11 ms·

I would have liked to see a cost breakdown. If it's significantly cheaper than OpenEVSE (somewhwere between $500-$600 depending on which kit you get), then mayb

I would have liked to see a cost breakdown. If it's significantly cheaper than OpenEVSE (somewhwere between $500-$600 depending on which kit you get), then maybe some people would be willing to take the risk (it seems like there are some safety concerns with his design that I don't feel qualified to adjudicate). But without some kind of a cost difference, I'd probably stick with something like OpenEVS, which seems to have some level of official certification [0]. I personally regret my decision to not go with them when I got my EVSE this year. I went with an EnelX Juicenet because A) it was on the list of devices my power company would mostly pay for and B) at the time it had API availability to integrate with HomeAssistant. They have since killed that second part, which I would have loved to still have. In the long term, the ~$500 I saved by having my power company reimburse was probably not worth the lack of control I now have. [0] https://openev.freshdesk.com/support/solutions/articles/6000113537-openevse-safety-features https://openev.freshdesk.com/support/solutions/articles/6000... -edit- He does have a total cost, I just missed it: he claims that it was ~£200. I leave it up to the reader if that's enough savings to give up the certification.

myself248 3y ago
OpenEVSE can be a codebase or a kit. If you want to source all your own parts, or maybe reuse significant parts of a non-open EVSE and just drop the processor into it or whatever, then it can be very cheap indeed. If you buy their branded kit, then full price applies. When OpenEVSE was new, I built one on perfboard, actually a very compact modular stack of perfboards, one layer per subsystem. Handed it off to my software-savvy friend, and he couldn't get it to pass self-test, despite both of us agreeing that every piece of the circuit appeared correct as per the published schematic. Eventually he gave up and ordered one of OpenEVSE's commercially-produced boards, and it failed self-test at the exact same place. Apparently there was some unclear documentation and it was supposed to fail there until something else was hooked up, I don't remember the details, but there was nothing wrong with my stack of perf in the first place. Argh. All of which is to say, it's not hard at all to build one of these things. if you want to lobotomize your closed-crap EVSE and lash an open board to it, do so! I'm pretty sure the JuiceBox is OpenEVSE-based in the first place so it might be as simple as reflashing some firmware.
- bigfudge 3y ago
  What I don’t understand is what the motivation for manufacturers to not just install the open software is… are they trying to harvest charge time data or something? It seems like it would be easier to change any logos in software but leave the rest as is. Non tech buyers wouldn’t know or care… and nerds would buy it in droves. This goes for a lot of the low end network gear too. Are they really adding much value in software?
  - MostlyStable 3y ago
    They must believe that there is, regardless of whether or not their _actually_ is. Manufacturers killing local control that is only used by extremely small numbers of their users is a pattern [0]. The only potential good reason I can think of is if the existence of these abilities allows users to get themselves into trouble when they don't fully know what they are doing, which could create a great deal of (expensive) support and so they decide to just remove the headache [0]https://www.theverge.com/23949612/chamberlain-myq-smart-garage-door-controller-homebridge-integrations https://www.theverge.com/23949612/chamberlain-myq-smart-gara...
- MostlyStable 3y ago
  oh wow, thank you. I'm going to look into this. I have been extremely unhappy with JuiceBox's app/etc the entire time I've had it. If I can reflash it (and/or relatively easily replace the brains), that would dramatically improve my charging situation). After reading this article, I had been considering getting one of their most basic kits that doesn't include the cables from the wall to the box or the box to the car (since I have both of those with my current EVSE), but if I can get the same functionality reusing even more of my current one, that would be ideal.

myself248 3y ago

OpenEVSE can be a codebase or a kit. If you want to source all your own parts, or maybe reuse significant parts of a non-open EVSE and just drop the processor into it or whatever, then it can be very cheap indeed. If you buy their branded kit, then full price applies. When OpenEVSE was new, I built one on perfboard, actually a very compact modular stack of perfboards, one layer per subsystem. Handed it off to my software-savvy friend, and he couldn't get it to pass self-test, despite both of us agreeing that every piece of the circuit appeared correct as per the published schematic. Eventually he gave up and ordered one of OpenEVSE's commercially-produced boards, and it failed self-test at the exact same place. Apparently there was some unclear documentation and it was supposed to fail there until something else was hooked up, I don't remember the details, but there was nothing wrong with my stack of perf in the first place. Argh. All of which is to say, it's not hard at all to build one of these things. if you want to lobotomize your closed-crap EVSE and lash an open board to it, do so! I'm pretty sure the JuiceBox is OpenEVSE-based in the first place so it might be as simple as reflashing some firmware.

bigfudge 3y ago
What I don’t understand is what the motivation for manufacturers to not just install the open software is… are they trying to harvest charge time data or something? It seems like it would be easier to change any logos in software but leave the rest as is. Non tech buyers wouldn’t know or care… and nerds would buy it in droves. This goes for a lot of the low end network gear too. Are they really adding much value in software?
- MostlyStable 3y ago
  They must believe that there is, regardless of whether or not their _actually_ is. Manufacturers killing local control that is only used by extremely small numbers of their users is a pattern [0]. The only potential good reason I can think of is if the existence of these abilities allows users to get themselves into trouble when they don't fully know what they are doing, which could create a great deal of (expensive) support and so they decide to just remove the headache [0]https://www.theverge.com/23949612/chamberlain-myq-smart-garage-door-controller-homebridge-integrations https://www.theverge.com/23949612/chamberlain-myq-smart-gara...
MostlyStable 3y ago
oh wow, thank you. I'm going to look into this. I have been extremely unhappy with JuiceBox's app/etc the entire time I've had it. If I can reflash it (and/or relatively easily replace the brains), that would dramatically improve my charging situation). After reading this article, I had been considering getting one of their most basic kits that doesn't include the cables from the wall to the box or the box to the car (since I have both of those with my current EVSE), but if I can get the same functionality reusing even more of my current one, that would be ideal.