Float a scope

[Greg65]

Hi folks I am wondering on how to float the scope properly. I was working on a broken bass guitar transistor amplifier and it took me days to investigate, identify the bad components and remove them, then source new components and finally install them. Ahhh. so finally I got them all installed then I decided to power up.
I used a light bulb in series with the mains live in, and I brought it up with a variac, I had my scope connected across the output with no speaker load, plus I had a 1Khz signial going in to the input jack with my function generator.
Everything seemed fine untill I turned up the master volume, then BANG! a couple of capacitors smoldered like blue lighting for a few seconds and finally another bang and then grey smoke and the glass of the lightbulb broke. 

My investigation was: I viewed Dave's video on "how not to blow up your scope" several times and found out that I created a ground loop on the negative terminal of the amps output which had nothing to do with true equi potental ground that I thought it was from the amp's schematic. 

Is it a reasonable question to float a scope in future to avoid this same scenario with audio transistor amplifiers?

[fivefish]

Probably time to invest in a proper differential probe.

[tautech]

If you made this simple error, I'd advise to forget floating your scope until vastly more experienced.

For simple needs use two channels for a differential measurement or buy a Differential probe.

[Greg65]

Thanks folks.
I also watched another youtube video by 'maxxarcade' an amplifier repair genuis, and he safely floats his scope on a huge 2000W power block PA amp so I thought I would ask here at the forum on how it should be done properly.
 
I do very well understand that I am a novice and the many dangers involved, but I thought that I'd ask anyway.

I would really like to become competant with the procedure of floating the scope for amp repair jobs, and I am very aware of the consequences.
A differential probe is more than likely the way I should will go as every amplifier is different plus it will give me and my equipment added protection.

Thank you for the advice, Greg

 

[nbritton]

Is it a reasonable question to float a scope in future to avoid this same scenario with audio transistor amplifiers?

You can float the scope, but you have to understand you're creating an unnecessary electrocution hazard by doing this. If you're in a basement, garage, or some other location where you're earthed I highly recommend not doing this. Generally you would float the device under test with something like an isolation transform or battery/inverter. You may want to think about connecting your scope to a GFCI outlet, this will keep you from blowing up stuff if you accidentally create a ground fault.

[Greg65]

Is it a reasonable question to float a scope in future to avoid this same scenario with audio transistor amplifiers?

You can float the scope, but you have to understand you're creating an unnecessary electrocution hazard by doing this. If you're in a basement, garage, or some other location where you're earthed I highly recommend not doing this. Generally you would float the device under test with something like an isolation transform or battery/inverter. You may want to think about connecting your scope to a GFCI outlet, this will keep you from blowing up stuff if you accidentally create a ground fault.

Great info nbritton, you mean float the equipment under test instead. That sounds even better.

I do have two Isolation transformers, one is rated at 1200Va and the other is 500Va.

Could I foat both the equipment and the scope with two different transformers or would just the equipment side be enough?

I have also watched toddfun's brilliant youtube video 'isolation transformers' and how to modify the isolation transformer. A great video! 

[Greg65]

Also, I do understand that a GFCI is the USA term for the ground fault protection outlet, over here it called an RCD outlet. I will get one or two of those also.

Thanks.

[rsjsouza]

Hi folks I am wondering on how to float the scope properly.

I am late to this thread, but when I read the subject title I thought you would need one of these:

Or maybe hire this guy:

[nbritton]

Thanks folks.
I also watched another youtube video by 'maxxarcade' an amplifier repair genuis, and he safely floats his scope on a huge 2000W power block PA amp so I thought I would ask here at the forum on how it should be done properly.
 
I do very well understand that I am a novice and the many dangers involved, but I thought that I'd ask anyway.

I would really like to become competant with the procedure of floating the scope for amp repair jobs, and I am very aware of the consequences.
A differential probe is more than likely the way I should will go as every amplifier is different plus it will give me and my equipment added protection.

Thank you for the advice, Greg

There isn't a technical reason why you can't float your scope, it all comes down to the fact that it's a electrocution safety hazard. If you float your scope and your device under test still references mains ground you could accidentally electrocute yourself while probing. If you leave your scope grounded and float the device under test there is no way for you to electrocute yourself in a accidental ground fault.

If you do float your scope you must make sure that you also float yourself. That is to say you must make sure you are not connected to mains ground in any way.

[Greg65]

Hi folks I am wondering on how to float the scope properly.

I am late to this thread, but when I read the subject title I thought you would need one of these:

Or maybe hire this guy:

Hey you,   get lost!

[Greg65]

Thanks folks.
I also watched another youtube video by 'maxxarcade' an amplifier repair genuis, and he safely floats his scope on a huge 2000W power block PA amp so I thought I would ask here at the forum on how it should be done properly.
 
