Ryobi Zero-Turn Mower SLA to LiFePo4 Conversion - (Updated - Build Complete With Pics!)

[AMDPower]

I got my cells! Just built my battery pack and will show pictures later but I have a question. All cells identify properly and are currently balancing with the overkill BMS. I see 16 cells. When I measure the OUTPUT of the BMS, it only measures 13V. The batteries themselves prior to the BMS measure 47.9V. All cells when I measured for assembly were between 2.96 and 2.98. They sat on a dock in Cali for a while I'm sure. My question is this...

Why am I only reading 13V on the BMS output? The BMS app says that says undervoltage error and is in the process of balancing. However, I can't charge because it appears that the BMS is shutoff. Do I just let it do its thing? Or do I manually "turn on" battery or what? What is the proper procedure here?

Thanks.

 

[AMDPower]

Hmm.. I just realized that the BMS software says Alert Cell Undervoltage Protection. I think this is a hardware cutoff. Reading through the BMS manual I might be able to lower the undervoltage release temporarily to "unlock" it which should then allow my charger to work as well. Maybe???

 

[AMDPower]

Yep! That was it! Cell must have dipped before during shipping. RTFM for the win! Charging up now!

 

[Short_Shot]

Yep! That was it! Cell must have dipped before during shipping. RTFM for the win! Charging up now!

Don't forget you need to charge *and* top balance the cells.

If you just slap them together you will have capacity issues with the bms cutting off inappropriately.

You may need to buy a cheap power supply to do this. There are top balancing videos around the net and tutorials here on this forum.

 

[AMDPower]

I just am glad I got them back in action. Doesn't the BMS handle the top balancing? What exactly is the process? They are all already within 1/100v.

 

[Short_Shot]

I just am glad I got them back in action. Doesn't the BMS handle the top balancing? What exactly is the process? They are all already within 1/100v.

It can do some of the top balancing but not a lot.

As for the voltage being close, that's just how lithium works. The voltage curve is pretty flat until you get to the extreme top and bottom of capacity, then it goes nuts.

You may find that when charged it's a LOT more than that.

 

[AMDPower]

It works!!!! Just got done with the first cut. Had 58% battery left after 1.25 acres! No bogging down or anything. I can do burnouts in my drive too! The weight of this rig is about the same as just one of my 4 SLA batteries. So pumped. Here are build pictures along with descriptions. It's my first build so if you guys see anything obvious that I should have done differently just let me know.

The biggest thing I would change is the enclosure obviously. I had planned on using a Pelican case and even purchased it but I just couldn't get it to fit right within the frame of the mower. I'm in the process of remodeling my kitchen and had a bunch of scrap maple cabinet grade plywood that was already sealed and finished on both sides. I used pocket screws and glue and then applied some adhesive roofing flashing I had around the edges. The hole is for a long bolt which goes through the existing tray to secure the battery to the chassis.



I got the cells from Michael Caro's group buy and used an Overkill 16S BMS. They are 100AH cells. The two larger groups are taped with Kapton. The cable that joins those blocks was used in my original SLA configuration. The reason for the layout is to allow the long bolt to pass through and secure it to the frame. I packed everything tight with foam as it's likely to get rocky in the mower. The bottom is lined with foam as well.



I used the existing quick connect cable and bolted the negative leads together and used shrink.


The box sits in a metal tray that can can be slid out on the existing rails. Here it is in the stripped down mower. Complete removal of the sides isn't necessary. I had some cleanup to do on the mower though.


Two bolts secure the tray in the rear and the one long bolt through the center. Drill twice, measure once..



Fully reassembled and charging.


BMS reading after full charge.


BMS reading immediately after full mow (1.25 acres).


Notes:
Because of the voltage and curve differences, the on-board battery level always reads 100%.

The charger had an error on it when charging was completed. It's designed for SLA though and is an intelligent charger. Might just be something I have to live with. Possibly because of the voltage level and impedence? Not sure.

I'm charging to full again and then plan to open up and inspect the pack and see if any errors or anything are thrown on the BMS during the process.

