Group Actions
Engagement
  • Author Author: dubbie
  • Date Created: Date Created
  • Views 426 views
  • Likes 7 likes
  • Comments 2 comments
Related
Recommended

Bluetooth Controller for LEDs #2 : The Output

dubbie

I have managed to connect up the output of the LED controller to my oscilloscope. This is the non-Bluetooth version of the controller because it is still fully shrouded so it is much safer, but it should be the same as the same LEDs are plugged in. I was expecting to see a 50 Hz sine wave of some description, but instead, I discovered a 66V peak to peak square wave at 25 Hz, see photograph below. Apologies for the fuzziness of the photograph, I have not yet worked out how to transfer my oscilloscope screen directly to my laptop - I'm sure it can be done.

 

image

 

 

So my sssumption about the output from the transformer being passed directly to the LEDs is obviously not correct. The transformer output must be being converted to a DC voltage and then switched. On a closer inspection of the Bluetooth PCB there is a square arrangement of transistors in one corner which is typical of a full H-bridge implementation. This makes sense as the LEDs would then be the load on the H bridge, with half of them facing one way and half the other way. Then the H bridge can be switched by turning on only two of the transistors at a time (diagonally opposite) to control the current direction and flow through the load (the LEDs). It also enables any frequency of switching to be used, hence the 25 Hz rather than the 50 Hz expected.

 

The video below shows the waveform when the controller is set to slowly fade on and off both halves of the LEDs. The square wave can be seen becoming longer and then shorter as the LEDs fade on and then off.

 

 

I'll have a look at the Bluetooth part next.

 

Dubbie

  • in reply to genebren

    Gene,

     

    You are correct, although it wasn't a typo error as such. I remember doing the calculation that 1000 ms divided by 4 ms didn't seem quite right at 25. I used my talking calculator to do the calculation (I know I should be able to do it in my head but I was just checking) and the segments are beginning to disappear so it is difficult to see exactly what the result is. I must have misplaced a decimal point somewhere. I can also remember at the time thinking that 25 Hz was a peculiarly low frequency for this type of activity. I also tried shaking the LEDs to see if the switching became visible - which it didn't, although if it had been 25 Hz it should have been visible. Obviously as it is 250 Hz that explains many things.

     

    Dubbie

  • That is looking like an even better deal than before.  I think you have a typo in regard to the switching frequency of the output (25Hz should be ~250Hz).  I like that they are using a slightly higher frequency, likely getting a lot less flickering in the LED at that rate.

     

    Keep up the good work!

     

    Gene