Looking for a motor for an EV conversion 3

[BenDorward]

I'm going to be converting a class b motorhome, most likely a sprinter, into an ev vehicle. I heard the new electric Mercedes have a 84kw motor (120hp) with 300 newtons (220f/lbs) torque so i thought I would model my needs off their specs. I found a few motors that were close on the hp but not the torque. Any help finding other brands would be greatly appreciated. Why would i need that torque if ev conversions for vw bus's are 71hp,120f/lbs.

https://www.evwest.com/catalog/product_info.php?cPath=40&products_id=226

https://www.evwest.com/catalog/product_info.php?cPath=8&products_id=176

(120hp,162f/lbs)

https://www.evwest.com/catalog/product_info.php?cPath=8&products_id=171

(165hp, 190f/lbs)

 

[BrianPetersen]

You need to figure the number of cells by the storage capacity you want to achieve, not necessarily the power rating, then back-calculate the current per cell to confirm that the power/current rating is within an acceptable range.

From Google, it appears that a normal 18650 lithium cell has a capacity of 3500 mAh = 3.5 Ah (amp-hours). Multiply by voltage 3.6, that gives 12.6 Wh (watt-hours). You said you wanted around 80 kWh capacity so that means it needs around 6350 cells in order to do that ... but it has to be an integer number of the number of cell modules. 300 V battery pack / 3.6 V = 83.3 i.e. 84 cells per module, as you stated. 6350 / 84 = 75.5 modules connected in parallel, pick something close to that depending on your battery pack layout.

When you are drawing 70 kW that's 70,000 W / 300 V = 233 amps. That's how much your inverter, power cables, contactors, etc need to handle continuously (more, intermittently). If you have 75 modules in parallel then it will be 233 / 75 = 3.1 amps through each battery, each series-connected module, etc. That is close to the 1C rating, which is good.

 

[BrianPetersen]

Doing some digging, the Panasonic NCR18650B cells appear to have a max discharge rating of 6.8 amps (continuous 2 amps), so you are ok on the power rating as long as you are not using 70 kW too much of the time. They weigh 47 grams, so your battery pack will contain about 300 kg of batteries ... less than I thought. (The structure that supports all these batteries and conducts current between them is going to weigh something, too.)

 

[BenDorward]

edit: nice, i didn't see your last post when i first read this. Yeah, definitely, i'm stoked to try and wire the batteries myself. I'm really concerned about the weight, especially with the current size of the engine. Yeah, I know the housing will take up some weight but i still think it'll be a better solution than buying the Nissan leaf batteries. Thanks for digging around, I'm feeling a lot better about where i'm at at least with the battery portion of the build.

https://www.ebay.com/i/223541997965...YerH33c11-ZRfPcud9XSLP8pasw3HReIaAseLEALw_wcB

edit2: ok i see where your going with this. 3200 mah/47 grams = 68 mah per gram. 80000 wh at 3.6 volts is 22,222 ah / .068amps = 326794 grams or 326 kilos. haha, nice! somewhere above i was like 1200 lbs or something, this makes me feel a lot better!

The other thing i'm thinking about is volume. 18650 is 18 mm by 65mm. With housing, if thats 20x70, even 5k batteries is 1 meter by 3.5 meters. that's like the size of a large coffin. haha, yikes!

Another thing i wanted to calculate was mileage on the flat with Air resistance. I want 200 miles range more than i have a strict battery size. i found air density times the drag coefficient times area all over two, and then multiply by velocity squared is

((at 4000 feet) .8194 X (typical suv drag) .4 x (height(8.3) x width(7.5) x .85))/2)x(80feetpersec^2)
*this is too much work* and i don't know if i'm in newtons per second or just...whatever

you said 20-30kw on air resistance earlier. if it's on the flat that's the only force acting so at 25kw and 4.2 volts thats 5952 amps so if i had 5952 batteries at 3.2 amps each i could drive for 3.2 hours at 55mph so 160 or so miles?

Just to confirm, 3.1 amps for a 84 cell module wouldn't be .04 amps per cell because their run in a series. The triple the c rating just means i have to run the engine at a third of it's max most of the time. I feel like if that's the case i should still play it safe and get the grid design holders instead of the honeycomb and also get the recommended heat shrink and insulators? Let me know if you think any of this is unnecessary.

