LMZM33606EVM: About Synchronization operation

Hi Yusuke-san,

1. Can you provide a probe of the external sync frequency and SW node on the EVM before and after synchronization. I see that you are providing a 5V high/ 0V low which is following the datasheet recommendation. Is the duty cycle between 10% and 90%? I am guessing that you have the RT resistors set to 400kHz frequency and the external sync frequency set to 350kHz? For 24Vin/5Vout, switching from RT mode of 400kHz to SYNC mode of 350kHz should not have any issues.

2. What is your load current? Depending on the load current, you may be in the transition phase between DCM and CCM which may be a bit jittery. You should not be seeing much jitter at around 1A and onwards.

Regards,

Jimmy

Hi Yusuke-san,

Do you have any updates on this? The EVM should be able to externally sync with a function generator and switch between RT mode and SYNC mode seamlessly with minimal propagation delay.

Regards,

Jimmy

Jimmy-san ,

Thank you for your kind support and message.
I haven't received a response from them yet.

By the way, have you confirmed this problem on EVM?

Jimmy-san,

Thank you for your kind support.
Customer again performed a synchronous operation test using EVM.

Test condition is:
Vin=24V、Vout=5V、"RT Mode = 600kHz → Sync Mode =400kHz"、Iout=No load

LMZM33606 external synchronization operation was successful.

However,There are cases with wide switching duty and narrow cases.
A wide duty and a narrow duty are repeated alternately.

Customer wants a fixed Duty behavior.
(Duty changes may appear in the image as noise.)

This phenomenon disappears when external synchronization is set to 420 kHz or higher or the input voltage is lowered.

What is the cause of this phenomenon?

LMZM33606 waveform.pdf

Best Regards,
Yusuke

Hi Yusuke-san,

I will have the check this with an EVM in the lab. Let me gather some data and get back to you by tomorrow.

Regards,
Jimmy

Jimmy-san,

Thank you for your response.
I'm looking forward to hearing from you.

Best Regards,
Yusuke

Hi Yusuke-san,

I took the LMZM33606EVM to the lab and tested for the following:

• VIN = 24V ; VOUT = 5V; IOUT = No Load ; RT Mode = 600kHz ; External SYNC = 400kHz, 350kHz, 315kHz

The EVM had J6 open and an external functional generator was connected to SYNC and GND. I did not see any intermittent pulse during the SYNC process. Attached is a zipped folder with the different test conditions: 2548.Orbison_24Vin_5Vout_SYNC_NoLoad.zip

Only when I went below the recommended SYNC frequency of 350kHz did the power module have issues with proper SW operation. The SW node started oscillating around 320kHz and below. The waveform of 315kHz was captured to show the intermittent failure seen by the narrow pulses.

From the waveform, CH 2 (SYNC) looks like it is roughly 1.25 divs which corresponds with 400kHz. Can you simply confirm this was the case from the customer?

I will need to talk to the design engineer to discuss this discrepancy.

Regards,

Jimmy

Jimmy-san,

Thank your for your strong support.
About Synchronization operation
The reason why the sync frequency did not match the setting was due to the SYNC voltage.
As a result of changing the SYNC voltage to 3.3V, it worked normally.

However, another problem has occurred.
Could you check the attached file? 
Let me ask another question.

LMZM33603 LMZM33606.pdf

Synchronous clock varies from 300kHz to 400kHz because of the trigger timing reset.
Synchronization clock and the switching frequency are shifted.
In the worst case, it takes 10 clock periods to synchronize.
This problem is happening in both LMZM33606 and LMZM33603.

Could you give me any advice on this cause and how to fix it?

Best Regards,
Yusuke

Jimmy-san,

Thank your for your strong support.
About Synchronization operation
The reason why the sync frequency did not match the setting was due to the SYNC voltage.
As a result of changing the SYNC voltage to 3.3V, it worked normally.

However, another problem has occurred.
Could you check the attached file? 
Let me ask another question.

1122.LMZM33603 LMZM33606.pdf

Synchronous clock varies from 300kHz to 400kHz because of the trigger timing reset.
Synchronization clock and the switching frequency are shifted.
In the worst case, it takes 10 clock periods to synchronize.
This problem is happening in both LMZM33606 and LMZM33603.

Could you give me any advice on this cause and how to fix it?

Best Regards,
Yusuke

Jimmy-san,

Thank you for your strong support.
I'm looking forward to hearing from you.

Best Regards,
Yusuke

Yusuke-san,

1. I'm not familiar with the trigger timing reset. Can you explain what that is and how you got CH 3 waveform? For my setup and waveforms that I've previously sent, my oscilloscope was always triggered off the rising edge of CH3 (SW node). 
2. Did you get a chance to remove the polymer aluminum capacitor (C13) on the LMZM33606EVM and retest? Ideally you would want the external synchronization clock to be fixed at your desired switching frequency so that the device can be externally synchronized according to the clock.
3. What would the waveform look like if you removed the C13 capacitor on LMZM33606EVM and have a minimum synchronization frequency of 350kHz?
4. In that same waveform where you saw the shift in channel 2(SW) can you zoom out? I'd like to see how long the synchronization was successful and how long it took for the SW to pop out of synchronization.

Regards,

Jimmy

Jimmy-san,

Thank you for your response and support.

"I'm not familiar with the trigger timing reset. Can you explain what that is and how you got CH 3 waveform? For my setup and waveforms that I've previously sent, my oscilloscope was always triggered off the rising edge of CH3 (SW node). "

→I apologize my explanation is insufficient.
　“Trigger timing reset” is a function on the customer application.
　 Depending on the application specifications, the clock frequency may change momentarily.
　 If the sync clock goes back and forth between 300kHz and 400kHz, the reported problem will occur.
　 If the clock fluctuates, will there be a loss of synchronization as sent?
　 Please advise about the speed of LMZM following clock fluctuations?

"Did you get a chance to remove the polymer aluminum capacitor (C13) on the LMZM33606EVM and retest? Ideally you would want the external synchronization clock to be fixed at your desired switching frequency so that the device can be externally synchronized according to the clock.
What would the waveform look like if you removed the C13 capacitor on LMZM33606EVM and have a minimum synchronization frequency of 350kHz?"

→I will try that.

”In that same waveform where you saw the shift in channel 2(SW) can you zoom out? I'd like to see how long the synchronization was successful and how long it took for the SW to pop out of synchronization.”

→I will contact the customer with the waveform acquisition request.

Best Regards,
Yusuke