[t0astm0nkey]

THISS must be a cheap Chinese cell manufacturer. I'm not able to find much more about them other than an address for imports/exports.
I'd be more interested in the housing and the circuit board. I've got a small pile of (almost) new Samsung 35E18650 cells that would be AWESOME to drop into this package and get a 7.5AH battery out of it.

 

[evluvseattle]

Hi all, I'm having a weird issue with one of these batteries (OP4015 circa 2013) and I couldn't find a solution earlier in this thread so I thought I'd add to the endless conversation. 😬 If a new thread is more appropriate, lmk.

The symptoms I'm seeing are:
1. when I press the test button on the battery, only the first LED lights up. (regardless of the cells' charge levels)
2. when I put a load across +/-, the battery provides power for about a second, and then nothing.

This pack was previously heavily discharged (to around .6v per cell), and I brought it up to full charge by charging the individual cells to 4.2 volts about a month ago. The cells are currently around 3.9 volts each. When I put the battery in the charger (OP403) I get a steady green light that eventually goes dark. I assume this means the charger thinks the battery is charged, which is strange since I believe full charge should be ~4.2 volts.

I have tried resetting the PIC by putting a resistor between the RST and GND headers, but that didn't seem to do anything.

Any suggestions?

Thanks!
Theo

 

[kennybobby]

Howdy Theo,
Does your circuit board have one FET or 2 on the heatsink? Does the board match with some of those shown in pictures. The 2013 version with 1.5AHr may be a different circuit board.

You may want to check if the LEDs on the check lights are burned out.

The + and - terminals on the pack do not connect directly to the cells--the - goes thru the FET(s), which must be ON in order to pass current. This is shown on some of the schematics posted on previous pages.

Maybe the cells are worn and won't go to 4.2 any longer?

 

[Treeific]

Great info here I really appreciate all the effort and time everyone has contributed to this thread.

I've rebuilt my 40v Ryobi battery and I am fairly satisfied. I don't have a standard to compare with as I only have used battery packs. I actually replaced cells during my rebuild, harvested from laptop batteries (tested and monitored of course) but my rebuilt lasts about twice as long as my other packs. (As I calculated, hoped for) it has some problems that I'm okay with, like the 4th led I'm not positive is working, and the pack will cut out under high load.

I use them in ryobi products like weed eater and mower.

Do y'all think the bms is cutting the power based on a current limit programmed in the MC? If so I would think I will experience the cut off from a new Ryobi battery of the same rating.
Or do you think my batteries do not supply enough current and the BMS can sense that, and that's what's makes it cut out.
I don't think it's heat related because I just need to start the mower up again (even with out a rest) and it works fine till I got another high spot of grass that puts a load in the pack.

I understand these BMS are poorly designed for us users, and probably designed with marketing and business in mind instead of reliability. I heard older batteries were designed worse, and newer one are more reliable. Have y'all heard this?

I want to ask this question, and I don't mean to disrespect all the amazing work yall have put into reverse engineering this product.

Can I simply buy a 10s BMS circuit (something like This ) and bypass all of Ryobi proprietary circuits?
It'd be great to use the same charger, but I don't expect it to be easily possible. I don't anticipate the tool having any problem with it. I know I will need to mod, rather alot, to make it fit and sink heat, etc, but it sounds smart to me... Do you guys see the same benefits as me in doing this?

Just asking because if I find a good quality 10s BMS that fits the specs, in my head I can get a much more reliable pack, probably simplified too.

What are yalls thoughts?
Thanks!

 

[kennybobby]

Could the mower motor be hot? or high temperature inside the battery?

The T1 and T2 lines are used by the battery, the charger and the tools to control the FETs, and we don't know the firmware, so it's hard to say how to defeat all that.

A separate control board that could respond to the charger and tools, and monitor the pack for OV, UV, OC, OT, etc would be an interesting project--i just wish we knew the present battery controller better. And the charger controller too. i've not taken any tools apart to see what they have either.

 

[Treeific]

I think I'll order one, it's really cheap. It's worth a try.

I'm also curious about the present BMS, but I just like throwing ideas so they can shake in someone's head, who knows what could happen.

 

[Halotem]

What kind of zener diode did you used in zener diode voltage regulator construction ?

 

[taylorb]

Thanks for those who help with this post! I used it to fix a Ryobi 36V power pack (BPL3650D 130302002DG9 10ICR19/66-2) which is identical to the 40V pack. Specifically Q3 & Q1 are NPN BJT marked "1D", MMBTA42, and, Q4 is the PNP compliment marked "2D", MMBTA92, all of which were faulty and replaced. (The link between batteries 5-6 was removed to safely effect the repair.)
Looking at the circuit, Q1, Q3 & Q4 is an obvious design fault providing a low impedance path between B- and Gnd. Obviously a 10k resistor should be fitted between Q3 collector and Q4 base just like there is one between Q5 collector and Q6 base.

 

[kennybobby]

That's an interesting observation, i will recheck my board to see if i missed it, like an "R3". Thanks for ID on those transistors.

