Floating the 'scope

[Fred Abse]

On Sat, 06 Mar 2010 15:16:07 +1100, John Tserkezis wrote:

On 6/03/2010 2:49 PM, Phil Allison wrote:

"George Hairoil"
I've been floating 'scopes for.... say since the early '80's. I
haven't been biten by it yet. (Though I've made plenty of other
mistakes.) What are the obvious safety reasons?

** My god you are stupid.

Welcome to an entirely new class of stupid, Phil.

At a place I used to work at, the cro on my bench was floated by the
previous "tech", he just forgot to tell anyone about it. And forgot to
label it too. No clues at all unless you measured it, or took it apart
to actually look.

Made worse by the fact it was a dual trace cro, and could have done the
job without floating it.

I can only presume these idiots have been doing for so long, they stop
caring about the fact you hadn't had to do it since two trace (or more)
CROs appeared on the market.
Hey, I don't care if they want to kill themselves, but their idiocy
will continue to kill others well into the future.

I'm starting to think that being dropped as a baby isn't enough to
explain this level of stupid.

Not such a new class.

I used to have a technician who had worked at a large equipment hire
outfit. He told me that checking for disconnected ground was SOP when
checking returned instruments.

They found several each week.


--
"Electricity is of two kinds, positive and negative. The difference
is, I presume, that one comes a little more expensive, but is more
durable; the other is a cheaper thing, but the moths get into it."
(Stephen Leacock)

 

[Fred Abse]

On Fri, 05 Mar 2010 19:38:49 -0800, John Larkin wrote:

On Fri, 5 Mar 2010 19:05:54 -0800 (PST), George Herold
ggherold@gmail.com> wrote:

Why do they make 'scope inputs ground referenced to the third wire of
the AC plug? Today I smoked the 10 ohm resistor that was separating
The AC ground from the 'ground' of my circuit. I'd forgotten to float
the 'scope and connected -15V to the ground clip of the scope probe.

George H.

That's a safety thing.

Our TPS2024 has four truly floating channels and floating trigger
input. You can hook a scope ground clip anywhere.

I use 7A13s.

A lot

--
"Electricity is of two kinds, positive and negative. The difference
is, I presume, that one comes a little more expensive, but is more
durable; the other is a cheaper thing, but the moths get into it."
(Stephen Leacock)

 

[Fred Abse]

On Fri, 05 Mar 2010 19:05:54 -0800, George Herold wrote:

Why do they make 'scope inputs ground referenced to the third wire of
the AC plug? Today I smoked the 10 ohm resistor that was separating
The AC ground from the 'ground' of my circuit. I'd forgotten to float
the 'scope and connected -15V to the ground clip of the scope probe.

Why not invert one channel and switch to "add".

Then use both inputs differentially.

Most, if not all worthwhile instruments should do that.

Probe compensation needs to be accurate for best CMRR.

Or buy a differential probe?

--
"Electricity is of two kinds, positive and negative. The difference
is, I presume, that one comes a little more expensive, but is more
durable; the other is a cheaper thing, but the moths get into it."
(Stephen Leacock)

 

[The Phantom]

On Sat, 06 Mar 2010 12:26:16 -0500, Jamie
<jamie_ka1lpa_not_valid_after_ka1lpa_@charter.net> wrote:

John Larkin wrote:

On Fri, 5 Mar 2010 19:05:54 -0800 (PST), George Herold
ggherold@gmail.com> wrote:


Why do they make 'scope inputs ground referenced to the third wire of
the AC plug? Today I smoked the 10 ohm resistor that was separating
The AC ground from the 'ground' of my circuit. I'd forgotten to float
the 'scope and connected -15V to the ground clip of the scope probe.

George H.


That's a safety thing.

Our TPS2024 has four truly floating channels and floating trigger
input. You can hook a scope ground clip anywhere.

John

So does my cheap crappie OWEN 20 mhz hand held. That also includes
isolation between the DMM inputs.

Is this a two channel scope? If so, what does an ohmmeter measure between the
shells of the two BNC inputs of the two channels?

Aren't the two BNC input grounds connected to the same internal ground
reference?

