Hello - considering using a GEVCU for my project
35 views
Skip to first unread message
Charles Galpin
Apr 18, 2021, 4:44:46 PM4/18/21
to gevcu-de...@googlegroups.com
Hi, my name is Charles, and I am considering using a GEVU for my project.
This is kind of tongue in cheek, but also mostly true :) I have been involved with the GEVCU project in the past, but literally haven’t touched it in 6+ years, and done very little on my EV in that period either, but I have some time right now, so I am trying to get back into it. I have a "use it or lose it” type of memory, so I don’t remember a lot of this stuff!
In all seriousness, I would like to re-evaluate my decision to use the GEVCU at this point, since I might have better/other options - a lot has changed in the past few years. For example I am now using a iBooster brake booster (from a Tesla) which wasn’t an options before.
I’d like to hear everyone’s opinions, and maybe if Collin or anyone associated with EVTV can give us an idea of the number of boards out there and long term availability. Here is where I am at:
- I have a GEVCU 2.x that used to have a working wifi adapter - no longer sure if that works still, because I pulled it to use on my v42 board, where i didn’t work iirc.
pros: i have it and it works
cons: if it fails I am screwed. Possibly no wireless, not likely to survive well in a rough/bouncy truck, i think this has a precharge circuit defect iirc
- i have a GEVCU 4.2 board. Failed to get wifi working, either with the connect one, or ESP32 alternative when I tried (but will re-try).
pros: I have it :)
cons: if it fails I am screwed. Possibly no wireless. Seems like it’s already not pin compatible with v2, Not sure if this has the precharge defect or not.
- i could buy a new v6.2 board (or two - see below)
pros: would have the latest? Community fractured, maybe not many using it
cons: not sure about long term availability. seems expensive ( could be wrong about this). different connectors so need to make this decision soon.
- I have general concerns about availability and cost. Although very grateful Jack started this project, since he has passed I am not sure how much longer EVTV will be in business and if they can/will continue to have these made. I know Collin is the main developer for them, but the community fractured (which sucks) and we now have 2 main forks, and I would likely work with Michael on his if I do use it since I know he’s using his and I like the work he has done (not a knock at Collin by any means). But if I used the GEVCU, I’d want to get 2 to have a backup, and $1200 seems a little steep to me.
What does everyone else think about this? There seems to be little to no use of it (I see little to no traffic here, and i just went to go look at the evtv forums and got a 404). I have not watched their videos in a while - just randomly had time once to watch one, and saw Jack had cancer. I don’t claim to have done a ton of looking, but it’s looking deadish to me.
I long for the days where we had a fairly small, but active community working on it, and the future for getting hardware looked bright.
Lost in Virginia,
charles
This is kind of tongue in cheek, but also mostly true :) I have been involved with the GEVCU project in the past, but literally haven’t touched it in 6+ years, and done very little on my EV in that period either, but I have some time right now, so I am trying to get back into it. I have a "use it or lose it” type of memory, so I don’t remember a lot of this stuff!
In all seriousness, I would like to re-evaluate my decision to use the GEVCU at this point, since I might have better/other options - a lot has changed in the past few years. For example I am now using a iBooster brake booster (from a Tesla) which wasn’t an options before.
I’d like to hear everyone’s opinions, and maybe if Collin or anyone associated with EVTV can give us an idea of the number of boards out there and long term availability. Here is where I am at:
- I have a GEVCU 2.x that used to have a working wifi adapter - no longer sure if that works still, because I pulled it to use on my v42 board, where i didn’t work iirc.
pros: i have it and it works
cons: if it fails I am screwed. Possibly no wireless, not likely to survive well in a rough/bouncy truck, i think this has a precharge circuit defect iirc
- i have a GEVCU 4.2 board. Failed to get wifi working, either with the connect one, or ESP32 alternative when I tried (but will re-try).
pros: I have it :)
cons: if it fails I am screwed. Possibly no wireless. Seems like it’s already not pin compatible with v2, Not sure if this has the precharge defect or not.
- i could buy a new v6.2 board (or two - see below)
pros: would have the latest? Community fractured, maybe not many using it
cons: not sure about long term availability. seems expensive ( could be wrong about this). different connectors so need to make this decision soon.
- I have general concerns about availability and cost. Although very grateful Jack started this project, since he has passed I am not sure how much longer EVTV will be in business and if they can/will continue to have these made. I know Collin is the main developer for them, but the community fractured (which sucks) and we now have 2 main forks, and I would likely work with Michael on his if I do use it since I know he’s using his and I like the work he has done (not a knock at Collin by any means). But if I used the GEVCU, I’d want to get 2 to have a backup, and $1200 seems a little steep to me.
