DIY Electric Car Forums banner

1 - 20 of 25 Posts

·
Registered
Joined
·
5 Posts
Discussion Starter #1
Hi all:

I've been fantasizing about building an electric car for years, and it's finally getting close to the time. I have built race cars before and certainly know my way around both the math and mechanics, but I haven't done an EV and my knowledge of how to do this is limited by the reading materials I can find.

So I have done a lot of research, read a lot of materials, including the "I want to build an EV! Where do I start?" thread right here.

And yet, I still have questions. Please indulge me or point me to the right resources, I'm not afraid to hear "All of the info you want is right here at XXXX, quit asking repetitive questions!"

Part of my problem is that I (perhaps mistakenly) believe that the state of the art is changing rapidly and that much of what I find to read might be older than the state of the art. Perhaps that's just hope.

Anyway:

To set the stage, the car in question (this month anyway, the fantasy changes from time to time) would be an MGA, circa 1960, front-engine RWD, solid axle. Light weight. The result doesn't have to be FAST, but it has to not be SLOW. I'm fine with a top speed (as in, maxing out the speed of the motor given the gearing) of 75mph. I would intend this to be a weekender that rarely sees freeway speeds, but I'd like it to be CAPABLE of it.

1) Sticking a complete EV drivetrain in there, like from a Leaf or a 500E or a Smart EV or something is too complicated, primarily due to packaging, and FWD vs RWD difficulties, not to mention too much reliance on the original electronics. I think the right answer is to use off-the-shelf individual components, not a complete drivetrain transplant.

2) Like everyone else, it would be awesome to do this with the simplicity of a single gear ratio, no real transmission, due to the space savings, lower drivetrain weight, and of course, elegant simplicity. But it seems that you STILL really can't do this with off-the-shelf components without a transmission, no matter what gear ratio you use in the final drive. I believe that to go with a single gear ratio you need 300V or more and a tremendous about of battery. Am I right, or has something changed?

3) Assuming one needs a transmission, I think one only really need a 2-or-3 speed (I'd love something tiny and built for EVs) but although I see a lot of product announcements, it seems there's still no market, and therefore still no products, so it's probably easier to find a semi-modern 5-speed that will fit. True?

4) Working out the numbers: for a conversion like this, it seems that the easy button is a motor like the AC-50 bolted to some sort of automotive transmission, going to a standard automotive differential. So, gear ratios in the "normal car range". Not missing anything, right?

5) Now, working on the batteries, and the disappointment: a motor like that is a 96V motor. Even several hundred pounds of LiFeMnPO4 batteries only gets a capacity of something like 6 kWh (total, and you can't even use all of it). Figuring out range, how many Wh/mile is a good, credible target number for a drivetrain like this? If it's like 500 Wh/mile (figuring all of this is way less efficient than the production EVs on the market), that yields a range of ... almost nothing, like 10-20 miles. Going to a higher voltage motor makes things worse. Going to a lower voltage motor isn't the right answer, is it?

So this is all very disappointing.

I thought that by now, here in late 2018, I'd be able to build a 2200-lb electric car that is usable and can go maybe 50 miles with off-the-shelf components. I hope I'm really missing something. Or maybe I need to wait another few years? Maybe for someone to make a decent hydrogen fuel cell? :)
 

·
Registered
Joined
·
5,050 Posts
2) Like everyone else, it would be awesome to do this with the simplicity of a single gear ratio, no real transmission, due to the space savings, lower drivetrain weight, and of course, elegant simplicity. But it seems that you STILL really can't do this with off-the-shelf components without a transmission, no matter what gear ratio you use in the final drive. I believe that to go with a single gear ratio you need 300V or more and a tremendous about of battery. Am I right, or has something changed?
You're right about the voltage, because that is what allows the motor to work well at high speed, eliminating the need for multiple transmission ratios. That does not mean that the battery must be enormous; the easiest sources of 300+ volt packs of a moderate size are plug-in hybrid cars.

3) Assuming one needs a transmission, I think one only really need a 2-or-3 speed (I'd love something tiny and built for EVs) but although I see a lot of product announcements, it seems there's still no market, and therefore still no products, so it's probably easier to find a semi-modern 5-speed that will fit. True?
I don't believe that there will ever be a significant market for a 2-or-3 speed transmission in longitudinal format (like the MGA) for EV conversions.

I agree that a lot of ratios are not needed, even with a low-voltage DC motor, so I don't see the need for even a 5-speed - the highest ratio seems particularly superfluous. Any sufficiently durable transmission that fits would work.