 

[The Phantom]

On Sat, 06 Mar 2010 07:03:56 -0500, Hammy <spam@spam.com> wrote:

On 6 Mar 2010 03:09:01 -0600, The Phantom <phantom@aol.com> wrote:



Tektronix has probes designed to solve these problems:

http://www2.tek.com/cmswpt/psdetails.lotr?ct=PS&cs=psu&ci=13471&lc=EN

They are specifically intended for floating use with the TPS2000 series scopes,
which, as John mentioned, have true isolated inputs.

Or, you can use these probes with an ordinary scope:

http://www2.tek.com/cmswpt/psdetails.lotr?ct=PS&cs=psu&ci=13415&lc=EN

Those are nice expensive but nice.

You should use an isolation transformer on any line powered device
your testing. You can usually find a 500VA one for under 100 bucks. I
got a TEMA 500VA one for 45 bucks from Newark. Well worth the money.

Its not really a good idea to float your scope.

A problem arises when you need to make measurements with a 2 or 4 channel scope
at locations in the circuit which don't have a common reference.

For example, suppose you need to look at the gate-source voltages of the two top
MOSFETs in an H-bridge configuration. Isolating the equipment doesn't allow you
to connect the ground clips of the two scope probes to different points in the
circuit that have a substantial voltage difference.

You need to use floating differential probes, or a scope with true isolated
inputs.

 

[The Phantom]

On Sat, 06 Mar 2010 07:03:56 -0500, Hammy <spam@spam.com> wrote:

On 6 Mar 2010 03:09:01 -0600, The Phantom <phantom@aol.com> wrote:



Tektronix has probes designed to solve these problems:

http://www2.tek.com/cmswpt/psdetails.lotr?ct=PS&cs=psu&ci=13471&lc=EN

They are specifically intended for floating use with the TPS2000 series scopes,
which, as John mentioned, have true isolated inputs.

Or, you can use these probes with an ordinary scope:

http://www2.tek.com/cmswpt/psdetails.lotr?ct=PS&cs=psu&ci=13415&lc=EN

Those are nice expensive but nice.

You should use an isolation transformer on any line powered device
your testing.

In days gone by, cheap consumer devices like televisions and table radios often
had a "hot" chassis; it was connected to one side of the line. And, in those
days, the line cord wasn't polarized; there wasn't even a third ground prong on
the outlet.

Using an isolation transformer when repairing such a device was a good idea.

You can usually find a 500VA one for under 100 bucks. I
got a TEMA 500VA one for 45 bucks from Newark. Well worth the money.

Its not really a good idea to float your scope.

 

[The Phantom]

On Sat, 06 Mar 2010 15:16:07 +1100, John Tserkezis
<jt@techniciansyndrome.org.invalid> wrote:

On 6/03/2010 2:49 PM, Phil Allison wrote:

"George Hairoil"
I've been floating 'scopes for.... say since the early '80's. I
haven't been biten by it yet. (Though I've made plenty of other
mistakes.) What are the obvious safety reasons?

** My god you are stupid.

Welcome to an entirely new class of stupid, Phil.

At a place I used to work at, the cro on my bench was floated by the
previous "tech", he just forgot to tell anyone about it. And forgot to
label it too. No clues at all unless you measured it, or took it apart
to actually look.

Why would anyone take a scope apart and disconnect the line cord safety ground
from the chassis inside the scope? Whenever I've seen anyone do this, they just
use a ground buster, easily obtainable from the hardware store, to disconnect
the safety ground at the outlet.

Made worse by the fact it was a dual trace cro, and could have done the
job without floating it.

I can only presume these idiots have been doing for so long, they stop
caring about the fact you hadn't had to do it since two trace (or more)
CROs appeared on the market.
Hey, I don't care if they want to kill themselves, but their idiocy
will continue to kill others well into the future.

I'm starting to think that being dropped as a baby isn't enough to
explain this level of stupid.

 

[The Phantom]

On Sat, 06 Mar 2010 12:18:17 -0800, Fred Abse <excretatauris@invalid.invalid>
wrote:

On Sat, 06 Mar 2010 15:16:07 +1100, John Tserkezis wrote:

On 6/03/2010 2:49 PM, Phil Allison wrote:

"George Hairoil"
I've been floating 'scopes for.... say since the early '80's. I
haven't been biten by it yet. (Though I've made plenty of other
mistakes.) What are the obvious safety reasons?

