TPS54618: TPS54618CQRTERQ1 for Vout issue

Hi user,

Thanks for posting your schematic and some waveforms.  For automotive applications, I would strongly recommend to use the newer, smaller, wettable flanks, higher frequency TPS62810-Q1 or TPS62816-Q1.

It looks like Vout starts rising before EN goes high!  Is there anything connected to the output, such as an electronic load?  Are there other rails in the system which could be back-feeding onto this rail?

Thanks,

Chris

Hi Chris,

thanks for your reply.

Due to resolving EMI issue, we adjust the switching freq.following WEBENCH Power Designer from 400kHz to 1.974MHz. However, we found that Vout at 400kHz is normal, but Vout at 1.974MHz is abnormal. Also, we looked the datasheet for TPS54618CQRTERQ1 and we think ON-time is too long, 75ns, which can only set at 1.3MHz. Do you think the switching frequency can operate at around 2MHz? or the datasheet has some wrong, hasn't it?

Hi User,

The TPS54618 can operate at up to 2 MHz for higher Vouts or lower Vins, where the minimum on time is met.  You need to meet both conditions: maximum (and minimum) frequency and minimum on time.

The TPS628xx-Q1 devices are designed to operate at these higher frequencies and have shorter minimum on times.

Chris

HI Chris,

Thanks for your reply.

We would like to know whether TPS54618CQRTERQ1 can operate these conditions, please see the table as below.

As we mentioned 5V to 1V@1A at 1.97MHz, the circuits for Vout rising is abnormal. (no electronic load, no back-feeding) . Also, 5V to 1.2V @0.5A at 1.97MHz has the similar problem, abnormal rising. However, we didn't see abnormal rising operating around 400KHz.

We have used TPS54618CQRTERQ1 on our products, so we would like to know any potential risks for telling our customers.

Can these designs based on the table cause any risks?

Many thanks!

Sincerely,

Woody

Hi Woody,

Do you have the Vout, Vin, and EN waveform for the 400 kHz circuit that did not show the abnormal Vout rising?

Do your previous TPS54618 designs use such a high switching frequency or are they all below 1 MHz?

Yes, both the 1V and 1.2V rails require on times below or in the range of the typical value of the minimum on time shown in the D/S.  This should be avoided.

Can you retest the startup waveform with a lower Vin, such as 3.3V?  This would create a longer on time and eliminate minimum on time concerns.

Thanks,

Chris

Hi Chris,

Thanks for your reply.

Actually, we have used TPS54618CQRTERQ1 @ around 2MHz in our products (as shown in the table below) , so we would like to know any potential risks for telling our customers. 

Can you explain any potential risks about TPS54618CQRTERQ1 @ around 2MHz (as shown in the table below) ?

Many thanks!

Sincerely,

Woody

Hi Woody,

Your design clearly violates the minimum on time.  Please read the D/S.

You need to check your system to see how it responds to this.  The IC might be unable to regulate the output voltage to the proper low level, may have startup oddities (as you show in this thread), or may not operate at such a high frequency.

Chris