If you can live without a clutch, can find a transmission with a removable bellhousing, and can get the input shaft made shorter, you could mount the motor much further back than a typical straight swap into the engine position, leaving more space ahead of it (for battery). This wouldn't be much different from an EV-specific transmission. I haven't seen anyone actually do this, perhaps since just getting a proper adapter plate (to mount the motor to the bellhousing) and working shaft coupler is enough of a challenge for most DIY projects.

4) Working out the numbers: for a conversion like this, it seems that the easy button is a motor like the AC-50 bolted to some sort of automotive transmission, going to a standard automotive differential. So, gear ratios in the "normal car range". Not missing anything, right?
Nope - that's a popular formula, and the gear ratios work. :)

5) Now, working on the batteries, and the disappointment: a motor like that is a 96V motor. Even several hundred pounds of LiFeMnPO4 batteries only gets a capacity of something like 6 kWh (total, and you can't even use all of it). Figuring out range, how many Wh/mile is a good, credible target number for a drivetrain like this? If it's like 500 Wh/mile (figuring all of this is way less efficient than the production EVs on the market), that yields a range of ... almost nothing, like 10-20 miles. Going to a higher voltage motor makes things worse. Going to a lower voltage motor isn't the right answer, is it?
Changing voltage makes no difference to this situation. For instance, if you used 96 LiFeMnPO4 cells with 20 Ah capacity each, you could hook them all up in series to get a 6.1 kWh 300+ V pack, or parallel in sets of three then 32 of those sets in series (still the same 96 cells) to get a 6.1 kWh 102 V pack. It's the same energy either way.

So this is all very disappointing.

I thought that by now, here in late 2018, I'd be able to build a 2200-lb electric car that is usable and can go maybe 50 miles with off-the-shelf components. I hope I'm really missing something. Or maybe I need to wait another few years? Maybe for someone to make a decent hydrogen fuel cell? :)
All you're missing is that battery technology hasn't advanced enough to keep a sufficiently large pack from making the vehicle heavier than a similar car with an engine.

Hydrogen fuel is a huge topic by itself. :rolleyes: Don't hold your breath for that to ever make sense for a short-range vehicle; the only advantage of fuel cells over battery-electric comes if you carry more fuel (because the filled hydrogen tank is lighter for the same energy than a battery, and the fuel cell size / cost / weight is the same regardless of range). While a fuel cell is wildly expensive, it's not the range/performance problem with current fuel cell vehicles - the problem is the bulk of the fuel storage.
 

·
Registered
Joined
·
7,838 Posts
...
I believe that to go with a single gear ratio you need 300V or more and a tremendous about{amount?} of battery. Am I right, or has something changed?
...
Hi GDGR,

Things always change. I don't see how changes relate to your statement which I quoted. How does voltage equate (relate) to a single gear ratio choice? And how does either, voltage or # of ratios, relate to amount of battery?

Welcome to the board,

major

Ps. Before I hit the submit button I noticed brian_'s reply. I don't agree with him necessarily about voltage and motor speed, when you have motor choice. There are high speed low/medium voltage motors. And also it is a fallacy to believe you need a high speed motor to run direct drive.
 

·
Registered
Joined
·
5,050 Posts
Before I hit the submit button I noticed brian_'s reply. I don't agree with him necessarily about voltage and motor speed, when you have motor choice. There are high speed low/medium voltage motors. And also it is a fallacy to believe you need a high speed motor to run direct drive.
It's true that you don't need a high speed motor to run with a single reduction ratio (no multi-speed transmission), but to be effective you do need a wide range of motor speed. For instance, the Chevrolet Spark EV had an unusually low-speed motor setup, with about half the gear reduction (and so half the motor speed) of typical modern EVs, such as the Chevrolet Bolt which replaced it. They needed about the same voltage to make that Spark EV motor work at its highest speed well enough while still having sufficient performance at low speed, and the motor is relatively large.

Look at the power or torque versus speed curves published for the common aftermarket AC motors for EV conversions designed to run at under 200 volts - they die off at speeds where higher-voltage production EV motors are maintaining constant power.

Toyota uses a voltage converter between the battery and the inverters in their non-plug-in hybrids because the small NiMH battery can only reasonably run 200 to 300 volts, and the cost and complexity of the converter is worthwhile to them to be able to run the motors at a higher voltage.

