Toro Workman MDE conversion

Batteries have arrived, I've made up a temp bus bar out of some aluminium flat bar for the initial balance charge. 3.2V 1600Ah, wouldn't want to drop a screwdriver on it!

https://imgur.com/a/6o6j5A8

Just waiting on BMS and charger. Charger might be a few weeks

Looking at mounting the box to the kart frame with these (or similar) to reduce damaging vibrations as I bounce over the old ex-vineyard humps on the property.
https://www.bunnings.com.au/goliath-30-x-40mm-dome-cylindrical-rubber-isolator_p4230094

Battery box built

https://imgur.com/a/IlTpuF9

Have configured as 4S4P temporarily and attached to a gel charger for a few days while I wait for the variable power supply to arrive. It's just hit the country so hopefully I'll see it soon. Will then reconfigure to 16P and finalize the balance charge.

Power supply has arrived and I have been performing a balance charge for the last few days. Have the power supply set at 3.45V and it has been hovering at around 8-10A output without dropping (bank is wired as parallel, ie 1600Ah). Pack is sitting around 3.35V.
Around 1.7kwh has been sent to the pack plus whatever I put in with the 12V car charger before I started metering it, may stop the charge soon and do a final 3.65V charge just before I finalize assembly (don't want to leave it fully charged for too long).
Pack is 5kWh and I'm not sure what percent charged the cells arrived at, but they arrived with 3.275V each, am worried about overcharging it, even at such a low input voltage.

https://imgur.com/a/uNbcTN5

Wired it all up yesterday with BMS etc and took it for a spin, runs like an absolute animal now, goes up hills like they aren't there, seems to peak at around 185A draw with just me in it on a moderate 30 degree hill. Drove it to a mate's place and we took it up a 45 degree hill together and hit 300A and it went up without breaking a sweat.
Continued on to the horse feed shop (5km round trip) and loaded around 100kg of feed bags in the back and drove it back home, still sitting on 79% charge.
I'm giving it another full charge to confirm it is actually full, BMS seems to be doing a good job of keeping cell balance in check.
Will put up some photos and videos soon once I have prettied things up a bit.

Some pictures

https://imgur.com/a/pH8l9tN

Just need to do a few tweaks to the voltage limits for charging etc, currently it is set a bit conservative. I'd also like to upgrade the white battery monitoring wires to thicker ones so that the current shunting works better, I think they are a bit thin.
All in all a successful project, thanks everyone for their advice!

Nice. That should stop the corrosion around the battery box too.

Cheers

I've picked up a 2000W pure sine 48V inverter (4000W peak), it appears to run my compressor, power tools, heat gun and other items easily, problem is it is so big I can't mount it anywhere on the kart!

But it could run the 650W server rack, or a fridge for many hours if there is a power outage, as it has a 5kwh pack.

https://imgur.com/a/jpPw2rc

Also mounted some 72W LED spotlights on top for when I need to cut firewood etc at night.

Finally mounted up the inverter, here's a quick video of the BMS and inverter doing their thing, with the air compressor running off it.

https://imgur.com/a/Kux2ZZB

Next step is to install some LED headlights, 6x4 inch. Something like this:
https://www.aliexpress.com/item/HJYUENG-led-work-light-4-X-6-Inch-Rectangular-60W-LED-Headlight-For-Peterbilt-Kenworth-FREIGHTLINER/32856513706.html?spm=2114.search0104.3.210.74dc299dkMP22W&ws_ab_test=searchweb0_0,searchweb201602_2_10065_10130_10068_10547_319_317_10548_10696_453_10084_454_10083_10618_10307_537_536_10131_10132_10133_10059_10884_10887_321_322_10103,searchweb201603_61,ppcSwitch_0&algo_expid=f07fe968-6d2c-4382-9815-f1d4621ecc02-33&algo_pvid=f07fe968-6d2c-4382-9815-f1d4621ecc02&transAbTest=ae803_4

My girlfriend parked up the kart yesterday and then went to use it later in the afternoon, it makes all the right noises (reverse beep etc) but the main contactor doesn't click on, and getting 4 flashes on the dash.
Tried bridging the 2 large terminals on the contactor to allow current flow but this didn't work.
Applied 48V to the contactor and it clicked over fine and reads 0 ohms when energized, so it appears OK. So it seems the Sevcon isn't outputting a contactor charge to energize it. It's now just giving 3 flashes which indicates it is likely faulty

So it appears that I have a faulty Sevcon Millipak controller. Does anyone know where I can get one of these affordably? Most seem to be around $700-1000US
It's a model 633T46301.
Are some of the other visually similar models interchangeable?
I think these work by switching things to ground, vs switching to positive, so not sure if I can change to another brand of controller.