What does everyone else think about this? There seems to be little to no use of it (I see little to no traffic here, and i just went to go look at the evtv forums and got a 404). I have not watched their videos in a while - just randomly had time once to watch one, and saw Jack had cancer. I don’t claim to have done a ton of looking, but it’s looking deadish to me.
I long for the days where we had a fairly small, but active community working on it, and the future for getting hardware looked bright.
Lost in Virginia,
charles
Collin Kidder
Apr 18, 2021, 5:16:58 PM4/18/21
to gevcu-development
Here's the story: Yes, basically nothing much has happened with GEVCU
from EVTV in the past 3 years. EVTV mostly moved on to solar panels
and the related inverters, battery controllers, etc for off-grid
solar. I think that Jack pretty much considered the EV fight to be
won. They're going mainstream, Tesla is blazing new trails, the EV
train has left the station. This left hobbyist builds somewhat of a
tough sell. After all, why build one when you can buy a Leaf, Tesla
S/X/3/Y, Bolt, etc and most of those would be cheaper and easier than
building one. So, EVTV has somewhat backed off on pushing DIY builds.
We still sell parts but the main drive just isn't with DIY builds any
longer. So, for the past few years not a lot has gone on with GEVCU6.
I'm appreciative that Michael is still doing some things with the
GEVCU4 based design. In fact, if you stick with your existing hardware
I'd recommend working with him. The old GEVCU4 code is as good as dead
to me honestly. His version is far better anyway.
Now. more recently, since Jack's passing, we've been looking to see
what we've got for items we produce and what we want to do with them.
Good news on two fronts. 1: EVTV is not planning to go anywhere. It
has new ownership (Jack's daughter) and I'm not aware of any plans to
sunset the business or anything. We're pushing for the long haul 2: In
taking stock of what we've got, it has been decided to see what we can
do with GEVCU6 and perhaps create a new, more versatile design. This
new design would have the capability to be used for our solar stuff
too which would then make it more lucrative for me to maintain the
code base and push it all forward. Like any business, EVTV has to
follow where the market leads. And, right now our market has led to
solar. But, DIY is not totally dead and seems to recently be making a
slight resurgence.
I was unaware that the forums broke. Recently there has been some
transition due to Jack's daughter taking over and there needing to be
a transfer of ownership of domains and such. Apparently something got
broken during this process. I'll let them know. But, the forums were
essentially dead anyway. The only reason I'd go there is to go look up
some old posts or download some old files when someone needed them. I
don't think basically anyone has been using the forums in several
years now, well before Jack passed. I experienced this same thing with
SavvyCAN where I made a forum but I never managed to cultivate it
properly and it just got weedy and hardly ever used. I've since
switched to discussions on GitHub which has worked better for
SavvyCAN.
One thing that hasn't changed is that I'm still here. I'm quite busy
these days so I don't always get to providing support for everyone but
I try. And, I am still building electric cars so I do still work on
GEVCU sometimes if only for my own uses.
TL;DR - If you stick with your existing hardware go with Michael's
code, it was much improved. GEVCU4 and below are basically deprecated
according to me. Otherwise, maybe watch to see what happens in the
next 1-2 months - there could be some good news.
> --
> You received this message because you are subscribed to the Google Groups "GEVCU-Development" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to gevcu-developm...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/gevcu-development/FD6D7A19-C2BC-46FF-BD41-B848F3E5D2F5%40gmail.com.
from EVTV in the past 3 years. EVTV mostly moved on to solar panels
and the related inverters, battery controllers, etc for off-grid
solar. I think that Jack pretty much considered the EV fight to be
won. They're going mainstream, Tesla is blazing new trails, the EV
train has left the station. This left hobbyist builds somewhat of a
tough sell. After all, why build one when you can buy a Leaf, Tesla
S/X/3/Y, Bolt, etc and most of those would be cheaper and easier than
building one. So, EVTV has somewhat backed off on pushing DIY builds.
We still sell parts but the main drive just isn't with DIY builds any
longer. So, for the past few years not a lot has gone on with GEVCU6.
I'm appreciative that Michael is still doing some things with the
GEVCU4 based design. In fact, if you stick with your existing hardware
I'd recommend working with him. The old GEVCU4 code is as good as dead
to me honestly. His version is far better anyway.
Now. more recently, since Jack's passing, we've been looking to see
what we've got for items we produce and what we want to do with them.
