Low AC Current Sensor Circuit (current sensor)

Thread Starter

HighVoltage!

Joined Apr 28, 2014
181
I am having issues finding a Variable current sensor switch that will precisely sense within the AC current ranges of 0-1 Amps (Test Range 0.001-0.2 Amps). Once the range of (0.001-0.2A) is sensed, I would like to trigger an alarm.

Any suggestions on what I could use to sense such low AC currents?
 

AlbertHall

Joined Jun 4, 2014
12,345
I would suggest a current transformer followed by a precision rectifier and a comparator.
The transformer would isolate the rest of the circuitry from the high voltage circuit and reduce the current.
Then the precision rectifier will produce a dc voltage proportional to the original AC current.
The comparator then compares and this DC voltage to a variable reference voltage and produces an output depending on whether the AC currrent is above or below your selected limit.
 

Thread Starter

HighVoltage!

Joined Apr 28, 2014
181
I would suggest a current transformer followed by a precision rectifier and a comparator.
The transformer would isolate the rest of the circuitry from the high voltage circuit and reduce the current.
Then the precision rectifier will produce a dc voltage proportional to the original AC current.
The comparator then compares and this DC voltage to a variable reference voltage and produces an output depending on whether the AC currrent is above or below your selected limit.

Whats company sells such comparator with relay or SSR output?
Do current sensing switches exist below 1 Amp? Had no luck finding?
 

ebeowulf17

Joined Aug 12, 2014
3,307
I am having issues finding a Variable current sensor switch that will precisely sense within the AC current ranges of 0-1 Amps (Test Range 0.001-0.2 Amps). Once the range of (0.001-0.2A) is sensed, I would like to trigger an alarm.

Any suggestions on what I could use to sense such low AC currents?
Am I reading this correctly - you want an alarm if there's current from 0.001-0.2A? So, if less than 1mA is flowing, it can be ignored, but anything 1mA and above triggers am alarm? That would be a fairly sensitive circuit, potentially tricky to avoid false triggers from spurious noise.

Or does the alarm just need to activate somewhere in that range, so it's potentially acceptable to have the alarm kick in at 200mA?

Anyway, it sounds like you're looking for a simple turnkey solution, and I'm not personally aware of one. However, if you're willing to do a little assembly and soldering, I can get you most of the way to the circuit you'd need. I recently designed custom PCBs and built a circuit that converts AC current, as detected with a current transformer, to DC voltage. You'd still have to add the comparator stage, but the rest is pretty much figured out. As built, one of the two ranges can measure roughly 0-2.5A, with 1mV of DC output representing 1mA RMS AC current. We could change a few resistor values and/or choose a different current transformer to increase sensitivity further, depending on what the max current will be.

@crutschow helped me out a lot with the circuit design, and there's a thread somewhere in these forums covering that discussion. If I can find it later, I'll share a link.

If you're interested, I can share all the Gerber files to make PCBs, or I probably have a spare PCB or two lying around that I could send you, depending on location and shipping costs.
 

Attachments

Thread Starter

HighVoltage!

Joined Apr 28, 2014
181
Honeywell has quite a range of non contact current sensors, e.g. https://sensing.honeywell.com/sensors/digital-inductive-current-sensors/CSDA-series
Max.
Long story short this is what I am ultimately what I am trying to achieve (should have mentioned it earlier). I currently have a 208 Wye circuit (Phases A-C and Ground). I WANT TO SOMEHOW DETECT/MEASURE THE "LOW" CURRENT WHEN ANY PHASE AND OR GROUND COME IN CONTACT WITH EACH OTHER:

1) I have current limiting resistors inline with each phase (A-C) except ground.
2) Based ALL POSSIBLE "shorting" combinations, the current range that is drawn is 1mA - 10mA (ACTUAL CURRENTS MEASURED)

EXAMPLE #1: IF PHASES A AND B SHORTED (ONLY WITH CURRENT LIMITING RESISTOR), I WOULD LIKE TO TRIGGER 2 RELAYS TO INDICATE THAT THESE TWO PHASES SHORTED.