If I was to buy a new Sevcon, would it need to be programmed also?
Does anyone maybe have a good one they are selling?

I've pulled it apart and it's full of that black potting epoxy, so I'm going to have a hard time getting to the innards to try and repair it otherwise. Has anyone repaired one successfully?
Might cross post this in a general area too.

I tried looking it up. All I could find was MILLIPAK, MOTOR CONTROL, 500 AMP, 48 VDC

Do you know if the motor is separately excited or permanent magnet? There are many compatible controllers, like Alltrax or Kelly.

But check all the sensors and switches before giving up on it. There should be a way to find out what the fault light is indicating.

It's a Sevcon Millipak 633T46301. Toro's heavily marked up version is 110-2569.

I'll check the motor brushes tonight, and double check the resistance of the contactor coil to confirm it is 126 ohms. The contactor does work when energized manually though. I've already removed the controller from the kart but I might reinstall it to facilitate testing.

Have a suspicion it is the diodes in the controller that manage the contactor, but it's going to be a major undertaking to open the unit due to the potting compound they used.
Pricing for a new controller is $2700Aus so unless I import one I'm up the creek

There is no magnet in the motor, it's a brushed DC model I believe. Has F1/F2 and A1/A2 connectors on it.

Just rechecked, only getting 1 volt between Pin 8 (contactor) on the Controller, and the 12V switched pole. The controller is supposed to drop pin 8 to ground when ready, and it doesn't seem to be doing this, it just floats up to positive voltage. Tried fooling it with a resistor but no luck.

Going to have to try and get the potting out and disassemble
Motor brushes look good and contactor reads 123 ohms on the coil, within spec.

Motor is a separately excited from the description. These are quite common on golf cars, so your local golf car repair place may have one that's reasonably priced and compatible.

If the potting is the hard epoxy kind, heating to around 80-100C will make it like rubber and you can dig it out with a small screwdriver.

Any luck with decoding the blinking LED code? There must be a manual that describes it somewhere.

I looked around a bit. Kelly does not make a sep-ex controller. Alltrax does though. XCT 48500 would be a 48v, 500A that would work with your motor. It would take some rewiring to fit the connector on the Alltrax.

Here's one I found on Amazon. $630.00usd. Still expensive but maybe less than the OEM Sevcon.
https://www.amazon.com/Alltrax-XCT-48500-Motor-Controller-XCT48500-IQ/dp/B016V10UZQ

Similar Sevcon $750.00usd:
https://www.electricmotorsport.com/...3t45303-millipak-sem-controller-48v-500a.html

But I found a manual for a similar Sevcon:
http://www.thunderstruck-ev.com/Manuals/PowerpaK SEM & MicropaK_M.pdf
10.2.5Diagnostic LED - This is mounted between the connectors on the front of thecontroller. It serves as a simple diagnostic tool as explained below:
Constant illumination - No fault, normal condition
LED extinguished- Internal controller fault
1 flash- Personality out of range
2 flashes- Illegal start condition or illegal steer switch inputs.
3 flashes- MOSFET Short Circuit
4 flashes- Contactor fault or Motor Open-Circuit
5 flashes- Not used
6 flashes- Accelerator, Steer Pot or Speed Probe wire off fault
7 flashes- Low or High battery voltage or BDI cut-out operating
8 flashes- Over temperature
9 flashes- Contactor coil s/c12 flashes- CAN bus fault

I'm pretty sure the codes will be the same on yours. The controller is sensing open circuit on the motor. You tested the contactor already so you might just have worn out or dirty brushes on the motor. Sometimes just spinning the motor by hand a few turns can get it going.

I bridged F1 to A1 one last night and then hooked up F2 and A2 to a bench power supply dialed to 50V 20A, and the motor spun, and upon inspection the brushes look OK, almost new judging by the amount of meat on them.
Measuring A1 to A2 is less than an ohm, same for F1 to F2, so everything seems normal there.
I've started removing the epoxy with a conical burr tool and some acetone. Tempted to try the oven method however the unit is rated to 70C (non operating) and I'm not keen to bake it at 150-200C. Steam is another method I have considered too.
I've ordered another identical unit from overseas for around $700, just waiting for them to confirm they have received my order. Hoping it doesn't need to be reprogrammed to suit any field characteristics for the motor, or any Toro-specific stuff etc.

