What is the purpose of a capacitor in a voltage divider circuit?

[Borek]

I need a voltage divider to make it possible to measure voltage on the 3S LiPo battery that delivers around 11 V using a device that measures up to 4 V (approx). I can make one, no problem, just solder two resistors with resistances in the more or less 1:3 ratio (measured output is sent as a single byte value and is later scaled by the RC receiver before being displayed, so it doesn't matter how precise the divider is, as I can change the scaling constant). I know from googling that the total resistance of about 10-80 kΩ works, so it gives me enough information to buy resistors, trivial and no problem.

However, I googled for schematics, and found that on some of them there is an additional capacitor added on the output side. What role does it play?

 

[rbelli1]

It acts as a filter and a snubber during battery removal. If there are long lines to the battery you can pick up noise and get false readings.You can also get a high voltage spike on battery removal due to the inductance. 10uF seems a bit large for this purpose.

BoB

 

[CraigHB]

It does more than that. It filters the output full time since typically it's measured by an ADC which can return jumpy values due to noise when unfiltered. Though for any application 10uF is large. For that resistor network a 1uF capacitor would be ample in providing any filtering that may or may not be required by the ADC. Typically they run filtering routines internally that can do the same job, but they don't always.

Another thing the cap can do is provide the burst current required by the ADC to take measurements. Usually ADCs can handle source impedance up to about 2kΩ, but if source impedance is higher, the cap acts as a tank to provide the current needed to charge the ADC when it takes a measurement.

Generally it's good practice to place a cap across the terminals of an ADC. Just a bit of insurance to cover some things that may cause a problem. Kind of the same idea as putting a decoupling cap across the power supply pins on a chip when designing PCBs.

 

[Borek]

Thank you guys.

Soldering time. In case I disappear for the next few days it means I have blisters on my fingertips.

 

[Borek]

As you can guess, soldering is not my forte. It is a bit longer than I wanted, but it will still nicely fit inside the model (after being put in a heatshrink). I didn't want to solder three free ends at the same time, which is why I have removed the isolation in the middle of the wire. It also helped to keep elements separated (only the capacitor required additional isolation).

There are probably many ways to make it better. Actually it can be even bought for about $4... but where's the fun in that?

 

[dlgoff]

... but where's the fun in that?

Nice divider. Now how about a multiplier?
image compliments https://www.ibiblio.org/

https://www.ibiblio.org/kuphaldt/electricCircuits/Semi/SEMI_3.html]A[/PLAIN] voltage multiplier of cascaded half-wave doublers of arbitrary length is known as a Cockcroft-Walton multiplier ...

 

[Borek]

The moment my model will run on AC and will require high voltage I will consider making one

 

[NascentOxygen]

A ratio of 1:3 and the total resistance of about 10-80 kΩ? — not quite!

 

[CraigHB]

There are probably many ways to make it better.

That should do the job.

If you want to start making your own circuits, you should look into making PCBs yourself using surface mount components. With some 0603 size components that small circuit would be a square not much wider than the wires soldered to it. Actually it's not all that hard to do, looks harder than it is.

More than anything else the real trick with doing small PCBs using SMD components is having a good magnifier and workstation. For example attached is a photo of a small dual ESC I made for a micro RC helicopter. I assembled that PCB by hand under a stereoscope.

 

[Borek]

A ratio of 1:3 and the total resistance of about 10-80 kΩ? — not quite!

Good enough - I know from googling that I am in the range people used. These are 1.5 and 7.5 kΩ (I know they look both identical), so the ratio is lower, but first - it makes place for stronger packets (like 4S), second - I was told conflicting information about the range measured by the AD, some say it is 0..3.3 V, some say it is 0..5 V. These resistors make me stay on the safe side.

And as you can see from the schematics I posted, even total of around 6 kΩ is OK.

 

[Borek]

That should do the job.

It does. Already tested.

 

[NascentOxygen]

These are 1.5 and 7.5 kΩ (I know they look both identical),

Ah, yes. ✔ (I see by enlarging the image it reveals the violet.)

 

[Borek]

Ah, yes. ✔ (I see by enlarging the image it reveals the violet.)

They are so similar I had to use a multimeter to check I wasn't sold two identical ones. In the artificial light in my room and using a magnifying glass, even knowing they are different, I had still problems finding which is which.

 

[NascentOxygen]

They are so similar I had to use a multimeter to check I wasn't sold two identical ones. In the artificial light in my room and using a magnifying glass, even knowing they are different, I had still problems finding which is which.

Quite. I don't understand why manufacturers don't make their band colours more distinct. With some it's difficult to discern whether a band is red or orange without having a second resistor with the other colour for comparison. Black vs. brown can be a problem, too.

 

[Planobilly]

I find I also have issues reading the color codes on resistors. I find a LED flashlight helpful. The colors appear clearer under that light source. At times when I get frustrated trying to divine the color code I just give up and measure the resistor with a DMM with a readout made for old guys...lol

Cheers,

Billy