Mains Isolation Transformer

[furnacemike]

I am kind of new to electronics work, and I am looking for a suggestion for a reasonable priced mains isolation transformer that I can use to safely work on hot equipment for the purpose of troubleshooting. Basically I want peace of mind for safety. I don't need a top of the line model, just something basic for beginners. I'm located in the USA.
Thanks!
Mike

[Ian.M]

Since the advent of cheap reliable GFCIs (RCDs), in almost all circumstances an isolation transformer *DECREASES* safety for a repair technician, as they prevent upstream or downstream GFCIs tripping if the technician accidentally becomes part of the circuit.

If that's not clear enough, without an isolation transformer, the GFCI in the feed to your bench^* will trip if you touch a 'hot' connection.  You'll get a shock, but the GFCI cuts power quickly enough to reduce the risk of death or permanent injury to near zero.   With an isolation transformer, the GFCI cant trip, the shock continues till you die or break free and meanwhile the current is cooking you like a hotdog in a Presto Hotdogger!

Additionally, most people get careless as they *think* it provides protection, which it will do as long as you only make contact with ONE point in the D.U.T (device under test), but that protection vanishes as soon as you make any connection from the D.U.T to anything ground-referenced, (including oscilloscopes,  signal generators, outdoor antennas etc.) so it increases the risk of you getting electrocuted.

There are still uses for an isolation transformer on the test and repair bench, but providing 'SAFETY' is *NOT* one of them.

* If your bench isn't on a GFCI protected circuit, or fed through a plug-in GFCI, you are a suicidal idiot.  Even if you don't care about your own life, get one for the sake of your family or the first responders who'd otherwise have to deal with your electrocuted corpse!

[bob91343]

Isolation transformers are expensive; I think they start at around $100.  However you can cobble one up with a pair of like transformers.  Just connect them back to back and you will get nearly the same output voltage as you put in.

Having said that, you didn't mention how much power you want to provide.  I do have an extra one if you want to PM me I can tell you about it.  However, these things are heavy and cost as much to ship as they are worth.  I am in Los Angeles.

[golden_labels]

Since the advent of cheap reliable GFCIs (RCDs), in almost all circumstances an isolation transformer *DECREASES* safety for a repair technician, as they prevent upstream or downstream GFCIs tripping if the technician accidentally becomes part of the circuit.

I am not arguing about which solution is safer, as that is a statistical statement and I have no data to support any of the three options, but above sentence looks a bit weird to me. How is a GFCI providing protection if the operator become a part of the circuit? Current on both lines remains balanced, so what is GFCI detecting in that scenario. Was that just poor wording or am I missing some risk?

[fourtytwo42]

Since the advent of cheap reliable GFCIs (RCDs), in almost all circumstances an isolation transformer *DECREASES* safety for a repair technician, as they prevent upstream or downstream GFCIs tripping if the technician accidentally becomes part of the circuit.

So why do we still have stringent primary to secondary circuit insulation standards for consumer equipment!

I think your statement is both misleading and dangerous particularly in a beginners section.

[golden_labels]

So why do we still have stringent primary to secondary circuit insulation standards for consumer equipment!

How is the manufacturer going to predict if the consumer has a GFCI-protected installation? The statement was also not made about the normal consumer use scenarios.

[Ian.M]

So why do we still have stringent primary to secondary circuit insulation standards for consumer equipment!

That's a completely different issue.

My problem with isolation transformers, is: Without proper training, or serious study of the issues involved, they give an *ILLUSION* of safety, and additionally remove the protection provided by any GFCI feeding the circuit.

A web forum is generally unsuitable for providing such training, as it should be done under the supervision of an instructor physically present who can see what the trainee is doing and (hopefully) intervene before they put themselves or others at risk.  We can of course use this forum to discuss the issues involved, but have no way of stopping the trainee putting themselves at risk if they act before they fully understand the situation.