EXAMPLE #2: PHASES A & B & GROUND SHORTED. HAVE 3 RELAYS INDICATE THAT PHASE A, PHASE B, AND GROUND SHORTED BASED ON CURRENT >= 1mA.
 

ebeowulf17

Joined Aug 12, 2014
3,307
This is one of the stranger applications I've heard described. If you don't mind my asking, what do you need 3 phase high voltage for if you'll never let it exceed 10mA?

Also, why do you keep letting it get "shorted?"

This is all very odd.

Anyway, are there times when you're allowing a load to draw power and you don't want these alarms to go off? If so, any idea how we should distinguish between acceptable loads and these "shorts?" If not, a current transformer, or other non contact current sensor, on each leg, plus comparators and indicators for each one.

Seems like a lot of work for a problem that doesn't really make sense to me, but it should be possible.
 

Thread Starter

HighVoltage!

Joined Apr 28, 2014
181
This is one of the stranger applications I've heard described. If you don't mind my asking, what do you need 3 phase high voltage for if you'll never let it exceed 10mA?

Also, why do you keep letting it get "shorted?"

This is all very odd.

Anyway, are there times when you're allowing a load to draw power and you don't want these alarms to go off? If so, any idea how we should distinguish between acceptable loads and these "shorts?" If not, a current transformer, or other non contact current sensor, on each leg, plus comparators and indicators for each one.

Seems like a lot of work for a problem that doesn't really make sense to me, but it should be possible.
I have tried simulating with a CT. Each phase would be passing through the secondary of a CT (step down side).

For example, using a 5:1 CT I simulated:

On the CTs secondary, I simulated with a resistive value and AC source voltage that generated a 1mA current loop. Since the CT has a 5 to 1 ratio, only a shorted insulated wire on the step up current side (primary - through donut opening) would step up current 5 times. Additionally, in series with insulated wire, based on the secondary current (1mA x 5), another resistor was used. Based on primary current loop (R=V/I) , I calculated so that 120VAC would be across resistor (120/0.005=24k-ohms), enough to pull-in a "standard" SSR relay. However, if I was to get an input current signal of 10mA on secondary, 120 x 10 would be across resistor = no good. Unless its done by calculating with highest possible input current (10mA). Unfortunately, the current will be 10 times less.

I ran into a slight issue in which the setup did not like a resistor in series with conductor loop on primary side. The ratio values would get thrown off. However, when I left only bare wire shorted, the ratios made sense.

Are no contact current sensors "sensitive" enough to sense at 1mA - 10mA range?
 
Last edited:

ebeowulf17

Joined Aug 12, 2014
3,307
I have tried simulating with a CT. Each phase would be passing through the secondary of a CT (step down side).

For example, using a 5:1 CT I simulated:

On the CTs secondary, I simulated with a resistive value and AC source voltage that generated a 1mA current loop. Since the CT has a 5 to 1 ratio, only a shorted insulated wire on the step up current side (primary - through donut opening) would step up current 5 times. Additionally, in series with insulated wire, based on the secondary current (1mA x 5), another resistor was used. Based on primary current loop (R=V/I) , I calculated so that 120VAC would be across resistor (120/0.005=24k-ohms), enough to pull-in a "standard" SSR relay. However, if I was to get an input current signal of 10mA on secondary, 120 x 10 would be across resistor = no good. Unless its done by calculating with highest possible input current (10mA). Unfortunately, the current will be 10 times less.

I ran into a slight issue in which the setup did not like a resistor in series with conductor loop on primary side. The ratio values would get thrown off. However, when I left only bare wire shorted, the ratios made sense.

Are no contact current sensors "sensitive" enough to sense at 1mA - 10mA range?
Being sensitive enough to detect 1mA is probably achievable. I only say "probably" because it may be so sensitive that various electrical noise causes false positives.

On the other hand you definitely can't extract enough power from a current transformer to operate a mechanical relay. You'll need a separate power supply in order to actually "do" anything, including lighting indicators or closing relays. A sensor or CT can generate a signal in response to current in the primary, but it can't magically harvest any meaningful power from it.

