SMPS and O-SCOPES

[Bigdrums]

I've heard a lot about the danger of damaging or destroying one's O-Scope when checking a SMPS. As I understand it, this is because of a "floating" ground common to the circuits in SMPS's.
I'm wondering the best way to eliminate the danger to the scope? Would using an isolation transformer to power the scope be an effective way to solve this potential serious problem? Are there other
and/or better solutions? What about a ground lift? BTW I'm working on a Mackie SRM450 V3 if anyone has a schematic that would be very helpful

[Enzo]

Use the iso to power the unit under test.

A typical SMPS has the primary side "ground" at something like -170vDC to earth. If your scope is earthed as most of them are, then you connect your scope probe ground to that circuit common and you are shorting across that 170v. You COULD isolate the scope, but then when you connect your scope probe ground to that, it works, but now the frame of your scope is sitting there at 170v. There are folks who do it that way, but I sure don't.

I power the UUT with the iso, and now that -170v common is not linked to an earth system, and I can safely "ground" my scope to it.

These concerns affect the primary side of the SMPS. Th secondary side has no such issues.

Someone may post your drawings, but you can probably get them by just contacting Mackie.

[Jon Snell]

Always use an isolating transformer on the device on test and earth your test gear.

[Greg Robinson]

Always use an isolating transformer!

But NEVER on the scope!

Only ever on the DUT (device under test).


Even better would be to use an active differential probe along with the DUT on the isolation transformer.

[Bigdrums]

I'm confused. If the ground (or common) on the primary side of the SMPS is sitting at -170v with or without the iso powering it, it's still -170v, right?
And if the scope in grounded, isn't there the potential for KAPOW when you attach the ground wire from the scope lead?
Wouldn't isolating the scope from ground prevent that? Or at least help?
As far as the schematic, Mackie tells me they don't provide schematics of current models. I am just hoping someone out there might be able to share
Thanks for the input on this somewhat confusing problem!

[Dave H]

I'm confused. If the ground (or common) on the primary side of the SMPS is sitting at -170v with or without the iso powering it, it's still -170v, right?

Use the ISO transformer between the mains supply and input to the SMPS. The primary side of the SMPS is now a floating 170V supply and you can ground its common togther with the common of the SMPS secondary side. It's now safe to use the scope on both the primary and secondary sides of the SMPS.

[nevetslab]

Years ago, I managed to pick up a Sony/Tektronix 326 10MHz Dual Trace Battery Powered scope, so I could safely work on SMPS's, though with extreme caution. With the scope probe's ground attached to the AC side of the SMPS, that is elevated to that dangerous potential, as stated above. Truth is, I never DID go working on them that way, and instead, will use an high current isolation transformer for the amp under test. All of the amps that we find commercially produced for back line gear that have SMPS and Class D amps, the output is still Earth Ground potential, as is the input. All the scary stuff has been done so we don't get ourselves and our test gear damaged or killed. AT least I haven't come upon one that was dangerous, but if I don't know the unit, I look and go digging for details on the internet to be sure.

[Enzo]

I'm confused. If the ground (or common) on the primary side of the SMPS is sitting at -170v with or without the iso powering it, it's still -170v, right?

But with respect to WHAT? In a plain old amp, that is -170v with respect to earth. On an iso, that -170v still exists between the amp chassis and that common, but not to the rest of the world.. The UUT is...well... isolated. So when my test gear is all earthed, my UUT chassis is floating.

You can get just a bare isolation transformer and wire it into your bench setup as you see fit. Or you can get commercial pieces of test gear isolation transformers. I had several of these RCA units in my shop.

https://www.worthpoint.com/worthoped...rmer-291163586

[g1]

I'm confused. If the ground (or common) on the primary side of the SMPS is sitting at -170v with or without the iso powering it, it's still -170v, right?
And if the scope in grounded, isn't there the potential for KAPOW when you attach the ground wire from the scope lead?
Wouldn't isolating the scope from ground prevent that? Or at least help?

Yes isolating the scope would prevent it, but it is too dangerous. Your scope ground and all parts connected to it will now be sitting at -170V. Like Enzo said, the scope frame and any ground terminals on the scope will now be sitting at high voltage. If you touch it the result could be fatal.

That is why we (when required) always isolate the unit being repaired and keep our scopes and test equipment earthed. Safety.

[Kevina]

I wrote this years ago for our techs when they were working on the A&H ZED power supply, it may help

[glebert]

This would be a much better discussion with a diagram and some common terminology. If I didn't know better and read this thread I would think that an SMPS can feed -170VDC back to AC mains. In the diagram from Kevina's doc the L GND is the one that might be at -170V, right?

I don't do much work on SMPS, mostly want to see if they are switching or not, and using a method similar to Kevina's service doc, where it is not necessary to ground the probe, is sufficient for me.

[Helmholtz]

A voltage only exists between two different points.
That's why a voltmeter needs 2 leads.

Voltage of a single point doesn't make sense and couldn't be measured.

So it always matters to what other circuit point the voltage refers.

[glebert]

A voltage only exists between two different points.
That's why a voltmeter needs 2 leads.

Voltage of a single point doesn't make sense and couldn't be measured.

So it always matters to what other circuit point the voltage refers.

RF probes don't necessarily need a ground.

[Helmholtz]

RF probes don't necessarily need a ground.

Because the scope and the DUT are grounded.
The scope probe always measures the difference between the probe potential and its own ground.

With RF some common capacitance may suffice to provide a ground reference.
But voltage measurement is always done between two points having different potentials.

Voltage always needs a reference, i.e. 2 points having different potentials.
A single point can't "have" a voltage.

[g1]

I wrote this years ago for our techs when they were working on the A&H ZED power supply, it may help

A reminder that for North American application, listed voltages need to be cut in half.
After you mentioned starting up on variac, I was scratching my head when you said after repair "set variac at 120V". Then I remembered you were speaking of 240V unit.