LMZM23601: Minimum value of maximum duty cycle

Hi Masashi, 

Could you please help me with the following information for better understanding?

 You are operating nearly at 84.25% duty. May I know for what reason you need the min duty information?

 How are you measuring the input current? I assume this is the average input current. For Vout=8V, Iout=2.5uA, and Vin=9.5V; Iin=4-5mA seems pretty high. Could you please share the waveforms of Iin and Iout as well with Vout for better understanding the issue?

Regards

Arpita

Hello Masashi,

Could you please probe the waveforms during this transition also: Vin 9.6V to 9.5V, Iin 400uA to 4-5mA in oscilloscope and share with us?

Hello Arpita,

We ask our customer to obtain the waveforms (Vout(9.5V/9.6V/9.6V to 9.5V), Iin and Iout ).

After I get these waveforms I will report back to this thread.

Regards,

Masashi

Hello Arpita,

We are currently requesting waveform acquisition, but do you know what value is set for IPEAK-MIN current level in the highlighted sentence below in the data sheet?

Regards,

Masashi

Hi Masashi,

I will check this data internally and get back to you within couple of days. 

Regards

Arpita

Hello Arpita,

Do you have any update?

Now, I am confirming the measurement data I received from the customer.

I will contact you as soon as I have it compiled.

Regards,

Masashi

Hi Masashi,

I have ordered the EVM and I will get back to you once the I get the results/data.

Do you have any update regarding the waveforms during the transition?

Regards

Arpita

Hello Arpita,

I have attached a PowerPoint file summarizing the waveforms for the problematic operation of the LMZM23601.

The customer is using a digital multimeter for the load current because the current is too small to be measured with an oscilloscope.

LMZM23601_waveform.pptx

The LMZM23601 that the customer measured this time works normally up to an input voltage of 9.5V, but when the input voltage drops to 9.3V, the input current does not drop.

There seem to be a variety of cases, including some that occur when the input voltage drops from 9.6V to 9.5V, and others that occur when the input voltage drops from 9.5V to 9.3V.

The results of this measurement are the latter.

Because the LMZM23601 is being used in a battery application, the customer wants it to operate with low current draw without abnormal conditions from input voltages ranging from 9.3V to 16.8V.

If you have any questions or need any information please feel free to contact us.

Regards,

Masashi

Hi Masashi,

I have tested and confirmed in the EVM. Please note that this behavior is expected in low drop-out mode. As Vin reduces nearly to Vout> initially Vout starts falling as the internal Error Amplifier initially tries to maintain previous duty >COMP increases> COMP min threshold is crossed, thus the mode shifts to FPWM, even at very light load. While operating on FPWM mode at low load, the efficiency decreases (Refer to datasheet).  Do you have any further concerns regarding this?

Regards

Arpita

Hello Arpita,

Customers want an operation in which the input current does not increase even when the input voltage decreases.

Is there a way to avoid this issue?

Regards,

Masashi

Hi Masashi,

May I know for how long your input is expected to be around ~9.3V-9.5V (where it shifts mode)? If it is for very less time (few ms), then overall efficiency should not be a concern. What is the nominal input voltage?

Meanwhile, I am checking other variants/buck controller also. I will get back to you in one or two days.

Regards

Arpita

Masashi,

Also please let me know what is your full-load output current? 

Regards

Arpita

Hello Arpita,

Normal input voltage is 12V, 9.5V to 16.8V.

full-load output current is about 30mA to 40mA. typically 10mA.

The customer said that there are individual differences in the input voltage that triggers this FPWM operation, and the customer was concerned that it could occur even at high voltages such as 10V.

So We would like to know the minimum value for the maximum duty cycle of this module.

Regards,

Masashi

Hello Arpita,

May I know for how long your input is expected to be around ~9.3V-9.5V (where it shifts mode)?

The time is short because it is the lower limit voltage of the battery.

Could you please let me know what is the concern with the mode-shifting behavior?

Customers are concerned about the timing of this mode at what input voltage.

Therefore, We would like to know what the minimum maximum duty cycle is expected for the device.

