DIY EV - Beginner's Guide - State of Affairs 2019 (contribute please!)

I'd like to go through and get up-to-date options (and maybe some historic options) for all major components and parts of the build process. The idea is that a newbie could come here and actually have a clue what to do. Current guides are like, 12 years out of date.

I'm not knowledgeable. I'm just going to spitball here and feel free to discuss and correct and add alternatives.

Donor Car:

- Not much has changed, same old advice is true. But something light and aerodynamic since it helps with range, though range isn't the issue it was with Lead acids.

Motor:

- Cheapest option is a DC forklift motor. Free-$200 anywhere in the world. This is still a viable build. There is a giant thread telling how, we should probably summarize and photo-journal those. These were popular 25 years ago to 5 years ago, and are still somewhat common today for converters.

- Impulse/Warp 9, 11, etc. DC still. Several thousand dollars. These are basically the same as forklift motors, there's little need to buy them, I don't think anyone's bought any for a new conversion in several years (?).

- AC forklift motor. I don't know anyone doing this other than me actually. Don't know how common this is.

- Hyper 9, etc. AC motor. Or other purchased AC motor like the Siemens? I don't know much about these at all or if anyone's actually using them.

- OEM EV motors. This is probably the biggest change since all the last guides were written. back then DIY EV was the only way to EV. Now it's just a way of making the vehicle you want to be electric into electric. What are the common choices?
- Nissan Leaf FWD.
- Tesla FWD (small)
- Tesla RWD (large)
- ??? ... and all these sections should have some links to some common guides and some quick notes on their suitability, power, cost-ish, complexity, etc.

Batteries:

- Lead Acid deep cycle batteries. Never do this anymore. They were the only option until ~2012 (?), everyone used these. Some used marine batteries, some golf cart batteries, some ordinary storage batteries. They are currently entirely outdated. There are zero reasons to use them, not performance, not cheapness, not durability, nothing. Don't even consider it.

- Large format Lithium-Iron-Phosphate (LiFePO4) batteries. These are recently, but now thoroughly outdated. They were popular from (? 2012-2016?). LiFe cells are lower voltage than other Lithium chemistries, and at the time were more durable and higher power. These were ordered typical by manufacturers like (? and ?), ranging in capacities per cell from (? to ?), and were physically large, like two hardcover books in size. You would generally have a single series string of these to add up to your total voltage. They turned out to not be as durable as expected, and are no longer cost effective in any way. Some DIY EVers will still buy them to increase voltage and storage of an earlier build, or to replace dead or dying cells, but no one converts with these in mind anymore.

- 18650-size Lithium Ion cells. This idea was popular in (year?) but almost never implemented or finished. Around this time 18650 format (thumb-sized) lithium cells became available for sale to ordinary consumers. Tesla uses this form factor for many of its vehicles because of their configurability. Small cells are more easily reoriented to match any shape. Typically dozens to hundreds of cells would be in parallel to make an equivalent large cell, and then many of those would be in series to create the pack. There is also a popular recycling movement to convert old laptop and tool pack cells into DIY Powerwall and off-grid storage, but for vehicles it's a long and difficult process. Very few people ever did this and it's no longer cost effective.

- OEM EV salvaged packs. While not available on the new market yet, salvaged packs from crashed or obsolete OEM EVs are a plentiful source of new packs, sufficient for the DIY EV community. Popular choices are:
- Chevy Volt (years?), large format cells with not great energy density, but having a reputation for high power (performance use) and durability. They're water-cooled.
- Nissan Leaf cells. First gen (years?) cells are prone to problems. Second gen (years?) cells are decent and commonplace. These are rectangular block cells the size of hardcover books, usually in cell series pairs (7.4v). They stack easily together and are clamped together with threaded rods inserted at their corners.
- Tesla packs. Tesla packs are difficult to disassemble groups of (? how many) cells in a block, with (number?) blocks in a typical car. Each block is fairly easy to separate but blocks are often ruined by attempting to access individual cells. They are typically more expensive than other OEM EV packs.
- ??? others, and, a moderate introduction should be available for each.

