Measure 500V DC with digital scope and 100:1 probe

[schmitt trigger]

Isolation transformers are recommended for input-transformerless tube amplifiers.

But that is not your case.
Therefore, as Nigel mentions, you can safely test it without an isolation transformer.

 

[Elerion]

But, I still don't understand why high voltage in the valves plates can't be considered live/hot.

Amplifier ground is tied to outlet's safety ground (which is at neutral potential).
So, there's a return path from any high voltage inside the chassis, to ground (earth potential), and shock hazard.
Am I missing something?

 

[Nigel Goodwin]

You mean in the scenario of the amplifier?
As far as I know, the amplifier transformer's secondary center tap is attached to the chassis, which is connected to safety ground (third outlet prong). So, any contact with any high voltage wire/point in the amplifier itself supposes a shock hazard. Wouldn't it?

That's not much of a concern, of FAR more concern is the fact that you can touch the bare metal chassis and a high voltage wire inside the amp - this gives a MUCH greater shock hazard (unless of course you work in your bare feet, and spread water on the floor before you start ).

You just need to understand the hazards involved, and be careful - a common old saying was to 'keep one hand in your pocket' - but it's pretty well impossible to work like that.

 

[KeepItSimpleStupid]

I have a few things to add.

Sometimes there is a spec that basically states how far above ground you can "Float" the test instrument. I have some that say 30 V.

Sometimes the scope can be put in differential mode. You use two probes and ground to make a measurement.
All scopes are different. You have A-B or A+B with B inverted.

Valve amps have a transformer. The secondary is isolated. There were tube TV's that had tubes whose filaments added up to 120 V. No transformer. So, an isolation transformer is required.

It's NICE to have a Variac when working on audio amps. Usually, you end up with an isolated variac for "The bench".

The "One hand in the pocket" rule works and so does "Make connections with power off....and discharged". On a system that takes 208 3 phase @ 90 A and can generate 15 kV DC at 1.5 A and requires keys and interlocks to open the door to the electronics, you have to be exceptionally careful.

 

[Nigel Goodwin]

There were tube TV's that had tubes whose filaments added up to 120 V. No transformer. So, an isolation transformer is required.

You're thinking far too restricted

The Philips G6 (probably the worst colour TV ever made ) had 240V of valve heaters

 

[Elerion]

There were tube TV's that had tubes whose filaments added up to 120 V. No transformer. So, an isolation transformer is required.

Why can't you just ground the scope or test instrument and measure those 120V volts? why is there need to "float" the instrument?

That's not much of a concern, of FAR more concern is the fact that you can touch the bare metal chassis and a high voltage wire inside the amp

That's for sure!
But not only the chassis, also, if touching ANY instrument which is also grounded. That's more probable than spilling water on the floor
With an isolation transformer, this last case won't give a shock hazard.

Isolation transformers are recommended for input-transformerless tube amplifiers.

Why will it be worse than an amplifier using transformer, like the one I posted above?

I think the problem is I don't fully undertand what you guys mean by "isolated".
Why is the valve amplififier using transformer, isolated, if grounds are shared?
Why "isolated", if there's a shock hazard between any voltage point and ground?
What are we really after, when seeking "isolation"?

I'd really appreciate if anyone could help me understand it

 

[Nigel Goodwin]

Why can't you just ground the scope or test instrument and measure those 120V volts? why is there need to "float" the instrument?

Because the chassis of the TV (or whatever) is live - connecting the scope to it is likely to result in a large BANG!.

It's an often discussed situation on here - and as a professional engineer I wouldn't consider using a scope that's earthed, it's far too dangerous.

People (including many on here) seem to think that earthing something makes it 'safe', this is totally untrue - it makes it 'safer' under some circumstances, but far more dangerous under many other circumstances.

What makes it 'safe' (for people who fully understand it), is understanding EXACTLY why earthing is and isn't used, and what you're dealing with.

To be fair it's pretty difficult to get a fatal shock from 240V, yet a number of people do it every year, and even from 120V as well.

I think the problem is I don't fully undertand what you guys mean by "isolated".
Why is the valve amplififier using transformer, isolated, if grounds are shared?
Why "isolated", if there's a shock hazard between any voltage point and ground?
What are we really after, when seeking "isolation"?