Anything else I should do? Thoughts? Also, thanks everybody for your help!

 

[Supervstech]

Very nice! I need to find me one of these.
Did you seal all the plywood edges?

 

[AMDPower]

Very nice! I need to find me one of these.
Did you seal all the plywood edges?

Thanks! I used that flex tape on all edges except for the top (as well as wood glue acting as seal). It's actual intent is for ledger flashing. It's from when I built my deck last year. I probably should have sealed the edges of the tops as well though.

 

[AMDPower]

One other question I have... Since I don't need anywhere near the capacity. Do you think I should change the max charge to something less than 100% to maximize battery life? I know this is the case with my car which uses Lithium Ion. Is LiFePo4 similar with max charge and longevity?

 

[Hedges]

Anything else I should do? Thoughts? Also, thanks everybody for your help!

Can you apply a test load (rather than driving the mower)?
Check BMS reported cell voltages under load. Check with DMM.
Ideally, check voltage from one cell terminal to the next which is shorted to it by busbar. That way DMM reads near zero volts for more digits resolution, rather than near 3.5V, but probably not accessible.

After a heavy load (e.g. test load for a period of time, or after a vigorous mowing run, check temperature of each busbar. Any hotter than others would mean poor contact.

One other question I have... Since I don't need anywhere near the capacity. Do you think I should change the max charge to something less than 100% to maximize battery life? I know this is the case with my car which uses Lithium Ion. Is LiFePo4 similar with max charge and longevity?

Typically LiFePO4 cells are balanced once up to 3.65V and some tail current, then charged each time to a lower voltage like 3.5V.
Lower voltage for storage would be less degradation. Holding at higher voltage allow cells to diverge enough to balance.

If your mowing session uses 50% +/- of capacity, I'd think that would be a good SoC for storage. Then recharge shortly before next use.


You didn't report on top balance experience. To what voltage do you charge? How far apart were the cells at that max voltage?
Goal is to have charge/discharge always stop without reaching limits where BMS disconnects (for either pack voltage or cell voltage & voltage differences)

 

[Short_Shot]

This makes me sad that I didn't buy one of these now.

I love my cub cadet but man. The quiet...

 

[Supervstech]

Since you now have a lifepo4 mower you could change the charger to solar, and have it power your garage, or house daily needs, a small 100W panel could easily trickle charge it in a week I'd think. For between mowings, and if more panel is needed, it wouldn't be difficult. Plug it into a setup in a shed or garage.

 

[AMDPower]

I didn't balance it correctly. I used too low of a voltage 3.2v and charged each cell manually. But now that I watched Will Prowse's video, I realize I should have connected them in parallel, charged at higher voltage and then disconnect when amps go down (I do have a 5A bench power supply). But now that it's already rigged up... Can the BMS be used to balance while off charge? What should I be looking for in the BMS. I did get a cell overvoltage error the first time. I set the full voltage on the BMS to 3.45 and when it hit that, I had a delta of 0.17. What is "normal"?

Thanks!

 

[Short_Shot]

I didn't balance it correctly. I used too low of a voltage 3.2v and charged each cell manually. But now that I watched Will Prowse's video, I realize I should have connected them in parallel, charged at higher voltage and then disconnect when amps go down (I do have a 5A bench power supply). But now that it's already rigged up... Can the BMS be used to balance while off charge? What should I be looking for in the BMS. I did get a cell overvoltage error the first time. I set the full voltage on the BMS to 3.45 and when it hit that, I had a delta of 0.17. What is "normal"?

Thanks!

If anything you'd want the bms to do that on charge as it most likely bleeds power to resistors rather than actively moving it.

But that's a question for the user manual of that bms, if there even is one.

You might not want to use that specific charger for the mower though if you can get a power supply on there for 3.5v per cell (56v) and just hold it there for a bit and see what the cells do.

It would be a pain to tear stuff back apart to balance it.

HOWEVER.

