TPS40210-Q1: TPS40210-Q1 extended discussion

Hi Stefan

Now the situation is...

Thr original design uses GMT G5322F61U PWMIC，and soft start cap use 200nF . It's work .
But I replaced GMT PWMIC with TI TPS40210,I've found that lighting is getting worse.

Oirignal IC(G5322)
→ Css= 200nf ， 80V/150mA can lighting up sucess

New TI TPS40210
→ Css= 200nf/100nf.. all fail
           = <15 pf... more better ，1pF it can lighting up 70V/150mA
              but... starting current increasing(side effect)

why ?
Can yout help me how to improve?

Thankyou

(I measure some waveforms , and I'll ask Boyan to attach them for me later)

Ming

20230825 Led Driver measure record.pptx

Update PPT

Thanks,

Boyan

Hi Ming,

when checking on page 2 the left lower scope plot:

Vout goes up to the target output voltage but without GDRV switching. For a boost the output voltage will be the same as the input voltage when the MOSFET is not switching.

So something is wrong with this measurement - can you please check again.

If you use a 15pF SS cap you have almost no softstart, so the circuit will properly detect overcurrent due to the charging current of the output cap.

In this case it will detect and error and restart - at that time you should see that SS pin going low again which is not the case in your plot - or can not be seen in the plot.

Best regards,

 Stefan

Hi Stefan

I re-measure the scope plot , Yes ,it does have the GDVR(left lower scope plot) . so I've updated PTT

I know if I use 15pF , I have almost no softstart function .  

But the question is.... how do I use 100nF cap? , the lightbar can't  light  up

Hi Boyen

   I just sent you the PTT , Pls attach it to this message . thanks

Ming

Hi Stefan

The PTT is updated here.

20230907 Led Driver measure record.pptx

Thanks,

Boyan

Hello Ming / Boyan,

Thanks for your feedback.

Stefan is already in the weekend.
He will look into it on Monday.

Best regards
Harry

Hello Ming / Boyan,

can you check the FB pin as well - if you are doing a constant current regulation with having the FB controlled by the current through the LEDs, this current will define the regulation and output voltage. If now the current is already reached at lower voltages this will not allow the controller to regulate further up.

Esp. something seems to happen ~ 120ms after power on which triggers the controller to further increase the output voltage. 

As the softstart time is already passed by - this will then trigger an overcurrent as beside the load current also the output capacitor charge current needs to be delivered by the power stage.

Note: it is not recommended to have two inductors in parallel as due the the variance of the two inductor the current can be distributed unbalanced.

Best regards,

 Stefan