LM5141-Q1: Is the low Iq feature available up to 30V input?

Hermann Briechle

Prodigy 55 points

Part Number: LM5141-Q1
Other Parts Discussed in Thread: LM5141

We have the switching regulator LM5141 in a sample structure as a controller:

- U in 18… 30V ( 24V truck electrical system)

- U out 13.2V

- I out 18A

- f 400kHz

The following problem has occurred:

- In no load operation Iq is 25 to 28mA (@ 22… 28V) and 1.7mA (@ 20V)

We have checked:

-Diode emulation mode is active

- no hidden load at the exit or somewhere ese

We have reenacted:

- With U out = 5V (FB on ground) and U in = 12V, Iq is approx. 100uA

- With U out = 5V (FB on ground) and U in> 13.5V, Iq increases significantly

- At f = 2.2 MHz, the result is slightly worse (Iq increases)

Our suspicion:

- The LM5141 can only comply with the low Iq feature with U out 5V or 3.3V and a U in <14V

Conclusion:

- The LM5141 is not suitable for U in> 14V or / and U out not equal to 5V or 3.3V as a low-Iq switching regulator

Did we do something wrong or is our conclusion correct?

over 5 years ago

0 Timothy Hegarty over 5 years ago

TI__Guru 73742 points

Hi Hermann,

Low IQ operation should be available for all input conditions. Please send the schematic and a completed LM5141 quickstart file (available by download from the product folder) for review.

Regards,

Tim

0 Terry Allinder over 5 years ago

TI__Expert 3545 points

Hermann,

Thanks you for your question about the LM5141 Iq. Can you send me a scope waveform of the SW node, and if possible the inductor current. A schematic would also be very helpful.

Terry

0 Timothy Hegarty over 5 years ago

TI__Guru 73742 points

Hi Herman,

Just checking on this - let me know if you need more help. Otherwise, we will close out this thread.

Regards,

Tim

0 Hermann Briechle over 5 years ago in reply to Timothy Hegarty

Prodigy 55 points

Hi Tim,

I am in contact with Terry, sent him schematic and scope captures. At first glance, we didn't do anything wrong. The problem is not solved yet. Should I send that to you too?

Regards,

Hermann

0 Timothy Hegarty over 5 years ago in reply to Hermann Briechle

TI__Guru 73742 points

Hi Hermann,

Sure, you can email me the files. You can also review the operation of the EVM.

Regards,

Tim

0 Hermann Briechle over 5 years ago in reply to Timothy Hegarty

Prodigy 55 points

Hi Tim

I just try to fix how much voltage swing the low Iq state is needing. There's no value mentioned in the data sheet. if it's too low, ringing with higher U in (>18V) may override the low iq function. Are fixed-by-design or proven data values available?

Regards

Hermann

0 Terry Allinder over 5 years ago in reply to Hermann Briechle

TI__Expert 3545 points

Hermann,

Here are the requirements to enter the low Iq state.

•The standby block includes 2 comparators to monitor the FB voltage and COMP voltage which are used to manage the standby mode. The controller enters standby mode when all of the following conditions are met:

•FB>VREF

•COMP<300mV

•No HO pulses for 16 consecutive CLK cycles.

Terry

0 Hermann Briechle over 5 years ago in reply to Terry Allinder

Prodigy 55 points

Hi Terry

If I'm right the three conditions are coming in sequence:

1. no load -> FB>Vref

2. Com falling below 300mV

3. Pulses will not happen as long as FB>Vref

but

what ensures that the stored energy will be high enough to avoid 16 cycles? Only a kind of hysteresis, I assume.

this is the missing data sheet value.

We will try to modify the EVA board for U out = 13,5V and we will see what happens when U in >20V.

Regards

Hermann

0 Timothy Hegarty over 5 years ago in reply to Hermann Briechle

TI__Guru 73742 points

Hi Hermann,

I will follow up with Terry to get the latest on this. Has there been any further development on your side?

Regards,

Tim

0 Timothy Hegarty over 5 years ago in reply to Timothy Hegarty

TI__Guru 73742 points

Hi Hermann,

Note that the current pulse in the DCM condition has a peak value of Ilimit/5 - this is the key to increasing the output voltage and extending the off time in excess of 16 clock periods.

Regards,

Tim

0 Hermann Briechle over 5 years ago in reply to Timothy Hegarty

Prodigy 55 points

Hi Tim

This means that Ilim5 charges the output capacitance via the inductor impdance. So, if the ouput cpactiance is rather high because of low ripple requirement, there might not occur enough "excess voltage" to blank the control circuit for 16 cycles. But, in my experience, this doesn't explain the extreme < >20V behavior: higher voltage in combination with fixed pulse time and fixed inductive ESR and fixed capacitve ESR may result in a higher capacitor voltage level. Only if there is a voltage feed forward parameter shortening the pulse width, this consideration is incomplete.

But we will do tests with the modified EVA board.

Regards

hermann

0 Timothy Hegarty over 5 years ago in reply to Hermann Briechle

TI__Guru 73742 points

Hi Hermann,

Is your schematic available? Also, please send a completed quickstart calculator file.

Regards,

Tim

Power management

Power management forum

LM5141-Q1: Is the low Iq feature available up to 30V input?