By the way, thanks for helping with all of this. It'll be cool if this ever comes to fruition. I'll be sure to take a lot of pictures.

panasonic 18650 batteries:

https://www.evwest.com/catalog/prod...ucts_id=483&osCsid=polp2sp45rmac13reu79758t00

grid plastic:

https://www.wish.com/product/59e5b8...7wEALw_wcB&source=related_products&position=2

honeycomb plastic:

https://www.wish.com/product/5b6940...iDeiAx5dGm8wZshnOEdnSB5RGp9OcpP4aApjLEALw_wcB

heat shrink:

https://www.aliexpress.com/item/328...XNBKqdxif-hZ8DpJtxkaAujrEALw_wcB&gclsrc=aw.ds

insulator cardboard:

https://www.wish.com/product/594203...X9TP7JgF23sXyFmsAvfXssm4UjQ8vPEIaAhXYEALw_wcB

more expensive insulators:

https://www.18650batterystore.com/1..._wbXaRIb2B9amoPpaiGVG0cF35uZmN1saAtUAEALw_wcB

 

[gruntguru]

Buying a battery pack from a wrecked EV is much simpler than building your own.

je suis charlie

 

[BrianPetersen]

A home-grown battery pack is unlikely to have the thermal management, cell charge-balancing, and safety features against fire and electrical shock and impact and penetration of an engineered battery pack. Once these things catch fire due to thermal runaway, there is no stopping them.

EV fanatics have touted the simplicity of EVs for years. They're not simple once you are done making them work properly. The Nissan Leaf doesn't have thermal management of its battery pack ... and they degrade. Better EVs (i.e. those developed after Nissan's experience ... I don't say that Nissan "learned" because the new/reworked Leaf still doesn't have it) all have battery heating/cooling systems with coolant pipes and pump and radiators and some sort of tie-in to the HVAC so that they can piggyback their thermal management on top of the HVAC system. And their batteries have generally been holding up better.

 

[VEBill]

"... I'm looking at buying a hell of a lot of 18650 cells and connecting them together..."

That doesn't seem to be a common approach.

[GG] "Buying a battery pack from a wrecked EV..." is the more common approach.

If you're going to build a battery pack from scratch, perhaps winding your own motor and building an inverter next?

Highly recommend you spend some time reviewing how many others have done similar projects. YouTube is a good starting point.

 

[RodRico]

BrianPeterson,

Unsprung weight... why not mount them inboard with a CV joint to the wheel?

Transmission... I believe motor torque is a function of rotor diameter and max motor RPM is a function of diameter and rotor stress. A high RPM motor will naturally have smaller diameter due to rotor stress, and that will limit its torque compared to a larger diameter motor turning at lower RPM. A larger motor with no transmission should have less friction loss and certainly has lower complexity. It would be interesting to plot the relations between RPM, diameter, torque, and rotor stress to determine whether a large diameter motor at low RPM produces more or less torque than a smaller diameter motor at high RPM with a transmission.

 

[RodRico]

For reasons of cost and quality, I would buy a battery pack rather than make one. There are dealers selling battery packs and management systems, but I doubt any would be more affordable than simply buying replacement battery packs from Tesla (or some other manufacturer). There are a good number of them with prices on E-Bay, but I think many of them come from wrecked cars, so I'd be very careful about buying there.

 

[BrianPetersen]

Inboard with a CV joint to the wheel is exactly what everyone building production EVs does!

With regards to your cost/weight/efficiency analysis, a large electric motor needs to contain lots of copper and iron to produce high magnetic field strength in order to have lots of torque. Gears (with a physically smaller but faster-spinning motor) contain a lot less metal. That's why everyone who is serious about this, does it with a centrally mounted (to the chassis/subframe) fast-spinning motor, gear reduction, and CV joints to the wheels. Whether there is a single motor on each axle with a conventional differential, or one motor dedicated to each wheel, is a design choice dictated by cost, the desire to provide an electronic limited-speed-diff or torque-vectoring effect, etc. Two separate motors also means two separate inverters plus enough smarts in the electronics to handle the torque splitting between them.

 

[BenDorward]

gruntguru - I've looked at EV batteries and I really wish they came with a warranty. To solder these batteries, looking at other videos of people making them, I'm looking at around 180 hours of work at around 15-20$ per hour; the price jump from the next cheapest pre-built option which are the nissan leaf batteries. I'm also looking at reducing a bit of weight doing it this way. With my budget and amount of time on my hands i can't afford not to take the work.

