"18650 circuit" for Verter 5V

Breakout boards, sensors, other Adafruit kits, etc.

Moderators: adafruit_support_bill, adafruit

Please be positive and constructive with your questions and comments.
Locked
User avatar
Cooker
 
Posts: 4
Joined: Sun May 14, 2017 4:06 am

"18650 circuit" for Verter 5V

Post by Cooker »

Sorry for my poor English.

I have bought a Verter 5V from Pimoroni for my RPi Zero W. I am thinking to power it by two 18650 in series. My problem is I want to unplug one 18650 (to swap a fresh one when battery low), and have the another 18650 keep my RPi running without stop during the moment. Then swap another one.

I have some very basic concept on electionics but tends-to-zero hands on experience. I know I cannot just unplug one 18650 because it will break the circuit. I cannot just put a wire in parallel with the 18650 which is a short circuit. I am not sure if using a diode (with a resistor?) on that wire can solve my problem.

Does the "18650 circuit" I need consists of only a diode? plus a resistor? what value? what kind of diode?
or
Such a simple circut won't work, diode here will cause other issues? A more complicated "18650 circuit" is required?

Also, will this action blow my Verter 5V or RPi zero w? I read from another post here that order of plugging in battery and load makes different.

Thanks for your help.

User avatar
adafruit_support_bill
 
Posts: 88096
Joined: Sat Feb 07, 2009 10:11 am

Re: "18650 circuit" for Verter 5V

Post by adafruit_support_bill »

I am thinking to power it by two 18650 in series. My problem is I want to unplug one 18650 (to swap a fresh one when battery low), and have the another 18650 keep my RPi running without stop during the moment. Then swap another one.
A couple of fundamental problems with that approach. One, which you have already identified, is that the circuit is broken when you remove one cell. The other issue is with the idea of running unbalanced LiPo cells in series (or in parallel for that matter).

Multi-cell LiPo packs need to be 'balanced' for safe operation. In general, this means using factory-matched cells, balance chargers and protection circuits designed to keep both cells charging and discharging evenly. Small imbalances can result in one cell doing a disproportionate amount of work, which can result in overheating and possible catastrophic failure of the cell.

A better approach would be to have two separate battery circuits with a switch from one to the other. Since each cell would be 3.7v, a PowerBoost would be a better board to use than the Verter. Your cut-over switch could be a 'make-before-break' type so that there is no loss in power during the cut-over. It would be OK for the cells to be operating in parallel for just the few millisoconds it takes to flip the switch.

User avatar
Cooker
 
Posts: 4
Joined: Sun May 14, 2017 4:06 am

Re: "18650 circuit" for Verter 5V

Post by Cooker »

Thanks Bill for your prompt reply.

I bought Verter 5V because "two 18650 in series" should already provide sufficient power for my RPi zero w on-the-road need. Swapping 18650 is something additional for just-in-case, and learning/fun.

I am timid. I have quite some 4054 & 4057 PCBs and I always charge LiPo cell one-by-one. They are always in room temperature when charging, just very slowly.
I also have a tiny device to monitor voltages of LiPo in series (up to six cells), it beeps loudly when one cell's voltage falls below a user pre-set level. I got it when I played RC helicopter before.
Basically I am put things on hand together for this.

I am not going to swap LiPo in parallel. I was warned that the fresh LiPo (with high voltage) may push back current to the dry LiPo (with low voltage) and ... just no good.

So, let me clarify my previous idea.
For each of the two 18650 in series, I will put a diode (in parallel with 18650) like this
   /-- 18650 --\ /-- 18650 --\
(-)---       ---       ---(+)
   \---diode---/ \---diode---/
     --> current flow direction on diode

When both 18650's are in position, diode prevent current from flowing back which makes a short circuit.
When I remove a 18650, the diode will keep the main circuit non-broken.
Does it make sense?

If the idea is viable, what kind of diode should I use? Do I need a resistor with it?

I have so many questions. I want to know more, so as to minimize the risk, avoid stupid mistake, before I start testing.