** My god you are stupid.

Welcome to an entirely new class of stupid, Phil.

At a place I used to work at, the cro on my bench was floated by the
previous "tech", he just forgot to tell anyone about it. And forgot to
label it too. No clues at all unless you measured it, or took it apart
to actually look.

Made worse by the fact it was a dual trace cro, and could have done the
job without floating it.

I can only presume these idiots have been doing for so long, they stop
caring about the fact you hadn't had to do it since two trace (or more)
CROs appeared on the market.
Hey, I don't care if they want to kill themselves, but their idiocy
will continue to kill others well into the future.

I'm starting to think that being dropped as a baby isn't enough to
explain this level of stupid.

Not such a new class.

I used to have a technician who had worked at a large equipment hire
outfit. He told me that checking for disconnected ground was SOP when
checking returned instruments.

As I asked John, why would anyone go to the trouble to get inside a piece of
equipment to disconnect the safety ground, when they can just use a ground
buster at the wall socket?

They found several each week.

 

[The Phantom]

On Sat, 06 Mar 2010 15:57:21 -0500, Jamie
<jamie_ka1lpa_not_valid_after_ka1lpa_@charter.net> wrote:

The Phantom wrote:

On Sat, 06 Mar 2010 12:26:16 -0500, Jamie
jamie_ka1lpa_not_valid_after_ka1lpa_@charter.net> wrote:


John Larkin wrote:


On Fri, 5 Mar 2010 19:05:54 -0800 (PST), George Herold
ggherold@gmail.com> wrote:



Why do they make 'scope inputs ground referenced to the third wire of
the AC plug? Today I smoked the 10 ohm resistor that was separating
The AC ground from the 'ground' of my circuit. I'd forgotten to float
the 'scope and connected -15V to the ground clip of the scope probe.

George H.


That's a safety thing.

Our TPS2024 has four truly floating channels and floating trigger
input. You can hook a scope ground clip anywhere.

John


So does my cheap crappie OWEN 20 mhz hand held. That also includes
isolation between the DMM inputs.


Is this a two channel scope? If so, what does an ohmmeter measure between the
shells of the two BNC inputs of the two channels?

Aren't the two BNC input grounds connected to the same internal ground
reference?



The 2 BNC's are isolated.

What's the model number of that scope?

I have no idea how they are doing it how ever, I can say this, I don't
like the BNC receptacles. It's a hard plastic shell instead of metal. I
guess that's part of the isolation. It does work.. I've used it to
measure signals around a resistor that is far from any common node with
one probe while the other has it's common attached to the device common.
It seems to work ..
He DMM inputs are standard DMM probes.. and are also isolated.

I don't think its that hard to fab 20 mhz isolated analog amps.

 

[Jamie]

The Phantom wrote:

On Sat, 06 Mar 2010 12:26:16 -0500, Jamie
jamie_ka1lpa_not_valid_after_ka1lpa_@charter.net> wrote:


John Larkin wrote:


On Fri, 5 Mar 2010 19:05:54 -0800 (PST), George Herold
ggherold@gmail.com> wrote:



Why do they make 'scope inputs ground referenced to the third wire of
the AC plug? Today I smoked the 10 ohm resistor that was separating
The AC ground from the 'ground' of my circuit. I'd forgotten to float
the 'scope and connected -15V to the ground clip of the scope probe.

George H.


That's a safety thing.

Our TPS2024 has four truly floating channels and floating trigger
input. You can hook a scope ground clip anywhere.

John


So does my cheap crappie OWEN 20 mhz hand held. That also includes
isolation between the DMM inputs.


Is this a two channel scope? If so, what does an ohmmeter measure between the
shells of the two BNC inputs of the two channels?

Aren't the two BNC input grounds connected to the same internal ground
reference?



The 2 BNC's are isolated.

I have no idea how they are doing it how ever, I can say this, I don't
like the BNC receptacles. It's a hard plastic shell instead of metal. I
guess that's part of the isolation. It does work.. I've used it to
measure signals around a resistor that is far from any common node with
one probe while the other has it's common attached to the device common.
It seems to work ..
He DMM inputs are standard DMM probes.. and are also isolated.

I don't think its that hard to fab 20 mhz isolated analog amps.

