Cheap differential scope probe?

[Speakerguy]

I have a grounded oscilloscope (Tek DPO3k series) and I will shortly be working with a full-bridge audio PWM amplifier chip (TI TAS5261). I know I cannot use my regular grounded probe and grounded scope with a regular power supply where it's GND is tied to neutral (as it should be in safely earthed amplifiers).

I need a solution so that I can look at the output of the H-bridge PWM amp section. A Tek probe is prohibitively expensive. Are there other brands of probes that would be OK? Can I use a 3 to 2 cheater plug to lift the scope ground? Or would I still have a problem?

I will be using an isolated switching supply (24V 320W) for the power amp output section, but it's (V-) may be connected to say the USB ground or some other ground that is earthed, and I don't want to count on everything being isolated from ground/neutral so that I could use my earthed scope.

Would an isolation transformer fix my problem, or do I need to find a differential probe? The Tek active probe is four figures! and I don't want to spend that. Need it to be 100MHz, would prefer 300-500MHz (scope is 300MHz).

 

[Bob Scott]

You can't just use a ground cheater adapter to unground your scope. If you must live dangerously, use an isolation transformer on the scope to troubleshoot hot wired equipment. I can't emphasize too much that this method is not safe and requires extreme caution.

Are you saying that your amplifier does not use an power transformer? Then add an isolation transformer and leave the amp ground floating instead. Leave the scope grounded.

Note: An autotransformer or Variac does not provide isolation.

If it is simply a situation where your amp is grounded and your amp output is bridged:
If you use a dual trace scope, connect a probe to each output. You can just use A-B mode to see the difference signal (differential mode).

 

[Mikebits]

Could you perhaps use your scope in differential mode. This from Tek notes.

Built-in differential amplifiers. Many scopes have the ability to make the simplest differential measurements built right in to them. This mode is referred to as "channel A - channel B" or "quasi-differential". While limited in performance, this technique may be adequate for some measurements. To make a differential measurement, two vertical channels are used - one for the positive input and one for the negative input. The channel used for negative input is set to invert mode and the display mode is set to "ADD Channel A + Channel B". For proper operation, both inputs must be set to the same scale factor, and both input probes must be identical models. The display now shows the difference voltage between the two inputs.

 

[Speakerguy]

So I can hook my scope up to an isolation transformer, and use one regular single ended probe across the differential output of an earthed circuit?

The two-probe method won't be high performance enough I don't think, I have to get in there with a single probe and a pigtail ground, I'm looking at PWM edges.

 

[kchriste]

I will be using an isolated switching supply (24V 320W) for the power amp output section, but it's (V-) may be connected to say the USB ground or some other ground that is earthed, and I don't want to count on everything being isolated from ground/neutral so that I could use my earthed scope.

Sounds to me like you'll be able to use your scope as is. It's OK to connect your scope probe ground clip to something that is also grounded. Don't try it on the input side to the power supply, but the output side it should be OK. To be sure, have everything turned on and running and then measure between the scope probe ground clip and where you intend to clip it with a voltmeter. Check with both AC and DC ranges. If it is less and a volt or so, you're good to go.
You can use an isolation transformer on the scope, but beware. If you connect your scope probe ground clip to 300V then 300V will be present on the case of the scope!

 

[Speakerguy]

Question: If an isolation transformer has its secondary neutral connected to earth ground, how is that protecting me?

 

[Nigel Goodwin]

Question: If an isolation transformer has its secondary neutral connected to earth ground, how is that protecting me?

It wouldn't be, and that isn't how a transformer should be used - as you say, it completely removes any usefulness of the transformer.

However, from what you've said so far, I fail to see any need for an isolation transformer anyway - the amplifer chassis is earthed, the amplifier isn't live - simply connect the scope to the grounded chassis as normal. Obviously this doesn't apply to the primary side of the power supply, but it doesn't on anything.

Why do you think you can't use your scope as usual?.

 

[Hero999]

Doesn't your scope have a differential setting?

 

[Speakerguy]

Hi Nigel,

Regarding the isolation transformer, I ordered one of these. That page says it provides an isolated ground, but it says something about 'secondary neutral to ground bonding' in its data sheet (linked to on that page). I was wondering if they meant earth ground, or just the ground lug on the isolated plug outlets. I guess I'll find out when it gets here.

Re: the scope measurement - I want to measure across the outputs of the H-Bridge. If I connect the ground clip of my scope probe to one side, won't it short it to ground when that side goes high? I'm trying to look at edge rise time / overshoot / ringing so I can optimize an RC snubber for it. When I did this at my last job they told me to use a differential probe.

Hero,

No, I don't think so. I will play around with it though, it is still new to me.

Thanks guys,

Mark

 

[Nigel Goodwin]

Hi Nigel,

Regarding the isolation transformer, I ordered one of these. That page says it provides an isolated ground, but it says something about 'secondary neutral to ground bonding' in its data sheet (linked to on that page). I was wondering if they meant earth ground, or just the ground lug on the isolated plug outlets. I guess I'll find out when it gets here.

Re: the scope measurement - I want to measure across the outputs of the H-Bridge. If I connect the ground clip of my scope probe to one side, won't it short it to ground when that side goes high? I'm trying to look at edge rise time / overshoot / ringing so I can optimize an RC snubber for it. When I did this at my last job they told me to use a differential probe.

Don't you have a double beam scope?, stick one probe on each output and set the scope to 'differential' mode - on my scope at work you set it to 'add', and invert one of the beams (by pulling the gain control knob out). This is a pretty standard function on double beam scopes.

Failing that, just scope one of the outputs, the other is just an inverted version of it - scoping one should do everything you need.

 

[Speakerguy]

Oh yes, I can simply subtract B from A no problem, or any manner of maths. Heck I can plug in a USB keyboard and type an expression with +-*/ on any of four channels. My worry is, is the timing going to be accurate with two probe leads in differential mode when I am looking at rise times of ~10ns? That's what I'm worried about with the two scope probes (A - B) solution. When I measured single-ended half bridge outputs at work with a single-ended probe I still had to use a pigtail ground probe right by the tip to get accurate waveforms, clipping onto the board ground clip made things look worse than they actually were.

 

[Nigel Goodwin]

Oh yes, I can simply subtract B from A no problem, or any manner of maths. Heck I can plug in a USB keyboard and type an expression with +-*/ on any of four channels. My worry is, is the timing going to be accurate with two probe leads in differential mode when I am looking at rise times of ~10ns? That's what I'm worried about with the two scope probes (A - B) solution. When I measured single-ended half bridge outputs at work with a single-ended probe I still had to use a pigtail ground probe right by the tip to get accurate waveforms, clipping onto the board ground clip made things look worse than they actually were.

Even if you had a 'diferential probe' there's still only a single ground clip - just clip ground as close as possible.

 

[Willbe]

I have a grounded oscilloscope (Tek DPO3k series)
will shortly be working with a full-bridge audio PWM amplifier chip (TI TAS5261).
an isolated switching supply (24V 320W) for the power amp output section, but it's (V-) may be connected to say the USB ground or some other ground that is earthed,
don't want to count on everything being isolated from ground/neutral so that I could use my earthed scope.
Need it to be 100MHz, would prefer 300-500MHz (scope is 300MHz).

If you don't need high impedance input or high CMRR but you do need high freq. response and slew rate, I'd think you could make up a circuit with BJT transistors, perhaps battery operated.

The Tek probe is high input impedance; a 500 MHz signal from a low impedance source doesn't need a high-impedance probe.

Can you post a simplified version of the circuitry this diff-input/single-ended output probe has to mate with?
What Vp-p levels will you be looking at?
How many signal paths need to be isolated? Just two? Just one?
What are your source impedance levels?

 

[nolfese]

When I was working in full time electronics design I used a Tektronix THS720 - the most fantastic bit of kit on the planet. It would solve your problem in two ways:

1) It will run off an internal battery so no earth issues (BTW, it's plastic - there are no metal parts that are touchable - at least not when using the plastic shrouded probes provided).

2) The two input channels, whilst referenced to the same 'pseudo ground', are independent of each other.

When designing SMPSs I could look at any part of the switcher with one channel whilst triggering off the driver with the other channel. That's just one example of the kind of thing I was able to do on a daily basis without ever giving a thought to shorting anything out or blowing the scope to bits - what a piece of kit.

The model has been discontinued - plonkers - but second hand ones are available on the web.

The Owon PDS series scopes are cost effective and have a battery option but don't have electrically independent channels - at least I don't think they do.
The day I left work was a sad one - more for having to leave my beloved THS720 behind than anything else. Oh well....

 

[The Electrician]

When I was working in full time electronics design I used a Tektronix THS720 - the most fantastic bit of kit on the planet.

The model has been discontinued - plonkers - but second hand ones are available on the web.

Tek now has the TPS2000 series:
**broken link removed**

Up to 4 channels and 200 MHz, color display, all channels and trigger input isolated, plus battery operated.

 

[MikeMl]

All of the posts about using two channels of the scope to create a differential (A-B) voltage trace missed one VERY important point. When using the two channels differentially, the highest frequency where that works reliably is very limited, typically 100Khz or less, even with a fast scope. RTFM!

Also, if you are trying to read a 1Vp-p A-B differential signal where the signals are riding on a 100V offset level, that wont work either, because neither channel has a 100V dynamic range (without saturating) when viewing a 1V signal.

 

[nolfese]

Wow! I bet they cost a bob or two. Sorry - I'm a bit out of touch these days...

 

[The Electrician]

The Tek active probe is four figures! and I don't want to spend that. Need it to be 100MHz, would prefer 300-500MHz (scope is 300MHz).

The cheapest one (new) I know of that will do what you want is this:

**broken link removed**

It's about $450.