Thanks again for your help.

User avatar
adafruit_support_bill
 
Posts: 88096
Joined: Sat Feb 07, 2009 10:11 am

Re: "18650 circuit" for Verter 5V

Post by adafruit_support_bill »

Swapping cells as you propose would guarantee that one cell is always less charged than the other. Not a recommended configuration.
I am not going to swap LiPo in parallel. I was warned that the fresh LiPo (with high voltage) may push back current to the dry LiPo (with low voltage) and ... just no good.
If you re-read my suggestion, you would see that is not what I proposed.

User avatar
Cooker
 
Posts: 4
Joined: Sun May 14, 2017 4:06 am

Re: "18650 circuit" for Verter 5V

Post by Cooker »

Bill, thanks again for your prompt reply.

I read you reply, couple of times. That's why I made a simple diagram to clarify what is in series and what is in parallel because your reply shows me that I may not have made myself clear enough on my first post.

You suggest me to use a 'make-before-break' switch to do the swap job. Thanks a lot for that. I will buy another battery box for another set of "two 18650 in series". Usage problem is 101% solved. Thank you.

But still, I want to have some learning and fun. Is it not what those makers looking for? Is it not what Arduino is supposed to achieve?

No two batteries are identical. Change two drying batteries with two full ones within 15 seconds won't make a significant change on the natural "unbalance". My concern is "surge" or other similar things I read from other posts but not quite understand at this stage. That is where I need you guys to advise.

Sorry I am not going to buy the PowerBoost you suggested at this moment, no need, and more important, no fun.

I am not looking for a perfect solution, I just don't want to blow either your Verter 5V nor my RPi when seeking fun.

Do you mind sharing some knowledge that I am looking for?

User avatar
adafruit_support_bill
 
Posts: 88096
Joined: Sat Feb 07, 2009 10:11 am

Re: "18650 circuit" for Verter 5V

Post by adafruit_support_bill »

You might be able to make it work that way. But I would not be surprised if the output voltage exhibits a substantial dip when you pull a cell from the circuit.
Change two drying batteries with two full ones within 15 seconds won't make a significant change on the natural "unbalance".
When you pull one cell out, you are expecting the other - presumably nearly drained - cell to suddenly shoulder more than twice the load. Keep in mind that boost conversion is inherently less efficient than buck regulation. And that the cell output voltage will inevitably drop with the increased load.

User avatar
Cooker
 
Posts: 4
Joined: Sun May 14, 2017 4:06 am

Re: "18650 circuit" for Verter 5V

Post by Cooker »

Thanks Bill for this reply. I almost walk away and never come back.

Thanks for reminding the voltage drop issue. That beeping tiny device draws power too. I don't know how to calculate so I will test from a higher beeping voltage, adjust down 0.1v for every test.

Regarding diode, all I know is that it block current from one direction and let go from another, physically some look like a light bulb (led) and some like a resistor, nothing more. Seems that you are pushing me to read more before experiment. It reminds me about my science teacher in high school. It's fine. Slow me down did help me compress my overall spending on this newly emerging hobby.

Best Wishes,

User avatar
adafruit_support_bill
 
Posts: 88096
Joined: Sat Feb 07, 2009 10:11 am

Re: "18650 circuit" for Verter 5V

Post by adafruit_support_bill »

Regarding diode, all I know is that it block current from one direction and let go from another,
In theory, an 'ideal' diode works that way. In the real world, diode behavior is somewhat more complex: https://en.wikipedia.org/wiki/Diode_modelling
Of interest in your circuit is the voltage drop across the diode. All diodes exhibit some voltage drop (Vf) that increases with current.

When you pull the LiPo cell out of the circuit, all the current has to flow through the diode. So, in addition to the voltage drop caused by the loss of one cell, you have the voltage drop across the diode. For a common silicon diode, that will be somewhere around 0.7v. All of which forces the boost converter to work even harder and demand more current - causing further voltage drops.

Locked
Please be positive and constructive with your questions and comments.

Return to “Other Products from Adafruit”