I'd really appreciate if anyone could help me understand it

Isolation means 'isolated from the mains' - not isolated from earth. An isolation transformer shouldn't break the earth connection.

The reasons for this are best explained historically, from back in the valve days. Old TV's and radios had just two core mains leads, and no transformer - the live and neutral weren't even marked!. The two wires entered the TV/Radio, one went to the rectifier (hopefully the live - 50/50 chance), the other was connected directly to the metal chassis (hopefully the neutral - again, 50/50 chance). So you had a 50/50 chance that your chassis (and every metal part of the unit) was connected directly to the live wire from the mains socket. So, earthed scope clip in one hand, TV/Radio chassis in the other - full 240V mains directly connected across your chest

In 'later' years, the electric companies complained about this practice (using half wave rectifiers), as if most were connected the right way round it mean people were only using the +ve half cycle of the mains, and this caused a DC offset on the mains. As a result manufacturers moved to using bridge rectifiers, this meant the chassis was at half mains potential regardless of the polarity of the wires - with similar consequences for your heart, or indeed your scope!.

So using an isolation transformer breaks the live and neutral from the unit, providing an isolated 240V (or 120V) supply, and in the case of an earthed chassis that should remain earthed.

A valve amplifier, using a mains transformer, is already isolated, so doesn't particularly need an isolation transformer.

As a professional engineer I'm perfectly happy to work with or without an isolation transformer, whichever the case you should consider EVERYTHING is live and act accordingly.

Interestingly I was talking to a plumber we use at work (for installing gas or electric cookers), and he was telling me he went to a house to fit a new gas boiler, and there was no earth connection ANYWHERE in the house He obviously advised her that she needed the house rewiring (urgently), but she wasn't prepared to do so - however, he couldn't install the gas boiler without an earth, so he had to install an earth spike, earth the electrical work for the boiler, and earth bond the gas and water pipes. While the UK probably has the best electrical system in the world, there are still a few VERY old installations that have never been updated from many decades ago when specifications were completely different.

 

[KeepItSimpleStupid]

I think the problem is I don't fully undertand what you guys mean by "isolated".
Why is the valve amplififier using transformer, isolated, if grounds are shared?
Why "isolated", if there's a shock hazard between any voltage point and ground?
What are we really after, when seeking "isolation"?

Nigel did a good job covering it. Bang (big spark) is a possibility w/o isolation on certain things.

Here's an example. Solid state amp. The primary had a short somewhere to chassis ground. Ungrounded plug. that meant, the chassis had say 70V from chassis to ground. Because of the isolation transformer, that chassis to EARTH path could not exist say for a person or any equipment connected.

A very rare event.

So, suppose I tried to connect a grounded signal source to the amp. You now have 70 VAC and ground that contact one another when you use the negative clip of the signal generator.

 

[Nigel Goodwin]

Here's an example. Solid state amp. The primary had a short somewhere to chassis ground. Ungrounded plug. that meant, the chassis had say 70V from chassis to ground. Because of the isolation transformer, that chassis to EARTH path could not exist say for a person or any equipment connected.

A very rare event.

As you say, a rare event - however, I had a VERY interesting fault on a friends TV a number of years ago. It was a 'modernish' CRT set, with an SMPSU, but without an isolated chassis. It also had a SCART socket, and that was isolated by opto-isolators, the aerial socket was isolated by capacitors. I can't remember exactly what the set was doing, but it was something really strange, and a fault I'd never seen (or even heard of). The actual cause was an artificial pine needle (from an artificial Christmas tree), it had fallen through the small slots in the top of the TV, and was bridging the safety barrier between the live and isolated parts of the chassis. It wasn't a serious short, as it was made of metallised plastic, but it seriously messed up the operation of the TV.

 

[Elerion]

Nigel, thank you very much for your detailed answer. I read it carefully twice before replying. Still, I don't understand:

Isolation means 'isolated from the mains' - not isolated from earth. An isolation transformer shouldn't break the earth connection.

This images are examples of isolation for testing and user safety. They both work because the earth connection is broken.


**broken link removed**


If there was an earth connection, I don't see the point.

Because the chassis of the TV (or whatever) is live - connecting the scope to it is likely to result in a large BANG!.

