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

[mswatosh]

Ah sorry, 55.8v not 58.8v, and I considered the main fuse, but if that was the case I wouldn't be able to get a voltage across the Relay right? The main fuse is the small box attached to the main positive battery lead, right?

 

[rio]

Yeah good point. Yes, fuse is on main positive lead.

I would also check that all the lug nuts are tightened. I can't imagine how the relay would suddenly go bad. You said you drilled through the panel without removing it. A metal shaving could be creating a short somewhere. Or like USJ suggested, maybe something got disconnected in there?

 

[mswatosh]

The metal shavings creating a short occurred to me, so I vacuumed out what I could, but still nothing. I can see voltage from the battery to the key terminals, as well as the driver selector terminals, so I think that confirms they're seated correctly? Nothing else looks disconnected from what I can see, but I'm also skeptical that the relay suddenly went bad. I'm wondering if maybe something went wrong in the key mechanism, but I'm not sure how I could test that.

 

[UltrasoundJelly]

The metal shavings creating a short occurred to me, so I vacuumed out what I could, but still nothing. I can see voltage from the battery to the key terminals, as well as the driver selector terminals, so I think that confirms they're seated correctly? Nothing else looks disconnected from what I can see, but I'm also skeptical that the relay suddenly went bad. I'm wondering if maybe something went wrong in the key mechanism, but I'm not sure how I could test that.

Try testing the key switch using this method (page 22).

Ryobi RY48ZTR75/100 Riding Mower Electrical Components Troubleshooting and Testing

 

[Jakeman]

Hi there, did you end up going with this route and if so, can you please share your experience and any tips or URLs of components you used? I got a RM480E for 800 bucks and 29 hours on it. I've mowed 2 times with it so far, but this morning when I was mowing I noticed that the blades would dis-engage and then start again, repeat. Was wondering if the batteries are starting to fault out. Shows that it's charged.

This just happened to mine. Blades would cut out even with a charge of 85% - 90%. Lifting the deck appeared to possibly affect it as well.

Over the winter I had the deck removed and upon re-installation I believe I mistakingly swapped the motor wires side to side. At the time it didn't seem like it would matter. Upon closer inspection it appears that the wires became stressed when the deck was moved up or down. I switched the wires and mowed yesterday with no issues.

I would check the wiring, especially since it was a used mower. I was ready to blame the batteries (original) but had a hard time believing it would cut out with a high level of charge.

 

[eleven_brews]

Thanks to all in this thread, I was able to complete the conversion of my RM480e to LiFePO4. I used new, US-sourced EVE 105Ah cells and an Overkill 100A 16s 48v BMS. I’m using the original DeltaQ SC-48 charger that came with the mower and have set the BMS with a cell high voltage cutoff of 3.600v. I’ll probably end up using an outlet timer so that the charger doesn’t keep continually erroring and cycling upon BMS cutoff.

Max current seems to be about 80A, which I can only achieve by engaging the mower blades while going up a small hill at full speed. Typical mowing current is in the 50A range.

I had a hard time finding 4awg m4 ring terminals or lugs so I ended up using dual 8awg between the 4 4-cell modules I created, the BMS and the terminal studs. I also ended up replacing the included m4 terminal bolts with stainless grub screws that I secured in place with red loctite. I did not want to chance stripping out these tiny m4 terminals.

We’ll see how the first mow goes.

 

[Hedges]

It would be preferable to limit charging voltage, and have BMS disconnect only as last-resort protection.
I don't know if you have any way to adjust charger output voltage. One idea I've had is using a few diodes in series to drop voltage.

Maybe you could rig up a circuit that senses battery voltage and disconnects A.C. to stop charging at target setpoint.

But voltage regulation so charge current drops to a trickle at lower voltage would give BMS more time to balance cells. How much is needed, I don't know. Probably matters more when battery is older rather than new.

We understand lithium batteries degrade faster when stored fully charged. So, recharging to 3.6V immediately after use wouldn't seem as good, better to recharge shortly before use.

 

[eleven_brews]

It would be preferable to limit charging voltage, and have BMS disconnect only as last-resort protection.
I don't know if you have any way to adjust charger output voltage. One idea I've had is using a few diodes in series to drop voltage.

You have a good point about using the BMS as a last-resort protection. I'm not sure I can just drop the voltage on this charger. It doesn't just charge with a constant current or voltage, it has a pre-programmed charging profile that is specific to 75Ah-100Ah AGM lead acid batteries.

Maybe you could rig up a circuit that senses battery voltage and disconnects A.C. to stop charging at target setpoint.