If you don't have issues with the BMS cutting off at whatever the highest voltage and full charge of your charger gives you, then chances are its close enough and you can just let the bms do the rest over time.


I'd consider a dedicated 48v lithium charger though.

 

[AMDPower]

Oh it's got a user manual! A very detailed 80 page one (https://github.com/FurTrader/OverkillSolarBMS/raw/master/Overkill_Solar_BMS_Instruction_Manual.pdf)!

Yeah, I think I'm just going to lower the full charge voltage and likely get new charger. The Overkill Solar BMS shouldn't need whatever BMS software is on that Ryobi charger. There is also thermistor that I can likely remove. that goes back to the charger.

I figure lowering the full charge to like 3.3 - 3.4 rather than 3.5 should be healthier for the battery and safer even if it is off slightly at the top. I just don't need the capacity. I got these batteries for the current rating mainly. And the lifecycle is definitely not going to be my issue. If I'm only doing 1/2 a charge cycle a week and the battery is rated for 3000 cycles...that's like...115 years?!!! I'm pretty sure something else will go wrong before then...like me an my offspring's death.

 

[Short_Shot]

Just make sure if the bms has a voltage cutoff for balancing that it's set just under your charger voltage and you should be golden eventually.

 

[Hedges]

Once your assembled pack is relatively fully charged through BMS, you can use a bench supply to top-balance by charging one cell at a time.
Set the supply to 3.65V and some current (e.g. 10A). You can observe CV while it is open circuit, then observe CC with the supply shorted. After confirming settings, connect to one cell and wait for it to charge in CC mode, switch to CV mode, then current tapers off to some target (e.g. 1A).
Do NOT turn up voltage setting while connected to the cell.
Top off each cell one at a time, and the pack is then top balanced.

But top balancing may or may not be needed. When your charger charges battery to its max voltage, at that point what are the cell voltages? If very different, needs balancing.

 

[AMDPower]

I'm running a full charge right now to check. Gonna check at a half hour and then full charge check the balance. I'll post. Thanks for all the advice!

 

[AMDPower]

OK. Here are the results. It reached full charge quicker than I thought. It's definitely not balanced but is this "way" off? Also, could I charge a single cell while in series with the pack or should it be completely physically removed?

 

[Hedges]

3.598 max, 3.381 min, 0.217V difference.
Kind of far apart.

No need to disassemble anything. Just stop charging pack, and charge each cell to 3.65V (stopping at some tail current) one at a time.

Bad leads and cheap alligator clips will drop significant voltage. Some good solid ones would be better.
Not assembled, people make leads with ring terminals and 16 awg to 12 awg wire.
You could use good clips, jumper cables? or whatever. Make sure they aren't going to short anything.

"56.25V max"
Is that the BMS disconnect? Would be 3.516V/cell average.
How high a voltage does the charger deliver?

 

[AMDPower]

My guess is 56V. I can't charge right now because it's at max. The Ryobi charger just says 48V-13A but it's obviously not charging at 48V max. Here is the rest of the information from the BMS readout. I really think I need to dump the Ryobi charger. Everytime I get a cell overvoltage protection and reach full charge, it send the charger into a fault state and I can hear a relay click. Also, does cell overvoltage protection in the BMS app just mean that it reached full or does it mean there is a problem?

 

[AMDPower]

Oh... to clarify I believe that IS the charge voltage.

 

[Hedges]

Image says "BMS overvoltage times 0"
Did it report cell overvoltage?

You want battery charger to stop at a voltage which doesn't cause BMS to disconnect.
Cell imbalance means some will get higher than others. You want them to stay close enough and max voltage low enough that no cell goes overvoltage, and no "imbalance" error causing disconnect.

Top balancing should help, then try to set up a battery charger which charges high enough for rebalancing to occur, but not too high.
Maybe a nice SCC and PV array. As people said, your lawn mower could power your house during power outages.

 

[AMDPower]

It reported cell overvoltage protection. You can see it at the top of the first image. Ahh... I thought the BMS just handled that. I gotta pull the specs on that charger. It looks like it is actually made by Delta-Q.