Why not just leave the probe ground clip unconnected, and just measure voltage relative to earth. Measure the chassis. Then the other "local grounds" inside the TV. Scope will be earthed, but not using the probe ground clip, there should be no short, and no bang. Isn't it?

Because of the isolation transformer, that chassis to EARTH path could not exist say for a person or any equipment connected.

This is in the case that earth connection is broken as in the example images above. Isn't it? If it isn't, there would be a path.

Right now I don't see the point of isolation transformer with a common safety earth connection... But I suppose there must be an advantage, or else, they wouldn't be used, sold or built!

 

[Nigel Goodwin]

Nigel, thank you very much for your detailed answer. I read it carefully twice before replying. Still, I don't understand:



This images are examples of isolation for testing and user safety. They both work because the earth connection is broken.

The scope example is measuring current, and NOT relative to chassis, so requires no common earth connection.

Right now I don't see the point of isolation transformer with a common safety earth connection... But I suppose there must be an advantage, or else, they wouldn't be used, sold or built!

As I said, it's to remove live mains (referenced to earth), secondary voltages aren't any particular concern.

From a personal point of view I prefer to work in a completely earth free environment, but various people on here prefer differently.

 

[Elerion]

The scope example is measuring current, and NOT relative to chassis, so requires no common earth connection.

Ok, let's talk about the other example of guy holding the shaver then
It is safe because the earth is not connected to the neutral of the secondary of the isolation transformer. Isn't it?? (hope so!)

Wouldn't then be safer to use an isolation transformer like in this image, when testing on a live amplifier?

 

[Nigel Goodwin]

Ok, let's talk about the other example of guy holding the shaver then
It is safe because the earth is not connected to the neutral of the secondary of the isolation transformer. Isn't it?? (hope so!)

Wouldn't then be safer to use an isolation transformer like in this image, when testing on a live amplifier?

An amplifier isn't likely to be 'live', as they pretty well universally have transformers.

Assuming you used a double transformer system like that example, then as soon as you connected your earthed scope to it you have added an earth again.

But as I've said all along, as long as you understand the reasons you can either use isolation transformers or not - just don't imagine that because you're using an isolation transformer that you are safe.

From the context of this thread, repairing a valve amplifier, there seems no need whatsoever for an isolation transformer.

 

[GromTag]

Essentially not a fan of the earth wire, it's a two way street, well, only unless I need to immediately hammer stored capacitive energy out of a system fast. The neutral in that event can switch the main off. Earth=Escape, as in escape velocity from....

 

[Elerion]

Thank you everyone, specially Nigel, for his very detailed explanations, but all of you for your contributions.

I'd like to leave a few links here, for reference, and maybe to be of help for someone who looks here in the future.

Please, if in doubt, read in order.

First. How isolation transformer can help:
http://electronics.stackexchange.co...g-the-ground-wire-in-an-isolation-transformer

"The whole point of an isolation transformer is to get a source of AC power that is completely isolated from (floating with respect to) the mains ground. Connecting the ground from one side of the transformer to the other completely negates this, and I don't know why they do it."

But wait before thinking this is the cure for everything...

Second. Isolation transformers, and reasons why can be dangerous for amplifier work (also about hot chassis scenario discussed in this post):
**broken link removed**


I will work without isolation transformer.
But as I don't have a Variac, I'll build a 100W bulb current limiter, for power supply testing. Then, just keep one hand in the pocket, and use crocodile probes when using a DMM. (and always discharge high voltage capacitors using a 10k 5W resistor hold by well insulated crocodile clip wires). Also, I've got a pair of 500V insulated globes with leather cover. Not very confortable, but can come handy. Safety first.

 

[KeepItSimpleStupid]

"The whole point of an isolation transformer is to get a source of AC power that is completely isolated from (floating with respect to) the mains ground. Connecting the ground from one side of the transformer to the other completely negates this, and I don't know why they do it."

The second part is done in some OneAC/ PowerVar power conditioners. I bought three of their conditioners for an electronics test system that ran daily 8 hrs/day typically. I combined it with a surge suppressor that they didn't like and it reduced electrical failures to zero. A hard drive was still working after 17 years. Still had dust, floppy and fan issues to contend with. it's actually done to "clean up" the ground rather than isolate it. The neutral/ground bond is done on the secondary side. I think the filters were on the primary side. Not sure. e.g. https://powervar.com/ground-guard-power-conditioners/