I found a couple different inexpensive charge controllers that would allow me to do just this. They basically control the connection to the charger with a 20A 6v-60v relay. It allows a user set high voltage cutoff and low voltage cutoff that would disconnect/connect the charger at set battery voltages. To your other point about not charging to 3.6v max cell voltage, I'd maybe set this at something like 52.0v (3.25v/cell) since 105Ah is way more than I actually need for 1 mow. According to the spec sheet on these cells 3.25v would be around 80% SOC @.5C (not sure exactly how that translates to idle voltage).

https://www.amazon.com/gp/product/B07F14QRZY

But voltage regulation so charge current drops to a trickle at lower voltage would give BMS more time to balance cells. How much is needed, I don't know. Probably matters more when battery is older rather than new.

We understand lithium batteries degrade faster when stored fully charged. So, recharging to 3.6V immediately after use wouldn't seem as good, better to recharge shortly before use.

Unfortunately I'd need some kind of adjustable charger for that. Does anyone have any recommendations on good, relatively inexpensive adjustable/programable chargers?

 

[eleven_brews]

First mow on the upgraded RM480e went well. From a full charge (to BMS max cell voltage cutoff of 3.65v), cell voltage went down to about 3.35v, which is higher than 80% SOC according to the EVE LF105 spec sheet. The previous 75Ah lead acid batteries seemed to barely make it a full mow when they were new, and recently wouldn’t make it more than 3/4 the way through our 1/2 Acre yard.

Here’s a few pictures of the build.

 

[wattmatters]

First mow on the upgraded RM480e went well. From a full charge (to BMS max cell voltage cutoff of 3.65v), cell voltage went down to about 3.35v, which is higher than 80% SOC according to the EVE LF105 spec sheet.

Except for when the battery is almost fully charged or almost fully depleted, voltage with LiFePO4 cells tells you almost nothing about their state of charge. The voltage is so flat from 20% to 80% SOC it's really not a good way at all to assess the SOC of a LiFePO4 battery.

 

[eleven_brews]

Except for when the battery is almost fully charged or almost fully depleted, voltage with LiFePO4 cells tells you almost nothing about their state of charge. The voltage is so flat from 20% to 80% SOC it's really not a good way at all to assess the SOC of a LiFePO4 battery.

I guess I didn’t realize voltage is a completely inadequate way to determine SOC in the 20%-80% range. Do the discharge curve charts on the datasheet only apply when under load?

 

[Hedges]

There is some series resistance, so under load or charging would change the voltage.
Maybe a small temperature effect, but not like for lead-acid.
A slight error in voltage measurement or curve data would cause you to make a larger error in SoC estimation.

Knee of the curve near full and near empty serve to recalibrate SoC; between those points you need a device that counts coulombs.
Battery shunt and a microcontroller integrating current x time does that.
Perhaps there are some clamp DC sensors (Hall effect) that would accomplish the same, possibly less accurate but non-invasive.

Yup:

Battery Monitor Hall Sensor 6in1 300V 200A Voltage Current Power Capacity AH SOC | eBay

Power Module. Battery Monitor. Resolution 0.1V, 0.1A, 0.1AH, 0.1W, 1WH. USB Monitor. Battery Monitor_UK. Battery Monitor_AU. Charging status indicator, looks like a smartphone. This Battery Monitor is designed to monitor performance most kind of battery systems with capacities of up to 999 Amp...

www.ebay.com

 

[gwang_hoops]

You have a good point about using the BMS as a last-resort protection. I'm not sure I can just drop the voltage on this charger. It doesn't just charge with a constant current or voltage, it has a pre-programmed charging profile that is specific to 75Ah-100Ah AGM lead acid batteries.

I found a couple different inexpensive charge controllers that would allow me to do just this. They basically control the connection to the charger with a 20A 6v-60v relay. It allows a user set high voltage cutoff and low voltage cutoff that would disconnect/connect the charger at set battery voltages. To your other point about not charging to 3.6v max cell voltage, I'd maybe set this at something like 52.0v (3.25v/cell) since 105Ah is way more than I actually need for 1 mow. According to the spec sheet on these cells 3.25v would be around 80% SOC @.5C (not sure exactly how that translates to idle voltage).

https://www.amazon.com/gp/product/B07F14QRZY

Unfortunately I'd need some kind of adjustable charger for that. Does anyone have any recommendations on good, relatively inexpensive adjustable/programable chargers?

does anyone have good numbers for SOC % (80%, 90%) versus a constant charge voltage for maximizing LiFePO4 longevity?

 

[Hedges]

We've heard 40%, for long-term storage.
That's why I suggest holding off recharge of the mower until just before use.

 

[Ampster]

I charge to 3.45 but they stay there or settle until sun goes down then inverter pulls them down. I agree with above for long term storage.

 

[Hedges]

My thinking is if you mow once or twice a week for 1 hour, the rest of the time counts as long-term storage.

I've taken to no longer putting my lithium tool batteries on charger as soon as they run down. I have at least two of each kind, so I used to always have one in the tool, one fully charged.

Cell phone I don't leave plugged in, only do that if it gets low. And I avoid using Apple charger most of the time, figure others charge more slowly.

