TPS54325: A question about the transfer function of On-time Timer.

Hi,

1. In my understanding, the delay transfer function stands for the control loop will re-start to monitor the Vout and do something to track the reference only after Ton. This is just personally understanding. Let's see if anyone else can give a better explanation.
2. Different parts use different ripple injection circuit, so the calculation of ripple injection poles and zeros will be different. Usually it's fixed in internal IC design. The zero locates at about tens of kHz, which is well below the typical bandwidth of D-CAPx.

Regards,
Sam

Hi,

The on time can not change during on time even though the output voltage change, the device should wait until the on-time is over, Therefore, there is a delay.

Shawn

Hi Sam,

Thanks for your reply and it helps a lot. However, I'm not clear that

1. The delay transfer function only considers on time but doesn't consider off time , why?

2. Why Hd(s) is e^(-sTon/2) but not e^(-sTon)?

3. Can we get the exact poles and zeros solution of Hd(s)? (e.g. to convert exponential form to (1-s/z)/(1-s/p)).

Best regards，
Tintin

Hi Tintin,

This is a average value, some papers use e^(-sDTon). e^(-sTon/2) is the delay transfer function, there are no poles and zeros.

Shawn

Hi, Shawn

Thank you for the reply.
During the on-time period, the high side MOSFET is turned on and the system is working. I don't understand which device is “the device”. Will you specifically tell me about the device? In addition, I wonder why Hd(s) equals to e^(-sTon/2) rather than e^(-sTon).

Best regards,
Tintin

Hi Tintin,

The device means the IC.

In my opinion, because the vout may change at any point  in the on-time. Ton/2 is the average value.

Shawn

Good morning, Shawn

Thank you very much for your prompt reply. I may understand what you said, then I made a little reasoning according to your prompt, please help me check it out.

I will wait for your reply. Thank you again for your patience.

Best regards,

Tintin