TPS61175: TPS61175

Shafeeq Muhammed

Part Number: TPS61175

Hi,

I am using tps61175 in my design. I am using this regulator for 5V to 12V conversion. I am using this regulator to drive an LED driver circuit for an LCD backlight.
The LCD backlight will be controlled by a PWM signal. So the load of the regulator will be continuously on and off.
I am getting an output ripple voltage around 1.5V when I am using PWM.
And I am getting 370mV for a continuous load.
Is there any way to reduce the ripple voltage on output side as I have to test the system for RE and CE.

Regards,
Shafeeq

over 4 years ago

0 Helen chen over 4 years ago

TI__Mastermind 40080 points

Dear Sir,

What is your maximum LED current and the minimum LED current during the PWM dimming. Try to add more caps at the output side to check whether it could be help. Please start from a 47uF cap.

0 Shafeeq Muhammed over 4 years ago in reply to Helen chen

Expert 1170 points

Hi Helen,

Thanks for your suggestion. I have added a 47uF capacitor instead of 22uF. And I am able to clip the positive spike when PWM turns ON. I have attached the CRO captures for 22uF + 1uF output capacitance and 47uF + 1uF for your reference. Is there any suggestion that you can provide to clip the negative spike also?

The maximum and minimum current while PWM dimming will be 600mA and 100mA respectively.

Also, I have checked with 100uF capacitor on the output side. But the regulator didn't work with 100uF capacitor on the output.
I am giving 5V input to the regulator. I am getting the same 5V on output with the 100uF capacitor.
Is this expected behaviour?
What is the maximum output capacitance supported for TPS61175?

Thanks & Regards,
Shafeeq

0 Helen chen over 4 years ago in reply to Shafeeq Muhammed

TI__Mastermind 40080 points

Dear Sir,

What kind of the 100uF cap you used? Is it an EL cap? What is the ESR of this cap?

0 Shafeeq Muhammed over 4 years ago in reply to Helen chen

Expert 1170 points

Hi helen,

I have used a 100uF ceramic capacitor.
Attached the ESR characteristics of the cap for your reference.

Regards<
Shafeeq

0 Helen chen over 4 years ago in reply to Shafeeq Muhammed

TI__Mastermind 40080 points

Dear Sir,

How about the layout file? did you put the cap far away from the IC? Could you please post the pcb photo.

From your second clean waveform. It is normal, because the PWM dimming similar to the load dynamic. When the load current suddenly gets high, output voltage drops quickly because the boost convert needs some time(30-50us) to response to the load change.

But the 100uF ceramic cap doesn't work really strange.

0 Shafeeq Muhammed over 4 years ago in reply to Helen chen

Expert 1170 points

Hi Helen,

I have attached the boost regulator section PCB photo for your reference.
C30 and C32 are the output capacitor. It is not too far from the IC.

Is there any technique we can implement to reduce the negative spike while load current gets high.

Regards,
Shafeeq

0 Helen chen over 4 years ago in reply to Shafeeq Muhammed

TI__Mastermind 40080 points

Try to increase the R35 to 10k to check whether it could be helpful.

To reduce the negative spike, one method is to increase R35, another method is to increase the output capacitance.

0 Shafeeq Muhammed over 4 years ago in reply to Helen chen

Expert 1170 points

Hi Helen,

I have tested R35 as 10K.
But the ripple is increased instead of decreasing.

Is there any other method which we can implement?

Regards,
Shafeeq

0 Helen chen over 4 years ago in reply to Shafeeq Muhammed

TI__Mastermind 40080 points

No, the only way is to increase the output cap. But increase the output cap doesn't work, which is very strange. Another method is to change your design, dimming from light load to the full load, instead of no load to full load.

0 Shafeeq Muhammed over 4 years ago in reply to Helen chen

Expert 1170 points

Hi Helen,

Could you please explain how to do dimming from light load to full load?

Thanks & Regards,
Shafeeq

0 Helen chen over 4 years ago in reply to Shafeeq Muhammed

TI__Mastermind 40080 points

light load means the minimum LED current is not too low!

0 Shafeeq Muhammed over 4 years ago in reply to Helen chen

Expert 1170 points

Hi Helen,

Could you please explain how to do that?