Of course you can get a 2-volt motor that will spin a shaft at 20,000 rpm... it just can't do anything useful at 100 rpm. You can even run a motor big enough for a bus at relatively low voltage and still manage with only a single gear ratio, but the motor is very large and heavy for it's power level.
 

·
Registered
Joined
·
5 Posts
Discussion Starter #5 (Edited)
Thanks to both of you. I realize after re-reading what I typed that I was too verbose and therefore not clear even in my own head.

Given that 400 lbs of battery is going to get only about 17kWh of capacity (unless there's something state-of-the-art that I'm hoping you'll all tell me about) ... and my requirements are:
1) 40 miles of "normal use" range
2) 0-60 in, say, under 9 seconds
3) Can reach and maintain 75mph

What I should have said, is ... how is it that after all this time, there is still not a motor that can do better than about 425 Wh/mile in a lightweight but non-aerodynamic car, with or without multiple gears? Or is there and I don't know about it? Or is there a better retrofittable battery tech that I don't know about that has higher capacity for the weight? (I think the answer to this part is no.)

More analytically, it's an interesting subject to understand what the effect voltage of the system has on the simple question of efficiency.

Part of the issue is that while it's easy to look at motor specs and understand their max output and even how much power they draw at max output, it's really hard to understand the efficiency when using them in this application (i.e., at varying speed, at varying output levels). We don't use "full throttle" that much ...
 

·
Registered
Joined
·
5,050 Posts
What I should have said, is ... how is it that after all this time, there is still not a motor that can do better than about 425 Wh/mile in a lightweight but non-aerodynamic car, with or without multiple gears?
I think there is a motor; there are lots of them. Production EV motors and inverters are more efficient than common aftermarket equipment - especially antique brushed DC stuff - but motor efficiency isn't the whole problem. Competitive efficiency in urban conditions requires regenerative braking (which some but far from all DIY conversions implement), and high speeds call for modern aerodynamics. Even a common economy sedan today is aerodynamically better than sports cars of a few decades ago.

This is also a game of many details. For instance, there is probably no production EV which uses a hypoid gear set, but almost every DIY-style conversion that keeps the original final drive of a front-engine/rear-drive car has a hypoid ring and pinion in that final drive. That hypoid drive throws away several percent of the power put through it, compared to the non-hypoid parallel gear set in any production EV (and any transverse-engined vehicle).

Part of the issue is that while it's easy to look at motor specs and understand their max output and even how much power they draw at max output, it's really hard to understand the efficiency when using them in this application (i.e., at varying speed, at varying output levels). We don't use "full throttle" that much ...
This data is available for serious EV hardware; real auto manufacturers wouldn't design a vehicle or procure components for it without this information in detail. For instance, the Nissan Leaf efficiency map - showing motor efficiency at all combinations of speed and load - is widely published. The stuff sold for amateur conversions is not documented nearly as well; a frustrating example is the PMAC motor that NetGain buys - the manufacturer publishes efficiency maps, but for the specific version that NetGain sells as the HyPer 9 NetGain publishes nothing but a peak output chart. Yes, that makes selection and optimal configuration (such as gear ratio choice) difficult.
 

·
Registered
Joined
·
7,838 Posts
Why do you blame the motor for the inefficiency and load of the vehicle? Even if the motor was perfect, you still need ~390Wh/mile.
 

·
Registered
Joined
·
5 Posts
Discussion Starter #10
This is also a game of many details. For instance, there is probably no production EV which uses a hypoid gear set, but almost every DIY-style conversion that keeps the original final drive of a front-engine/rear-drive car has a hypoid ring and pinion in that final drive. That hypoid drive throws away several percent of the power put through it, compared to the non-hypoid parallel gear set in any production EV (and any transverse-engined vehicle).
I totally get it. In this presumed project I'd be willing to replace 100% of the drivetrain; but it would all have to be made to fit under a classic chassis (something between a '33 Packard and a '66 Morgan!). Basically it'd be a complete restomod, but with an electric motor instead of an ICE.

But in any of these cars I've considered, it would be pretty much impossible to do a transverse drivetrain. I think it could conceptually be possible to fit something really small under there with two back-to-back electric motors (so no diff), but to do a transplant from a transverse EV would require complete fabrication of a independent suspension, and that would leave precious little room for the drivetrain, never mind the height issue.

And these cars were not made to take a few hundred extra pounds of batteries. Their old cast iron drivetrains were heavy, but the math says the new electric system would still have to be considerably heavier.

But to your point about the hypoid gearset, a modern final drive & differential is a lot less lossy than the originals from any of these cars and I'd replace that.
 

