LP5562: Large current noise

Hi Mrianas,

Current accuracy is average current value, not noise.

Could send me the current waveform and let me check your current noise?

Thanks!

Summer

Hi Summer,

please see attached "noise" scope traces at 1 mA, 10 mA and 60 mA. These were recorded by directly detecting LED light with an optical sensor. As you can see the noise is very high (~20%) when the current is 1 mA and then progressively decreases at larger current values. Note, that the LED driver is pulsed using a high side switch at 200 Hz as the attached schematics shows. Any advice on how to reduce the noise at low current values would be appreciated. We have tried to connect 1 uF capacitor in parallel with the LED which reduces the noise by x2 but we cannot increase it any further due to distortions of the pulse shape.

Thank you. Mirianas

Hi Mirianas,

This overshot comes from your external PWM dimming circuit, May I konw the reason why you didn't use the LP5562 intrenal dimming?

Thanks!

Hi Summer,

the overshot is not a problem to us as its contribution can be completely eliminated by reading light intensity with a small delay. The problem is the current noise after the overshot which in turn makes LED light noisy. We are looking on how to reduce the noise after the overshot so that the current profile looks flat (like at large current values).

We used external switch to allow for more flexible design so we can used something else instead of LP5562 .

Mirianas

Hi Miranas,
If LP5562 was used the typtical circiut, there should be no so larger noise in the output current.
Would you mind to try powet the LED with the DC supply and use the intrenal PWM dimming?
Thanks!
Summet

Hi Summer,
can you please clarify what do you mean by PWM dimming? Would we be able to have the pulses being synchronized to the external clock?

Hi Mirianas,

LP5562 has internal oscillator, just write internal PWM registor can generate the PWM dimming duty, but it can't synchronized to the external clock.

If you have to sychronized with the extrenal clock, the external circuit is necessary or you need to change solution with simple PWM dimming.

Let me know what is your thinking.

Thanks!

Summer

Hi Summer,

we do need to synchronize with an external clock, that's why we have the "high side switch" (I provided the circuit diagram in my earlier post). We originally decided to use LP5562 because it provided single-package solution for variable current source. However, if it's current noise is so large, then we cannot use. Is there anything else we could to do bring that noise down?

Mirianas 

Hi Mirianas,

I got from schmatic, you only need to drive one LED, may I know max current of LED.

I think LP5524 will be easy for your application.Could you check the spec in ti.com?

Any more support please let me know, thanks!

Summer

Hi Summary,
the problem is we need to have a current control via I2C which LP5524 does not have. Any other suggestions?

Hi Mirianas,
May I understand your application more?
The synchronize is for pwm frequecy or for pwm duty?
If you can accept the PWM duty is same but the PWM frequency, I would like to suggest LP5569.
Let me your thinking.
Thanks!
Summer

Hi Summer,

this is used for optical detection device. We have a FPGA that controls LED pulsing and performs ADC reads at specific time points when LED is ON and OFF.  In addition we need to be able to have software control of the magnitude of the LED current during ON cycles i.e. that's why we originally chose LP5562 since it has IC2-controlled current. Would LP5569 be suitable replacement for this application with less noisy current?