Battery circuit

crutschow

Joined Mar 14, 2008
38,503
Looks like it was done by their rooky designer.
I also see no reason for the long string of capacitors in descending decade values.
 

Thread Starter

hoyyoth

Joined Mar 21, 2020
528
Looks like it was done by their rooky designer.
I also see no reason for the long string of capacitors in descending decade values.
This is a battery powered product.
May I know can I replace all those capacitors with a single supercatacitor
 

Ian0

Joined Aug 7, 2020
13,131
This is a battery powered product.
May I know can I replace all those capacitors with a single supercatacitor
I would keep two or three of them, depending on what I was powering, but a supercapacitor will have far too much series inductance to decouple fast switching circuits.
 

Thread Starter

hoyyoth

Joined Mar 21, 2020
528
I would keep two or three of them, depending on what I was powering, but a supercapacitor will have far too much series inductance to decouple fast switching circuits.
Will keep decoupling capacitors close to the IC.Will that work.I mean supercap near the battery and decap near the IC
 

Art Vandelay

Joined Nov 1, 2024
140
Looks like it was done by their rooky designer.
I also see no reason for the long string of capacitors in descending decade values.
According to this graph, a capacitor of a given value has a minimum impedance at a specific frequency. Putting several in parallel creates a lower impedance path that would otherwise not be present. Not sure how effective this actually is but it's an interesting idea.

On second glance each capacitor is functioning as a bandpass filter. I think this is because of parasitic inductance forming a resonant circuit unique to the construction of each capacitor.
 

Attachments

Last edited:

MisterBill2

Joined Jan 23, 2018
27,508
Given that we have no hint as to the application, all of those caps help provide a low source impedance "DC toDAYLIGHT" as one expression goes. And the mosfets can serve as a low loss fuse for when things go wrong.
 

meth

Joined May 21, 2016
304
Didnt you get an answer on another forum, that, due some of the safety standards in electronics there must be at least 2 MOSFETs, in case one of them fails in short circuit?
 

Thread Starter

hoyyoth

Joined Mar 21, 2020
528
Yes..in that also some contradictions are there...but still as a learner I am trying get views of experts from different forums
 

Ian0

Joined Aug 7, 2020
13,131
Given that we have no hint as to the application, all of those caps help provide a low source impedance "DC toDAYLIGHT" as one expression goes. And the mosfets can serve as a low loss fuse for when things go wrong.
They usually fail short, unless you can get enough current through it to melt the source bond wire.
I think that they copied the circuit from a BMS schematic which have two back-to-back MOSFETs, and it was easier to turn one symbol round than delete it.
 

MisterBill2

Joined Jan 23, 2018
27,508
They usually fail short, unless you can get enough current through it to melt the source bond wire.
I think that they copied the circuit from a BMS schematic which have two back-to-back MOSFETs, and it was easier to turn one symbol round than delete it.
THAT is entirely possible. Probably the command string for delete was unknown.
 

AnalogKid

Joined Aug 1, 2013
12,128
Guess #1: The small capacitors are spread out on the pc board, one power supply decoupling capacitor per IC. They are grouped in the schematic for convenience and to reduce clutter in the signal sections. On page 2 of the schematic there are no decoupling caps at the opamp power pins, but there are on page 11 of the reference design document.

Guess #2: Error. If Q2 is reversed, the MOSFET pair provides reverse-polarity protection. An error of this kind is unexpected from TI, but the entire design has a inexperienced / undisciplined character. Exhibit 1, pages 1 and 2 of the pc board assembly file. The reference designators are placed in ***5*** different orientations. Gold star for creativity, black dot for WTF.

Also, each mounting hole has two different reference designators. That's a new one on me. Two different shapes/sizes? Sure. Two different nets, or one side is connected to a net and the other is not, because the hole is not plated? Strange, but OK. But two different refdes?

ak
 
Last edited:
Top