...

Anyway that's a start. If anyone wants to include missing options, fill in missing details, make corrections, etc. That'd be appreciated. I'll lump it together and start to edit the first post to fill that content in.

Oops, forgot controllers.

DC controllers:

- Forklift controllers. Cheap and readily available, but low voltage.

- 2 or 3 purpose build controllers generally from forklift or golf cart manufacturers.

- DIY community options
- Paul & Sabrina's (status?)
- Someone here was building a controller very simply a while ago, and then changed later to AC, I forget the name and this is all off the cuff or I'd be swamped in research.
- Who else?

AC Inverters:

- JHeuber's OpenInverter.org?

- Someone else built that DC controller that became AC?

- Someone today posted their open-source 4-man team that just finished an inverter?

- OEM EV as designed. Bring the whole EV brain and ECU and re-use the existing inverter.
- Leaf
- Tesla, various options that work or don't work, I don't know much about them.

- OEM EV hijacked. Damien is working on one of these for a Prius and Lexus inverter. In development. Open source.

... brain is mush. I'd like this to be a community effort, I'm just collecting the info and trying to make it beginner-friendly. I'm not actually much of an expert of what's done or why or what the options are.

major said:

I take issue with much of the second post. Please delete it. This is diy. It is not do it Matt's way.

Click to expand...

Hey, I'm sorry you hate me so much, but, you can either contribute in a useful way, or leave the discussion to people who want to have it rather than trying to dictate that other people aren't allowed to have it.

No, I'm not going to delete an entire post that is a loose, roughly-assembled beginner draft specifically for people to critique, because you "take issue with it" and somehow thing it's me trying to tell people to build things "my way", which... I don't even know wtf you're talking about, as I described 20 different options.

If you can't be bothered to at least be detailed about your criticism, and contribute useful corrections, then all you've done is show up and introduce negativity and a smug authoritarian denigration of someone else's efforts. It's not helpful, it's not useful.

Frankly, I think you hate me so much, and you're so biased, that it doesn't matter what I say, you'll find a way to say it's worthless.

It's fine if I'm wrong, it's fine if anyone's wrong. It's a discussion. I'm sure things will be corrected 100 times here, but no, I'm not going to be bluntly censored by a curmudgeon with nothing more than an opinion and a declaration.

Be a civil human being. You could use practice.

Been researching this recently, my contribution, which is welcome for fact checking:


Tesla Motor unit Options
Now at least 3 dealers selling them (EVWest, HSRMotors, Zero-ev.co.uk):

Large Drive unit (132kg, 475+ HP), 9.73:1, open differential
Rear Small Drive unit (88kg 260+ HP), rear refers to placement in Model S Dual motor model. Rear unit is wider. 9.34:1, open differential internally installed.
Front Small Drive unit (93kg 260+ HP), front refers to placement in Model S Dual motor model. Front unit is taller. 9.34:1, open differential internally installed.
Model3 drive unit not yet commonly available and different type of motor from Model S.

Comes with Invertor, Open Diff and Motor as one unit. Quaiffe LSD available from Zero-EV).
All models requires controller.

Controllers available:
HSR Motors/057 Controller board, replicates Tesla CAN and therefore same performance as stock.
EVBMW (Damien Maguire) has open source boards that can run above stock HP, characteristics.

Big gotcha so far is that the Tesla units are designed for mounting between wheels (eg in place of stock diff) which given their size, weight and deviation from ICE standard (engine in front, driveshaft to rear etc) is a legalisation impossibility in many regions around the world. Only Polykup (an independent conversion shop) has demoed a front engine bay install (in an Audi S5) by separating a large Tesla motor unit into its 3 major components (motor, inverter and diff).

brian_ said:

There is a wiki within this forum

If someone wanted to, they could try to fix the wiki articles (clean them up, then update them), rather than creating reference-style forum entries scattered through other categories.