 

[Paul E. Schoen]

"Fred Abse" <excretatauris@invalid.invalid> wrote in message
newsan.2010.03.06.11.18.10.148083@invalid.invalid...

On Fri, 05 Mar 2010 19:05:54 -0800, George Herold wrote:

Why do they make 'scope inputs ground referenced to the third wire of
the AC plug? Today I smoked the 10 ohm resistor that was separating
The AC ground from the 'ground' of my circuit. I'd forgotten to float
the 'scope and connected -15V to the ground clip of the scope probe.



Why not invert one channel and switch to "add".

Then use both inputs differentially.

Most, if not all worthwhile instruments should do that.

Probe compensation needs to be accurate for best CMRR.

Or buy a differential probe?

This will not work for reading a voltage of 1 or 2 volts or less, which is
floating as much as 400 volts above ground. For instance, reading the gate
voltage of an SCR on a 480 VAC mains system. You must float the scope, and
I've done it safely and effectively, with a hand-held scope. There may be
some differential probes that can handle this; I don't know.

Paul

 

[The Phantom]

On Sat, 06 Mar 2010 07:17:49 -0500, Hammy <spam@spam.com> wrote:

On Sat, 06 Mar 2010 07:03:56 -0500, Hammy <spam@spam.com> wrote:

On 6 Mar 2010 03:09:01 -0600, The Phantom <phantom@aol.com> wrote:



Tektronix has probes designed to solve these problems:

http://www2.tek.com/cmswpt/psdetails.lotr?ct=PS&cs=psu&ci=13471&lc=EN

They are specifically intended for floating use with the TPS2000 series scopes,
which, as John mentioned, have true isolated inputs.

Or, you can use these probes with an ordinary scope:

http://www2.tek.com/cmswpt/psdetails.lotr?ct=PS&cs=psu&ci=13415&lc=EN

Those are nice expensive but nice.

You should use an isolation transformer on any line powered device
your testing. You can usually find a 500VA one for under 100 bucks. I
got a TEMA 500VA one for 45 bucks from Newark. Well worth the money.

Its not really a good idea to float your scope.

Anyone use any of PICO's differntial probes?

This ones only $350.00 Canadian.

http://canada.newark.com/pico-technology/oscilloscope-x10-x100/active-differential-oscilloscope/dp/02M0836

Also available are these:

http://www.caltestelectronics.com/www/Cat1GetSubCategory.asp?PN=General%20Purpose%20Differential%20Probes&ID=1.c.1&subcat=Differential%20Probes


I think they may be made by the same folks that make the Yokogawa differential
probes.

I've used the 100MHz Yokogawa probes, and they are very handy.

 

[Michael A. Terrell]

The Phantom wrote:

As I asked John, why would anyone go to the trouble to get inside a piece of
equipment to disconnect the safety ground, when they can just use a ground
buster at the wall socket?

Most idiots just cut the ground pin off the power cord.


--
Greed is the root of all eBay.

 

[Jamie]

The Phantom wrote:

On Sat, 06 Mar 2010 15:57:21 -0500, Jamie
jamie_ka1lpa_not_valid_after_ka1lpa_@charter.net> wrote:


The Phantom wrote:


On Sat, 06 Mar 2010 12:26:16 -0500, Jamie
jamie_ka1lpa_not_valid_after_ka1lpa_@charter.net> wrote:



John Larkin wrote:



On Fri, 5 Mar 2010 19:05:54 -0800 (PST), George Herold
ggherold@gmail.com> wrote:




Why do they make 'scope inputs ground referenced to the third wire of
the AC plug? Today I smoked the 10 ohm resistor that was separating
The AC ground from the 'ground' of my circuit. I'd forgotten to float
the 'scope and connected -15V to the ground clip of the scope probe.

George H.


That's a safety thing.

Our TPS2024 has four truly floating channels and floating trigger
input. You can hook a scope ground clip anywhere.

John


So does my cheap crappie OWEN 20 mhz hand held. That also includes
isolation between the DMM inputs.


Is this a two channel scope? If so, what does an ohmmeter measure between the
shells of the two BNC inputs of the two channels?

Aren't the two BNC input grounds connected to the same internal ground
reference?


The 2 BNC's are isolated.


