Leakage Voltage Across the Medium Voltage SCR in "OFF" State

Leakage current I can understand...
A leakage voltage is pretty meaningless unless current flows....
If current flows in the OFF state then surely it's a dead SCR.

Beware of voltage readings on high impedance meters they can be very missleading.
Del

Pls, see the following detail to batter understanding the Case.

• We have a "SCR" connected at 7.89kv.AC rated system (Line voltag).
• Line voltage are available at ANODE Terminal.
• No Gate firing yet ( SCR is in OFF state).
• we have measured the voltage at the LOAD terminal ( between Phase to Ground) around 500v at all three SCR .
• These voltage suppose not to be pass before the firing.

I hope that now its clear to understand the case/condition.

What instrument did you use to measure this voltage? Was the load connected when you made the measurement?

FLUKE-287 & Load was Isolated/Disconnected.

I suggest you repeat the measurement with a dummy load - a 12V lamp will do nicely.

_{Pity you didn't use a Fluke 289 with the LoZ setting, which reduces the meter input impedance from >10MΩ to about 3.2kΩ.}

we have connected an Incandicent Lamp( 220vac , 300W) across the Phase to Ground which is "busted" within seconds.

<..."busted" within seconds....> That's not very surprising, if it comes from a higher voltage supply!

An SCR is simply an electronic version of a switch.

• If the SCR is off, then the voltage across it could be whatever is on the supply terminals to the equipment. At that time, the voltage across the load will be very low.
• When the SCR turns on, expect the full line voltage to be across the load. If the load is not rated for the full line voltage, then <..."busted" within seconds...> would be the anticipated outcome.

The concept of "leakage" applied to the label "voltage" is abstruse.

Be sure you have a sensible low load resistance value when you take these leakage-current measurements. If you measured 500V with your intended operating load resistance, perhaps your selected SCR part is too large. What part are you using?

Not sure how you have the SCR connected, but having the load in the anode circuit would be normal.

With the SCR off, if you read the voltage between phase and ground from either side of the load or from the anode, you should get full line voltage.

To check for leakage through the SCR when off, either measure current through the load or voltage directly across the load.

With the SCR off, there should be no voltage impressed across the load but full voltage across the SCR. If there is a small voltage, then that voltage divided by the resistance of the load will give you a fair indication of the leakage current.

With the SCR conducting, you should get almost line voltage across the load and almost zero across the SCR. There will be a small voltage drop across the SCR when conducting.

In the event that you have the load in the Cathode circuit, try disconnecting the gate to see if you are getting some form of feedback via the gate

I believe we have another case of ESO here.

A SCR is a self latching electronic switch that cannot be actively turned OFF. It can only be turned ON. When the current drops below a near zero critical value or an anode to cathode reverse voltage bias happens and then the SCR turns itself OFF. By replacing the load with only a the high impedance load of a digital volt meter (DVM), the SCR will turn OFF and only the leakage current through the SCR will propagate through. Lets not forget that in this OFF state, the circuit leakage current will be larger than just the SCR leakage current because all alternate paths (wire insulation, circuit board contaminants, etc.) will sum together. Now the high impedance load of the DVM becomes a detriment because this current produces a voltage across the DVM.

Definitely "SCR" was turned into "OFF" state, otherwise we should received 100% voltage( i.e 7.89kv) insted of 500v at Cathode/load point.

While, during this measurement load( Motor) cable are isolated.

Hmmm,
Post #1.
I covered all the points subsequently raised, (despite not being told it was high voltage) but I don't get a GA?
Del
<shambles off a broken kitty shaking head in dispair>

Cheer up. There's a tin of tuna in the post.

Del the cat.

Unfortunately our friends understand wrong, Basically the question I raise is that,

What is the limit of leakage Voltage at Cathode/Load ( as per design/manufacturer limits) for a Medium Voltage( 13.8KV/7.89KV SYSTEM VOLTAGE) "SCR" while the "LINE" side is healthy but, SCR is still in OFF State ?

We noticed that the metering device (digital voltmeter connected at LOAD side secondary side of an open Delta P.T) was measuring around 500vac at the primary of PT even the SCR was in OFF state that suppose to be Zero volt. After that ,we start invastigation of this LEAKAGE VOLTAGE & posted this to CR4 friends.

Thanks

#10↑ and #11↑ explain all, which is why the replies have attracted so many GAs. The voltage on the terminals is simply a function of the resistance of the device in the off state and the resistance of the connected load, which seems to be a digital voltmeter with high input impedance. The voltage could still be anything from zero to full line voltage, which explains why the low-voltage filament lamp <...busted...> so readily.

The term <...leakage voltage...> is still abstruse.

_{<unsubscribes>}

Since you do not seem to be grasping the concept here, I'll try it this way:

Q: "What is the limit...? "

A: 100% line voltage, because under the worst case scenario, the SCR may not block voltage at all.

You are asking if there is a design spec on an allowable limit to the voltage leakage through the SCR, we are continually trying to tell you that the basic premise of your question is flawed. There is no such thing.

There will ALWAYS be a voltage potential on the load side of the SCR and what you read is TOTALLY subjective and dependent on what is connected to it, including the meter and any wires etc. Also in your case, you apparently have a PT connected to the load side, that alone provides additional paths for your leakage current to flow. So if you are attempting to use a volt meter on the load side of an SCR controller in order to determine if the SCRs are on or off, the concept is flawed if you are looking for the voltage to be some pre-defined number. The best you can do is to determine, empirically, that the value has not significantly CHANGED from some known value indicating the desired state.

And by the way, if you put your meter on the PRIMARY side of the PT, you may as well have put it on the SCR itself. You still performed a VERY dangerous procedure with a hand held device that was not rated for the task.

Why do I hear a... "Psssst... Boom" happening soon?

Because this OP measured 500V. Then connected a 220V device to the 500V terminals and was surprised that it it died.

LOL...

That was a rhetorical question.

When you block a voltage with a semiconductor, you have a leakage current. If you push it, it may define a leakage current path. But, there is no leakage voltage as far as I can see.

Explain your assertion.

How about this:

Leakage voltage doesn't exist until you try to measure it.

Kind of like Schrodinger's cat.