VEBIll - I've looked at 18650 soldering and it doesn't look too difficult. I imagine building my own motor would require me to machine some parts, which i don't have the tools for. It would also be a lot less of an issue if a module failed than if my motor had something wrong with it. The motor i'm looking at has a 1 year warranty. If the EV batteries had the same warranty i would consider them a lot more.

rodrico - thanks for doing the search! Those packs, are around 10g for 30kw hours. That's a little more expensive than the reclaimed Tesla batteries. It's good their new, but i didn't see a warranty. The Panasonic batteries I'm looking at are 8 watts per dollar. Tesla batteries are about half that, at 4.5 watts per dollar. The cheapest pre-made batteries are the nissan leaf modules at 6 watts per dollar but they'll weigh a little over 11 hundred pounds whereas the Panasonic batteries will only weigh 700 plus plastic housing, nickel, and whatnot.

ugh, don't talk to me about cv joints. the rubber casing for the joint on my civic has split and my mechanic said best fix is just to replace the axle. all that work taking off the brakes just to slide on some rubber. for now i've resolved to just avoid dirt roads.

being that this is a rear wheel drive, I'd connect the motor directly into the driveshaft unless there was a reduction gear or some type of transmission? I'm doing research as we go but i know i'm still under educated. I have a friend that's been a mechanic for four years and is looking into his own EV project this spring so we'll work together and i'll probably pay him to consult. I still haven't heard back from ev west about whether they recommend the Tesla model s drivetrain in a Mercedes Sprinter. The model s drivetrain has a 1 year warranty, i think through ev west, but i'd probably need to get their OK on it first or it they might void the warranty. I wouldn't know where to start changing my gear ratio but i could probably live with going only 55 on the highway if the van could get up hills ok.

https://www.evwest.com/catalog/index.php?cPath=44&osCsid=9dp2ms0mnpfgcc88hsc7v9vqg1

 

[EdStainless]

A lot of trucks used to have two speed rear ends built into the differential.
Often in low range your max speed would be about 40-50mph with lots of grunt, and in high you could reach 1.5x, or 1.66x that speed depending on ratios.
I would buy batteries with the cooling and charging systems worked out. Trying to get that all to balance and work might be fine if you were just building a little commuter car, but if you want to go long range reliability needs to be there.
And to get range you are going to need a lot more battery.
Even on flat ground with no wind this van would likely need 25-30 hp just to maintain 70mph. That takes a lot of juice.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy

 

[VEBill]

"...18650 soldering..."

If you insist on building your own pack, then be aware that spot welding is actually the preferred approach. They sell (e.g. eBay) the appropriate strips of metal, and inexpensive spot welders for this purpose. Many kits and projects on this exact topic. It's the usual approach. Again -> YouTube is a good starting point to see what is the common approach.

To be clear, the spot welds are just a few mm apart on the same terminal, so the cell isn't even 'aware of' what just happened (almost no heating). As opposed to soldering, where the heating required to achieve a good joint is not really a good idea from the safety point of view. It's also faster and easier.

There are also 'clamp together' packs, but the pressure contacts are presumably not intended for high current applications.

--

There are vendors that rebuild battery EV packs, and would provide something like a one year warranty.

 

[moon161]

^ Many thousands of joints later, your unscarred fingertips and un-rosin vapored lungs will thank you so much for spot welding.

 

[moltenmetal]

Absolutely DO NOT make your own battery pack out of 18650s. That would be, frankly, a completely idiotic move.

Also refrain from purchasing what I purchased, which were LiFePO4 prismatic cells. They are selling for virtually the same cost today as I bought mine for back in 2014. You can in contrast buy a Chevy Volt pack, complete from a wrecker, for 1/4 as much money. Those cells have a higher power to weight ratio by a lot, come with a BMS that other people have already hacked/cracked so you can make use of it, comes in nice "bricks" of 24-48 V that you can recombine into new series/parallel arrangements that best suit whatever motor/inverter you end up using etc.

DO NOT attempt to run any conversion without a BMS. A BMS, with at minimum an alarm trip on high and low voltage on every cell (or every group of cells in parallel) is minimally necessary safety equipment for working with Li ion high energy density cells. If you go without a BMS, expect to have a fire, because YOU will end up being the BMS and the likelihood that one day you screw up is very high indeed.

As to re-purposing OEM EV motors/transaxles and inverters- it can be done. It's not for the faint of heart, and it doesn't go well in all types of vehicle.

I'm not familiar with drivetrains or suspension arrangements in sprinter vans so will leave it to others to comment.

A trip to

http://www.diyelectriccar.com

would be more useful to you than a trip to YouTube in my opinion, as someone who did his own conversion and who is about to undertake another.