What's the model number of that scope?


http://www.saelig.com/PSHA/PSSA017.htm

HDS1022MN

and the 100 Mhz version

http://www.saelig.com/PSHA/PSSA023.htm

 

[Paul E. Schoen]

"Michael A. Terrell" <mike.terrell@earthlink.net> wrote in message
news:4B92D17E.A66A9652@earthlink.net...

The Phantom wrote:

As I asked John, why would anyone go to the trouble to get inside a piece
of
equipment to disconnect the safety ground, when they can just use a
ground
buster at the wall socket?


Most idiots just cut the ground pin off the power cord.

Sometimes when I have seen this, I complete the job by cutting off the plug.
Makes it safe again, but not very useful.

Paul

 

[Fred Abse]

On Sat, 06 Mar 2010 14:39:01 -0600, The Phantom wrote:

As I asked John, why would anyone go to the trouble to get inside a piece
of equipment to disconnect the safety ground, when they can just use a
ground buster at the wall socket?

The one can be done in five minutes for free, the other involves going out
to the hardware store and costs money?

--
"Electricity is of two kinds, positive and negative. The difference
is, I presume, that one comes a little more expensive, but is more
durable; the other is a cheaper thing, but the moths get into it."
(Stephen Leacock)

 

[Fred Abse]

On Sat, 06 Mar 2010 16:05:28 -0500, Paul E. Schoen wrote:

"Fred Abse" <excretatauris@invalid.invalid> wrote in message
newsan.2010.03.06.11.18.10.148083@invalid.invalid...
On Fri, 05 Mar 2010 19:05:54 -0800, George Herold wrote:

Why do they make 'scope inputs ground referenced to the third wire of
the AC plug? Today I smoked the 10 ohm resistor that was separating
The AC ground from the 'ground' of my circuit. I'd forgotten to float
the 'scope and connected -15V to the ground clip of the scope probe.



Why not invert one channel and switch to "add".

Then use both inputs differentially.

Most, if not all worthwhile instruments should do that.

Probe compensation needs to be accurate for best CMRR.

Or buy a differential probe?

This will not work for reading a voltage of 1 or 2 volts or less, which is
floating as much as 400 volts above ground. For instance, reading the gate
voltage of an SCR on a 480 VAC mains system. You must float the scope, and
I've done it safely and effectively, with a hand-held scope. There may be
some differential probes that can handle this; I don't know.

The OP was talking about 15 volts common mode, where the technique will
work perfectly well.


56 dB CMR with a common mode voltage of 480 RMS is beyond the
capability of even a 7A13. There are differential probes that will do it,
but they have 4 figure price tags.

I've never needed to go to that extreme. Most of what I do is IGBTs with a
DC bus of around 280V, and the (now occasional) SCR bridge at 200V RMS. A
7A13 will handle those, as will a fairly cheap differential probe (few
hundred bucks).

I hope the handheld you used was battery powered. I wouldn't like to stick
670-something volts peak across the PSU isolation of a floating
line-powered 'scope.

--
"Electricity is of two kinds, positive and negative. The difference
is, I presume, that one comes a little more expensive, but is more
durable; the other is a cheaper thing, but the moths get into it."
(Stephen Leacock)

 

[Fred Abse]

On Sat, 06 Mar 2010 14:26:01 -0600, The Phantom wrote:

On Sat, 06 Mar 2010 07:03:56 -0500, Hammy <spam@spam.com> wrote:

On 6 Mar 2010 03:09:01 -0600, The Phantom <phantom@aol.com> wrote:



Tektronix has probes designed to solve these problems:

http://www2.tek.com/cmswpt/psdetails.lotr?ct=PS&cs=psu&ci=13471&lc=EN

They are specifically intended for floating use with the TPS2000 series
scopes, which, as John mentioned, have true isolated inputs.

Or, you can use these probes with an ordinary scope:

http://www2.tek.com/cmswpt/psdetails.lotr?ct=PS&cs=psu&ci=13415&lc=EN

Those are nice expensive but nice.

You should use an isolation transformer on any line powered device your
testing. You can usually find a 500VA one for under 100 bucks. I got a
TEMA 500VA one for 45 bucks from Newark. Well worth the money.

Its not really a good idea to float your scope.

A problem arises when you need to make measurements with a 2 or 4 channel
scope at locations in the circuit which don't have a common reference.