The big risk with using an isolation transformer is if the user introduces any ground to any circuit in the D.U.T. that is on the primary side of any isolation barrier it may contain.  e.g. clip a scope probe ground onto the negative side of the main reservoir capacitor, which results in much of the primary side of its PSU becoming a high voltage DC touch hazard, with nothing to break the circuit except the input fuse or other over-current protection, that may well not act until the  victim is well cooked!

'Back in the day', we did use isolation transformers as one component of an isolated ground-free safe working zone, but that was before GFCI protection was commonly available.

[VooDust]

So why do we still have stringent primary to secondary circuit insulation standards for consumer equipment!

Because consumer equipment may not have it's device ground connected to earth grounding and exposes conducting parts, such as headphone jacks, antennas.

[bdunham7]

Basically I want peace of mind for safety.

There have been endless debates here on the issue of isolation transformers for service bench use.  I use one routinely and advocate their use where appropriate, but your statement is the opposing view in a nutshell.  Peace of mind and safety are two diametrically opposed concepts here.  The isolation transformer simply removes the ground reference from your mains voltage.  Why you might want to do this depends on the situation, but there is still a potentially harmful voltage present that can still fry you.  The use of an isolation transformer should not cause you to let your guard down.

So what do you hope to gain  (specifically) by using an isolation transformer? 

To answer your specific question, perhaps the BK Precision 1604A would be a good start.  They are about $140 new and are specifically intended for this purpose.  You can also get various medical-grade units on eBay, but then you'd have to test them to make sure they work and so forth.

[bdunham7]

How is a GFCI providing protection if the operator become a part of the circuit? Current on both lines remains balanced, so what is GFCI detecting in that scenario. Was that just poor wording or am I missing some risk?

The idea is that if you grip a hot line with each hand and current is flowing through your arms, at least a little bit of it might leak out to ground through your feet and trip the GFCI/RCD.

[rob77]

If that's not clear enough, without an isolation transformer, the GFCI in the feed to your bench^* will trip if you touch a 'hot' connection.  You'll get a shock, but the GFCI cuts power quickly enough to reduce the risk of death or permanent injury to near zero.   With an isolation transformer, the GFCI cant trip, the shock continues till you die or break free and meanwhile the current is cooking you like a hotdog in a Presto Hotdogger!

if you touch neutral and live then the GFCI will not trip and it will cook you anyways... so stop spreading misinformation ! if you touch two parts of a circuit at different potentials then you get shocked (and or cooked) regardless of GFCI or isolation transformer !

if you touch live while your body is grounded - then you get a shock but the GFCI will break the circuit once the current to ground reaches the GFCI's rated current (usually 30mA or 10mA for doemstic use )
but if we consider the same scenario (grounded body and touching "live") behind a isolation transformer then you will not get shocked at all because the "live" is not ground referenced anymore.

[VooDust]

if you touch live while your body is grounded - then you get a shock but the GFCI will break the circuit once the current to ground reaches the GFCI's rated current (usually 30mA or 10mA for doemstic use )

but if we consider the same scenario (grounded body and touching "live") behind a isolation transformer then you will not get shocked at all because the "live" is not ground referenced anymore.

In your first scenario, your mistake will have immediate consequences and you'll receive a shock of 30mA that is considered non-lethal.

In your second scenario, you are spared of consequences, but face the possibility of receiving a larger, possibly lethal shock should you repeat the same mistake with your other hand (remember, you did not learn from your first mistake - there was no shock).

Maybe all of this is a matter of individual evaluation of these risks. Personally I'd favor the first scenario.

Keep in mind that for the second scenario, the number of previously-save-now-high-potential touch points can be greatly expanded in unexpected ways as explained by Ian.

[nightfire]

From reading all of the above comments, I also want to add my 0.02 Eurocents to that...

