Can you put an oscilloscope on the D+ and D- rails with a capacitor to isolate DC. That is, check those rails for AC.

Is there a safety ground in the LED string?

Is there a safety ground in the LED string? --- No
capacitor to isolate DC - This i will try and let you know the result

 

Hi jpanhalt,
I think that you assume that D+ and D- are the outputs from the LED driver to the LEDs. They are in fact the control inputs (Dim+ and Dim -) to the driver. (See the link to the driver datasheet in post #13 There are no page numbers but the block diagram is about a third of the way down on the scroll bar.)

Les.

 

Hi Saiteja,
I don't think there is a device that can be used to directly replace the relay. The only solution I can see to switch the output of the driver would be to us a mosfet in one leg of the output of the driver. This would require a small floating output power supply to provide gate voltage to the mosfet and an optical coupler. This is because we don't know the internal schematic of the driver so we could not safely connect the mosfet source to your arduino ground. An variation on this would be to use a isolated DC to DC converter to drive the mosfet gate. A solution that would probably work would be a solid state relay on the AC input to the LED driver. This would probably give a delay in the LEDs turning off fully due to energy stored in capacitors in the driver.

Les.

 

Is there a safety ground in the LED string? --- No
capacitor to isolate DC - This i will try and let you know the result

I tried those two solutions but still there is no change in the output

 

Hi Saiteja,
I don't think there is a device that can be used to directly replace the relay. The only solution I can see to switch the output of the driver would be to us a mosfet in one leg of the output of the driver. This would require a small floating output power supply to provide gate voltage to the mosfet and an optical coupler. This is because we don't know the internal schematic of the driver so we could not safely connect the mosfet source to your arduino ground. An variation on this would be to use a isolated DC to DC converter to drive the mosfet gate. A solution that would probably work would be a solid state relay on the AC input to the LED driver. This would probably give a delay in the LEDs turning off fully due to energy stored in capacitors in the driver.

Les.

Mosfet i tried but in that also same result...
Thank You.

 

Hi jpanhalt,
I think that you assume that D+ and D- are the outputs from the LED driver to the LEDs. They are in fact the control inputs (Dim+ and Dim -) to the driver. (See the link to the driver datasheet in post #13 There are no page numbers but the block diagram is about a third of the way down on the scroll bar.)

Les.

That is correct. Based on this:


I did not assume D+ and D- were Dim+ and Dim-, respectively.

 

Hi Saiteja,
I don't think there is a device that can be used to directly replace the relay. The only solution I can see to switch the output of the driver would be to us a mosfet in one leg of the output of the driver. This would require a small floating output power supply to provide gate voltage to the mosfet and an optical coupler. This is because we don't know the internal schematic of the driver so we could not safely connect the mosfet source to your arduino ground. An variation on this would be to use a isolated DC to DC converter to drive the mosfet gate. A solution that would probably work would be a solid state relay on the AC input to the LED driver. This would probably give a delay in the LEDs turning off fully due to energy stored in capacitors in the driver.

Les.

Is there any solution by playing with OR,AND,NOR gates etc...

Can we get those IC's with t 48V DC and 6A capacity

 

No. Logic ICs only handle up to 15 volts and a few mA.

Les.

 

No. Logic ICs only handle up to 15 volts and a few mA.

Les.

OK.... Thank You

 

After reading post #18 again I suspect the problem could be caused by stray capacitance together with the output of the driver not going all the way down to zero duty cycle. With only one side of the LEDs connected the unconnected side could have some capacitance to earth producing very short pulses to the LEDs at every transition of the low duty cycle PWM waveform. I think you are going to have to start looking at the problem using an oscilloscope. (Or try the solid state relay on the AC input to the LED driver module.)

Les.

 

How did the DPDT switch solve the problem? Show how it was connected.

Bob

 

Hi Bob,
He uses both poles of the DPDT switch to break the connections between the driver output and the LEDs. See the picture in post #16.
I have seen similar problems when mains voltage filament lamps have been replaced with LED lamps due to the capacitance between the conductors in the cable going to the switch. I have solved this by adding capacitance in parallel with the LED lamps. I use capacitance rather than resistance to avoid the heat that would be generated by the resistor when the lamps were switched on. I actually used contact suppressors such as this as they had flexible leads which made them easier to fit.
Les

 

Hi Bob,
He uses both poles of the DPDT switch to break the connections between the driver output and the LEDs. See the picture in post #16.
I have seen similar problems when mains voltage filament lamps have been replaced with LED lamps due to the capacitance between the conductors in the cable going to the switch. I have solved this by adding capacitance in parallel with the LED lamps. I use capacitance rather than resistance to avoid the heat that would be generated by the resistor when the lamps were switched on. I actually used contact suppressors such as this as they had flexible leads which made them easier to fit.
Les

Capacitance in parallel to both D+ and D- or anyone?. i tried by connecting parallel capacitor to D+ but problem still remains same

 

I was NOT suggesting the capacitor as a solution to your problem as it would not work with the PWM drive to the LEDs. I was just using it to illustrate the effect stray capacitance even at low frequency sinusoidal AC can have. Is there a reason why you MUST solve the problem by creating an electronic version of a double pole relay ?

Les.

 

I was NOT suggesting the capacitor as a solution to your problem as it would not work with the PWM drive to the LEDs. I was just using it to illustrate the effect stray capacitance even at low frequency sinusoidal AC can have. Is there a reason why you MUST solve the problem by creating an electronic version of a double pole relay ?

Les.

Reason is to eliminate the sound while switching

 

But why can't you use a normal solid state relay on the AC input to the LED driver ?
Solid state relays do not make any noise.
Les.

 

But why can't you use a normal solid state relay on the AC input to the LED driver ?
Solid state relays do not make any noise.
Les.

I can do that but here i am not turning off supply every time. sometimes for turning off the light what i will do is i will change PWM to 0 at that time LED is on a little bit

Previously i tried with constant current circuit in that there is no issue like this but 1 or 2 IC failure issue is there. If i choose dimming driver my cost will reduce and solution will be robust.

 

While reading this thread this morning "crutschow" mentioned a device in post #3 that I have never heard of before. I think this device and a mosfet in each leg of the driver output may solve your problem. It's output current is very low so the switching speed will be quite be quite slow due to the time taken to charge the source gate capacitance of the mosfet.

Les.

 

While reading this thread this morning "crutschow" mentioned a device in post #3 that I have never heard of before. I think this device and a mosfet in each leg of the driver output may solve your problem. It's output current is very low so the switching speed will be quite be quite slow due to the time taken to charge the source gate capacitance of the mosfet.

Les.

Thank You. Will check regarding the component

 

Hi Everyone,

With continuation of my testing what i found is When i am keeping the light on the table where oscilloscope, Wattage meter etc... were kept and when i am touching the light LED's are ON if i lift light by holding cable connected to LED it is off what i am feeling is it is because of ESD suppression for eliminating ESD suppression i tried TVS Diode but issue not solved can any one have idea how to eliminate ESD suppression with low cost