Good news on two fronts. 1: EVTV is not planning to go anywhere. It
has new ownership (Jack's daughter) and I'm not aware of any plans to
sunset the business or anything. We're pushing for the long haul 2: In
taking stock of what we've got, it has been decided to see what we can
do with GEVCU6 and perhaps create a new, more versatile design. This
new design would have the capability to be used for our solar stuff
too which would then make it more lucrative for me to maintain the
code base and push it all forward. Like any business, EVTV has to
follow where the market leads. And, right now our market has led to
solar. But, DIY is not totally dead and seems to recently be making a
slight resurgence.
I was unaware that the forums broke. Recently there has been some
transition due to Jack's daughter taking over and there needing to be
a transfer of ownership of domains and such. Apparently something got
broken during this process. I'll let them know. But, the forums were
essentially dead anyway. The only reason I'd go there is to go look up
some old posts or download some old files when someone needed them. I
don't think basically anyone has been using the forums in several
years now, well before Jack passed. I experienced this same thing with
SavvyCAN where I made a forum but I never managed to cultivate it
properly and it just got weedy and hardly ever used. I've since
switched to discussions on GitHub which has worked better for
SavvyCAN.
One thing that hasn't changed is that I'm still here. I'm quite busy
these days so I don't always get to providing support for everyone but
I try. And, I am still building electric cars so I do still work on
GEVCU sometimes if only for my own uses.
TL;DR - If you stick with your existing hardware go with Michael's
code, it was much improved. GEVCU4 and below are basically deprecated
according to me. Otherwise, maybe watch to see what happens in the
next 1-2 months - there could be some good news.
> You received this message because you are subscribed to the Google Groups "GEVCU-Development" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to gevcu-developm...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/gevcu-development/FD6D7A19-C2BC-46FF-BD41-B848F3E5D2F5%40gmail.com.
Jason Arnold
Apr 18, 2021, 5:22:44 PM4/18/21
to GEVCU-Development
Welcome back, Charles. I've seen your name in the annals of GEVCU's history books, what was your role back when you were involved?
Like you, I do appreciate that Michael and Collin have kept this project alive, such as it is, but have also found myself wondering about its future. I've been following an eerily parallel effort currently being spun up on the OpenInverter community known as the ZombieVCU (https://openinverter.org/wiki/ZombieVerter_VCU), with a fair bit of interest. It's open-source, proposes can control of a host of setups via CAN, has a Wi-Fi interface, and, maybe most importantly, is under active development - in a lot of ways it feels like the next generation of GEVCU, even though there's no direct link between the two projects (that I know of). I really appreciate the effort that's gone into GEVCU over the years, but given the current pace of development, wonder if it wouldn't be served to hitch this project to a bigger wagon. Zombie VCU started out as SAM3X-based as well and has since been ported to STM32, so (in my naive, not software developer opinion) moving the GEVCU software over is doable. One thing I'd love to see ZombieVCU get into its features is the very trick throttle mapping that GEVCU has. I think it's an ideal marriage, does this idea appeal to anyone else?
As an alternative for you particular situation, Charles (assuming you're using an AZD setup), something else to consider is ditching the DMOC entirely and using a gen 3 Prius inverter to run the Siemens 1PV5135. Damien did this in his latest re-conversion of his E39 "Land Yacht" - more here: https://www.youtube.com/watch?v=2ISZffhCs4I
Angelum Band
Apr 18, 2021, 8:56:09 PM4/18/21
to gevcu-de...@googlegroups.com
Hi everybody I am not sure I agree totally with how the hobby diy market is seen. Around the corner from where I live there's a mechanic who collects old cars. Amongst them there's a Ferrari Testa Rosa that can't be driven to it's full potential... legally. He has them, shows them off and goes to Moroso race track every now and then.
Second, I did consider buying a Tesla until I realize you don't own the car: Tesla Motors Inc. does. Not only that buying a Tesla implies giving up your privacy and... well even the dna of ones farts. Thank Elon we love you but until you let me tinker, fix and modify your car I pass.
Last some years ago Chip Yates broke the record for speed in an electric modified Long Ez. Hey who knows maybe I want to build an electric Mosquito helicopter. Is Jack's daughter the one that is a helicopter pilot?
I'm rambling but you get my brief point.
Alcibíades
To view this discussion on the web visit https://groups.google.com/d/msgid/gevcu-development/CABBbThCNNnn6iG8RQRiBKSBTw1KjeM_czyRC3rkmYDUU__afVg%40mail.gmail.com.
Collin Kidder
Apr 18, 2021, 9:13:48 PM4/18/21
to gevcu-development
Yes, I know about the ZombieVerter project. I'm a member on that forum
too. I know Damien pretty well but not some of the other people. I can
tell you this much - if/when I make a GEVCU refresh it isn't going to
be using the STM32 chip they're using. That chip is in basically all
ways worse than the SAM3X chip. I think they switched to that chip
only because more people there have experience with it.
I think that ZombieVerter and GEVCU occupy somewhat different places.
One or the other might be better for a given project. And, really both
are open source. Yeah, the GEVCU6 hardware really isn't but the
software is. So, a certain amount of code could be swapped around
between the two. Both are ARM based and Arduino adjacent so I think
some things could be merged back and forth between the two.
too. I know Damien pretty well but not some of the other people. I can
tell you this much - if/when I make a GEVCU refresh it isn't going to
be using the STM32 chip they're using. That chip is in basically all
ways worse than the SAM3X chip. I think they switched to that chip
only because more people there have experience with it.
I think that ZombieVerter and GEVCU occupy somewhat different places.
One or the other might be better for a given project. And, really both
are open source. Yeah, the GEVCU6 hardware really isn't but the
software is. So, a certain amount of code could be swapped around
between the two. Both are ARM based and Arduino adjacent so I think
some things could be merged back and forth between the two.
> --
> You received this message because you are subscribed to the Google Groups "GEVCU-Development" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to gevcu-developm...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/gevcu-development/b0b8f362-1828-4b5e-9da1-a424c78c35a9n%40googlegroups.com.
> You received this message because you are subscribed to the Google Groups "GEVCU-Development" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to gevcu-developm...@googlegroups.com.
Charles Galpin
Apr 19, 2021, 8:54:16 AM4/19/21
to gevcu-de...@googlegroups.com
>
> On Sun, Apr 18, 2021 at 5:22 PM Jason Arnold <arnold...@gmail.com> wrote:
>>
>> Welcome back, Charles. I've seen your name in the annals of GEVCU's history books, what was your role back when you were involved?
>> Like you, I do appreciate that Michael and Collin have kept this project alive, such as it is, but have also found myself wondering about its future. I've been following an eerily parallel effort currently being spun up on the OpenInverter community known as the ZombieVCU (https://openinverter.org/wiki/ZombieVerter_VCU), with a fair bit of interest. It's open-source, proposes can control of a host of setups via CAN, has a Wi-Fi interface, and, maybe most importantly, is under active development - in a lot of ways it feels like the next generation of GEVCU, even though there's no direct link between the two projects (that I know of). I really appreciate the effort that's gone into GEVCU over the years, but given the current pace of development, wonder if it wouldn't be served to hitch this project to a bigger wagon. Zombie VCU started out as SAM3X-based as well and has since been ported to STM32, so (in my naive, not software developer opinion) moving the GEVCU software over is doable. One thing I'd love to see ZombieVCU get into its features is the very trick throttle mapping that GEVCU has. I think it's an ideal marriage, does this idea appeal to anyone else?
>>
>> As an alternative for you particular situation, Charles (assuming you're using an AZD setup), something else to consider is ditching the DMOC entirely and using a gen 3 Prius inverter to run the Siemens 1PV5135. Damien did this in his latest re-conversion of his E39 "Land Yacht" - more here: https://www.youtube.com/watch?v=2ISZffhCs4I
Funny story - since Lars was having problems I though it would be a good idea for me to hook everything up and get back to my last known working state, which was a bench setup using a rectifier as a power source and be able to spin the motor mated to the gearbox (transfer case in my case). It would have helped narrow down his problem and help me get back into this. But with the motor/gearbox installed in the vehicle, I can’t even position the DMOC to be able to connect the cables without lengthening them. Sigh :)
I’m going to go back to plan A, which is use one of two battery packs I have accumulated (neither of which I want to use in this project as a final solution) - a gen 1 volt, and a gen 1 leaf, and get the charging side of things working, since that’s where I left off. I assume I can do this without the DMOC having a motor attached.
> On Apr 18, 2021, at 5:16 PM, Collin Kidder <col...@kkmfg.com> wrote:
>
> So, for the past few years not a lot has gone on with GEVCU6.
> I'm appreciative that Michael is still doing some things with the
> GEVCU4 based design. In fact, if you stick with your existing hardware
> I'd recommend working with him. The old GEVCU4 code is as good as dead
> to me honestly. His version is far better anyway.