Also, we have heard that the TPSM33625 does not enter this mode.

Regards,

Masashi

Hello Arpita,

I apologize for taking so long.

If it is difficult to identify the threshold at which FPWM operation occurs when the input voltage drops, the customer would like to know if when the LMZM23601 has an input voltage of 10V and an output voltage of 8V, it will remain in PFM operation and will not operate in this type of FPWM mode.

Regards,

Masashi

Hi Masashi,

I will check on the EVM and let you know the result in couple of days.

Thank you

Hello Arpita,

I would like to confirm something from your past questions.

Yor said "May I know for how long your input is expected to be around ~9.3V-9.5V (where it shifts mode)? If it is for very less time (few ms), then overall efficiency should not be a concern. What is the nominal input voltage?"

What I meant in my answer was that as the battery voltage gradually decreases, from 10V to 9.5V to 9.3V, PFM operation cannot be maintained and the controller switches to FPWM, which increases power consumption and shortens the operating time.

Is the intent of your question consistent with my answer?

By any chance you were asking in terms of momentary low voltage of the battery and time to recovery after low voltage?

We only need to know the input voltage at which PFM operation will not change to PWM operation when the output voltage is set to 8V and the load is extremely light, but we would like to know if this problem can be avoided if the input voltage is 10V.

Regards,

Masashi

Hello Masashi, 

I have checked in EVM. I do not observe the mode shifting at Vin=10V, Vout=8V, at very light load in EVM. However, I will confirm with design team for the margins. I will get back to you once I hear from them.

I would like to know "in terms of momentary low voltage of the battery and time to recovery after low voltage"? If the time to recovery after low input voltage is very less/momentary, I don't think the increased power consumption will affect the overall system efficiency. 

Thanks

Arpita

Hello Arpita,

Do you have any update?

Below are the waveforms that were investigated to determine the drop in input voltage in response to your answer.

You can see the difference in operation when Vin=9.5V and when Vin=9.3V.

Since it is difficult to read the current value from an oscilloscope, the values ​​shown are those obtained using a digital multimeter.

Regards,

Masashi

Hi Masashi,

Apologies for the delay. I am yet to get the data. It might take one-two more days.  As I have said previously it is very difficult to calculate and dependent on other conditions.

However, I asked   "in terms of momentary low voltage of the battery and time to recovery after low voltage"?  It would be helpful for me in understanding the concern about efficiency if you can let me know this clearly.

Regards

Arpita

Hello Arpita,

The waveform obtained by adding the input voltage (battery voltage) to the waveform in the previous thread is shown below.

The input voltage and output voltage are set to AC so that voltage fluctuations are easy to understand.

And one division of the input and output voltage is 100mV.

Please contact us if you have any information you would like.

Regards,

Masashi

Masashi,

Thanks for the complete waveform. Can you please share your schematic for double-check? May I know what are the input capacitors you have used? What is the slew rate of load transient?

Regards

Arpita

Hi Arpita,

Our customers provided us with information.

1.

Can you please share your schematic for double-check? May I know what are the input capacitors you have used?

[Ans]

Input capasitor is 10uF/50V ceramic capasitor.

2.

What is the slew rate of load transient?

[Ans]

Channel 4 (Red) is an enlarged waveform of the input current.

The time axis is 1ms/div, and the current value is 10mA/div.

Input current slew rate is about 35mA - 15mA / 0.5ms = 40mA / ms.

I request the customer to confirm the output current transient.

Regards,

Masashi

Hi Masashi,

Thanks for the information. The schematic looks good. As I have told previously, it is very difficult to calculate the entry point (where mode shifts) of vin exactly, and the behavior of mode changing is quite expected behavior, let me check internally if I can get the data (if any).

Regards

Arpita

Hello Arpita,

The load change transient response waveform is attached below.

Channel 4 (Red) is an enlarged waveform of the output current.

The time axis is 200us/div, and the current value is 100mA/div.

Input current slew rate is about 400mA / 160us = 2.5mA / us.

Regards,

Masashi

Masashi,

Your previous waveforms and the latest waveforms looks like they have taken in different instants/test conditions.