Laptop is always plugged in, so I'm at the mercy of their battery management software. I have an old Dell M70 I just pulled out of retirement for PV equipment monitoring; it is asking for a new battery. Picked up another recently to be our juke box. 20 year old PC, pretty reliable.

 

[gwang_hoops]

NVM

does anyone have good numbers for SOC % (80%, 90%) versus a constant charge voltage for maximizing LiFePO4 longevity?

ignore this stupid question...for those who want the perfect charging voltage from their charger, I finally found an awesome thread on LiFePO4, longevity, charge settings, etc. my takeaway as a Ryobi owner for me (and not in general):

- install an accurate coulomb counting battery monitor
- charge only when needed (watch the Ahr, keeping the number of "full" charges down)
- maybe I'll charge switch to a reduced charging voltage of 3.4V/cell (54.4V, it'll take longer)
- every 6 months (just to be safe) kick off a rebalance by recharging my 12V batteries "fully"

one could argue that with a extremely reduced longevity of say 2500 cycles (even though most of us won't even care at 80% capacity), that'll be 48 yrs at once/week mowing -- don't care.

https://diysolarforum.com/threads/b...mum-battery-stress-5-000-10-000-cycles.34813/

 

[wattmatters]

I guess I didn’t realize voltage is a completely inadequate way to determine SOC in the 20%-80% range. Do the discharge curve charts on the datasheet only apply when under load?

I don't know what those charts say but LiFePO4 does not have the same clear relationship between voltage and SOC, in say the manner that lead acid or even lithium ion batteries do, which have a more distinct linear slope. With LiFePO4 you really need to count the charge flowing in and out, and every so often recalibrate that measurement at 100% SOC when you fully recharge them (with a charger designed for charging LiFePO4).

Here's a constant discharge curve example for LiFePO4 at different C rates. You likely are not looking at >1C, at least not often, so check the curves for say 0.2C to 1C and note just how flat they are. The lower axis is relative SOC (1=100%).

https://www.researchgate.net/publication/244669937_Approaching_Theoretical_Capacity_of_LiFePO4_at_Room_Temperature_at_High_Rates

In a mower application the C rate will naturally be quite variable as the machine moves about at different rates, stops and starts and deals with different cutting loads so voltage will bounce a little depending on the load.

The main thing to note is how little the voltage varies over a very wide span of SOC.

As result of the very flat discharge (and charge) curve, the smallest error in voltage measurement can result in a large error in estimated SOC.

Fortunately this battery chemistry shows much sharper changes in voltage at each end of SOC (<5% and >95%) and this is what allows the charge and discharge controls and battery management systems to protect the battery from over charge and over discharge. But in between lies much opportunity for guesstimation errors if relying on voltage alone.

 

[gwang_hoops]

I guess I didn’t realize voltage is a completely inadequate way to determine SOC in the 20%-80% range. Do the discharge curve charts on the datasheet only apply when under load?

my owners manual says "The voltage needs to be tested at rest (with zero current) after 30 mins of disconnecting from the charger & loads."

I think the resting voltage gives you a gross estimate of SOC to tell if it's empty, full, or can be used some more. Really need to have a shunt battery monitor and reset the Ahr counter whenever you charge anywhere near full.

 

[eleven_brews]

There is some series resistance, so under load or charging would change the voltage.
Maybe a small temperature effect, but not like for lead-acid.
A slight error in voltage measurement or curve data would cause you to make a larger error in SoC estimation.

Knee of the curve near full and near empty serve to recalibrate SoC; between those points you need a device that counts coulombs.
Battery shunt and a microcontroller integrating current x time does that.
Perhaps there are some clamp DC sensors (Hall effect) that would accomplish the same, possibly less accurate but non-invasive.

Yup:

Battery Monitor Hall Sensor 6in1 300V 200A Voltage Current Power Capacity AH SOC | eBay

Power Module. Battery Monitor. Resolution 0.1V, 0.1A, 0.1AH, 0.1W, 1WH. USB Monitor. Battery Monitor_UK. Battery Monitor_AU. Charging status indicator, looks like a smartphone. This Battery Monitor is designed to monitor performance most kind of battery systems with capacities of up to 999 Amp...

www.ebay.com

My biggest concern right now is being able to safely and reliably charge to a reasonable SOC between mows and I’m still not sure how to best do that. Are there any inexpensive chargers that can be paired with a coulomb counter to charge to a SOC of <80%? With 105Ah at my disposal and the occasional use a mower gets I’m not concerned if charge current is relatively low.

Using the existing DeltaQ charger with an external charge controller would work to charge to a specific pack voltage and have the BMS act as a fail safe in case the external charge controller failed or an individual cell went over-voltage. However as has been alluded, going by pack voltage is going to be a rough estimate at best. Probably fine in this case, since I just need enough charge to get through a mow, but not ideal or accurate.