Brett's Arduino 8x Smart Charger / Discharger

brettwatty

Member
Joined
Jan 14, 2017
Messages
81
Hey all,

Just thought I would start a new thread.

Finally I have had some success on building my newprototype 18560 charger / discharger. I am using 3.3ohm resistors for the discharge,TP4056 modules to charge and it is controlled by an Arduino Megamicro controller. My first manufacturedPCB version almost done check out the photo's. I have 1 out of 8 module setup.

I have already found some mistakesin my design:

9v power jackhasbackwards orientation. Soldered a cable instead
The mounting holesfor the Rotary Encoder module are wrong. Might be able to drill them.
The RobotDyn Arduino Mega Mini Module I used in easyEDAhas the ICSP header 1.27mm to the left. only using 1 wire for the SD Module so can jumper wire it.
Also thinking of adding a relay switch to the 5V 10A charger power input.

Still ironingout some issue in the code at the moment.

The sequence is:
Check for a battery to be inserted
Get the battery barcode via a barcode scanner
Charge the battery
Check the milliohms of the battery
Rest the battery for the set duration (if the battery wascharged)
Discharge the battery
Recharge the battery
Competed cycle uploads the battery information to my database wait for battery to be removed then it restarts the cycle back to check for a battery to be inserted.


Checkout my database stats page of currently processed batteries: http://vortexit.co.nz/battery_statistics.php

My youTube channel:https://www.youtube.com/channel/UC3F60pBuzm2kRsHjqQOAO9Q

So I thinking hopefully next revision releasing the PCB design and code once it isfully working.


image_cgwnzz.jpg

image_lozfmu.jpg

image_ebmfqy.jpg
 
Nice job! Im nearly finished a very similar project to this. How important is it that you let the cell rest after charging before discharging for capacity checking? I have mine set to immediate as of right now
 
StephenRyan said:
How important is it that you let the cell rest after charging before discharging for capacity checking? I have mine set to immediate as of right now

I would probably wait about 5 seconds just so the voltage settles. Other wise, you might read the cell when it's slightly higher and it'll throw your capacity readings off a little bit.
So, theend of charge the cell could be at4.21V, and the settled voltage is 4.19V. This doesn't seem like a lot, but it could make the difference into which bin you put your cell. (ie. 2200mAh vs 2100mAh as the cell could be 2230mAh vs 2160mAh, respectively with all numbers). Granted, it might not matter in the end overall, but just saying ;)
 
StephenRyan said:
Nice job! Im nearly finished a very similar project to this. How important is it that you let the cell rest after charging before discharging for capacity checking? I have mine set to immediate as of right now

Awesome mate. Yeah I rest for 1min but I can change it to whatever I want. LikeKorishan said it is good for them to settle also lets the temperature drop a bit for the discharge cycle.
 
Hey brettwatty,

thank you very much for the presentation of this project.
In a small group we also imagined something like this, our layout is very similar: 8x charger / discharger, scanner, database!

We would like to support your project. How can we help?
 
Hi.
Nice !! and very finished looking :) cool

I'm in stage of upgrading my 20x tester to do pre-charge and i need that cells ,after 2-3 week rest after first charge, coul'd be recharged to max. But .. TP4056 are made that way, that they do not charge until <4V so .. i need to pre-discharge prior the pre-charge :) Did you use this kind of stages in your code ? :)
I'm thinking of calculating the drop while in 1 amp and then discharge the cell until it is voltage+ drop < ~3.9 then the tp4056 would start charge... and then run real discharge test
 
EGOksy said:
Hi.
Nice !! and very finished looking :) cool

I'm in stage of upgrading my 20x tester to do pre-charge and i need that cells ,after 2-3 week rest after first charge, coul'd be recharged to max. But .. TP4056 are made that way, that they do not charge until <4V so .. i need to pre-discharge prior the pre-charge :) Did you use this kind of stages in your code ? :)
I'm thinking of calculating the drop while in 1 amp and then discharge the cell until it is voltage+ drop < ~3.9 then the tp4056 would start charge... and then run real discharge test

Hey mate, I am not currently doing that but can do. I might look into that. Thanks


tango said:
Hey brettwatty,

thank you very much for the presentation of this project.
In a small group we also imagined something like this, our layout is very similar: 8x charger / discharger, scanner, database!

We would like to support your project. How can we help?

No worries mate. Yes all I need is time and money either I have much of hahaha. Hopefully will have something presentable in the next few weeks and I can do some demo's. Cheers Brett
 
I got the other 7 modules in last night. Got the whole charger / discharge working. I found some small bugs in the code will need to fix them.I had an issue with the 10W 3.3ohm resistors getting so hot that they were melting the wire insulation. I will need to order some silicone 20awg wire. I the interim I wrapped some kapton tape around the wire insulation.
Has anybody had issues with the TP4056 chargers over chargingthe batteries I have 1 charging at 4.27V another 4.30V luck I have a voltage read out during the charge cycle. I can easily replace them.

