteThe place is nearly 65 years old.
I am 67 and my house is 102 years old. We both are showing our age. The front of the house still has the original wiring and the back half was re-wired in the 70's (still no grounds). The original wiring is thick (maybe 10 ga) covered with heavy cloth. It is very hard to work with. In the 90s I connected some ceiling fans and really hate to touch this wiring. So I will do anything to avoid touching an outlet (like running extension cords all over the place)
If I get an isolation transformer, it will also be an old one (no ground) so there is no question about a ground. -
A few thoughts on the isolation transformer issue.
1. I would be very cautious when it comes to lifting the ground in the isolation transformer. Only lift it on the outlet if you can, if you cannot look for a short extention cord you can remove the ground pin from. The chassis of the isolation transformer should always be at ground potential.
2. As has been already pointed out placing the chassis of a piece of test gear above ground is a bit of a shaky practice, and a good way for someone with a limited amount of experience to get hurt.
3. There are some things that get done in electronics that qualify as "if you have to ask; you most likely need to learn a bit more to make yourself completely aware of what is going on and of the potential dangers.", This is one of those I believe.
-
I've used a fully isolated variac (NOT an autotransformer) over the past 10 years mainly as an isolation transformer when testing mains powered devices.
I always wondered if I would get zapped from its secondary if I only touched one of the wires, but never dared to test it for real, until today.
First I used a low input impedance DMM (3 kOhm) to measure the voltages between isolated and unisolated mains, and as you might expect, these were very low (much higher in the normal high Z position, of course). Then I tried a neon screwdriver on the secondary side. Hmmm, it lit, does it need so little current that the capacitive leakage between primary and secondary is enough for it to light?
Another thing I didn't expect was the fact that the secondary side showed a hot and a cold side when using the neon light screwdriver? One one side it lit, on the other it stayed dark. How can that be?
Then I incremented the voltage from 0 V to 230 V in steps of 10 V and each time touched the "live" side of the secondary (floating) winding (the one that made the neon lamp light up). Never felt a thing, despite being in an ESD safe surrrounding (ESD-safe floor, table, chair, shoes, labcoat etc.) which should have better conduction to earth than your average surroundings.
Of course theory predicted that this would be the outcome, but still... I wouldn't want to be doing this work, though:
-
Incorrect.
I don't know what you're measuring mate, but an isolation transformer it is not!
It always amuses me when someone has just enough time to call foul, but never enough time to explain their reasoning. Thanks for the laugh MATE! -
Incorrect.
I don't know what you're measuring mate, but an isolation transformer it is not!
It always amuses me when someone has just enough time to call foul, but never enough time to explain their reasoning. Thanks for the laugh MATE!
He may not have explained his reasoning,but I did back in reply # 11.
A moment's reflection should bring you to the realisation that what you suggested could not happen with a !:! transformer,or any other ratio of simple transformer,for that matter.
It could happen in the special case of a 1:2 transformer,with the secondary centre tapped,& that point earthed---but that was not the situation we were talking about.
Also,in that case,the voltage across the whole secondary would be 2X the Mains voltage! -
Quote
I've used a fully isolated variac (NOT an autotransformer) over the past 10 years mainly as an isolation transformer when testing mains powered devices.
I always wondered if I would get zapped from its secondary if I only touched one of the wires, but never dared to test it for real, until today.
First I used a low input impedance DMM (3 kOhm) to measure the voltages between isolated and unisolated mains, and as you might expect, these were very low (much higher in the normal high Z position, of course). Then I tried a neon screwdriver on the secondary side. Hmmm, it lit, does it need so little current that the capacitive leakage between primary and secondary is enough for it to light?
Yes, it's enough. There are isolated variacs, but there were never made with electrostatic shields. Medical isolation transformers on the other hand are (almost always? Probably just not possible to achieve the specs called for by the standards without it) always fitted with electrostatic shielding. So they have about the best specs regarding isolation you can get.