I do very well understand that I am a novice and the many dangers involved, but I thought that I'd ask anyway.

I would really like to become competant with the procedure of floating the scope for amp repair jobs, and I am very aware of the consequences.
A differential probe is more than likely the way I should will go as every amplifier is different plus it will give me and my equipment added protection.

Thank you for the advice, Greg

There isn't a technical reason why you can't float your scope, it all comes down to the fact that it's a electrocution safety hazard. If you float your scope and your device under test still references mains ground you could accidentally electrocute yourself while probing. If you leave your scope grounded and float the device under test there is no way for you to electrocute yourself in a accidental ground fault.

If you do float your scope you must make sure that you also float yourself. That is to say you must make sure you are not connected to mains ground in any way.

Thanks so much nbritton. 

[nbritton]

Great info nbritton, you mean float the equipment under test instead. That sounds even better.

I do have two Isolation transformers, one is rated at 1200Va and the other is 500Va.

Could I foat both the equipment and the scope with two different transformers or would just the equipment side be enough?

Yes, connect your equipment under test to one of the isolation transformers. It's sufficient to leave the scope grounded and only float the equipment under test with the isolation transformer... in fact, weird things (stray voltage) might happen if you float both your scope and equipment using two separate isolation transformers.

[C]

Hi folks I am wondering on how to float the scope properly.

You don't !!!
Floating a scope is trying to protect the scope while RISKING YOUR LIFE.
A bench scope was not built to work this way.

Use two probes in differential mode
Use a differential probe
Use a portable scope

"GFCI" Residual-current device
Residual-current device

is an electrical wiring device that disconnects a circuit whenever it detects that the electric current is not balanced between the energized (line) conductor(s) and the return (neutral) conductor

While a good idea would not have protected anything in this case.
It would trip with a fault in the primary side power of the scope or the amplifier's primary power.

C

[Greg65]

Great info folks, as I said earlier, I'm still learning the ropes as a novice. Cheers for all input. 

And,,,,,     How about this scenario then?   I bought a CRO scope recently and the ground/earth wire was disconnected at the plug   .   Is that another procedure? or was the last owner living dangerously?   

[nbritton]

"GFCI" Residual-current device
Residual-current device

is an electrical wiring device that disconnects a circuit whenever it detects that the electric current is not balanced between the energized (line) conductor(s) and the return (neutral) conductor

While a good idea would not have protected anything in this case.
It would trip with a fault in the primary side power of the scope or the amplifier's primary power.

Hmm, yeah you're right. You would be shorting the device under test to ground, not the scope. How come they don't build in a residual-current device into the scope to protect you from short circuiting a device under test?

[C]

NEVER disconnect a GROUND

Think about a circuit that has an AC output like your guitar transistor amplifier.
When you float it and tie one of the outputs to ground you now have the power rails moving or trying to move at the  AC output rate. You just changed the circuit.

When you connected your  function generator, you are now trying to change the  function generator at the ampilifier's AC rate also..
If the  function generator was grounded, you could have huge voltage connected to the inputs which are small signal inputs.

C

[Greg65]

I am with you C,

I never used the CRO scope with the ground disconnected at the plug on any circuit yet and since then I have fitted a new plug that has it.
I used another scope that I own, which is the famous Rigol 1025E to test the amp's output.

I knew immediately that the CRO with the ground disconnected set off the alarm bells so I sorted that one out.

Thanks.

[C]

How come they don't build in a residual-current device into the scope to protect you from short circuiting a device under test?

The BNC connector is connected to Ground, It's doing it's job.

The testing procedure is WRONG.

Before you connect your scope to anything, measure the difference between the scope's ground and where you want to connect the probe's ground clip.
You could use a DVM the best choice
 or
 you could just use the scope probe with out a ground clip.
How hard is it to touch the probe tip to where you want to connect the ground clip to see if there is a signal?
Need to remember that a DVM is more protected for overloads then the scope. Scopes have limits that can change based on the volts/div setting.

C
 

[Armxnian]

If the circuit you're probing has a ground that is referenced to mains earth, then you can attach the scope ground clip to that point and probe away.

If the ground is not mains earth referenced, attaching the scope ground clip could cause problems, because you've now mains referenced that point and the circuit. If you connect a second scope ground clip to anywhere other than the same point, you've now shorted the two ground clips and everything in their path. You should generally be fine probing a floating circuit.

It is not a good idea to float any device that is grounded. It is grounded for a reason. If you float your scope and the chassis gets charged with a voltage, you're screwed. Same goes for the thing you're probing. Once again, if you float your circuit, and you connect a scope ground clip, then you've grounded the circuit again, this time through the probe instead of the wire leading out the back of the chassis. So careful if using multiple ground clips.

