TPS40170: Adaptive Dead Time

Sanel Hirkic

Part Number: TPS40170
Other Parts Discussed in Thread: LM25145, LM5145

What is the minimum deadtime for the TPS40170 controller? Also, do you have any general information about how the adaptive deadtime works for this part? I'm looking to get an answer on the first question today if at all possible.

3 months ago

+1 EricLee 3 months ago

TI__Mastermind 48471 points

Hello Sanel

There is no guarantteed minimum deatime. The mimum and the maximum values are blank in the EC table. However, there is typically ~ 10-20ns minimum propagation delay.

About the adaptive deadtiem scheme, please refer the page 10, 12, 13 of this datasheet https://www.ti.com/lit/ds/symlink/tps51604.pdf . TPS40170 deadtime circuit is very similar with this device. The adaptive dead-time control circuit turns on HDRV after LDRV voltage falls down below a proper threshold, and also turns on LDRV after HDRV-SW voltage falls down below a proper threshold.

-EL

0 Sanel Hirkic 3 months ago in reply to EricLee

TI__Intellectual 1000 points

Hi Eric,

Appreciate the quick response. The customer is wondering if there is a fixed time delay between sensing LDRV falling below a threshold and HDRV being turned on? To be more specific, the propagation delay between PWM going high and DRVL starting to go low is not what I'm after. Per the description trpd-drvh should be dependent on when DRVL drops below 1V, not based on when PWM goes high:

The customer is interested in the timing of the window.

0 EricLee 3 months ago in reply to Sanel Hirkic

TI__Mastermind 48471 points

Hello Sanel

There should be 10ns fixed delay but no data is aviable now because the device is old (released in 2008).

-EL

0 Sanel Hirkic 3 months ago in reply to EricLee

TI__Intellectual 1000 points

My apologies -- so the time in the red window above should be 10ns? How does it differ from the following in the datasheet:

+1 EricLee 3 months ago in reply to Sanel Hirkic

TI__Mastermind 48471 points

Hello Sanel

No, the red window should be typically 60ns.

The minimum value is not guarantted and not mentioned in the EC table, but since 10ns fixed delay is also designed inside the IC. The minimum value should be greatert than/equal to ~10ns. The adative deadtime (the red window) cannot be smaller than this 10ns fixed delay.

-EL

0 Sanel Hirkic 3 months ago in reply to EricLee

TI__Intellectual 1000 points

Which operating conditions would be most likely to produce the minimum deadtime, high or low temp? If you have any additional details on lot-to-lot variation and aging, please share that as well.

+1 EricLee 3 months ago in reply to Sanel Hirkic

TI__Mastermind 48471 points

Hello Sanel

It should be low temperature.

Production data is not allowed to disclose in public.

-EL

+1 Timothy Hegarty 3 months ago in reply to EricLee

TI__Guru 52142 points

Hi Sanel,

Note the preferred controller series here is the LM5145 (75V) or LM25145 (42V). These have better performance and are more cost effective than the TPS40170. The voltage-mode control loop architecture is the same.

Regards,

Tim

Power management

Power management forum

TPS40170: Adaptive Dead Time