DIY Electric Car Forums banner

1 - 20 of 21 Posts

·
Registered
Joined
·
9 Posts
Discussion Starter #1
Hi,
I'm looking into buying an Orion 2 BMS, which has native SAE J1772 support, so does this mean I can wire it up with a CCS type 2 port? Or am I going to have to get a normal AC type 2, and a CHAdeMO?
 

·
Registered
Joined
·
59 Posts
Hi,
I'm looking into buying an Orion 2 BMS, which has native SAE J1772 support, so does this mean I can wire it up with a CCS type 2 port? Or am I going to have to get a normal AC type 2, and a CHAdeMO?
Apparently, Orion is working on offering CCS support (possibly an add-on module?); however, I haven't heard anything definite yet. Basically, I think a number of us are waiting on it.
 

·
Registered
Joined
·
12 Posts
For CCS you need a special module that can communicate over IEC15118 or DIN70121 compliant protocol using GreenPHY as physical layer. I have used one such module from Auronik (they were bought by Akka a while back) but it worked terribly and the price was about 1500 EUR for <10 pcs. I do have access to IEC15118 specification and have a good insight as how it works under the hood and was actually thinking of making an low cost version of it a while back, but didn't have the time. Just out of curiosity, how much would you be willing to pay for such module?
 

·
Registered
Joined
·
9 Posts
Discussion Starter #4
I have no clue, most controller replacements etc, are around €300, so I'd say anywhere from €100 to €300.

Also, since you know a bit about the protocol, I'd like to get about a 100V battery, since that's what the Hyper 9 motor wants (and it's safer). But CCS spec says it goes down to about 200V. So would it be possible to simply install a DC/DC converter between the port and battery? Or would this mess with the BMS communication?
 

·
Registered
Joined
·
12 Posts
Well 200V is the minimum for the standard, doesn't mean that there are no stations that support lower voltage charging. It means that in order for charger to be compliant with the standard it must support those ranges. The reason it doesn't say from for example 10V is because designing converters that have such a wide operating range is hard, if not impossible, while also maintaining efficiency. Typical converters have about 1:2 to 1:5 minimum to maximum voltage ratios. Also the limiting factor for power is usually the current, so by lowering voltage you also lower the power available for charging. And since the EVs usualy have 200-450 VDC battery pack voltages, this was a logical tradeoff, while also allowing higher voltages for some high-performing EVs.

As for your question regarding the DCDC converter between charger and battery, two things you should keep in mind. 1. the DCDC must be able to work with full charging power 2. you should be able to regulate the output current from DCDC OR you should have a DC transformer module which basically keeps constant ratio between output and input voltages. Either way i think it would be very expensive.
 

·
Registered
Joined
·
9 Posts
Discussion Starter #6
You're right, but if I do go with DC charging functionality, I don't want to show up at a DC charger only to find out it doesn't support a low enough voltage.

What I think I'll do when CCS type 2 is better documented and supported, is upgrade my battery pack to 200-300v anyways, and try to run the controller off a DC-DC converter. Which should be much cheaper, especially considering the Hyper 9's relatively low power usage.

Thanks for helping me out!
 

·
Registered
Joined
·
12 Posts
Do you know what is the status of the project, or at least the timeline? I can't seem to find anything about it.
 

·
Registered
Joined
·
39 Posts
+1 to have an accessible CCS controller.
Seems like every post I read - anywhere on the web - regardless of the year posted, there's always some company "currently developing" their own solution, but no update afterwatds. :(
 

·
Registered
Joined
·
138 Posts
Came from the other CCS thread...
CCS is a gigantic pain in the ass (I started reading through a free download from some RISE-V2G website).
Getting the hardware is bad enough, but the protocol is ridiculously complicated -- in the name of safety and 'security' aka protecting their own IP. There's a Github repository with some Java that apparently implements it, but I don't know where to start. https://github.com/V2GClarity/RISE-V2G

There's been some discussion of using an EU Tesla port, since that would possibly be controlled over CAN. But nobody's really on it yet.

I'd pay $300 for a CCS controller. I've done CHAdeMO, it's not bad, but CCS is beyond my level.

-Isaac
 

·
Registered
Joined
·
12 Posts

·
Registered
Joined
·
138 Posts
Looks like someone beat me to it...
.

Although it does say COMING SOON :\
Well... It doesn't look to be very easy to interface...
Option 2 seems to implement the full ISO15118 stack, with a simple API for control over CAN, SPI, USB, or Ethernet. That would probably be the most useful.
 