·
Registered
Joined
·
5,050 Posts
... anything I'm interested in EV-ifying is old.
Old but still capable of cruising at 75mph and doing 0-60mph in 9 seconds?

Are we talking 1960's muscle car or a 1930s hotrod?
Neither...
To set the stage, the car in question (this month anyway, the fantasy changes from time to time) would be an MGA, circa 1960, front-engine RWD, solid axle. Light weight. The result doesn't have to be FAST, but it has to not be SLOW. I'm fine with a top speed (as in, maxing out the speed of the motor given the gearing) of 75mph. I would intend this to be a weekender that rarely sees freeway speeds, but I'd like it to be CAPABLE of it.
This is a classic sports car... but with the performance of an ordinary (not high-performance) modern car.
 

·
Registered
Joined
·
5,050 Posts
But to your point about the hypoid gearset, a modern final drive & differential is a lot less lossy than the originals from any of these cars and I'd replace that.
While I'll accept that a new final drive is probably better, in this layout of vehicle they're still hypoid, so there's still additional loss. Even with longitudinal component (engine/transmission or just a propeller shaft), a right-angle drive can be done with a simple bevel gear rather than a hypoid, but in the classic front-engine/rear-drive configuration a hypoid set is used to keep the pinion shaft (and so the propeller shaft) lower. It's not a big deal in itself, but everything adds up...
 

·
Registered
Joined
·
5,050 Posts
In this presumed project I'd be willing to replace 100% of the drivetrain; but it would all have to be made to fit under a classic chassis (something between a '33 Packard and a '66 Morgan!). Basically it'd be a complete restomod, but with an electric motor instead of an ICE.

But in any of these cars I've considered, it would be pretty much impossible to do a transverse drivetrain. I think it could conceptually be possible to fit something really small under there with two back-to-back electric motors (so no diff), but to do a transplant from a transverse EV would require complete fabrication of a independent suspension, and that would leave precious little room for the drivetrain, never mind the height issue.
I agree. In many cases, especially with smaller cars, the desired drive-unit-at-rear approach is just not very practical. That was my conclusion when I considered an EV conversion of our Triumph Spitfire.
 

·
Registered
Joined
·
5,050 Posts
And these cars were not made to take a few hundred extra pounds of batteries. Their old cast iron drivetrains were heavy, but the math says the new electric system would still have to be considerably heavier.
Eliminating the transmission would certainly help with that, and it can work to drive the stock final drive without the stock transmission, but it is a challenge in motor sizing and performance.
 

·
Registered
Joined
·
1,494 Posts
I achieved everything you seem to want with my E-Fire, with an 18.5 kWh 105 V pack and an AC50/Curtis- except my 0-60 was likely a little longer than 9 seconds. I had to limit the rate at which I allowed my motor torque to rise from zero to protect the weak differential. 0-30 mattered much more than 0-60 and it was PLENTY fast enough for my tastes, especially compared to the car pre-conversion. My ex-pack performance, mostly highway, averaged over at least 10,000 miles, was 235 Wh/mile- that gave me a comfortable range of 60 miles with 75 miles being possible on a wing and a prayer (and no bad cells- I had one Sinopoly go bad on me, gradually lose electrolyte (which I could smell in my trunk) and drop from 180 to 130 Ah for no apparent reason, but only the one)

It was expensive, though. Way too expensive in hindsight, now that my car was destroyed by an inattentive pickup truck driver after only 16,000 miles, and my insurance compensation wasn't what I put into the project in parts much less paying for the car or my labour. If I got another 5 years out of it, or could re-build it without paying $4100/yr for Facility Association insurance, it wouldn't feel quite so bad given how much I spent!

Mind you, you can get a Chevy Volt pack for less than 1/4 what I paid for my 18.5 kWh worth of LFP prismatic cells- and there are people who have helpfully cracked its OEM BMS. Those cells have a higher energy density per unit mass if not volume, and don't have any problem with power transfer given Yabert and Duncan's experiences with their big DC motors. That makes things a lot more affordable than what I spent.

Personally I think people make too much noise about wanting to get rid of the transmission. When I took my car to car shows, I got way more excitement from people who LIKED the fact that the car had selectable gear ratios- and it made a big difference to the performance and driving comfort. The tranny was well worth its weight and the little bit of loss. And though you can easily do without the clutch with an AC50, using a CanEV stock transmission mounting plate and coupling made the installation fast and worry-free. To me, it was pushing the easy button and well worth the money.