I don’t recall what board Michael has, but afaik there isn’t anything stopping his working on a 6.x board (but just guessing) but if you made another branch, i’m also guessing bringing them back together would be a pain.
But as you can tell, I am not thrilled about two independent efforts for essentially the same thing, and IMHO this shouldn’t be needed or desired by *anyone*. And if wanting to be on the most recent and available hardware conflicts with the software side of things, it makes my decision even more difficult.
Also, are the hardware differences documented somewhere in laymans terms? I know the connectors changed and you added bluetooth, but there was some issue with it Lars said? Even between the v2 and v4, which have the same connector, some of the pinouts have changed, and something affecting the throttle, although I haven’t looked into it yet - i just know I can’t plug in my v4.2 board and have my throttle work like it does on the v2. I think Michael's branch has the pinouts documented through 4.2 and showed no differences for the throttle pins.
I hope my dilemma is clear - I am faced with having to go with unsupported hardware with different connectors/wiring harness that if fails my truck is dead, and who knows what it would take to repair/replace, or buy the latest unsupported hardware with slightly better likelihood of getting a it repaired/replaced, but can at least buy a spare, and it sounds like use your branch, and lose all the wonderful contributions Michael brings.
Hey Michael, any plans to upgrade your hardware to v6+? :D
> Now. more recently, since Jack's passing, we've been looking to see
> what we've got for items we produce and what we want to do with them.
> Good news on two fronts. 1: EVTV is not planning to go anywhere. It
> has new ownership (Jack's daughter) and I'm not aware of any plans to
> sunset the business or anything. We're pushing for the long haul 2: In
> taking stock of what we've got, it has been decided to see what we can
> do with GEVCU6 and perhaps create a new, more versatile design.
> But, the forums were
> essentially dead anyway. The only reason I'd go there is to go look up
> some old posts or download some old files when someone needed them.
> One thing that hasn't changed is that I'm still here. I'm quite busy
> these days so I don't always get to providing support for everyone but
> I try. And, I am still building electric cars so I do still work on
> GEVCU sometimes if only for my own uses.
> TL;DR - If you stick with your existing hardware go with Michael's
> code, it was much improved. GEVCU4 and below are basically deprecated
> according to me. Otherwise, maybe watch to see what happens in the
> next 1-2 months - there could be some good news.
Lars Rengersen
Apr 22, 2021, 3:31:31 AM4/22/21
to GEVCU-Development
This has become a useful topic for those who still have and want to keep using the DMOC.
Hope the EVTV forums will be back online. It contained a huge amount of useful information.
After a failed attempt in repairing my GEVCU 6.2 I am now also having a GEVCU 4 to work with.
Since I have heard this is not a very stable version and Collin confirming it is deprecated I am keen on replacing it.
The options I'm considering are:
1- Buying another GEVCU 6
2- Buy another VCU an have it support the DMOC
3- Make a GEVCU 6 hardware board myself
4- Get a microcontroller and develop a new VCU
1- Buying another GEVCU 6
Not really an option, is quite expensive and has it's flaws.
2- Buy another VCU an have it support the DMOC
I have an option for a VCU with ECE certification (officially required due to regulations in Europe).This is developed and supported by a third party.
Disadvantage is that it is overkill (also does contactor control etc) and very expensive (3x GEVCU) plus customization costs of supporting the DMOC.
3- Make a GEVCU 6 hardware board myself
The Eagle schematic and board files of the GEVCU are open source. Furthermore if I understand correctly also the GEVCU6 is bottomline just a Arduino Due.
So by aligning the pinout with the latest firmware (from Michael) this is very feasible for me to do.
While looking at the PCB design of the current GEVCU6 I did see many improvement opportunities.
My challenge is that I'm not a software coder so I need to rely on others for that.
4- Get a microcontroller and develop a new VCU
Myself I just need a VCU that controls the motor/DMOC. So basically this is about throttle, brake pressure, drive and regen profiles and limits (temperature/SOC).
All other functions such as contactor control and thermal management are being handled by other systems.
Within this scenario I see/have two options:
a) Use my CAN display which is very capable and can be programmed using an easy to use graphical environment.
b) A CC16 controller by MRS Electronics
From these two I prefer option b) but I need to hire a developer to accomplish this.
Currently I plan to start driving using the GEVCU4 and then after the summer start the upgrade project and go for option 4b.
Regards,
Lars
Op maandag 19 april 2021 om 14:54:16 UTC+2 schreef Charles Galpin:
Reply all
Reply to author
Forward
0 new messages