I contacted Alltrax yesterday but they said they didn't have anything suitable. I think the link you just sent is not suitable for series wound motors. I was looking at the Alltrax SR-48500 but it would involve adding a big 2-way contactor to handle forward/reverse, after taking a detailed look at the wiring diagrams. The Sevcon handles that internally.
https://www.cartszone.com/alltrax-sr-48500-500-amp-36-48-volt-speed-controller

Was initially getting 4 flashes on the controller, but this progressed to 3 flashes once I manually energized the contactor. Suspect I fried something in the circuit when I did that as the magic smoke was released. I'd say there are dead diodes in the contactor circuit, and dead mosfets which would have been the original fault.

New controller has arrived at work, can't wait to get home and fit it! It's a generic one with the same model number, hopefully there is no Toro-specific programming in the old one.

Installed the controller in the kart last night finally, great to have it up and running again! Had to adjust the accelerator stop button to eliminate a 2-flashes issue.
For some reason this new controller appears to draw a bit more current than the old one, so maybe it has some different motor parameters in it. It also seems to have less regen, likes to coast a bit more gently to a stop, but maybe it will self adjust over time.

Great you got it running again. No risk in trying to tear apart the old controller. My experience is most of that potting stuff gets like rubber when heated to 100C and that should not be hot enough to damage any components. Boiling water is one way. A heat gun and IR thermometer is another way.

Yeah might give that a try sometime, nothing to lose now.
The controller seems to have recalibrated itself and now drives the same as the old one. Must be some pretty clever electrickery going on inside.

Please be sure and report back with any new findings...I have a Toro MDE also and am going thru the same exact trials...
I would surely appreciate any insight or info you can help with...Currently I have the 3 flashes of death on my controller...
Also VERY interested in details of what it took to swap over the batteries since I'll probably be replacing them soon too...
I commute on an ES inspired/built E-bike everyday and love the thought of doing the same techno swap to the Toro!
Thanks in advance...Andrew

3 flashes might be the same issue mine had. I'm hoping it was just old age (2600hrs) vs the marginally higher lithium voltage.

The details of the conversion are on the previous few pages, but basically:
Remove lead acid batteries
Remove rust remnants
Spray the bay
Put in a Formply floor (or whatever you want to use)
Buy the cells - Leaf cells might be a cheaper option for you if you can find some, or 18650's. I chose the set of 16x 100Ah CALB LiFePO4 cells from Lithium Lab as I wanted warranty and minimal dramas, with good stability/safety. Cells were $2500 alone.
Build a "compression" battery box, as per: http://nordkyndesign.com/category/marine-engineering/electrical/lithium-battery-systems/
I used Formply for mine, with threaded rods and large washers to hold the 2 ends on. Used L brackets to attach the sides.
You'll then need a battery management system and 2 contactors. Zeva sells the BMS16v2 which works very well. I'd recommend getting the optional screen, the 500A shunt, and the 2 contactors from them.
Assemble it all, wire it all up with suitable gauge cable (Jaycar etc have this).
A suitable charger would be advantageous also. I used a Delta-Q with a custom lithium profile loaded into it. If you go LiFePO4 I'm happy to share it with you.
Preferably balance all the cells in parallel before assembling the pack, but if they are in a batch of sequential serial numbers you might get away with just assembling it as is, and let the Zeva do it.

Feel free to contact me if you want more info, or if you are in Australia you can always ring me, PM for number.

So I was driving up a hill with the pack around 70-80% charged over the weekend, and the Zeva BMS started beeping at me and cut the kart off from the pack, complaining about a low cell. Reset the BMS and took the remainder of the hill more slowly to not stress the pack, and checked it all over.

Appears cell #16 (the one the negative terminal goes to) is sagging heavily under load.
I recharged the whole pack using the charger, and then manually topped off cell 16 using my bench/lab power supply, dialed to 3.6V 20A.
Once current dropped down to near zero I immediately cut power and went for a drive, but the battery is still sagging far below the others under load. Idle voltage matches the other cells though. Seems it has developed a high internal resistance, or lost capacity.

Would there be any specific reason the last cell in the string/series would fail, or just coincidence? Vibration from banging around the farm? Would I have any warranty comeback with the seller, as I only bought these cells in March? The cells are 100Ah and typically see less than 50-100A of load, 200A if I give it a good hit up a hill.

Here's a quick video of what I am talking about.

https://i.imgur.com/WwY0E3G.mp4

I'd certainly contact the seller.

You could switch positions of cell #16 and say #15. See if the problem follows the cell and is not in the BMS.

Regards,

major

Cheers, I shot them an e-mail this morning, will see what they say.
Might do a juggle of the pack tonight or over the next few days, I'd say the cells have 20-30 cycles on them.

I rearranged the cells in the pack, and moved #16 to #15. All seems to be working well now, so maybe there was a loose connection on the eyelet terminal that was doing the voltage measurements, or the bolt to the cell link was loose. Will keep an eye on it over the next few weeks.