Voltage isolation for scope to test incoming service power?

[fastline]

We want to use our scope for testing line voltage. the same voltage that is supplying power for the scope. I have learned in a brief connection years ago that this is a no-no. Will a simple isolation transformer take care of this or do I need something else?

 

[Nigel Goodwin]

An isolation transformer will - or just a full understanding of what's involved (but I'm not going there again).

 

[MikeMl]

240V or 120V? USA or Europe?

 

[fastline]

input power to the scope is just 120v, 1ph, US power. Will be testing 3ph

 

[crutschow]

Since you don't really want to connect the scope common to the AC common even if you have an isolation transformer on the scope, a preferred way is to use a differential (A-B) setting on a dual-trace scope and use two 10x probes to measure the voltage.

 

[MikeMl]

input power to the scope is just 120v, 1ph, US power. Will be testing 3ph

That makes it complicated, because then I would have to know if your 120V is derived from a Wye or Delta. To be safe, get a 240V-240V or 240V-120V transformer with two isolated windings.

 

[fastline]

I guess I am confused by that. I was unaware a load could tell if it is connected to a wye or delta connection. Obviously a corner grounded delta will have a high leg but that is sort of irrelevant in a split phase 120/240 configuration.


Crutschow, I think I am also a bit confused on your reply. How would the 10x leads fix this? I think the concern was that the neg lead on the scope is directly coupled to the input ground so if I connect that neg lead to a common or ground that is a mutual connection with the scope input power, bad things happen. Surely there is a common way to islolate a power supply for the scope so it can test grid power?

 

[MikeMl]

...Surely there is a common way to islolate a power supply for the scope so it can test grid power?

Yep, it is called a an isolation transformer

Rather than floating the scope, and creating a death trap because the scope case gets connected to one of the 240V lines, use the transformer at the input of the scope.

 

[Reloadron]

input power to the scope is just 120v, 1ph, US power. Will be testing 3ph

There are active differential scope probes designed for taking measurements like this. I frequently measure 480 VAC 3 Phase Delta using probes like these probes. They aren't cheap probes. There are other less costly than the Tektronix line. If the power is Delta and you want to measure A to B, B to C and C to A all at the same time for comparison you need to get your vertical inputs from single ended to differential.

You do as Mike suggest, need to find out the 3 phase power configuration as to Delta or WYE.

Ron

 

[crutschow]

Crutschow, I think I am also a bit confused on your reply. How would the 10x leads fix this? I think the concern was that the neg lead on the scope is directly coupled to the input ground so if I connect that neg lead to a common or ground that is a mutual connection with the scope input power, bad things happen. Surely there is a common way to islolate a power supply for the scope so it can test grid power?

I understand the problem with connecting the grounds.

I was suggesting the use of a differential connection with two probes similar to what Ron mentioned. If you use two probes, one on each of two channels and use the A-B setting on the oscilloscope, it will measure the voltage difference between the two probes. You connect one probe to one side of the line and the other probe to the other side.

The 10x probes are just to reduce the voltage to a range that the oscilloscope can measure (I think most scopes only go to about 50V maximum). The scope common is connected to the scopes input power safety ground and nowhere else.

As Mike mentioned, this may not work with a delta connection.

 

[canadaelk]

A number of years back I designed a distributed paging/music system (12 locations, 15 miles of 1 pair of a 100 pair telephone cable). I needed to know frequency response and noise levels of that pair at each location. I had not budgeted for a probe so I built one.
I just tested it with the resistors on the input (mine was only for audio levels) and it works just fine. The voltage should be determined with a RMS meter anyway.
Hope that helps! E

 

[fastline]

As I understand it, the ratio probes have a built in isolation transformer so pretty much cover the voltage reduction and isolation requirement in one shot?

What we need to do is look specifically at 3ph VFD outputs. THD, PF, phase shift, phase angle, etc. Looking to filter harmonics to achieve a more sinusoidal output. I was thinking a 3ch scope would be needed here but maybe more. We have been looking for a scope solution for some time. So far, it looks like a "math scope" is what we are looking for. Found one but price was 20 grand and just cannot spend near that amount. Certainly open to ideas to test these criteria. For this app, the IGBT processor operates at only 20khz.

 

[crutschow]

As I understand it, the ratio probes have a built in isolation transformer so pretty much cover the voltage reduction and isolation requirement in one shot?