If you're unsure if a point is a ground or a mains earth referenced ground, then use a meter to find out.

As Tautech mentioned, if you need to, remove the ground clips from the probes and use the scope as you would a multimeter by using the difference operation in the math menu of the scope. 

Otherwise, get a discrete differential probe.

[Matje]

Is it a reasonable question to float a scope in future to avoid this same scenario with audio transistor amplifiers?

No, as has been pointed out.

You can use an isolation transformer to power the amplifier.

Or you use a battery powered scope (but usually *not* USB powered ones!). E.g. Owon has some of these. Of course you should not probe around inside the high voltage part of the amplifiers power supply, the scope being battery powered won't help you there. Plus do not forget to remove the power supply cord of the scope in this scenario  ;-)

And always remember: no matter what, the ground terminals of the scopes channels are still connected and can cause a short.

[tggzzz]

Get this wrong and the first thing will be loud noises (human and/or electrical), followed by deathly quiet.

Have a look at the "Praxis - high voltage" references in https://entertaininghacks.wordpress.com/library-2/scope-probe-reference-material/

I refer you to that page since if you think disconnecting/floating the scope is a good idea, you will probably also benefit from many of the other references on that page.

Don't rely on things you pick up on a bulletin board. Do rely on high-quality professional material.

[Greg65]

Yip, you's are all dead right and I can see how wreckless I was by not understanding the scopes hidden bite.    In the first place I misunderstood the amplifier schematic too and broke the first rule in electronics,   'Assume'!   
So I connected everthing up to what I understood was good, and 'assumed' that the negative connection of the amps output was ground/earth, but I realised only after re-studying the schematic that it was circuit ground from the zero center tap of the amps transformer secondary side wrere up it went in smoke when I turned up the master volume as there was + & -70Vdc either side. 

I have learned a lot from that mistake and thank you for referring to use the DVM before making final connections with the scope clip, plus yes I should have only used the probe tip for that stage of the test as all I wanted to see was an amplified signal.

Ok so, it looks like never float the scope and get the isolation transformer out into action plus invest in a differential probe.

Thanks everyone. 

[Shock]

Floating Oscilloscope Measurements And Operator Protection
www.tek.com/dl/51W_10640_1.pdf
The Three Facets of "Floating" Measurement Solutions
http://www.tek.com/document/application-note/three-facets-floating-measurement-solutions

[LukeW]

remove the ground clips from the probes and use the scope as you would a multimeter by using the difference operation in the math menu of the scope.

As long as the common-mode voltage (between the scope probe and the scope ground, which is tied to mains earth) isn't higher than it's allowed to be.

[jlmoon]

There isn't a technical reason why you can't float your scope, it all comes down to the fact that it's a electrocution safety hazard. If you float your scope and your device under test still references mains ground you could accidentally electrocute yourself while probing. If you leave your scope grounded and float the device under test there is no way for you to electrocute yourself in a accidental ground fault.

If you do float your scope you must make sure that you also float yourself. That is to say you must make sure you are not connected to mains ground in any way.

Once upon a time I had a professor who made this recommendation:  For any reason you have to float a device, always try to float the device that you will habitually touch the least.  This implies that knobs and switches (on test equipment) are made for setting to various positions, not to mention the probe & ground leads.  So the temptation is pretty great to reach up and connect yourself physically to the chassis of that scope, sig gen.. etc.   Best to float your device under test (with a isolation supply or transformer with variac) or use the proper differential probe... period!!!

[Richard Crowley]

Can anyone explain WHY guitar amps and other dodgy consumer gadgets are NOT properly grounded?
Other than the fact that they are so cheap, they don't want to spend the extra pocket-change it takes to do it right?

Instead of spending hours repairing a guitar amp and then kludging a dangerous setup to test it,
why not simply install the proper grounded power mains cord and bring it into the 20th century?

I just don't get it.   

[helius]

For certain types of audio gear "ground lift" switches are common, either from the factory or bodged on by users. The reason they are there is as one way to break ground loops. Ideally the ground current should be minimized so as not to cause audible interference, but in setups that are slapped together quickly and need to work right away (familiar to musicians) this is not always possible. It's much safer to break ground loops using transformers but that is an additional cost and device that needs to be carried.

Intro to ground loops: http://www.bestaudio.com/s/generic_grounding_seminar.pdf

[LukeW]

There's not necessarily anything wrong with the DUT equipment design, it may have the mains earth connected to the chassis for safety where it should be, but the OP just wants to use the scope to measure a voltage at some point relative to some other reference other than mains earth - which means the scope ground can't be connected directly there without shorting it to mains earth.