If you get a big DC motor, there's no need for the tranny- if you've got enough voltage to allow you to get to high enough speeds with whatever final gear ratio you choose. The voltage is necessary to battle back EMF which builds at higher speeds, if I recall Duncan's discussion on the matter.

Anyway- the big issue in my mind, aside from insurance which seems to be the big local issue, or RFI/EMI/compliance issues in some places in Europe, is still the cost of a conversion. It's fairly easy to be green, but it still takes far too much "green"...For most people, the best option is just to buy a used Leaf. Remember that most of the conversion community was doing this back before there WAS a Nissan Leaf available for sale new, much less a used one for $10k.
 

·
Registered
Joined
·
217 Posts
I skimmed most of your original post - and most of what has been posted, so I may be giving you some repetitive information. The biggest challenge that remains rarely spoken about is cost.

Off the shelf components are great - but costly, especially when looking at doing batteries. For your vehicle to get the range you are looking for and top speed, you would likely need 10-12kwh. I suggest this be done by obtaining a salvage Chevy Volt pack, and paralleling between two 6kwh banks of cells to give you ~150V and 90Ah. You will also need a 36S BMS to keep from burning your house down. A person could always go higher voltage, but from my experience as voltage increases so does cost.

Having the 150V will allow you enough RPM before you hit the back EMF of a brushed DC motor (forklift, ME1002, warp 9, something like that) so you can likely get by with a single speed and still get to 75mph (driveshaft from motor to straight rear axle). It won't accelerate extremely hard, but 1000A controllers are out there that would make it quick off the line. Kelly and ZEVA make examples that won't break the bank.

If you are able to optimize your final drive ratio in that solid rear axle, that makes life much easier. A person could always swap out another rear end that either has many different ratios (ford 9", 8.8".. etc). I'm not here to spoon feed every step but I like to try and help steer people on a reasonable path forward!

The electrical setup as described is what I plan on using for my next build - which will be tube chassis and extremely light using miata front/rear suspension.

Just my $.02, good luck!
 

·
Registered
Joined
·
1,151 Posts
how is it that after all this time, there is still not a motor that can do better than about 425 Wh/mile in a lightweight but non-aerodynamic car, with or without multiple gears?
I feel like the conversation skipped over this a bit. Your question has a bit of a false premise. I think that you're imagining driveline losses to be more significant than they are, and are thinking "Gee why haven't we put more effort into having better drivelines?"

The energy it takes to move a vehicle through air, on a road, at a certain speed, is the result of physics.

Let's say we have perfect materials science and a million years from now the most amazing engineering.

1 - We eliminate driveline/transmission losses to 0%. 100% efficient motors, bearings, shafts, gears, or whatever. Heck maybe we replace those things with different technology that is 100% efficient. Zero losses.

2 - We eliminate rolling resistance (tires squishing) to 0%. Maybe we don't use tires, we use anti-gravity. Or we make weightless vehicles.

3 - We eliminate air friction by having a magical transparent infinitely slippery paint that we can cover the car in (even the tires, while somehow still allowing them to grip the road!).

Unless I'm missing some, that is every possible improvement there could ever possibly be to the vehicle.

Even in that situation, as it is still by far now, the dominant impact of the power required is that you have to push air out of the way of the vehicle. The faster that you want to do that, the harder it is to do (and, non-linearly, it's a cubed function).

That is, say we apply these technological improvements to your car. If you have already selected the shape of the object you are pushing through air (classic car of some kind?), and you have decided on the speed you want that object to push the air out of the way (75mph), then other than change the density of the atmosphere itself, you will still have nearly the same power requirements as before.

Pushing an object through air a speed takes a certain power. Nothing you can do about it.

About the only thing left is to change the shape and size of the vehicle to make it move through air with lower power requirements, but your premise forbids that.



That's for a bicycle, which has very low weight and very high tire pressures (and therefor low rolling resistance), and a very efficient drivetrain (and therefor very low drivetrain losses).

But, you can see that areodynamic drag just dominates and is accelerating the faster you want to go.
 

·
Registered
Joined
·
27 Posts
I have a '78 VW bug that has been fully restored and converted to electric. But it uses lead-acid batteries and only goes 25 miles on a charge. But it would be a great platform for you to upgrade to better batteries, etc. See pics here: https://photos.app.goo.gl/wrtsr3PydwAieu9K8 I've other projects and need to sell the PlugBug. Interested?

Kerry
 
1 - 20 of 25 Posts
Top