So far, it looks like a "math scope" is what we are looking for. Found one but price was 20 grand and just cannot spend near that amount. Certainly open to ideas to test these criteria. For this app, the IGBT processor operates at only 20khz.

The ratio (differential) probes that Ron mentioned have everything included to perform high voltage isolated differential measurements.

If your budget is limited, you might consider some type of PC oscilloscope adapter such as one of these or these, since the PC would give you all the math processing you need.

Also most modern oscilloscopes allow downloading of the waveform data to a PC where you could then do the number crunching.

 

[fastline]

Yes, those PC scopes are what I was looking at and since we rarely would need mobile testing, it seems viable BUT I was rather concerned about the speed and accuracy of the PC based scopes. Maybe they do great, just not sure. I was also a little turned off on the 1 grand price tag and got to wondering if I can get about the same from a used scope on fleabay. Many decisions here I guess...

 

[fastline]

Only having one scope around here, I did look at the difference in single ended and differential systems. I guess I was under the impression that ALL scoped had 2 leads per channel. It looks like a single ended system actually is designed to use a mutual ground side which would be for maybe testing on a mutual grid. Our scope has the discrete or differential channels with 2 leads per channel. maybe I have that right or maybe I need to read some more.... I would sure think you could switch from differential to single ended but maybe that is a "special" feature.

 

[crutschow]

Most scopes have a single-ended input with a common ground between all inputs, not differential. What make and model is yours?

 

[fastline]

Scope is a BK Precision 1477. Ancient so that is why we are looking to replace it. It did what we needed at the time but now needing better triggering, digital recording, etc. I think I might be off base with my previous post as well. Shows how much I use the thing.... We did use it extensively for examining discrete AC save pulse peaks but that is about it.

If we are looking to buy a scope, would it be best to just get a differential scope to reduce costs with isolated leads and such?

 

[crutschow]

If we are looking to buy a scope, would it be best to just get a differential scope to reduce costs with isolated leads and such?

The problem is finding such an oscilloscope. I don't know if anyone makes such a unit.

It might be better just to get a standard oscilloscope with the features you need and then buy a differential probe that will measure the high voltages such as one of these or this. That likely will be cheaper, especially if you buy a used oscilloscope.

 

[The Electrician]

In your first post, you said "We want to use our scope for testing line voltage." Why can't you just use a voltmeter, a handheld DVM for example?

Perhaps you need to observe the waveform of the line voltage, but you didn't say if that is the case. Assuming you really need to use a scope, and if you're looking to buy a new scope, you could get a scope with isolated inputs and which is intended for measurements on grid power service. Such a scope is the Tektronix TPS 20xx series:

**broken link removed**

You don't need to use differential probes with this scope; the 4 inputs are all isolated from each other, and the scope can be battery operated for greater safety. You need to use the right kind of passive probes for safety, and for voltages over 120 VAC you will probably need to use 100:1 probes.

With a 4 channel unit, you can observe all 3 phases at once with no further isolation other than what the scope itself provides.

Alternatively, rather than using an isolation transformer to power the scope you could use a transformer to isolate the voltage you want to measure. To get good accuracy you should use a potential transformer which is specified for the purpose with known accuracy, for example:

**broken link removed**

 

[Reloadron]

I have no idea what they cost but the Tektronix TPS2000 series does have isolated differential inputs with up to 4 vertical channels. They can be found here. I use an older 2 channel version of a similar Tek Scope that has the same feature but can't recall the model? Maybe 5 to 7 years ago I want to think it was about a $2K scope? With 2 channels you won't be looking at 3 phase signals. Not at once anyway.

Something else I like and use is one of these units however, a base unit runs about $3,000 for starters. Eight differential channels capable of 1,000 volts on the highest range. I like the record and playback features. There are much cheaper units like this out there. I grab a laptop and I am on my way. These are NOT a scope, but can behave like one.

Now if you want to experiment cheap then one of these in a $200 flavor will get you 4 channels differential input with a max range of +/- 64 volts. The high speed version sells for about $343. The ***** here is the FS Voltage of 64 volts. However, if you could live with a few cheap transformers it would give you some ideas maybe.

<EDIT> Beat by the Electrician to the finish line.... </EDIT>

Ron