• State TI Thinks Resolved
  • Locked Locked
  • Replies 2 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 305 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TPS734: AC coupling capacitor construction 5100uF in How to measure LDO noise SLYY076

Neil Allison
Prodigy 10 points
Part Number: TPS734
Other Parts Discussed in Thread: TLV62130A

Hi,

Please could you share any construction details of the shielded 5100uF AC-coupling cap (Figure 3 in white paper "How to measure LDO noise", SLYY076)?  I'm interested in both the component capacitors you chose and the physical layout within the shielded box.

I need to evaluate some LDO options for a GNSS application.  We currently have TLV62130A pre-regulators (Vin 4.75 - 14 V, Vout 3.3 V) followed by TPS73401DDC LDOs to generate 2.8 V.  I wish to compare the noise performance of LP5907MFX-2.8 to the TPS73401 and the approach described by Kyle Van Renterghem in SLYY076 is helpful.  Before I rush off to construct a coupling capacitor of approx 3 - 5 mF with a minimum 16 V rating, say constructed from approx 50 100uF X7R MLCC or possibly solid/polymer electrolytics, I'd appreciate any pointers you have on how you assembled your capacitor and any potential traps for the unwary.

Thank you

Neil

  • 0
    TI__Expert 4350 points

    Hi Neil,

    Kyle will be back in the office on Monday, 16-Sep. I'll ask him about the AC-coupling cap then.

  • 0
    TI__Genius 13085 points

    Hi Neil,

    You are on the right path. The AC Coupling capacitors that were used for that app note and for the majority of our noise measurements for the last few years was constructed using a large number of MLCCs (multi-layer ceramic capacitor). There are a few reasons for this including; MLCCs are fairly easy to get, they have low ESR/ESL, they are easy to work with and they aren't polarized.

    As for potential traps, the main one most people forget about but you alluded to is the voltage rating of the caps. In general MLCCs derate pretty badly when a voltage is applied to them so it is important to have a good idea of what the maximum voltage is that will be applied across the terminals of your AC Coupling Cap so you can select caps with a high enough combination of the capacitor value and voltage rating to get the capacitance you calculated. 

    Since we were building these by hand and were working with MLCCs we were able to shrink the x-y size of the filter board by stacking caps vertically. This is how we went from Figure 2 in the app note to Figure 3.  If you choose to stack the caps for your build that means you'll have to do it yourself since a board house won't assemble them that way and you should be a bit careful not to overheat the caps while soldering so you don't damage the internal dialectic.

    I will say that as shown in Figure 3 we built a shielded box around the capacitors to help shield any potential environmental noise from coupling onto either side of the AC Coupling Cap, but I can't say I ever saw a noticeable difference in our measurements so this is probably overkill for most applications. 

    Let me know if you have any other questions.