Thanks & Regards,
Shafeeq

0 Helen chen over 4 years ago in reply to Shafeeq Muhammed

TI__Mastermind 40080 points

Hi Shafeeq,

Actually it is hard to change, it is based on your spec. 200mA-600mA load step is better than 100-600mA load step.

0 Shafeeq Muhammed over 4 years ago in reply to Helen chen

Expert 1170 points

Hi Helen,

Currently, I have set the frequency as 1MHz.
Whether changing the frequency will impact on the ripple voltage?

Thanks & Regards,
Shafeeq

0 Helen chen over 4 years ago in reply to Shafeeq Muhammed

TI__Mastermind 40080 points

Hi Shafeeq,

I'm sorry to tell you that operation frequency doesn't help with the dynamic ripple. load dynamic ripple only relates with the output capacitance and the compensation network.

0 Shafeeq Muhammed over 4 years ago in reply to Helen chen

Expert 1170 points

Hi Helen,

Now I have tested the compensation networks as 4.42K and 15nF and 10K and 15nF.
Are there any other values I can experiment?

Regards,
Shafeeq

0 Helen chen over 4 years ago in reply to Shafeeq Muhammed

TI__Mastermind 40080 points

You can keep 4.42K and 15nF, then what is the Cout you have tried before?

0 Shafeeq Muhammed over 4 years ago in reply to Helen chen

Expert 1170 points

Hi Helen,

As I informed earlier, when I tried with 100uF, the regulator didn't work.
When I tried to debug it, the regulator was gone bad. I couldn't make out how trhe regulator gone bad.

Any way, I decided to test in one more borad with 100uF and I have tested.
And the ripple reduced to 388mV. And this was a huge reduction for me.
And also I have expirimented with different PWM frequencies.
(The PWM is used to vary the brightness of LED which will decide the transient load characteristics).
And I have got the ripple as 164mV.

Further I have tried with one more 100uF in parallel. (total o/p cap= 100uF*2 and 1uF).
And with this o/p capacitance and the previous PWM frequency the ripple is reduced to 76mV.

I think I cannot reduce the ripple further.

Thanks for all your support.
And also let me know if any countermeasures are there to reduce the ripple further.

Thanks & Regards,
Shafeeq

0 Shafeeq Muhammed over 4 years ago in reply to Shafeeq Muhammed

Expert 1170 points

Hi Helen,

As I informed in the previous post, I have tried with 2 nos of 100uF capacitance on the output of the regulator.
And I am using 32.5KHz PWM to drive my LED driver.

In the above condition, I am getting the least ripple.

Can I consider the above configuration as legal?
I am asking about the reliability of the design
For long term usage, any issue will be there?

I am asking because of the following reasons.
• Too much capacitance on the output of the regulator.
• We will be using PWM with a 50% duty cycle in normal operation. It may vary in some condition.
But anyhow the load will be turned On and off continuously in a frequency of 32.5KHz.

I want to confirm whether the above conditions will affect the reliability of the design in long term use or not.

Thanks & Regards
Shafeeq

0 Helen chen over 4 years ago in reply to Shafeeq Muhammed

TI__Mastermind 40080 points

Dear Shafeeq,

I'm very glad to know that finally the TPS61175 could work well with 200uF output cap. With 200uF output cap, you can actually change the Rcomp to 10-15k and Ccomp=10nF.

One fact for the load transient is that: the output voltage drop during the the load step from the light load to heavy load is unavoidable, because the boost converter need some time to response. In my experience, <300mV drop during the load transient is already a very good performance.

What are you worry about for the voltage drop during the load transient? Is a 200-300mV drop effect the performance?

For your current setting, it is OK from the reliability point of view, the only concern is you need to check the current capability of the TPS61175 main inductor. Becuse with big Cout, the inrush current during turn on will gets higher.

0 Shafeeq Muhammed over 4 years ago in reply to Helen chen

Expert 1170 points

Hi Helen,

The inductor part we have used is XAL5050-682ME.
the saturation current for the inductor is 6A.

Please let me know whether this will be ok or not?

Regards,
Shafeeq

0 Helen chen over 4 years ago in reply to Shafeeq Muhammed

TI__Mastermind 40080 points

Dear Shafeeq,

6A rating inductor is OK.

Power management

Power management forum

TPS61175: TPS61175