What you'll need to do is build a circuit that monitors a current transformer or some other non contact current sensor, and activates whatever outputs you want in response to the input signals. It still doesn't necessarily have to be terribly complicated, but it will certainly need a separate power source and almost certainly a little bit of custom circuitry.
 

Thread Starter

HighVoltage!

Joined Apr 28, 2014
181
Being sensitive enough to detect 1mA is probably achievable. I only say "probably" because it may be so sensitive that various electrical noise causes false positives.

On the other hand you definitely can't extract enough power from a current transformer to operate a mechanical relay. You'll need a separate power supply in order to actually "do" anything, including lighting indicators or closing relays. A sensor or CT can generate a signal in response to current in the primary, but it can't magically harvest any meaningful power from it.

What you'll need to do is build a circuit that monitors a current transformer or some other non contact current sensor, and activates whatever outputs you want in response to the input signals. It still doesn't necessarily have to be terribly complicated, but it will certainly need a separate power source and almost certainly a little bit of custom circuitry.
Hmmmmm. Im unfortunately out of ideas at this point.

Any suggestions on something I could try?
 
Last edited:

ebeowulf17

Joined Aug 12, 2014
3,307
Hmmmmm. Im unfortunately out of ideas at this point.

Any suggestions on something I could try?
Sorry for slow response here. Busy long weekend travelling. I'll take a look and do some playing around in simulation. The circuit I shared in an earlier post would get you 95% of the way there, but it's probably overkill for your situation.

I'll see how much simpler I think I can make it and post the results. No guarantees on time frame, but hopefully within the next few days.
 

ebeowulf17

Joined Aug 12, 2014
3,307
If you short phases/ground the current won't be 'low': it will be enough to blow fuses or trip circuit-breakers!
The thread starter claims to have current limiting resistors in series with each leg of power, such that the maximum "short circuit" current is 1-10mA:
1) I have current limiting resistors inline with each phase (A-C) except ground.
2) Based ALL POSSIBLE "shorting" combinations, the current range that is drawn is 1mA - 10mA (ACTUAL CURRENTS MEASURED)
I've asked once, but haven't yet received an explanation as to why such a system is in place, or what useful circuit could exist on a system with such current limiting resistors in place, but that's how it has been described so far.

Apparently, we've got three phase 208V supply, but we never want more than 1mA to flow?!?! Or at least we're supposed to trigger an alert when it does flow. It's all very strange to me. I feel like making a detector circuit should be pretty manageable, but understanding it's real purpose is beyond me.
 

Thread Starter

HighVoltage!

Joined Apr 28, 2014
181
The thread starter claims to have current limiting resistors in series with each leg of power, such that the maximum "short circuit" current is 1-10mA:

I've asked once, but haven't yet received an explanation as to why such a system is in place, or what useful circuit could exist on a system with such current limiting resistors in place, but that's how it has been described so far.

Apparently, we've got three phase 208V supply, but we never want more than 1mA to flow?!?! Or at least we're supposed to trigger an alert when it does flow. It's all very strange to me. I feel like making a detector circuit should be pretty manageable, but understanding it's real purpose is beyond me.
Depending of the "short combination" of 3 phase wye circuit, you can pull between 1mA (min) - 10mA (max). I would like to detect when the minimum current of 1mA is detected. 1mA or greater in any phase shall "immediately" trigger its respective relay to indicate which phases and or ground touched. Be assured that nothing will trip or "blow up".
 
Last edited:

ebeowulf17

Joined Aug 12, 2014
3,307
Depending of the "short combination" of 3 phase wye circuit, you can pull between 1mA (min) - 10mA (max). I would like to detect when the minimum current of 1mA is detected. 1mA or greater in any phase shall trigger its respective relay to indicate which phases and or ground touched. Be assured that nothing will trip or "blow up".
So, just to be clear, in the final implementation of this project, there will never be more than 10mA flowing through any of the conductors?

I ask this because it effects what sort of input protections we need to include in the detection circuits.
 
Top