Here are some photo's

image_bcixih.jpg

image_exogwd.jpg

image_ixgdgt.jpg
 
brettwatty said:
I got the other 7 modules in last night. Got the whole charger / discharge working. I found some small bugs in the code will need to fix them.I had an issue with the 10W 3.3ohm resistors getting so hot that they were melting the wire insulation. I will need to order some silicone 20awg wire. I the interim I wrapped some kapton tape around the wire insulation.
Has anybody had issues with the TP4056 chargers over chargingthe batteries I have 1 charging at 4.27V another 4.30V luck I have a voltage read out during the charge cycle. I can easily replace them.

"Preset 4.2V Charge Voltage with 1.5%
Accuracy " - so it could be 4.2V+- 0.07245V
better test .. and get some tp4056 IC's to replace or replace whole pcb

No heat-sink on irlz44 ? they must be toasting ? :) mine with the small heat-sink are very hot ... i use a fans to keep everything to touchable temperatures.. and 1ohm10W resistors to..
Tried just now after code update to not use the fans, but the heat is quite high :)

P.S: liked your decision to add the irlz44 as polarity protection ;) Rely nice idea ... i keep doing this mistake of adding in wrong polarity's .. and after that MEGA gets crazy .. the load on all slots starts jumping from 400mA to 1.5A :)
 
EGOksy said:
brettwatty said:
I got the other 7 modules in last night. Got the whole charger / discharge working. I found some small bugs in the code will need to fix them.I had an issue with the 10W 3.3ohm resistors getting so hot that they were melting the wire insulation. I will need to order some silicone 20awg wire. I the interim I wrapped some kapton tape around the wire insulation.
Has anybody had issues with the TP4056 chargers over chargingthe batteries I have 1 charging at 4.27V another 4.30V luck I have a voltage read out during the charge cycle. I can easily replace them.

"Preset 4.2V Charge Voltage with 1.5%
Accuracy " - so it could be 4.2V+- 0.07245V
better test .. and get some tp4056 IC's to replace or replace whole pcb

No heat-sink on irlz44 ? they must be toasting ? :) mine with the small heat-sink are very hot ... i use a fans to keep everything to touchable temperatures.. and 1ohm10W resistors to..
Tried just now after code update to not use the fans, but the heat is quite high :)

P.S: liked your decision to add the irlz44 as polarity protection ;) Rely nice idea ... i keep doing this mistake of adding in wrong polarity's .. and after that MEGA gets crazy .. the load on all slots starts jumping from 400mA to 1.5A :)

Yeah the bloody TP4056's I have are not even close to+- 0.07245V.
Thinking of adding a fan on the side.
I am usingIRF540N N-Channel FETS rated at28A max never get hot for me I only used them because a have a bag of 100 left over from another project.
I haveSI2306DS SOT23 for the polarity protection rated for2.8A @ 4.5V VGS I will eventually useIRLML2502TRPbF-1 SOT23.
My board isdesigned for either THT or SMD components for all the resistors and FETS.
Polarity protection is a must for me so many TP4056 fried because of dumb mistakes and last thing you want is a -4.20Von an Analoginput :D .
 
brettwatty said:
Yeah the bloody TP4056's I have are not even close to+- 0.07245V.
Thinking of adding a fan on the side.
I am usingIRF540N N-Channel FETS rated at28A max never get hot for me I only used them because a have a bag of 100 left over from another project.
I haveSI2306DS SOT23 for the polarity protection rated for2.8A @ 4.5V VGS I will eventually useIRLML2502TRPbF-1 SOT23.
My board isdesigned for either THT or SMD components for all the resistors and FETS.
Polarity protection is a must for me so many TP4056 fried because of dumb mistakes and last thing you want is a -4.20Von an Analoginput :D .

Cool thx ..will look int to adding the protections i think .. even tho i have only 700 cellsleft to test :) for now

Have you tried to cut power with still cells in slots ?
mine was powering everything from those cells(used only 1 for test .. not to fryup the MEGA and multiplexers)... yes i use multiplexers for inputs of analog..
Constantine had the same issue in his build, and he removed multiplexers i think.
I use dividers to limit inputs, because i need 20 cell tester not 8 ;):)

maybe we all could add the backups from 18650 and boost/buck converters for emergency situations :D as we all have enough of 18650 :D
 
Yeah EGOksy I have the same issue. From some reading I found that the power is flowing from the analog pins, through the internal clamping diodes, to the Vcc line. So will either need to find a way to disconnect the analog pins when there is no power from the 5V regulator (transistor maybe) or I can put some 10K resistors in between the battery + to the analog pins. Apparently this is bad as those diodes are not rated for much current. Yeah the backup power is a good idea. I the future I am also thinking of making a Master module and Slave modules with multiplexer and shift registers. But for the moment I want refine my code.
 
Great project! I suggest rest the battery for one minute too.
 
Re the FETs getting hot for EGOksy posted earlier, check the gate "on" voltage for the FETs you're using is high enough & not in the threshold region wheer the FET is only partly on. FETs like these should be pretty much stone cold with several amps going through them when properly "on". You might have to pick a different FET with "logic level" gate threshold.
 
Back
Top