tps53319 LOOP SIMULATION

Sam Bu

Other Parts Discussed in Thread: TPS53319

The phase margin of the DCAP meaning something or not?

In one case, the phase margin of DCAP (TPS53319)is only 30deg as below, but there is still no big ringing on the transient. So does the phase margin of DCAP meaning something?

over 11 years ago

0 JohnTucker over 11 years ago

TI__Guru**** 191445 points

It is not unusual to have low phase margin for DCAP or DCAP2. The topology is inherently stable as PH cannot go below 0 deg. until above Fsw/2. Because of this, the concept of gain margin is not applicable to DCAP control. I think your design performance looks acceptable.

0 Sam Bu over 11 years ago in reply to JohnTucker

TI__Prodigy 40 points

Thanks for your reply! But why the dcap has inherently stable in low PM, you know that there is big ringing in current mode and voltage mode if the PM is under 30 deg.

I have 1 more question.

ref to the paper"DCAP Mode with allpceramic output capacitor application" there are equotion shown below

I dont understand why the equation (1)(3)(4) can make the DCAP loop stable.

Do you have any material which can help me understand that, thanks for your help.

0 JohnTucker over 11 years ago in reply to Sam Bu

TI__Guru**** 191445 points

There are two things to consider primarily. Under no circumstance do you want the crossover frequency to approach Fsw/2. That will violate the Nyquist criteria. So you need to be sure the ESR zero of the output capacitor is well placed (Eq1). DCAp uses hysteretic off time control. You need to have some amount of ripple on teh output voltage for proper operation. That is the requirement of Eq2. If the ESR is high enough, you can generate the required ripple without the additional ripple injection. If you add the ripple injection of Eq3, then the injection components should be chosen as shown. If you follow those equations, you should not have any issues.

0 Sam Bu over 11 years ago in reply to JohnTucker

TI__Prodigy 40 points

Could you pls tell me the derivation of the equ3? or what paper can i reffer to get the derivation?

thanks a lot ! sometimes i need to explain it to curious customer

0 JohnTucker over 11 years ago in reply to Sam Bu

TI__Guru**** 191445 points

I did not derive that personally. It is likely that person is not with TI any longer. I'll see what I can find out, but it will not be this week. You may just have to trust the equation to be true.