Click to expand...

I've organized "what's the best way to get the best information to people the easiest way" in other communities a few times. How to transition out of where a place is and into what it could be.

Wikis are great, if there is a large enough group of people updating them with new information regularly.

The problem with this community is that we don't have that active team archiving things. And we probably won't. When there is new discussion or new things we learn (which is sometimes hard to identify, as it occurs gradually enough that the point where it is now "a thing" is never clear), no one adds them to the knowledgebase.

Further, many people with the most knowledge, by virtue of having the most knowledge, aren't *using* the wiki or archive or whatnot, because they don't need to. So they don't notice on a regular basis that it's out of date.

With wikis, it's very much that it *must* be active and up to date, because the answer to any question has to be "Check the wiki, it's always up to date". As soon as the answer becomes "You could check the wiki but it's old info so search for..." then there's no point. You might as well just let people keep asking the same questions over and over.

Another pet peeve is "Oh it's on my github". That's great for people who are capable of replicating your work by themselves (other EEs/programmers), and are only taking advantage of the fact that you've already done the work. It's utterly worthless for someone who's two tiers of knowledge below your and just wants to learn how to do something.

Another awful one is "Just go read this thread". Where the info gradually changes and you have to keep track of a conversation. That's garbage. A concise summary is what's useful, not "go replay an entire conversation".

Almost none of the info in many places is written the way that is useful to person it's supposed to help, typically because the people who can help aren't good at putting themselves in the shoes of the people who need help.

Anyway, a thread like this is a terrible way of archiving information, but it's the best way I could think of, for collecting that information. Maybe it'll end up in an article or pushed to the wiki or something like that in the future. But in terms of community participation, I think a discussion thread is going to work best so that's what I'm trying.

Most importantly, actually doing something rather than complaining about it is what gets things done, so, it's great to see people stepping up and chipping in where they can.

You REALLY want to help newbies at this? Post up a correlation between electric motors and ICE setups. They are NOT the same HP or performance, but there MUST be a correlation to help people get an idea on what they need to select to get the performance they want.

I'm an ICE gearhead from way back, and I've been researching a conversion for years now but I had all but given up. I worked out the details for a supercharged 427/8 speed for my project, but decided to take one more stab at electric. The only way I've been able to get my head even halfway wrapped around a conversion correlation was when I stumbled across a Tesla S motor up for sale that listed its power specs. From there I was able to get a point baseline for performance so I finally have some sort of an idea on what I want. Example, a Tesla S has a 0-60 time of around 3 seconds and a quarter mile time in the 10-13 second range, weighs in around 4300lbs and uses (according to the site selling one) a 400KW motor. I know the numbers aren't completely accurate, just what I remember them being. At any rate, I finally felt confident about what I think I need in an EV motor that I finally moved into the next phase, asking manufacturers for information about specific motors.

People can read the HP and TQ for an ICE vehicle, and know what sort of performance to expect. If there was a way to read the HP and TQ for an EV and know what to expect, more people might be willing to pull the trigger. So, for those of you who blazed a trail, please, go a step further and make a Rosetta stone. You built it so you know what's in it. A chart that lists everyone's car and its drivetrain specs and an analysis of how it drives would be great. Something in a standardized format with defined parameters to compare one to another in an apples to apples style format would make this a lot easier. A visit to your local dyno shop for a couple of runs, followed by a trip to the drag strip for times would be an even bigger help. This sort of information would bridge the gap between ICE and EV for the mechanially minded, who are the only ones who would be attempting a conversion. Those who aren't mechanically inclined would be more likely to buy an OEM. I realize this is both asking for a lot of work on everyone's behalf while also asking for a lot of money to be spent, but it would help drive more conversions. I can even suggest a few fields:
- Car weight before conversion
- Car weight after conversion
- Car cd (available online)
- Drivetrain specs
- Power specs
- Range
- Power consumption at WOT
- Power consumption on level ground at 10-100MPH in 10MPH increments
- Power consumption when pulling a hill (steep, moderate, shallow) at city/highway speed
- Towing capacity
- 0-60, 1/8 mile, 1/4 time (1/4 mile tracks can provide all 3)
- HP/TQ rated on a dyno. The dyno will read your EV like it is an ICE, and will convert it into a graph an ICE minded person can understand.
- Any other data fields you think would be useful

This information can also give you a baseline record for your own use, or if you want to see how someone else's controller or battery works with your motor so it's not all one way here. Everyone can benefit.

" There are zero reasons to use them [lead-acid], not performance, not cheapness, not durability, nothing."

Actually, there *are* good reasons to use them.

The problem with this thread is that you are pretending solve a fictitious/hypothetical problem, making design compromises, which is called "engineering", without defining the constraints and the problem set.

Example. I live in a cold climate. My usage model is to go into town twice a month for supplies, the vehicle sits there the rest of the time...outside, getting charged by [your favorite energy source...for me, it'd be a Plutonium-driven TEG, because a NASA buddy had one lying around, unused]. A lithium battery is probably the WORST choice and lead-acid is favored. Even for merely traction in the snow...

Or not needing to have an f'ing heater running 24/7, which makes an F-350 start to look economical with my mission profile.

Another example - the guy who got a free DC forklift motor, total noob. You threw him out with the bathwater by eroding his confidence through "obsolete" namecalling.

You can't win on this thread. Evar.

All you're doing is declaring yourself to be a winner, bolstered by calling any critic or counter-arguer a loser.

If there was one panacea, *everybody* would be doing it. Somebody would offer a complete kit (because, according to you, each block only has one optimal choice, cost be damned).

But there isn't - everyone's needs, mission profile, resources, and the fill level of their bags of money are different. If they weren't you could pick this and that and put together a "beginner's cookbook". Or offer a kit. EV West would have a long time ago. But they haven't.

There's also the Heisenberg Principle...mess with something and it changes. Two guys using an EV West motor, or battery, are very different than 400 guys seeking out that same motor from EV West, or even from a Tesla Model S, for example. Again -- resources. Availability. Cost. Speed. Torque. Living in Illinois vs Oregon. Alaska vs California. Room in the vehicle itself. Weight & balance. Duty cycle ("driven by a little old lady on Sundays, a quarter mile at a time"), etc etc etc

While you might think you're god's gift to a universal EV build, it's weapons-grade bullshit to even think there is one.

There's no such thing.

All you can do is put the choices out there and their specs/capabilities. Going beyond that, even in the best interests of a total rookie, by making the selections, or even listing strengths/weaknesses (like you did with lead-acid) without understanding each builder's constraints, problem set, mission profile, resources, skillset, etc, is doing them a huge disservice.

remy_martian said:

Actually, there *are* good reasons to use them. Example. I live in a cold climate.

Click to expand...

I understand your argument, I just don't think it has much merit. If other people agree strongly with you, I'm open to reconsider, but this is exactly the kind of thing I think should be nipped in the bud for beginners. That thinking that Lead-acid batteries in 2019 for an EV is a viable, useful option worth considering. I think it would make sense to include an aside or a link about why Lead-Acids used to be popular, and why no one in the community choose to use them anymore.

But, that's the takeaway. No one chooses to use them anymore.

I mean, there's technically viable reasons to include a horse and buggy as your method of transportation, but that's so unlikely to be useful information to anyone that "What vehicle should I drive" summaries don't include it. I suppose equestrian fanatics may be upset, but that does't make their advice useful.

Finland (?) is leading the world in EV sales. Finland is cold.

What do literally 100% of OEM EVs use for batteries? Lithium.

The case for Lead-acid in EVs is dead when the combined knowledge and engineering forces of the entire planet has concluded there is no reason to use them in EVs. To say otherwise is to think you've come up with something that the rest of the planet could not. There are some circumstances where the DIY solution differs from the new market, but I don't think that's the case you're making here.

