Sufficiently Isolated scope

[treez]

Hello,

We have increased staffing levels and now wish to buy more scopes. We are currently making 4KW H-bridge inverter drives for brushless DC motors and also all the associated control, auxiliary and also CAN comms circuitry.

We currently use Tektronix TDS 210, TDS 2012 and TDS 2014 scopes...

http://www.tek.com/products/oscilloscopes/selection_chart.html

-This TDS range of scopes is not isolated. The TPS range is isolated but the degree of isolation is not considered to be enough so it is thought to be best to just use non-isolated scopes...and where isolation is needed, just use a differential probe.

I have searched high and low but can't find a scope with enough isolation to probe for example the high side FET gate-emitter input to the upper IGBT in the inverter drives.

I am surprised nothing appears available and wonder if any reader may know of some scope with a good degree of isolation..(since our DC bus is 350V).

I am sure there must be a scope which is ideal for our application but can't seem to locate it.

 

[Skogsgurra]

A differential probe is what I use.

Even if the scope is well isolated, there's usually enough capacitance on the low side of the input to cause problems. Especially if you deal with fast edges. And, in an H bridge, you certainly are.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[LionelHutz]

600Vrms Cat II provided with the TPS series optional probes is not enough isolation?

The THS series scopes have good isolation too, but Tektronix seems to have quit making them. Maybe they'll bring out a replacement?

 

[Skogsgurra]

The cat thing is for power frequency. The low side of most (all) oscilloscopes I have met has a considerable capacitance to ground. When edges are 50 - 100 ns, there will be lots of influence that you do not want. Mostly inductive (ringing) voltage drop in the gnd lead.

The THS is decent. Tried a battery Fluke, the 199 on such a circuit. It went completely wild. Yes, with original probes.

Isolation amplifiers or differential probes are the way to go.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[boBK7IQ]

I use the Yokogawa differential probe. It works great.
boB
K7IQ

http://www.yokogawa.com/tm/dl/700924/tm-700924_01.htm

 

[electricuwe]

Basicly there are four different soulution for this problem

- use differential probe
- use 2 probes and 2 channels for each signal and have the scope calculate the difference
- measure current instead of voltage using current transformers
- haver an isolated supply to the equipment under test (only suitbale for one floating signal and if isolation in the supply has low parasitic capacitance)

 

[gigger]

hi
if you want to scope a high side FET gate with a Tektronix TDS scope....can't you just cut the earth wire in the plug?

surely that will solve any isolation problem.

 

[ScottyUK]

That is a remarkably stupid thing to do, causing avoidable risks of killing yourself or damaging your scope. In the old days it was almost acceptable under carefully controlled conditions, before differential probes with decent bandwidth became available. There's no excuse now other than penny pinching.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[Skogsgurra]

And, in addition, it will not work. Simply because there are internal capacitive paths that will disturb the signal. That's why the battery powered scopes are inadequate. And a mains powered one even more so - even if you cut the PE.


Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[treez]

Hi,

I was just about to write in and disagree with all of you and say that isolated scopes were ok...then i remembered.......

Some years ago i was testing 40 off 6KW prototype electric drive units at a big company.

Thay had a three phase inverter using 6 igbt’s...the DC bus that it switched was 600V.

We used to do “double pulse” tests on each igbt...this is where we would scope the collecter-emitter of each igbt in turn when it’s gate was pulsed on for 10us, then off for 10us, then on for 10us then off again.

While the igbt was pulsed , the opposite igbt in the same leg had an RL series across its collector-emitter. (so that to get some current flowing when it turned on)

Each igbt had a driver which did desat detection (detects if CE voltage goes too high and turns off igbt).

Anyway.....when we scoped the CE when the igbt was turned on......we saw the on state voltage rising up to about 100V then going back down again....

...this voltage could not have been “real” else the igbt would have blown and the desat protection would have kicked in.

..from the above posts...i must say that it must have been due to capacitive current flow in the impedance of the scope probe which caused this strange voltage.

Do you agree?

the thing is , what are isolated scopes for.

are isolated scopes useless for power electronics engineers?....and why does the RS catalog, nxt to the isolated THS720 scope(RS2389361), sat that it has HAS "INDEPENDENT FLOATING ISOLATED CHANNELS FOR SAFE HIGH VOLTAGE MEASUREMENT" (apologies for capitalisation!)


from the above posts...independent isolated channels are worthless.

 

[Skogsgurra]

I agree totally. The THS720 and the Fluke 199 are very good as long as you stay with voltage edges below a few tens of volts per microsecond. So, there are some areas where they can be used. Thyristor drives for instance. But not IGBT stages.

Glad that you got the message. Most guys don't, "It is Isolated - isn't it?"

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[boBK7IQ]

Unfortunately those Tektronix TDS isolated scope channels are only specified to be used with somewhere around 60 or 70 volts of probe differential voltage, although before I read the specs, I used them for phase-to-phase measurements of a few hundred volts. That's why I use those Yokogawa differential probes. The Tektronix isolated channels are nice for lots of stuff, but they are somewhat limited in usage for high speed power circuits because of all their common mode noise susceptibility. I almost always have to put the probe wires through ferrite toroids or similar to help the noise.

boB