·
Registered
Joined
·
59 Posts
Looks like someone beat me to it...
.

Although it does say COMING SOON :\
It's not a product until someone can buy it.
 

·
Registered
Joined
·
12 Posts
In last two weeks I downloaded latest RISE-V2G code and successfully compiled it. I also managed to run a test on PC with dummy example where you have one software instance for SECC and one for EVCC and it worked. The code is structured in a way that the core of the IEC15118 protocol is fully implemented, you are responsible for implementing the access to the HW layer such as setting CP pins and also to implement control interface from the EV side for sending the target voltage, current, etc.

My idea was to use BeagleBone Black single board computer for running the RISE-V2G stack and create cape with STM32 MCU for interfacing with CP/PP/Lock motor electronics that uses UART for communication with RISE-V2G stack.

For the EV comms i was thinking of using one of two CAN interfaces on BBB itself.

Currently i am writing a PC simulator code that would simulate the UART data being sent from MCU so that i have more clear requirements for the hardware and also to simplify transition to actual hardware.

Also i have ordered two QCA7000 modules for testing.

Stay tuned...
 

·
Registered
Joined
·
59 Posts
In last two weeks I downloaded latest RISE-V2G code and successfully compiled it. I also managed to run a test on PC with dummy example where you have one software instance for SECC and one for EVCC and it worked. The code is structured in a way that the core of the IEC15118 protocol is fully implemented, you are responsible for implementing the access to the HW layer such as setting CP pins and also to implement control interface from the EV side for sending the target voltage, current, etc.

My idea was to use BeagleBone Black single board computer for running the RISE-V2G stack and create cape with STM32 MCU for interfacing with CP/PP/Lock motor electronics that uses UART for communication with RISE-V2G stack.

For the EV comms i was thinking of using one of two CAN interfaces on BBB itself.

Currently i am writing a PC simulator code that would simulate the UART data being sent from MCU so that i have more clear requirements for the hardware and also to simplify transition to actual hardware.

Also i have ordered two QCA7000 modules for testing.

Stay tuned...

Thanks for the update!
 

·
Registered
Joined
·
18 Posts
Also i have ordered two QCA7000 modules for testing.

Stay tuned...
This is great. I was considering going another route on the Tesla side by using some of what they did on the CHAdeMO adapter, but I haven't gotten that far. Glad to see the reference code seems to work. Running a Java server stack definitely sucks, but the flip side is being able to use 'standard' Linux tools makes debugging easier for those well versed in that sort of thing.

Which module did you order? They looked to be a pain in the butt to solder by hand...

-Matt
 

·
Registered
Joined
·
12 Posts
This is great. I was considering going another route on the Tesla side by using some of what they did on the CHAdeMO adapter, but I haven't gotten that far. Glad to see the reference code seems to work. Running a Java server stack definitely sucks, but the flip side is being able to use 'standard' Linux tools makes debugging easier for those well versed in that sort of thing.

Which module did you order? They looked to be a pain in the butt to solder by hand...

-Matt
I'm not particularly thrilled using Java stack due to high overhead (it takes about two minutes to boot just the Java VM on BBB, or at least that is my experience) and as of recent JDK licencing mess. But it does seem to be well maintained and worked on constantly for last couple of years.

I ordered Yellow beet PEV, I'm thinking of trying to solder them using through hole pads from the botttom.

I'm still not sure how I'm gonna test it all once and if i get it running :\

-Mario
 

·
Registered
Joined
·
18 Posts
Well, 58 Euro + ~80? Euro for beagle is much better than 329 Euro for the white beet! But the flip side is more work on the software front.

That's a long boot, but if you do something like a sleep mode then its not awful to wake up from. At least that's how I'd do it if it is possible.

I look forward to seeing where it goes.

-Matt
 

·
Registered
Joined
·
138 Posts
Mario, this is great news! I'm currently talking to In-Tech about one of their CCS control modules, which has Linux onboard and ISO/DIN software. If their pricing is too high I might go the way of the Beagle and Beet.
Charge Control M

I've got a CCS/CHAdeMO station a few minutes away that has been good for CHAdeMO testing -- if I go your way I'd be glad to test software.
 

·
Registered
Joined
·
134 Posts
subscribing, consider me extremely interested, as high speed charging is a must for my project and chademo is a dying standard, plus capped at 50kw
 
1 - 20 of 21 Posts
Top