For example, suppose you need to look at the gate-source voltages of the
two top MOSFETs in an H-bridge configuration. Isolating the equipment
doesn't allow you to connect the ground clips of the two scope probes to
different points in the circuit that have a substantial voltage
difference.

You need to use floating differential probes, or a scope with true
isolated inputs.

7904, two 7A13s, job done.


--
"Electricity is of two kinds, positive and negative. The difference
is, I presume, that one comes a little more expensive, but is more
durable; the other is a cheaper thing, but the moths get into it."
(Stephen Leacock)

 

[The Phantom]

On Sat, 06 Mar 2010 15:25:03 -0800, Fred Abse <excretatauris@invalid.invalid>
wrote:

On Sat, 06 Mar 2010 14:26:01 -0600, The Phantom wrote:

On Sat, 06 Mar 2010 07:03:56 -0500, Hammy <spam@spam.com> wrote:

On 6 Mar 2010 03:09:01 -0600, The Phantom <phantom@aol.com> wrote:



Tektronix has probes designed to solve these problems:

http://www2.tek.com/cmswpt/psdetails.lotr?ct=PS&cs=psu&ci=13471&lc=EN

They are specifically intended for floating use with the TPS2000 series
scopes, which, as John mentioned, have true isolated inputs.

Or, you can use these probes with an ordinary scope:

http://www2.tek.com/cmswpt/psdetails.lotr?ct=PS&cs=psu&ci=13415&lc=EN

Those are nice expensive but nice.

You should use an isolation transformer on any line powered device your
testing. You can usually find a 500VA one for under 100 bucks. I got a
TEMA 500VA one for 45 bucks from Newark. Well worth the money.

Its not really a good idea to float your scope.

A problem arises when you need to make measurements with a 2 or 4 channel
scope at locations in the circuit which don't have a common reference.

For example, suppose you need to look at the gate-source voltages of the
two top MOSFETs in an H-bridge configuration. Isolating the equipment
doesn't allow you to connect the ground clips of the two scope probes to
different points in the circuit that have a substantial voltage
difference.

You need to use floating differential probes, or a scope with true
isolated inputs.

7904, two 7A13s, job done.

TPS2024. No plugins. 4 channels. Job done, and battery operation for ultimate
portability.

 

[Fred Abse]

On Sat, 06 Mar 2010 17:37:01 -0600, The Phantom wrote:

On Sat, 06 Mar 2010 15:25:03 -0800, Fred Abse
excretatauris@invalid.invalid> wrote:

On Sat, 06 Mar 2010 14:26:01 -0600, The Phantom wrote:

On Sat, 06 Mar 2010 07:03:56 -0500, Hammy <spam@spam.com> wrote:

On 6 Mar 2010 03:09:01 -0600, The Phantom <phantom@aol.com> wrote:



Tektronix has probes designed to solve these problems:

http://www2.tek.com/cmswpt/psdetails.lotr?ct=PS&cs=psu&ci=13471&lc=EN

They are specifically intended for floating use with the TPS2000
series scopes, which, as John mentioned, have true isolated inputs.

Or, you can use these probes with an ordinary scope:

http://www2.tek.com/cmswpt/psdetails.lotr?ct=PS&cs=psu&ci=13415&lc=EN

Those are nice expensive but nice.

You should use an isolation transformer on any line powered device your
testing. You can usually find a 500VA one for under 100 bucks. I got a
TEMA 500VA one for 45 bucks from Newark. Well worth the money.

Its not really a good idea to float your scope.

A problem arises when you need to make measurements with a 2 or 4
channel scope at locations in the circuit which don't have a common
reference.

For example, suppose you need to look at the gate-source voltages of
the two top MOSFETs in an H-bridge configuration. Isolating the
equipment doesn't allow you to connect the ground clips of the two
scope probes to different points in the circuit that have a substantial
voltage difference.

You need to use floating differential probes, or a scope with true
isolated inputs.

7904, two 7A13s, job done.

TPS2024. No plugins. 4 channels. Job done, and battery operation for
ultimate portability.

No comparator / voltage slideback function.

--
"Electricity is of two kinds, positive and negative. The difference
is, I presume, that one comes a little more expensive, but is more
durable; the other is a cheaper thing, but the moths get into it."
(Stephen Leacock)