Electrically Isolated Light indicator for a current carrying conductor

[hisystems]

I have a current carrying conductor for a motor pump and require to have an electrically isolated lighting indicator on this conductor. The light indicator should come on when current is flowing in the conductor and off when there is no flow of current. I need a practical simplest of circuits. I was thinking of something like a current transformer connected to a small indicator bulb. I would appreciate your kind contributions. Thanks great guys!

 

[Skogsgurra]

Highly doable. But will not work on a cable with all three phases, you need to split the Cable and sense one phase.
Then, it is mostly a question of what intensity you want the lamp to have and what current range the motor works with.

If current changes a lot, a series PTC/R combination can be used to keep light output at reasonable level, it is also possible (actually better) to use an NTC in parallel for the same purpose. Using a LED (or two because of AC) will give you the most lumens for the buck.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[hisystems]

Thanks so much Skogsgurra. Its a single phase 1.5HP/230V motor, so its easy to sense the live or neutral line.The motor running current is about 5Amps. Could you throw more light on a practical PTC/R combination and an NTC in parallel, in achieving this. Are you referring to Positive and negative temperature coefficients?

 

[Skogsgurra]

Yes, thermistors.

You may need that if current varies a lot. But I think that a 1.5 HP pump motor won't change vary much. Perhaps the starting current can hurt the lamp/LEDs.

Do tis:
1. Find a small non-potted transformer with some space left between core and winding.
2. Put one of the wires through the hole.
3. Connect a couple of LEDs to the primary of the transformer. Connect them in "anti-parallel" so that one LED lights on positive half-cycles and the other on negative ones. A LED usually works over a rather broad current range - from 1 mA to 50 mA should be possible if you run it on half-waves.
4. Short out the LEDs when you start the pump. That will allow a "Life after start".
5. Open the short to see if there is any light at all. If it looks OK, and not too Bright, you should try and stop/start the pump several times to see if the LEDs survive.

If they don't light brightly enough or if they are too bright, you will need to do some measurement and perhaps some calculations. But try it out first - you may be lucky the first time...

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[benta]

Instead of the non-potted transformer I'd select a toroidal inductor and stick either the phase or the neutral through the hole in the middle.

Benta.

 

[Skogsgurra]

I was looking in my scrap bin - no toroids. A toroid is, of course, so much better. Then you can use several primary turns, if necessary.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[hisystems]

Thanks so much guys! Wonderful!

Shorting out the LEDs when starting the pump may not always be a practical idea for my end user . Whatever device I use here should be able to survive at start and after start.However, I am thinking that 3000V rated LED should survive if that is available in the market. Moreover, what voltage and current level would we expect at the terminals of the toroid? Enough to power an LED? My ammeter/voltmeter would not read anything with an improvised toroid of about 20 turns.

 

[Skogsgurra]

Voltage is not the problem. It is current. Normally, an LED is rated at 20 mA DC, that would correspond to 50 mA half-wave AC. You may need to swith your meter to AC (if you didn't do that already) it will not show anything on DC.

The shorting out was precautionary the first time you start. Just to make sure that the LED doesn't burn out directly. If the LED lights "normally" - whatever that is, you can skip the shorting out.

Twenty turns and one primary turn would produce 5A/20, which is much too much and at too low a voltage. Try 500 turns instead, that should give you around 10 mA.

But DO NOT use a ferrite toroid - you need transformer plate here. So, a non-potted transformer might serve you better after all. The winding is already there.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[waross]

Off the shelf:

http://www.johnsoncontrols.com/content/dam/

http://WWW/jci/be/integrated_hvac_s...urrent Monitoring Insert_CSDandCTD_lowres.pdf

There are other similar devices on the market.

motor current monitoring device

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[hisystems]

ok ,I am gonna try this out and possibly with different turns until I get something good. whats the risk or otherwise of using a ferrite transformer core?

 

[Skogsgurra]

A ferrite core will simply not work. Saturation flux is low and permeability is low. Ferrites are used for switchmode and HF applications. No risk involved. Other than it won't work. Go with Bills tip - I didn't know there were readymades.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[jraef]

Or just buy one of these:

http://www.nktechnologies.com/current-sensing-switch/as1-current-switch.html

They have an LED indicator built-in, but they also have a relay output that changes state when current flows and you can use your existing 120V to power a pilot light.

"Will work for (the memory of) salami"

 

[jraef]

Ah, sorry to step on your post Bill. I see you already showed that route, but when I surf on my iPad it jumps too much and I miss things.

"Will work for (the memory of) salami"

 

[hisystems]

@Skogsgura. your advice to use transformer plate and not a ferrite core, probably explains why I have not gotten anything reasonable with all my trials! in short,I have not cared about the core I was using.
Thanks great guy!

 

[waross]

There was at one time a cheap and dirty version of the current relay on the market for refrigeration compressor use. I can't find it now. It may be worthwhile to spend a little time Googling yourself.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[itsmoked]

Ebay

Digikey

CR Corp (very cool)

The one I use most

Keith Cress
kcress -

http://www.flaminsystems.com

 

[SceneryDriver]

+1 on CR Magnetics:

http://www.crmagnetics.com/Products/CR2550-P13.aspx

Can't beat the price at $12.35, either.


SceneryDriver

 

[waross]

Great Google skills Keith and SceneryDriver.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[LiteYear]

Second Keith's experience. We happily use the CR2550, though my experience is at lower AC currents where we use multiple turns of a thin cable. With a running current of 5A I'd be uncomfortable about exceeding the 20A max of the CR2550 rating during startup/stall, but I'll leave the details up to you ;-)

 

[itsmoked]

Always helps when you know exactly what you're looking for.

Keith Cress
kcress -

http://www.flaminsystems.com