 

[taylorb]

There is definitely no 10k resistor fitted - there is a direct connection between Q3 collector and Q4 base. Also it seems that Ryobi engineers put in high voltage transistors (200V and 300V) assumedly to stop the problem from happening - however Vebo (reverse voltage) is always low for any transistor, and when a transistor is on it is definitely low impedance.
Anyway your circuit was instrumental in locating the failed transistors and replacing them. So thanks very much for working all that out - much appreciated! I actually used a MMBT5401 for the PNP, which is 150V, but it worked fine.

 

[kennybobby]

i haven't found my bare board yet, too many projects on my bench.

i noticed that the T2 circuit is missing the base resistor such as R16 on T1, and also the same missing resistor that you pointed out. The 200k from the supply to the base allows it to be pulled lower than the emitter to turn the pnp ON, but i would have liked to see a 10k on the base as a current limiter.

The voltage of Gnd will be higher than B(-), but would it be enough to cause a current leakage if the Main FETs were turned OFF. [i also wonder if the FETs are powered OFF during "sleep" mode, and what turns them back ON?]

It would seem that the Gnd voltage would need to exceed B(-) voltage by the sum of the reverse PN junction voltage of Q1 E-B, plus the forward PN junction voltage of Q4 C-B, plus the reverse breakdown voltage of Q3 C-E, before any leakage would occur--does that appear to be a correct analysis or is there another path to consider?

 

[taylorb]

Yes the missing “R16” in the T2 circuit, or for the matter if it was missing in the T1 circuit, is less likely to cause a failure because Vcbo for Q29 (Q6) is quite high. Again given it is in one circuit and not the other indicates to me that the Ryobi engineers were trying to fix an actual or potential problem.

Yes when the FETs are off, Gnd > B- and is obviously designed to isolate the battery. As you imply there must be another circuitry that continues to power the controller (from B-). Without seeing the whole circuit we can only assume that there is a low impedance path (e.g. a capacitor) between B+ and Gnd in order to cause the failure.

So yes you are correct on the likely failure path: Q1 Vebo = 5V, Q4 Vbc(sat) = 0.9V, and Q3 Vce(sat) = 0.5V, the total of which is well short of 36/40V excluding the assumed controller supply voltage (noted above). This assumes that Q3 is on (which is dependent on the controller) in order to cause the failure. We can only assume there is a set of intermittent circumstances that cause Q3 to be on while the FETs are off (along with other possible conditions). The C6 capacitor on the base of Q3 along with C12 tells me that the engineers were trying to prevent this “race” condition, but this is a very short time (I calculate around 20uS for C6 and in my experience capacitors on the base of a transistor are not much use). The 10k resistor between Q3 and Q4 would be a better idea.

Of course, much of this is conjecture without knowing more information (e.g. a full circuit diagram). It should be noted also that I have found that these types of level shifting circuits are difficult to design and perfect, and there are ICs that do a better job.

 

[kennybobby]

Did you happen to see the schematic in post #8 . That plus the T1 and T2 circuits covers most of what's on there. i may have a more complete version in my kicad folder.

 

[taylorb]

Well you have been busy! C26 and C27, the power supply capacitors, would have to form the final path to the failure circuit. So I think we can definitively say what happened: a "glitch" caused Q3 to turn on while the FETS were off and this shorted Q1, Q3 & Q4 through the power supply capacitors from B- to B+. However as to what caused the sequence of events one can only guess and of course the transistors may actually survive the fault current.
In any case thanks for your work on the circuit which allowed recipients to fix their battery packs! May Ryobi pay attention and fix the problem!

 

[Oric 1]

A faulty battery for $ 10:

Nothing is happening.
All cells below 0.1 V. Pull up to 3.6V with a power supply, but always nothing.
No light when push on the capacity button, and flashing on the charger.

The U1 marking is 3710LN 090.4G 1850B
The U3 marking is Q822J or O822J Z34 06 18511A

Hi everybody,
Sorry for not being present enough, but i have a lot of work at home.

I have a half a dozen Ryobi batteries , all have capacity losses.

Else, for this new model, without success to repair it , i was about to unsolder the pcb.
I had an idea : unsolder the Power link, apply the short to RST, resolder power link and re-apply the short to RST.
All cells was above 3.0 V

Miraculously the battery has woken up !
The first led flashing when pushing on the test button.

Currently the battery is on the charger, and it's charging....

 

[d.vencius78]

hi. could any one please tell me how to reset it. Is it RST+ground. if yes for how long. The PCB is with one FET and on the side is 5 small holes the RST and oters. thanks

 

[Oric 1]

hi. could any one please tell me how to reset it. Is it RST+ground. if yes for how long. The PCB is with one FET and on the side is 5 small holes the RST and oters. thanks

Hi, yes on your PCB the Reset is made when apply a link between the RST contact to GND contact. One second is enough.

 

[Oric 1]

Hi everybody,
Sorry for not being present enough, but i have a lot of work at home.

I have a half a dozen Ryobi batteries , all have capacity losses.