 Please get back to us once you have vin,vout, iout, Iin captured in single waveform , as I have asked in my previous replies. It will be helpful to debug further and rule out any other issue.

Regards

Arpita

Hello Arpita,

Until now, I had only been observing the current on the input side, but when I maximized the waveform of the load fluctuation on the output side (Iout), I found that it was the waveform I have attached.

Because it is a fast, instantaneous fluctuation, it may not be possible to observe it on the input side.

Regards,

Masashi

Hello Arpita,

When you say slew rate, what timing load change slew rate do you mean?

(a) Load change during normal operation

(b) Timing for driving the relay before transitioning to standby mode

The waveform we are sending is a load fluctuation waveform under normal operation (a).

The customer's normal operation is approximately 0 to 50mA load variation.

In other words, the red frame in the sent waveform corresponds.

The slew rate is 50mA/200us = 250uA/us.

Does this mean that the output current fluctuation waveform under condition (b) is absolutely necessary?

Similarly, regarding the single waveform you mention, which timing waveform do you need?

Regards,

Masashi

Masashi,

As the concerned is regarding the instant (b), It is better if you can share waveforms of Vin,In,Iout,Vout captured during the instant (b) in single waveform at 1) Vin=9.5V 2) Vin=9.3V. Iout waveform during instant (b) is necessary to check what is the exact Iout during the instant (b) when Vin is 9.5V and Vin =9.3V

Hello Arpita,

Please let us confirm your request.

Since you wish to obtain individual waveforms, does that mean you need a waveform like (B) instead of (A) as before?

Regards,

Masashi

Masashi,

Please try to capture like below and share if you can get the waveform like this. However, I will get back to you once I hear from design team internally. 

Hello Arpita,

If there were two current probes, it might be possible to fit the waveform on one sheet, but since there is probably only one current probe, it may be two sheets, Iin and Iout.

I will ask the customer to retrieve these waveform.

Regards

Masashi

Hello Arpita,

Because there is only one current probe, there will be two waveform images, Iin and Iout.

(A-1) Vin = 9.5V (Yellow), Vout (Blue), Iin (Red)

(A-2) Vin = 9.5V,(Yellow), Vout (Blue), Iout (Red)

(B-1) Vin = 9.3V (Yellow), Vout (Blue), Iin (Red)

(B-2) Vin = 9.3V (Yellow), Vout (Blue), Iout (Red)

We feel that there are no major differences in the above waveforms.

The waveform shown below is an enlarged version of the light blue part of the B-2 waveform.

(B-2-1) Vin = 9.3V (Yellow), Vout (Blue), Iout (Red)

(B-2-1-1) A waveform that is further expanded from B-2-1, The current division is 50mA/div, which is very different from the others.

(B-2-2) Vin = 9.3V (Yellow), Vout (Blue), Iout (Red)

(B-2-3) Vin = 9.3V (Yellow), Vout (Blue), Iout (Red)

Please understand that due to the resolution of the current probe, it may be difficult to observe currents on the order of microamperes.

When the input voltage is 9.3V, the waveform appears to show an improved transient response, but please note that this is FPWM rather than PFM.

When the input voltage is 10V and the output voltage is 8V, are there any indicators that the current will not switch to FPWM operation if the fluctuation from XXmA to YYm A is less than ZZ mA/us?

Regards,

Masashi

Hi Masashi,

Apologies for delayed response as I was OOO. I will check and get back to you by tomorrow IST.

Hello Arpita,

Our customer has already completed the design and is therefore unable to change the module.

So We just want to know what input voltage the LTZM23601 can maintain PFM at when its output voltage is 8V.

As far as we can tell from checking with a digital multimeter, when the standby (idle load) is around 2.5uA, the input current is around 400uA when the input voltage is 9.5V, but we know that the input voltage is around 4.5mA when the input voltage is 9.3V.

Is it possible to tell us whether PFM can be maintained under these conditions when the input is 10V, regardless of load fluctuations?

Regards,

Masashi

Hello Arpita,

I would like to thank you for your long-term support regarding this issue.

We will close this case for now.

Regards,

Masashi