- In todays world, every live circuit (especially in domestic use) should have a RCD present. Here in germany the usual RCD for the normal wall outlets is 30 mA by 230 VAC nominal voltage. This means that the tripping range according to VDE standards is between 15mA...30mA, and tripping time is allowed to be as high as 0.2 sec- usually within 20-30 ms with fresh gear.
- As I understand, a GFCI protected outlet according to UL is required to have 5mA +/- 1mA tripping current- which would loosely correspond with the german 10mA RCD (tripping between 5mA and 10mA). As in the USA approx. 110 VAC are used, and the resistance of the human body about 1000-2000 Ohms, depending on path of the current is constant, a lower voltage would drive lower currents through the human body.
(Please corrrect me, if these data I had to look up are not correct)
- above 10mA you really feel the current flowing to you, so even if 30mA is not directly and immediately lethal within short exposure, it can do other nasty things like blood clogging up and causing thrombosis some hours later
- If you have a live circuit and touch L-N, it really depends on the circumstances of enough current is able to flow through your body, shoes and floor to ground/earth connection to make a RCD trip. a well insulated floor (wood etc.) and shoes with thick rubber soles probably would make sure the current is not high enough to cause a standard RCD (30 mA usually, so in reality between 20mA and 25mA range) to trip


This said, I acquired myself some used isolation transformer some days ago, that needs some minor repairs- and given the situation that I am currently doing some repair work on a lab power supply and want in the near future to be able to do some ripple measurements with my oscilloscope, an isolation transformer is one way to go and deal with the situation.
(As I do not have a differential probe for it- and with the price point of this entry level scope a decent quality differential probe probably would be more expensive  than my scope...)

And most important: As previously stated, knowledge can (and should) not only be acquired through internet forums- the whole situation on-site has to be taken into account to prevent disasters and accidents. Also some other good practices can and should be employed to minimize the risk of getting accidentally zapped.
These can include working on wooden benches, wearing rubber gloves, clothes that cover the arms and are quite tight, or standing on a rubber mat- and making sure that the workbench or table in the hobby-room is nowhere near where central heating or plumbing is in the reach with grounded metal pipes...
Depending on knowledge and preferences there are the options to ground everything in the hope(s) of having the grounding and RCD/GFCI getting into place when something goes south, and leading the currents away from the body, or isolating the whole enchilada...
Isolation is in most cases a bit easier to do and check, whereas grounding has to be frequently checked to ensure everything is working well...

[rcjoy]

eBay is a good source for isolation transformers.

Topaz or Xentek brands are nice. 

[rob77]

if you touch live while your body is grounded - then you get a shock but the GFCI will break the circuit once the current to ground reaches the GFCI's rated current (usually 30mA or 10mA for doemstic use )

but if we consider the same scenario (grounded body and touching "live") behind a isolation transformer then you will not get shocked at all because the "live" is not ground referenced anymore.

In your first scenario, your mistake will have immediate consequences and you'll receive a shock of 30mA that is considered non-lethal.

In your second scenario, you are spared of consequences, but face the possibility of receiving a larger, possibly lethal shock should you repeat the same mistake with your other hand (remember, you did not learn from your first mistake - there was no shock).

Maybe all of this is a matter of individual evaluation of these risks. Personally I'd favor the first scenario.

Keep in mind that for the second scenario, the number of previously-save-now-high-potential touch points can be greatly expanded in unexpected ways as explained by Ian.

seems everyone is forgetting the "golden rules" or don't even know them...

when probing mains powered circuits:

1. no jewelry , no rings , no necklaces
2. second hand in your pocket or behind your back

if you follow the rules, you'll not get killed behind a isolation transformer - worst that can happen is that you get shocked between two points on your hand => no path through body => no danger to your heart.

so the conclusion is... no and no and no... GFCI is not a preferable replacement for an isolation transformer !

p.s. try the 30mA through your body to ground and then suggest the GFCI as a viable option  preferably live to left hand through body to grounded feet.

