detecting break in current flow

[aliyesami]

Reviewing your posts, I feel that a clear specification is still missing. You want to detect a) failure of the power supply b) LED failure

l want to detect failure of power supply since as I mentioned earlier the LEDs could have been manually dimmed to zero light emission. If I check for LED current it would give me false reading .

I said find out, not ask Meanwell.

how can I find out the internal circuit working of this LDD driver ?its a sealed box.

Any example?

in my sun server if any module goes bad including the power supply a warning led lights up.

 

[chuckey]

Your LED driver generates a current which is PWM'd, i.e. it consists of unidirectional current pulses. So in series with the earthy end of your LED chain insert a transformer, so every time a current pulse occurs you get a voltage spike on the secondary. Use this spike to charge up a capacitor via a diode then amplify the voltage to a convenient level. PWM circuits do not normally produce zero volts , just very narrow pulses.
Frank

 

[aliyesami]

Your LED driver generates a current which is PWM'd, i.e. it consists of unidirectional current pulses. So in series with the earthy end of your LED chain insert a transformer, so every time a current pulse occurs you get a voltage spike on the secondary. Use this spike to charge up a capacitor via a diode then amplify the voltage to a convenient level. PWM circuits do not normally produce zero volts , just very narrow pulses.
Frank

I really like this idea. please see the attached picture. . something like this ?

 

[andre_luis]

The diodes at the primary side of transformer should be removed, otherwise the core will saturate.

 

[aliyesami]

The diodes at the primary side of transformer should be removed, otherwise the core will saturate.

those are not diodes , those are LEDs
regards

 

[aliyesami]

I measured the voltage across diodes in the full dim position and its zero .meaning no current is flowing through the diodes . so in this situation the proposal by Chuckey will also not work.
how does DMM measure current in micro Amps? if I can do that I will be fine since I have to detect only two states , current lower than 1mA and current >=1 mAmp.
or am I thinking wrong?

 

[BradtheRad]

how does DMM measure current in micro Amps?

It inserts a resistance, and measure volts across it. I have a VOM where the resistance is 100 ohms for the 60 uA range. It is 1 ohm for the 300 mA range.

- - - Updated - - -

Perhaps a VOM has a different arrangement from a DMM, however.

 

[aliyesami]

so why can't i apply the same principle in detecting if the current is below or above 1 Amp ?

 

[BradtheRad]

A sense resistor in the range of 1/20 or 1/10 ohm ought to develop sufficient voltage across it, so you can easily amplify it with an NPN operated in common-base mode.



Voltage gain is high. This arrangement is not so well-known as the typical ways of using a transistor. It may require some effort to make it work properly.

Notice none of your led current passes through the transistor itself. The relative B-E voltage is altered by current going through the sense resistor.

 

[chuckey]

As I said the current pulses are very narrow, when you said the current is zero , how did you measure it?
Frank

 

[aliyesami]

As I said the current pulses are very narrow, when you said the current is zero , how did you measure it?
Frank

I didn't measure the current I measured the voltage and it drops to zero so the current must be zero as well ?

- - - Updated - - -

A sense resistor in the range of 1/20 or 1/10 ohm ought to develop sufficient voltage across it, so you can easily amplify it with an NPN operated in common-base mode.



Voltage gain is high. This arrangement is not so well-known as the typical ways of using a transistor. It may require some effort to make it work properly.

Notice none of your led current passes through the transistor itself. The relative B-E voltage is altered by current going through the sense resistor.

since iam not very good in analog electronics I don't know how to tweak this circuit, anyway someone can test it ?

 

[BradtheRad]

since iam not very good in analog electronics I don't know how to tweak this circuit, anyway someone can test it ?

My common-base configuration works on a similar concept as your schematic in post #23. It controls the transistor via emitter voltage, rather than bias voltage.

The main idea is to choose the sense resistor so it develops .2-.3 V when the led's conduct. This is sufficient to operate the transistor.

By experimentation you adjust bias to the transistor, so it turns on when the led's are off, and shuts off when the led's are conducting.

- - - Updated - - -

The led turns on when the source (Amperes being measured) is off: