TPS3808 Delay Min/Max Calculations for given cap

Brian Antheunisse

Other Parts Discussed in Thread: TPS3808

Hello team,

I found a lot of posts about this and I have a feeling there is no precise answer here, but my customer is using the TPS3808 and is trying to get the worst case min/max delay time for a given Ct. There is an equation in the datasheet to approximate this, but we are wondering how to bound the min/max.

Looking at the spec table, it says for a Ct of 180nF, the delay could range from .7s to 1.7s. Then if you load the equation with 180nF you get 1.03s.

It's nice we have a bound for a couple of extreme capacitor choices, but is there a way to bound the min/max for a cap value of our choosing?

(I understand Cap tolerance will play a factor here, I'm talking about just part to part variance)

Thanks!

Brian

over 7 years ago

0 Ryan Eslinger over 7 years ago

TI__Guru 55940 points

Hi Brian,

Please refer to the following post on how to estimate the delay min/max:

e2e.ti.com/.../291414

Very Respectfully,
Ryan

0 Brian Antheunisse over 7 years ago in reply to Ryan Eslinger

TI__Expert 5215 points

Hi Ryan,

Thanks for the response. I noticed that post before, and it was helpful. However, according to this post, it looks like that equation is just an approximation, and using the method described in the post you linked gives a min max that is only dependent on the capacitor tolerance. The fact that you have a spec in the datasheet that varies an entire second for the same capacitor value makes one think that there is some part-to-part variance that comes into play as well.

Can you truly rely on that equation to give you a 100% certain min/max, or is the equation just a close approximation?

Thanks!

Brian

0 Ryan Eslinger over 7 years ago in reply to Brian Antheunisse

TI__Guru 55940 points

Hi Brian,

The equation is accurate to how the device was designed; however, it is idealized and does not account for every variable such as capacitor tolerance. The post I pointed to is the best approximation available. Supervisors are not designed to be precision timers.

Very Respectfully,
Ryan

0 Brian Antheunisse over 7 years ago in reply to Ryan Eslinger

TI__Expert 5215 points

Thanks Ryan. I don't think my customer wants to use the part as a precise timer, they just wanted to know the worst case tolerance of the silicon, and add it to the worst case tolerance of their capacitors, so they could arrive at a bounded min max delay time (regardless of the accuracy between the two).

Looks like since capacitor tolerance seems to dominate the delay time variation, it makes sense that you would only give a close approximation.

Thanks for the help.

Brian

0 Brian Antheunisse over 7 years ago in reply to Ryan Eslinger

TI__Expert 5215 points

Ryan,

Quick question. How was this spec in the datasheet tested? Characterization on a board with a Ct attached, or is that number tested on ATE somehow?

0 Ryan Eslinger over 7 years ago in reply to Brian Antheunisse

TI__Guru 55940 points

Hi Brian,

As with all min/max specs in an Electrical Characteristics table that I am aware of, these are tested at final test.

Very Respectfully,

Ryan

Power management

Power management forum

TPS3808 Delay Min/Max Calculations for given cap