[madires]

Since the advent of cheap reliable GFCIs (RCDs), in almost all circumstances an isolation transformer *DECREASES* safety for a repair technician, as they prevent upstream or downstream GFCIs tripping if the technician accidentally becomes part of the circuit.

Neither an isolation transformer nor a GFCI does protect you in the sense of 'nothing bad will happen'. In both cases you need to know about the things which can go wrong and how to avoid them. For example, do you check your GFCI each time before working on a device?

[madires]

seems everyone is forgetting the "golden rules" or don't even know them...

when probing mains powered circuits:

1. no jewelry , no rings , no necklaces
2. second hand in your pocket or behind your back

Rubber gloves help too.

[TERRA Operative]

So does using clip-on probes and standing back when you cackle maniacally and throw the Frankenstein style knife-switch.

[BrokenYugo]

For example, do you check your GFCI each time before working on a device?

I do. It has a built in test button for that reason. All GFCI receptacles do, at least in the US. It shouldn't be something you lean on, if you accidentally trip one often you're doing it wrong, but they do work.

About the only way you can screw up with a GFCI that I am aware of is not understanding how it works and why it will not sense a fault after a transformer.

[themadhippy]

whilst hitting the test button is a good idea,dont rely on it.some of the lesser models dont apply a "fault" instead its just a mechanical test.if your confident the polarity is correct a short on the load side between neutral and earth will cause it to trip without a bang ( a switch wired to a plug this way caused much amusement during my collage days) or apply a load between live and earth large enough to operate the device.

[BrokenYugo]

Interesting, I always just assumed it connected a resistor from the protected live output to the neutral input (or protected neutral to input live) upsetting the output live/neutral current balance enough to trip it if everything is working properly. I know the internal test doesn't involve the earth/ground connection, as in the US it is legal to float the ground on a GFCI provided it is appropriately marked, I have installed and inspected a few like that and they pop when the test button is pressed.

[james_s]

- In todays world, every live circuit (especially in domestic use) should have a RCD present. Here in germany the usual RCD for the normal wall outlets is 30 mA by 230 VAC nominal voltage. This means that the tripping range according to VDE standards is between 15mA...30mA, and tripping time is allowed to be as high as 0.2 sec- usually within 20-30 ms with fresh gear.

That depends on where you are, in the USA it is certainly not the norm, GFCI (RCD) protection is used only on outlets in damp locations or around a water source. Bathrooms, kitchens, garages, outdoors, ect. As far as I know you *can* install one anywhere but they are not normally found and not required to be used in most household locations.

[golden_labels]

Ian.M:
In absence of a direct answer, I assume that was just poor wording and what you meant was a circuit that is no longer properly isolated due to accidently getting grounded. If this happens, the transformer is certainly making things worse. No opinion regarding which option is safer, for reasons given earlier. I just thought I am missing some risk in the isolated scenario.

nightfire:
A rule of thumb regarding AC power is that 1mA causes pain, 10mA makes you unable to release the conductor from your hand, 100mA is causing fibrillation. In case you observed one can survive much more: those are values used for safety purposes. They are at the lower end of the distribution. The interpretation is “stay below X for your safety”, not “going above X certainly causes the effect”.

As for GFCI presence, you should be aware that most of the humanity is not that lucky to even get earth connection.

[Terry Bites]

Dont try popping a GFCI through your body- what an idiotic suggestion!


Because the secondary is no longer referenced to ground a GFCI (RCD in UK) gives you no protection on your isolated power.
You wont get a severe shock to ground because the only path from the secondary to ground is via the transformer's leakage resistance and capacitance.
These leakages will not trip the GFCI!
You will get a killer shock by touching across the secondary terminals.
So care is to be exercised.

A power tool site transformer is a lot cheaper and there are usually plenty of second hand units to be had. It won't meet all the leakage and insulation specs of a "real isolation transformer" You must lift the secondary CT ground!

[themadhippy]

A power tool site transformer is a lot cheaper and there are usually plenty of second hand units to be had.

and will only kick out half the voltage you chuck in