CT under load and not connected to emonTx

Hello,

I discovered too late that CT clips should be plug firstly to the emonTx... It has been under load during 1 minutes, emitting a little noisy vibration. Now it is connected to the emonTx.

Should I worry ? Does it destroy the sensor very fast ?

Best regards,

Fabrice

 

Robert Wall's picture

Re: CT under load and not connected to emonTx

Which CT is it? The YHDC ones from the shop have zener diodes fitted for protection, which should prevent damage. If it is a different one to that, it is anyone's guess what has happened to it. At worst, it may have become permanently magnetised, or the insulation could have broken down under the very high voltages that can be generated, or both. That is the reason why you should never open-circuit a current transformer. Short-circuiting it is fine.

fabrice's picture

Re: CT under load and not connected to emonTx

It is the YHDC SCT-013-00 so it should be ok :-)

Thank you

Robert Wall's picture

Re: CT under load and not connected to emonTx

What you heard was either the cores shaking in the magnetic field because they were not held together tightly, or magnetostriction - this is where the dimensions of the material change under the influence of the magnetic field, so making a noise.

bob teatow's picture

Re: CT under load and not connected to emonTx

I have a CT with no internal protection of any kind - just wire on a ferrite core - Besides a burden resistor, is it necessary or advisable to put zeners or tvs-diodes or other protection devices into the circuit that will be monitored by the Arduino ADC?

I will be using this to monitor a house circuit:  HWCT004 (available on ebay http://www.ebay.com/itm/HWCT004-AC-Micro-Current-Detection-Module-50A-Current-Sensor-Module-e-/261439788221?pt=LH_DefaultDomain_0&hash=item3cdf0694bd )

Tx for your expert input!

Robert Wall's picture

Re: CT under load and not connected to emonTx

I can't find any sensible information about that. It should not be necessary to add anything in parallel with the burden, unless you have a very large CT (one with a big VA rating) and you expect a downstream fault that will generate a high transient voltage. If you have a small CT that is not excessively rated (say 100 A CT on a 60 A or 80 A supply) then it's likely to saturate before much harm will be done. To protect the Atmel processor inputs, you could always limit the current with a 1 kΩ series resistor between the burden and the input, as has been done with the emonTx V3. It should make no discernible difference to the measurements.

bob teatow's picture

Re: CT under load and not connected to emonTx

Dear Mr. Robert Wall,

Thanks for your reply.   Now I see you had a theoretical discussion of this in http://openenergymonitor.org/emon/node/171 but in the end went with a 1K ohm current limiter and omitted the 100pf bypass.  

(( For my project I am using an Arduino Due board, which like the EmonTx V3 uses an Atmel 3.3volt processor.  So I'm going to guess there is similar internal protection on its ADC inputs.  (I couldn't find info in the Sam3X data sheet...)  I will follow the TxV3 analog input design, adjusting the burdens for my several different CTs (30a, 50a, 100a, and 200a) and adjusting the voltage divider for whatever AC transformer I pull from my "junk box".  I plan to bench test the CTs and burdens using an electric heater as a load and several loops through a CT to simulate multiples of the current, so easy to test 100a or more and do the fine calibrations in software.  I have an autoranging voltmeter so I can sanity check each analog circuit before hooking into the Arduino.  I haven't messed about with electronic breadboarding for some years - and this is my first analog to microprocessor project, so please be tolerant of my questions and ramblings. 

BTW - I have my Due talking via USB to my MK802 running Linux, waiting for some other components to come in from Asia. ))

Tx again, YT, BobT

phermansson's picture

Re: CT under load and not connected to emonTx

Did you have any success with the HWCT004? I think of buying one myself as it is small and easy to fit.

calypso_rae's picture

Re: CT under load and not connected to emonTx

On a handful of occasions, I have unintentionally had current flowing through an SCT-013-000 CT without the other end being connected to anything.  The result is a characteristic buzzing noise which goes away when the connection is correctly made.  I very much doubt whether the CT will have been permanently damaged by this operation.  The processor has some internal protection for its input pins but I generally provide some additional protection diodes in the interface circuit too.

When an emonTx V2 is mounted inside an outer enclosure, it can be difficult to ensure that a good contact is made when the CT's jack plug is inserted.  An alternative approach is to mount a jack-socket in the wall of the box, and to extend the wiring to the PCB as necessary.

Robert Wall's picture

Re: CT under load and not connected to emonTx

The SCT-013-000 has 22 V zener diodes internally to limit the output voltage. However, in the emonTx V2, the impedance of the current limiting components in the input circuit of the ADC (the voltage divider resistors and their bypass capacitor) given that voltage is marginal and the current could be enough to damage the ADC input. Hence the warning not to open-circuit the CT under load.
The emonTx V3 has an additional series resistor that should provide that extra protection to be reasonably certain that the current will be within the limits allowed for the internal protection diodes.

[Edit]
I should add that all this applies only if you have a small CT with a correspondingly small VA rating. If you have a larger CT, you MUST ensure it has a burden, or is short-circuited, at all times, otherwise damaging and possible dangerous voltages might result.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.