Author Topic: Can someone explain the values of the resistors?  (Read 2325 times)

0 Members and 1 Guest are viewing this topic.

Offline vk4ffab

  • Regular Contributor
  • *
  • Posts: 230
  • Country: au
Re: Can someone explain the values of the resistors?
« Reply #25 on: November 14, 2023, 08:10:12 am »
As said, that circuit is crap. Not that mine is great, but some TL431 will do a much better job.
A Li-ion battery is mostly discharged at 3.7-3.6V... pushing it further greatly shorts lifespan while providing very little extra energy.


(Attachment Link)
People seem to be assuming that the three different colour LED all require the same current for equal subjective brightness.

1.  Is that true?


For a given series of LED's this should remain true. If you look in the datasheet of most LED's they will plot Luminous Intensity verses forward current. So having the 4 LED's using the same current should equal similar luminous intensities and thus a different resistor value for each led, as long as that is what the datasheet suggests. It is for parts I regularly use, but might not hold true for others.

You know, its electronics and everything is filled with caveats.
Equal objective luminous intensity does not mean equal subjective brightness.

https://www.researchgate.net/figure/The-Relationship-Between-Eye-Sensitivity-and-Colour-Wavelength-and-Colour-Luminosity_fig4_269300186

https://www.nde-ed.org/NDETechniques/PenetrantTest/Introduction/lightresponse.xhtml

Luminous intensity is a measure of the wavelength-weighted power emitted by a light source. So it should also mean that for a given current, LEDs of different colours within the same series of LEDs producing a similar luminous intensity should appear to have the same relative level of brightness because the measure accounts for it. The key words being wavelength weighted.
« Last Edit: November 14, 2023, 08:16:18 am by vk4ffab »
 
The following users thanked this post: wasedadoc

Offline Siwastaja

  • Super Contributor
  • ***
  • Posts: 8173
  • Country: fi
Re: Can someone explain the values of the resistors?
« Reply #26 on: November 14, 2023, 08:42:06 am »
A Li-ion battery is mostly discharged at 3.7-3.6V... pushing it further greatly shorts lifespan while providing very little extra energy.

No, 3.65V is most definitely close to 50% SoC. 50% in charge and maybe 40% in energy in my opinion is not "very little". And there is absolutely no reduction in lifespan by deeper discharge, although one might want to prevent the last 10-15% i.e. below 3.50V OCV, but that could be as low as 2.5V under load.

Quite the opposite, it's most harmful to fully charge, especially at combination of high currents and low temperatures, so if you choose to cycle between 100% and 50% instead of, say 75% and 25%, you are doing yourself a disservice.

You don't need to make this stuff up, just look how it's done normally. Normal cut-off voltage is somewhere around 2.0V to 3.0V. Also see datasheets for discharge curves.
« Last Edit: November 14, 2023, 08:54:28 am by Siwastaja »
 

Offline MrAl

  • Super Contributor
  • ***
  • Posts: 1440
Re: Can someone explain the values of the resistors?
« Reply #27 on: November 14, 2023, 08:44:09 am »
As said, that circuit is crap. Not that mine is great, but some TL431 will do a much better job.
A Li-ion battery is mostly discharged at 3.7-3.6V... pushing it further greatly shorts lifespan while providing very little extra energy.


(Attachment Link)

Hi,

That's not a bad idea for an Li-om battery cell checker.
 

Offline MrAl

  • Super Contributor
  • ***
  • Posts: 1440
Re: Can someone explain the values of the resistors?
« Reply #28 on: November 14, 2023, 08:47:21 am »
Why were the 220 \$\Omega\$ and 47 \$\Omega\$ chosen and in those particular spots? Will they be the same for any other 18650 3.7V battery?

This circuit looks like it was meant to be a very cheap Li-ion battery cell tester.  Depending on how the LEDs light up, you can get some idea of what the battery voltage is on the cell.
It's about as cheap as you can get, and should be used at the same temperature for every measurement.
We could probably improve it greatly by just changing some resistors and maybe added a few more diodes.  It will not be super accurate, but if used indoors it might be ok.
A better unit would use more accurate set points such as you might get from using a few comparators.

I myself have to check the status of AA batteries and Li-ion cells and I use a cheap commercial unit that takes any size cylindrical cell.  You can get them on sites like Amazon pretty cheaply.


 

Offline Siwastaja

  • Super Contributor
  • ***
  • Posts: 8173
  • Country: fi
Re: Can someone explain the values of the resistors?
« Reply #29 on: November 14, 2023, 08:57:42 am »
This circuit looks like it was meant to be a very cheap Li-ion battery cell tester.  Depending on how the LEDs light up, you can get some idea of what the battery voltage is on the cell.
It's about as cheap as you can get

I would suggest an even simpler one with only two parts: a blue LED with series resistor (e.g. 220 ohms), nothing else.
Bright -> maybe 30% - 100%, depending on brightness
Dim -> empty
Very dim to the point of not being clearly visible in room illumination -> cell overdischarged, check using a separate meter.
 

Offline Nominal Animal

  • Super Contributor
  • ***
  • Posts: 6264
  • Country: fi
    • My home page and email address
Re: Can someone explain the values of the resistors?
« Reply #30 on: November 14, 2023, 12:31:23 pm »
I'm actually very surprised nobody has made a Li-ion/LiFePoâ‚„ fuel gauge using Padauk PMS171B yet.  It is programmable using the Free PDK toolchain, and costs only about 8 cents in sets of five at LCSC.  A Diodes Inc. AZ431AN-ATRE1 2.5V 0.4% shunt voltage reference costs three cents.  If you run the PMS171B from the Li cell voltage, the reference will read 152 at 4.2V, 156 at 4.1V, 168 at 3.8V, 183 at 3.5V, 200 at 3.2V, and 213 at 3.0V; i.e. 8-bit ADC reading N corresponds to voltage 640/N volts.  (Resolution 0.025V around 4V, 0.016V around 3V.)
With three output pins, you can light one of four different LEDs at a time (without high-Z), each with their own resistor.  You can do PWM/PDM to compensate for the varying current if running from the 3.0V - 4.2V Li cell voltage.  In the SO8 package, there is even a pin free for doing a low-side switch using an N-channel MOSFET, so that a button press turns connects the circuit to the real negative rail, with the MCU pin keeping it connected so it can turn itself off after a time limit, not consuming any current while turned off.

Or, you can run the circuit off the shunt voltage reference, and measure the cell voltage for example via an opamp that subtracts the 2.5V from the cell voltage, yielding about 0.01V precision.  The LEDs would then run off the 2.5V I/O levels, too.

By my count, the components for that would cost under 30 cents in sets of five at LCSC, add maybe 15-20 cents for a suitable TI RRIO opamp for the more precise version.

I like the idea of having the most significant LED blink at say 2 Hz, with duty cycle indicating the fraction (or constant on time, but off time and period varies).
With the SO-16 version of PMS171B, you could even use a two-segment LED display to show the charge as percent between 00 and 99.
 

Online DavidAlfa

  • Super Contributor
  • ***
  • Posts: 5912
  • Country: es
Re: Can someone explain the values of the resistors?
« Reply #31 on: November 14, 2023, 02:40:42 pm »
IP stolen or not, I prefer Puya ARM 1000 times over a weird mcu architecture and custom programming interfaces.
Hantek DSO2x1x            Drive        FAQ          DON'T BUY HANTEK! (Aka HALF-MADE)
Stm32 Soldering FW      Forum      Github      Donate
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf