Future of Tesla Boards

[Jack Bauer]

Thought I'd put the question out there. My desire from day one was to make a simple and affordable solution to allow people reuse Tesla hardware. Specifically drive units and chargers. All the while I've looked to reducing cost, increasing supply and making support material more wide ranging and accessible. Now, recently I've observed there seems to be little desire within the community as it stands now to take up the support role. With almost 300 boards out there as of now I did think the knowledge base and desire to help each other would be stronger. The support threads are little more than a public email chain between a customer with a problem and myself. So the question becomes what now?

[Kevin Sharpe]

So the question becomes what now?

Given you are supporting the boards today I think three options exist;

(1) Offer support as an hourly billable item in your store so that those who need it can buy it. We should still try and encourage the community to step forward but in reality this clearly does not work today.

(2) Stop developing and selling the Tesla boards because If you do not support them then in reality a lot of people are going to have a very bad experience. You've already published everything on github as opensource so in reality anyone who's up to the task can get boards built for themselves and take on the burden of learning how to use an Open Inverter board. Maybe in the future someone in the community will step forward and support the boards if you step back.

(3) Encourage people to buy from the commercial vendors like Zero EV. The open source solution is not an end user product.


** Note ** I think the future is very clearly the Lexus/Toyota development because it can deliver very low cost conversions on a standardised platform. I think focusing on that development and the training courses is where you'll have most impact and fun

[P.S.Mangelsdorf]

Thought I'd put the question out there. My desire from day one was to make a simple and affordable solution to allow people reuse Tesla hardware. Specifically drive units and chargers. All the while I've looked to reducing cost, increasing supply and making support material more wide ranging and accessible. Now, recently I've observed there seems to be little desire within the community as it stands now to take up the support role. With almost 300 boards out there as of now I did think the knowledge base and desire to help each other would be stronger. The support threads are little more than a public email chain between a customer with a problem and myself. So the question becomes what now?

My two cents: It may be that the community is not helping yet. Perhaps we need a bit more time to build up experience.

As an example, my board is in the mail, but I likely won't have a functioning car until August. My budget and schedule restrict how fast I can build. Until I have a car that can at least power on, I won't have any experience that I can use to help with.

Out of the 300 boards, how many projects are "done" or at least far enough along to help?

[Kevin Sharpe]

My two cents: It may be that the community is not helping yet. Perhaps we need a bit more time to build up experience.

The first Tesla inverter board design was published on GitHub in June 2017.

[tmotion]

A few years ago I attended to a ecu motormanagement and chip tuning course in The Netherlands of a guy who sells his own custom ecu. You could buy these with any sort of support for a price similar to the tesla control boards you sell.

Next to this 0-support ecus he offered courses:

- 1day introduction course in a group 250€
- 2day advanced course in a group 500€
- 1day private project specific consult 500€
- 1day in the car setup and live tuning 500€ (Ex travel)

Only if you attended to the majority of these courses support for the hardware was given.

His courses were wel-attended and great fun. People with these kind of projects were fine in spending the additional money (compared to total project costs it was minimal, like with using tesla parts in your ev conversion).

And for the owner of the company it was nice as well: he enjoyed giving courses, could easily earn some money with just selling hardware and in case he got any support questions, he knew the project, the owner and he was more or less sure that no stupid repeating questions would occupy his precious time.

I would be more than happy to support and pay for such a setup.

Another point: if I look around in car-diy-project-land average projects take at least 2 years. My ev conversion will take at least 2 years as well. I hope to buy a board somewhere this summer, but expect that it takes at least another 1 or 2 years before the car will be on the road.

[Kevin Sharpe]

So the question becomes what now?

Given you are supporting the boards today I think three options exist;

(1) Offer support as an hourly billable item in your store so that those who need it can buy it. We should still try and encourage the community to step forward but in reality this clearly does not work today.

(2) Stop developing and selling the Tesla boards because If you do not support them then in reality a lot of people are going to have a very bad experience. You've already published everything on github as opensource so in reality anyone who's up to the task can get boards built for themselves and take on the burden of learning how to use an Open Inverter board. Maybe in the future someone in the community will step forward and support the boards if you step back.

(3) Encourage people to buy from the commercial vendors like Zero EV. The open source solution is not an end user product.


** Note ** I think the future is very clearly the Lexus/Toyota development because it can deliver very low cost conversions on a standardised platform. I think focusing on that development and the training courses is where you'll have most impact and fun

I thought of a fourth option

(4) see if Johannes will take on the supply and support of the Tesla inverter boards given he already does that for 'his' boards

[P.S.Mangelsdorf]

My two cents: It may be that the community is not helping yet. Perhaps we need a bit more time to build up experience.

The first Tesla inverter board design was published on GitHub in June 2017.

I've got a project car I've been working on since April 2012 thats still not on the road. Thats not uncommon. I know my 1 year build is seen by alot of people (at least stateside) as pretty damn fast

I still wonder how many projects are "done" or at least driving.

[Jack Bauer]

Ok. I'll put up a few paid support options on the webshop.

[Kevin Sharpe]

I've got a project car I've been working on since April 2012 thats still not on the road. Thats not uncommon. I know my 1 year build is seen by alot of people (at least stateside) as pretty damn fast

I think that's changing fast as we get more 'lego' parts to use... three of the eighteen participants from last years courses will have cars on the road in six months... we also have a one day conversion target for Lexus and Prius cars

[Kevin Sharpe]

Ok. I'll put up a few paid support options on the webshop.

Your time is valuable and most people are happy to pay

[tom91]

Ok. I'll put up a few paid support options on the webshop.

I even gotten offers from people purchasing the SimpBMS to be reimbursed for my time. I have only taken them up on the offer when the required support/development was either beyond what I was planning on doing and if it was support that would be 'hand holding'.

In the end people have to realise, some already do, that ones time is their most valuable asset. You should be compensated for you spending time to reduce the amount of time they are going to spend.

[Kevin Sharpe]

In the end people have to realise, some already do, that ones time is their most valuable asset. You should be compensated for you spending time to reduce the amount of time they are going to spend.

Spot on!

[xp677]

I'd continue to offer the boards for sale and allow the community to provide the support. If they won't do that, then people will either buy the boards or not, it's their choice.

What I would suggest to the community, is that when answering questions, instead of answering here, they put it in the OpenInverter Wiki, and link to that. By ramming the Wiki full of information, people can be asked to check that first, and after a short while, they will be able to find 99% of their answers in there, rather than asking for 1-on-1 support. If they still need support, then it makes sense to commercialise it.

[57 Chevy]

I'd like to get involved with this project but as it stands the hardware doesn't support dual throttle input and I need that for compliance. It also needs the capability to operate dual-motor and provide torque limiting for traction control.

The reality is that I'm not going to have the time to head these as projects myself and I don't have coding skills to just jump in and get busy. I do however have the means to pay somebody else to do these which brings up an odd scenario- I could buy a proprietary system for far less than what it would cost to have a competent engineer make an open-source design- assuming a proprietary system will have the necessary functions that is. An experienced employed engineer should earn around 10k/month give or take and it is probably a 1-2 month project to complete the modifications I'd need. That buys a lot of other useful stuff too.

So, how much would I have to pay somebody to produce hardware and firmware to improve on what is there already? That would be my contribution to open source inverter

[Bratitude]

I'd continue to offer the boards for sale and allow the community to provide the support. If they won't do that, then people will either buy the boards or not, it's their choice.

What I would suggest to the community, is that when answering questions, instead of answering here, they put it in the OpenInverter Wiki, and link to that. By ramming the Wiki full of information, people can be asked to check that first, and after a short while, they will be able to find 99% of their answers in there, rather than asking for 1-on-1 support. If they still need support, then it makes sense to commercialise it.

This. I’d imagine a good portion of the problem is organization. We should be building the wiki not forum threads.

[Kevin Sharpe]

So, how much would I have to pay somebody to produce hardware and firmware to improve on what is there already?

Like any project you'd need to scope exactly what you need in a requirement specification. Once you have that spec i'd recommend submitting it to Johannes for a quote. Obviously lots of other engineers could do the work but it would take time to get them up to speed on both the hardware and software.

While you're working on that spec you could also support both Johannes and Damien on Patreon if you have some spare cash each month

https://www.patreon.com/evbmw/posts

https://www.patreon.com/openinverter/posts

[Kevin Sharpe]

I'd continue to offer the boards for sale and allow the community to provide the support. If they won't do that, then people will either buy the boards or not, it's their choice.

What I would suggest to the community, is that when answering questions, instead of answering here, they put it in the OpenInverter Wiki, and link to that. By ramming the Wiki full of information, people can be asked to check that first, and after a short while, they will be able to find 99% of their answers in there, rather than asking for 1-on-1 support. If they still need support, then it makes sense to commercialise it.

This. I’d imagine a good portion of the problem is organization. We should be building the wiki not forum threads.

Agreed

[MattsAwesomeStuff]

TL;DR - Urgency should cost money. Most people are no good as volunteers. Forums for discussion, wikis for information. Try to share so there's no reinventing the wheel.


In agreement with what some people said above...

I've said for, I dunno, 20 years, that discussion forums are great places to discuss things, but terrible places to archive things. Heck, even when I know that an exact post was made about an exact issue because I've repeatedly read it in the past, and I know who posted it... I have trouble finding it in the search. That's actually how I ended up at this thread. I'm trying to help contribute some support for the Toyota boards, and I've been looking for a specific post of Kevin's in the search for a half hour (this topic seemed interesting, so I started reading). To someone who doesn't know the right question to ask, or if their question has been asked, or doesn't have the context to know their answer when they see it, it's hopeless.

In other communities, any time volunteers have complained about getting the same questions over and over again, I've countered with "We're solving this problem wrong. 'Go search the forums' is a terrible solution. If we get the same questions so often, the solution is to put all the answers in the same place. If we're writing the same answers repeatedly, why not just write it up really well once, and post it?"

That does 2 things:

1 - It means fewer repeat questions because people will find the answers on their own. And,

2 - Because people just don't read anyway, a quick "Here's the link to that answer already written up in the wiki" is the only effort needed, not writing out the solution again.

So, let's try to make sure that we have the solutions archived in the wiki.

...

Now there's a slightly bigger problem. The people who create a solution and the people who use the solution, are different groups of people. It doesn't make sense to say "You bought a board, you benefited from this community, so stick around and answer some questions." Some people are going to buy, ask their questions, and then leave. Most in fact. We should expect that. We should expect most buyers to be drive-byes.

We had the same dynamic when I helped grow a local Makerspace. Maybe 10% of the community wants to make the Makerspace better. Add and fix tools, do administrative work, teach classes, give tours to newcomers, etc. The rest just want to show up and use it. In our community, we controlled who could join, and forced a (very) minimal level of volunteering to join. I.E. Before you join, you have to do more than zero volunteering. Sweep 1 room? Good enough. Schedule that you'll teach a welding class every month this year? Outstanding. Carry one bag of garbage to the dumpster? Better than the 30 people who walked past it today. Just to cross the line between "'Someone' should do that. That is an other person thing" and "I am a someone, and I can do that."

An open source movement doesn't have that level of control, nor does it make sense to. And, our freeloaders still paid membership dues so, they're still contributing. Most people you wouldn't want to help, they don't have anything to contribute. If they tried, they'd just give more headaches.

For example:

My SUV had a problem. So I went to the forums of that SUV. I asked complicated questions, I had guys troubleshooting my vehicle, then walking me through repairs. Took about a week, a hundred posts at least. And at the end I told them basically "I'm going to sell this when I'm done. I'm not passionate about this vehicle like you guys are. I'm just using you to get my answers. I'm not going to hang around and be part of this community. I can't give advice to anyone else. I'm going to take some good pictures of the process so you guys can reuse them, and where can I donate to keep the lights on?" I gave them like, 5% of what they saved me by taking it to a mechanic, and everyone seemed happy with that.

There's a small group of people here who actually want to help with these open source projects and it seems to be just because they want to. Few of them actually need to. Almost all of them have an EV already, they're not benefiting much. Everyone else just wants to pay some money and get some benefit.

The only solution I can see, is to just get them to pay for it.

...

Frankly, the companies that benefit from Damien and Johannes' designs, that turn around and mark them up 1000%... are the real ones freeloading on the community. If I was selling something that someone else volunteered to make, I'd be ashamed of myself to not share some of those profits and chip in as a thank you. And I'd be ashamed of asking a question without throwing money in the hat. They should at minimum be tithing back to the community as a thank you. But "Nearly Open Source - For Christsakes Look At How Cheaply I'm Releasing These Designs For" is not a tidy category like "Open Source" is.

...

I basically ignore most Tesla related threads here because I'm not using Tesla parts. But, in the Toyota forums my take is that the community is pretty damned helpful. If the answers are out there, people are willing to share them (lots of times the answers aren't out there). Maybe Tesla people are more pushy and demanding for answers?

So I have a solution for that...

Charge them.

If their questions are things that are documented that they won't look up? Charge them.
If they want modifications that suit them? Charge them.
If they are pushy about wanting a solution or a response right now? Charge them.

In contrast to the bums and grifters using Toyota parts, Tesla project people seem to me to be performance/racing minded, willing to spend money, and want things done right now. So just let them spend money to get quick answers. And really charge them. Charge them enough that you're not stressed and happily trade your time for their pushy answers.

If people are asking for help and don't need to interrupt anyone and can wait a few weeks to figure it out? Freeload away. Urgency is expensive.

...

The last thing I'll say is that, there's no point in the community reinventing the wheel. So anytime anyone ends up having information, share it. If you found a tech spec or a pinout or whatnot, share it. Don't make everyone else start over at zero.

And for those of you doing the reverse engineering making these notes by hand? Snap a shot of your notes and post it here, and just straight up ask a volunteer to add it to the wiki.

That's my long-winded two cents, having struggled at a coordinator level with volunteer-run communities for a couple decades.

[xp677]

Excellent reply, well said, and I think I agree with all of it.

If you're reading this, go log into the Wiki and spend 5 minutes adding in something that you know, but isn't in there.

[57 Chevy]

Like any project you'd need to scope exactly what you need in a requirement specification.

Yup. The basic spec is dual pedal inputs same as OEM Tesla has which would mean a board revision and some code to determine what was an acceptable signal and what to do if one or both signals went out of range. That would mean I can use this board and comply it here. The traction control could be developed later once I have hardware to test it on so that is considerably less important. Main thing is I can't use this board as-is without adding some sort of hacky throttle switch system.

One thing to note is what has happened to other open source projects. I am fairly confident Linux-kernel devices have been launched off the planet by commercial space operators and when I go to the electronics shop here they have run Linux on their sales system for at least the last 15 years. It is human nature to take what is free and put it to use, we have been doing this since the beginning of time. The other side of that is opening an EV conversion shop is a way to make a small fortune- on the proviso that one starts out with a much bigger fortune. Most of the EV conversion operators will be gone within ten years and the ones who will be making the profits will be running trade shows and training courses. That is where the real money is.

[Jack Bauer]

Thanks for the feedback folks. Much appreciate it and the help. I do want to address this point about the dual throttle. As I have said before, and this can be found by spending a few minutes looking through the parameters,the current firmware and hardware does support a dual channel throttle. Like Johannes I am a believer in simplicity. What he has achieved with a little base STM32 processor is quite amazing. IF that does not appeal to anyone for whatever reason then please feel free to work away on an alternative.

[johu]

I was gonna write a lengthy reply later but also want to address dual throttle. Zero EV sponsored that like 2 years ago. Thus there are pot and pot2 and you can choose to use that for redundancy by setting throtmode to dual.

[Kevin Sharpe]

I was gonna write a lengthy reply later but also want to address dual throttle. Zero EV sponsored that like 2 years ago. Thus there are pot and pot2 and you can choose to use that for redundancy by setting throtmode to dual.

I do want to address this point about the dual throttle. As I have said before, and this can be found by spending a few minutes looking through the parameters,the current firmware and hardware does support a dual channel throttle.

I guess the wider point is it's hard to know what the hardware/software supports and I guess that's an argument for the wiki (which I will now put effort into)

Like Johannes I am a believer in simplicity. What he has achieved with a little base STM32 processor is quite amazing. IF that does not appeal to anyone for whatever reason then please feel free to work away on an alternative.

I suspect many people don't understand that financing feature development is also an option (like the Zero EV dual throttle support)

[Kevin Sharpe]

The only solution I can see, is to just get them to pay for it.

I've made a great deal of money out of open source and I intend to do the same again. We should not be afraid of this.

[johu]

I do however have the means to pay somebody else to do these which brings up an odd scenario- I could buy a proprietary system for far less than what it would cost to have a competent engineer make an open-source design- assuming a proprietary system will have the necessary functions that is.

I think this is a logical fallacy. You can't compare the development cost with product cost. Of course it will be more expensive to have something tailormade ONCE. But as it is with open source you and everybody else can profit from the development afterwards. If you buy an off the shelf product you pay the higher price again and again. Still, to answer your original question: making little improvements (such as adding dual channel throttle support) took me like 10 hours, including a test drive. So say 10h * 70€ = 700€. Had you decided to sponsor FOC development that would have weighed in at about 300h or 21,000€. Annual Patreon donations payed for like 10% of that. Annual hardware sales maybe for another 25% of it. I'm cool with that because in turn I'm using Linux and LibreOffice and whatnot free of charge.

I think that is an issue that spawns in various threads. Active developers throw in hundreds of 1000s of € of unpayed development time and yet some people expect to get free support on top of that because we dared to publish something that is not as easy to use as an iPhone.

But that is only one way to look at it. In reality, what I do as a user is copy and paste error messages on google and see what comes up. We're not quite there with openinverter - try search for "openinverter parameters not loading" - all that comes up is the "Parameters" page on the wiki. So this would be something the community can pick up on - spot recurring questions and add them to the wiki + keywords. Admittedly I will be the only one here to precisely answer most software questions (example: dual channel throttle). And my answers are spread throughout the forum. It would be a HUGE relief if someone would just copy and paste those answers to a simple FAQ wiki page. Most conveniently it would be the person who asked the question.

But even that isn't written up anywhere: how to say thank you if your problem was solved. This should be prominently placed on the forum or maybe the "Forum Police" could remind the original questioner what to do.

We need to solve the redundancy issue!

I'd be happy to sell the Tesla hardware on the web shop if Damien is tired of it.