LM3550: Slow charge and weak drive

Fred P

Part Number: LM3550
Other Parts Discussed in Thread: CSD17310Q5A

I have a circuit board with a LM3550 that has been validated and working good.

Now I have another board that does not work so well but has the same circuit. I wonder if a wrong part has been installed or the IC itself is damaged.

The supercap charges ok up to ~4.5V then takes forever to go up to 5.3V as configured in the registers. We are talking several minutes long. Input current is barely nothing during this abnormal long charge.

Also, the FET_CON drives FET only at 2V while it normally goes up to 5V (Vin=5V).

If I disconnect the FET, it does go up to 5V. FET is a CSD17310Q5A.

I did not try replacing the IC since a brand new damaged IC very rarely happens.

Can this be related to a wrong charge pump capacitor? How can I check for that?

Any other idea?

Regards

over 3 years ago

0 Frank Qiu over 3 years ago

TI__Genius 15776 points

Please replace the good board device to the issue board to check whether it's a issue about device first,

0 Fred P over 3 years ago in reply to Frank Qiu

Expert 2100 points

Well, the working board is not available right now (and for a while...).

I'd rather have an idea of what the charge pump waveform should be. That would be very helpful in the current situation.

Is that possible?

0 Frank Qiu over 3 years ago in reply to Fred P

TI__Genius 15776 points

Hi Fred, you can refer the wave of Vout on datasheet.

0 Fred P over 3 years ago in reply to Frank Qiu

Expert 2100 points

Vout is not behaving as it should and I already know that. That IS my problem.

I need help about where to look to understand why Vout is wrong.

Regards

0 Frank Qiu over 3 years ago in reply to Fred P

TI__Genius 15776 points

Sorry, i can't help you only on your description. Could you offer some key waveform.

0 Fred P over 3 years ago in reply to Frank Qiu

Expert 2100 points

Just tested another freshly baked board.

It appears to have the same wrong behavior. One part may be wrong in the assembly...

I'll get you few waveforms sometimes next week.

0 Fred P over 3 years ago in reply to Fred P

Expert 2100 points

Recently added ferrite beads (MPZ2012S601AT000) on both VOUT and BAL are to blame for unexpected charging behavior. It comes as a surprise. Whatever...

As for the wrong FET drive level, it is because of a short-circuit in the LEDs assembly. LED current is regulated properly and not much Vgs is required to get current flowing into a short...

Case closed.

Power management

Power management forum

LM3550: Slow charge and weak drive