TPS568230EVM: Spur detected on device upon cold-crank but recovers upon increased load with device under FCCM mode

Leo Chow

Part Number: TPS568230EVM
Other Parts Discussed in Thread: TPS568230

Hi,

We have some questions with regards to the evaluation of TPS568230 while evaluating the TPS568230EVM.

As of now we have TPS568230 operate in FCCM mode with Vin 7.5V & Vout 3.6V tested on EVM & we've observed the following outcomes. Hence will need advise on

end equipment started with non-transmit mode @ Iout ~200mA & spur was detected
end equipment begin to transmit with output load @ Iout ~1A, device operate without spur. So we would like to know if this is normal behavior of this device?
additionally are there any conditions required to meet so that such occurrence can be avoided?
we also bbserved ~1MHz of switching frequency @ Vin but output is normal, what may be the potential root cause?

Additionally are there any app-notes with regards to DCAP3 technology which we can read on? Thanks.

Regards,

Leo

over 5 years ago

0 Yuchang Zhang over 5 years ago

TI__Mastermind 18510 points

What do you mean about “Spur” and “non transmit”? Do you mean DCM status? Please upload your waveform if you have.

Yuchang

0 Wai Mun Lee over 5 years ago in reply to Yuchang Zhang

Prodigy 40 points

Non transmit- u can consider as idle with 0.1A at reg output

Question: If we have set it to FCCM, will the regulator goes into DCM at light load after start up? I assume if we set to FCCM, it will run in constant PWM?

0 Yuchang Zhang over 5 years ago in reply to Wai Mun Lee

TI__Mastermind 18510 points

Hi, Yes, if FCCM is selected in mode pin, Forced constant PWM will be asserted.

Yuchang

0 Wai Mun Lee over 5 years ago in reply to Yuchang Zhang

Prodigy 40 points

One question, if we set to FCCM mode, it will not enter into eco mode so the switching freq will be per what we have set like 1MHz?

In datasheet section 7.3.1, it states that there is on time control and pseudo frequency.Would u be able to explain what is this pseudo freq?

0 Yuchang Zhang over 5 years ago in reply to Wai Mun Lee

TI__Mastermind 18510 points

Yes, if set the mode 1MHz FCCM, IC will work at fixed 1Mhz at steady state.

As for pseudo frequency, firstly in Dcap family mode there is no internal oscillator for PWM generation like VCM/PCM mode, the Ton timer is made up with a constant current source(related to Vin...) and an internal cap, Ton is fixed when the internal cap is charged to Vout, and from Ton=D * T to get Ton=D*1/Fsw, here we can see Fsw is related to many parameters, internal cap and current source, Vin, Vout.., so generally Fsw has ~10% variation and even around 20% for some part is normal, this is why we called it pseudo-fixed frequency on time control.

Yuchang

0 Wai Mun Lee over 5 years ago in reply to Yuchang Zhang

Prodigy 40 points

Using the EVB as example, can we estimate the range of Ton and Fsw? I found this in one of the DCAP doc and the freq could rely on the output cap ESR and C. Wonder if this is related.

There are in general 2 problems.

1. The regulator seems to have low freq ~15kHz before we load it to ~2A. After we have done that, the regulator input/output no longer has this 15kHz frequency.

2. When the regulator is running at 1MHz freq at the output, we noticed this 1MHz can be seen at the input of the regulator (Vin). Have u seen this before?

https://e2e.ti.com/blogs_/b/powerhouse/archive/2015/05/07/dcap3-a-sequel-better-than-the-original

0 Yuchang Zhang over 5 years ago in reply to Wai Mun Lee

TI__Mastermind 18510 points

Generally you can use Fsw to calculate Ton with Ton=D/Fsw. Seems you misunderstood the description in the article, ESR and Cout will generate a zero which is used to compensate the Output LC double pole, in Dcap mode it needs ESR compensation, rather than using it for Calculating Fsw.

For question 1, Do you mean IC works into eco mode even the mode is set FCCM? Check the mode voltage to make sure the setting is correct.

For question 2, Fsw is 1Mhz, It is normal Vin ripple is 1Mhz, not fully understand what are worrying about..

Yuchang

0 Wai Mun Lee over 5 years ago in reply to Yuchang Zhang

Prodigy 40 points

Q1: Sure will check the voltage to confirm

Q2: What is the ripple you typically seeing at the input VIn? I checked through the doc, it seems no measurement on Vin so far.

0 Yuchang Zhang over 5 years ago in reply to Wai Mun Lee

TI__Mastermind 18510 points

Generally <hundreds of mv, depends on vin voltage/external power source ability..., if you think it is abormal with your test, please upload your waveforms for review.

Yuchang

0 Kee Sin Thong over 5 years ago in reply to Yuchang Zhang

Prodigy 10 points

Please see below oscilloscope plot

Yellow - Vout

Green - Vin

0 Yuchang Zhang over 5 years ago in reply to Kee Sin Thong

TI__Mastermind 18510 points

From your waveform, the Vin ripple is smaller than 100mv, that's ok.

Yuchang

Power management

Power management forum

TPS568230EVM: Spur detected on device upon cold-crank but recovers upon increased load with device under FCCM mode