[Richard Crowley]

For certain types of audio gear "ground lift" switches are common, either from the factory or bodged on by users. The reason they are there is as one way to break ground loops. Ideally the ground current should be minimized so as not to cause audible interference, but in setups that are slapped together quickly and need to work right away (familiar to musicians) this is not always possible. It's much safer to break ground loops using transformers but that is an additional cost and device that needs to be carried.

Intro to ground loops: http://www.bestaudio.com/s/generic_grounding_seminar.pdf

Yes, audio is my primary engineering field. I am well acquainted with ground loops. I also know that decent modern audio gear is either equipped with a proper grounded mains cord, or else is completely floating (as gear powered from a wall-wart, etc.)  The dodgy design of guitar amplifiers (in particular) with their suicide capacitors has always baffled me.

[Bassman59]

Can anyone explain WHY guitar amps and other dodgy consumer gadgets are NOT properly grounded?
Other than the fact that they are so cheap, they don't want to spend the extra pocket-change it takes to do it right?

Because back in the 1950s and 1960s, there was no real notion of grounded, or even polarized, outlets, and mains wiring was as good as the training of the electrician who did the work.

So Leo Fender designed his amps with a "ground" switch, which connected one or the other leg of the amp's mains transformer to chassis via a capacitor (called "death caps" in the trade). This mitigated noise but wasn't safe.

Once proper standards were introduced and safety agencies started to require proper grounding of all equipment, designs changed to meet the requirement. You won't find an amp built from the 70s on with a two-prong power cord.

Instead of spending hours repairing a guitar amp and then kludging a dangerous setup to test it,
why not simply install the proper grounded power mains cord and bring it into the 20th century?

ALL competent amp techs will suggest to their clients that their vintage amps be properly grounded, and it's a simple fix. Or they'll suggest that the amps be retired to a glass case, unused. All of mine (including a '62 Bandmaster, a '62 Deluxe, a '64 Bandmaster, and a '66 Bassman) all shipped with ungrounded mains cords, and all have been fixed.

Back when I was a regular club sound guy, bands would roll through with amps in all states of disrepair. A lot of guys would have ground lifts ("death cubes") on their amps, and I would ALWAYS just take them away. "Why did you do that?" "I don't like people dying on my stage." The worst were the guys who purposely cut off the ground pin on the mains cord.

There is a whole other sordid tale about audio gear installed in studios where the install techs would regularly cut off mains ground pins in a futile attempt to solve buzz and hum problems caused by poorly-designed gear ("pin 1 problems") ...

[jlmoon]

Back when I was a regular club sound guy, bands would roll through with amps in all states of disrepair. A lot of guys would have ground lifts ("death cubes") on their amps, and I would ALWAYS just take them away. "Why did you do that?" "I don't like people dying on my stage." The worst were the guys who purposely cut off the ground pin on the mains cord.

You should never underestimate the power of an Electron!   

[nbritton]

It's really not that hard to stay safe if you understand how split phase power works. Electricity always takes the path of least resistance. Figuratively speaking, if your in the electrician business long enough you learn to probe live circuits with one hand in your pocket, because you never want to accidentally touch something else that causes you to become part of the circuit. Currents as low as 30 milliamps can trigger fibrillation in the heart; a shock entering from one hand and exiting through the other hand can cause current to go directly through the heart, very deadly. In every workshop I've had I've gone out of my way to make sure every outlet is GFCI protected, in fact in my last shop I installed a sub panel that was supplied by a 50 amp GFCI breaker in the main panel. These things really do save lives and equipment.

[knks]

What experts think about inserting a fuse into the probe ground lead?

Sure it will create some inductance, but this is all about audio frequencies

[nbritton]

What experts think about inserting a fuse into the probe ground lead?

Sure it will create some inductance, but this is all about audio frequencies

I'm no expert, but I don't see a major problem using an inline fuse on the ground lead. You would not want to fuse the ground coming out of the oscilloscope chassis through the power cord though, that would be a safety hazard. The internal resistance in a scope is usually 1 Mohm and then you have a 9 Mohm resister inside a 10X probe. Assuming you have 120V mains you would use 120V and ohms law to get the current rating of the fuse.

1X Probe: 120V @ 1 Mohm = .12 mA
10X Probe: 120V @ 10 Mohm = .012 mA

However, if you were working with a 12V circuit you would need to use that voltage in your ohms law calculations because power changes with voltage. I think in practice you would still blow up your device under test or constantly blow the fuse because you'll have to deal with varying voltages.

I think (haven't tested this) a better approach is to put all of your equipment (scope + device under test) on a single GFCI receptacle. These devices have circuitry inside of them that constantly monitors how the current is distributed on the hot and natural wires, it compares the current going through each wire and if they are not equal it trips the breaker. Class A GFCI trips when ground fault current exceeds 5 milliamps and Class B GFCI trips when ground fault current exceeds 20 milliamps. 5 mA @ 120V = 0.6W