Another way of thinking about it... Lead-acids are so non-durable that gasoline is more cost effective, even if electrical energy is free. Just in replacement cost over their expected lifecycle.

Another way of thinking about it... when's the last time someone build a DIY EV using lead-acid? And, if they did, was it foolish of them to do so? Of the people who may read this, what percentage of them is it better advice to get them to consider lead-acid, than it is to tell them it's almost certainly not worth considering and give them a footnote if they're really curious to read more?

Another example - the guy who got a free DC forklift motor, total noob. You threw him out with the bathwater by eroding his confidence through "obsolete" namecalling.

Click to expand...

Huh?

What guy?

Did you read what I wrote about DC forklift motors? Scroll up and re-read it, because you're arguing against something I didn't say. Here's what I actually wrote for my notes on the topic, in a draft. "Cheapest option is a DC forklift motor. Free-$200 anywhere in the world. This is still a viable build. There is a giant thread telling how, we should probably summarize and photo-journal those. These were popular 25 years ago to 5 years ago, and are still somewhat common today for converters."

Who's confidence did I erode? Where did I use the word "obsolete"? If I had (I didn't), how is calling a technology "obsolete" namecalling?

I don't understand your criticism.

I am generally writing observationally, based on what I see people building in the end, after they slog their way through all the options, and trying to make that path shorter for them.

You can't win on this thread. Evar. All you're doing is declaring yourself to be a winner, bolstered by calling any critic or counter-arguer a loser.

Click to expand...

Huh? Where are you reading that? This seems like you're arguing with a completely different person.

I'm not trying to win. I'm trying to kickstart writing a summary. I invited and welcome criticism. I never called anyone a loser for being critical. What I actually said was to be "usefully critical", rather than just demanding someone delete content with no explanation given.

If there was one panacea, *everybody* would be doing it. Somebody would offer a complete kit (because, according to you, each block only has one optimal choice, cost be damned).

Click to expand...

That is not how DIY works on large or expensive projects, so I disagree with your assertion.

There are hundreds of car options available. There is no sense in a generic "kit". The mechanical constraints of each donor car is fairly unique in terms of a solution.

Further, economic constraints matter too. What someone can find in their area and for what price is widely variable.

Just because there is not a kit, does not mean that a summary of the state of affairs of the meta of the current DIY EV community has no value. Quite the opposite. It means that there are some choices to make, that most people are leaning in a certain direction, and here's how to think about the choices available to you based on your preferences for performance, budget, suitability, etc.

While you might think you're god's gift to a universal EV build, it's weapons-grade bullshit to even think there is one.

Click to expand...

Honestly you keep putting words in my mouth and insulting me over things I didn't say, don't think, haven't implied, etc. I invite criticism and instead you're openly hostile.

What I did was kickstart a community project to address what I think is the biggest hole in our knowledgebase, and be the easiest way to fill that gap for people who join the community.

Telco said:

You REALLY want to help newbies at this? Post up a correlation between electric motors and ICE setups.

Click to expand...

That's something I'm happy to collimate but I'm too far out of the loop.

DIY EVs are at 2 extremes, generally:

1 - The frugal.
2 - The performance-minded.

I see a lot more of the frugal here, partially I think because performance-minded people are more secretive due to their competitiveness and due to the money involved. Damien ran into that a few months ago, two different groups of professional racers had the same problem with something he built, that he helped tweak to fix, but I think some of them took most of his solution and then finally figured out a result, and, despite being open-source, didn't want to share it. They wanted their competitive advantage.

I see a lot of value in someone contributing things from a performance perspective, that's important. But to be extra-specific goes a bit against the point of this guide. I think a solid overview of the basics is important to know what kinds of ballparks to expect (though apples-to-oranges is going to be a bit of an issue).

I could write up a very loose overview but I wouldn't trust my finger being on the pulse of that part of the community much. Happy to collect and incorporate whatever anyone else contributes though.