Else, for this new model, without success to repair it , i was about to unsolder the pcb.
I had an idea : unsolder the Power link, apply the short to RST, resolder power link and re-apply the short to RST.
All cells was above 3.0 V

Miraculously the battery has woken up !
The first led flashing when pushing on the test button.

Currently the battery is on the charger, and it's charging....



View attachment 122332

View attachment 122333

After several balancing cells with my T240 li-ion charger, the battery increased up to 4.4 Ah. I used an electronic load to verify the capacity.

My process is :

• Charge the battery with the original Ryobi charger.
• Discharge the battery at 6A with my electronic load, until the BMS stop the discharge
• monitor each cells with a multimeter , and verify the delta voltage.
• if the delta voltage between cells discharged is more than 50 mV, the battery need to be balanced.
• i solder the wire balance to the half of the battery, on the differerent test points. The battery is assemblied with 2 packs of 5S2P.
• I charge and balance the half of the battery with my T240 charger
• When balancing finished i solder the wire balance to the other half part of the battery, and i charge and balance the other side.
• At the end charge, i discharge the battery with the electronic load.
• And i check the voltage cell at the end of discharge.
• If delta a always present, i try to rebalance with an other max cell voltage ( 4.20 instead 4.15 )
• After 2 or 3 charge discharge cycles, the battery recover capacity.
• For the last test , i desolder all wires and charge with the original Ryobi charger.
• This battery was 3.2 Ah before balancing, and 4.4 after

 

[d.vencius78]

Hi, yes on your PCB the Reset is made when apply a link between the RST contact to GND contact. One second is enough.

Thanks. Should I use the resistor? or just wire?

 

[City]

That is quite a puzzle; i would have thought it was the mower shutting off using the T2 possibly due to over-temperature, or the pack shutting off for over-temperature. But when the problem disappears just due to cell change, it somewhat excludes the over-temp fault.

Possible theory: Maybe there is a low-voltage cutoff level, and the weaker cells let the pack drop too low under high load. When the load was removed the cell voltage bounced back up above the threshold, but then tripped again when under high load.

The 5AHr pack appears to still have good capacity when tested at 1C, but i suspect that the mower has a higher current draw and will bring cells down faster to hit the mower's LV shutoff point, maybe..?

i have a couple of "bad" packs that will charge to 4 bars, then run the mower or chainsaw for a period and cut off. The scale shows 1 bar; then i let it set for a bit and shows 3 bars, and starts back up to run for another period. Similar to your experience? i call them "bad" packs because they were discarded for recycling due to not working; i opened and charged the low cell(s) individually, and then they work and charge as normal except those low cells are likely weak and causing the LV cutoff.

Another idea: these cells have vents and internal protection devices with a thermal cutoff. A spec sheet that i found indicated no explosion and no fire requirement. Maybe that device is turning ON and disables the pack for a brief time, then resets itself?

That is quite a puzzle; i would have thought it was the mower shutting off using the T2 possibly due to over-temperature, or the pack shutting off for over-temperature. But when the problem disappears just due to cell change, it somewhat excludes the over-temp fault.

Possible theory: Maybe there is a low-voltage cutoff level, and the weaker cells let the pack drop too low under high load. When the load was removed the cell voltage bounced back up above the threshold, but then tripped again when under high load.

The 5AHr pack appears to still have good capacity when tested at 1C, but i suspect that the mower has a higher current draw and will bring cells down faster to hit the mower's LV shutoff point, maybe..?

i have a couple of "bad" packs that will charge to 4 bars, then run the mower or chainsaw for a period and cut off. The scale shows 1 bar; then i let it set for a bit and shows 3 bars, and starts back up to run for another period. Similar to your experience? i call them "bad" packs because they were discarded for recycling due to not working; i opened and charged the low cell(s) individually, and then they work and charge as normal except those low cells are likely weak and causing the LV cutoff.

Another idea: these cells have vents and internal protection devices with a thermal cutoff. A spec sheet that i found indicated no explosion and no fire requirement. Maybe that device is turning ON and disables the pack for a brief time, then resets itself?

The first Ryobi I got was a mower that someone put out at the end of their driveway. I was excited when I saw that the battery was included, but no charger. Fortunately the battery showed 4 bars full charge!!! yahoo! I took it home and it had the same symptoms as yours. It would run for 5-10 mins then turn off. I would release and retry the switch and get a few more seconds. If I let it sit for a bit it would work for a few more minutes and die again. I found the problem to be a few cells that were getting hot. I was able to power the mower with my e-bike battery no problem, so I pulled apart the battery and found a few cells in the ryobi pack were defective/fubar. I believe that one cell of the two parallel cells was defective/fubar and eventually caused its mate to become less affective as it was doing all the work and getting hot. If you are getting hot cells while the rest of the pack is relatively cool the battery pack probably has bad cells. I recycle the defective cells and am reusing the other cells in other projects. I was working on a project of rebuilding the battery packs with a different BMS and salvaged cells.

Here are some of my projects.

https://www.youtube.com/channel/UClzPIppZHLRt1j0muxKdaZQ/

City