Energy Controller including CCS interface

[jrichl]

It is running. Thank you very much for your advice. It wouldn't have been possible without your great work and installation description. Next step will be the lightbulbs, then we will see how long the i3 is happy.

wonderful, the lightbulbs are burning since more than 15 minutes - stopped it after that

[uhi22]

Ali delivered a quite solid CCS plug. Adding 10A 1000V DC PV fuses directly in the plug.

[uhi22]

Progress with the demo box:
- Adjustable HV, consisting of a bunch of small DCDC converters, which are controlled in groups by 5 FETs.
- HV measurement using the muehlpower board
- supply by two 18650 cells and a circuit board from a power bank to get 5V
- relay to connect/disconnect the bulbs

The plug is more bulky and more heavy than the box itself

On software side implemented some kind of cable check: applies HV and measures it back. Stop the session in case of shortcut.

[uhi22]

Feature update of the demo box:
- pyPLC controls the precharge voltage
- pyPLC sends the measured HV to the display in all states
- display shows the log file number (so that for testing it is easier to match the pcap file to the actual test situation, just by making foto/video of the test session)
- added block diagrams to the github https://github.com/uhi22/DiDeBoCCS

[uhi22]

Success on the Ioniq (not a big surprise ).
But how to stop it? Suddenly noticed that I have no stop button, neither in the car nor on the discharge box. Finally pulled the banana plug of the CP line, this worked as intended.

[Jogi]

Hello all.

Just stumbled in here from goingelectric forum, by interest for this project.

I'd like to test the setup with a hybrid inverter on my car, a Mustang MachE, wich successfully was tested with a chinese CCS-discharger, so i think (hope) it will work.

I heard, someone here runs such a system in the middle of Germany, i will contact...

[johu]

Oh sorry, your account is legit. Welcome Your other post looked like a spam bot trying to get in.

Kassel is about the middle of Germany. Where are you?

[Jogi]

It's about 300km south of you, in the north of BaWü.

Would be a day trip for a test, no problem for me.

Let's contact via PM for adresses...

[uhi22]

Progress: Integrated the stop-button in pyPLC and in the setup description.
https://github.com/uhi22/pyPLC/commit/7 ... 75eede3fd5
https://github.com/uhi22/DiDeBoCCS/comm ... 625da08ff3

[DVD3500]

It's about 300km south of you, in the north of BaWü.

Would be a day trip for a test, no problem for me.

Let's contact via PM for adresses...

Rhein-Neckar Kreis here....

[vbarrier]

Hi,

I received my focccicape and trying to use it

The 5V output should work? I don't get power from it.

I can measure the 5v power until the d11.

[uhi22]

This depends on the board version. I only have a quite old board, this has the "VCC" of the phoenix connector connected to the 5V line. (Also my board has manual patches, not sure whether all boards have this). As I understand, the newer boards do not have a 5V on the phoenix connector, they have a supply input for 12V there.

Edit: Added a chapter "Board Versions" to the wiki: https://openinverter.org/wiki/FoccciCape#Board_Versions

[vbarrier]

I have the v1.1 version. It has written 5v on the phoenix connector.

For now I made the connection on P9_5 but I want to be sure it is intented to be use like this.

[uhi22]

Ah, das ist die Variante mit dem Buck-Converter. Der Aufdruck "5V" ist falsch. An dem Pin kann man extern Spannung einspeisen, 7 bis 20V (die geht dann über D11 zum Buck-Converter, der 5V draus macht). Wenn man die ganze Baugruppe aber eh mit 5V versorgt, also 5V an P9_5, ist die Einspeisung an Phoenix Pin 2 überflüssig.

Edit, sorry, now in english:
Ah, that's the version with the buck converter. The "5V" marking is incorrect. You can supply an external voltage to that pin, 7 to 20V (which then goes via D11 to the buck converter, which converts it to 5V). However, if you're already supplying the entire assembly with 5V, i.e., 5V at P9_5, then supplying power to Phoenix pin 2 is unnecessary.

[vbarrier]

Well I made progress (learning curve!) but now I'm stuck on a cable check process:

Code: Select all

[15422ms] [EVSE] Received CableCheckReq. Extracting SoC parameters via DC
[15450ms] [EVSE] responding (27bytes) = 01 FE 80 01 00 00 00 13 80 9A 02 00 40 80 C1 01 41 81 C2 10 20 00 02 00 00 04 00 
[15453ms] [EVSE] from 4 entering 4
cableChecker evaluating results and cleaning up
measured voltage: 0
cable check error: voltage too low
[15491ms] [EVSE] In state WaitForFlexibleRequest, received (24bytes) = 01 FE 80 01 00 00 00 10 80 9A 02 00 40 80 C1 01 41 81 C2 10 11 40 06 40 
[15516ms] [EVSE] {
"msgName": "CableCheckReq",
"info": "16 bytes to convert", 
"error": "",
"result": "",
"schema": "DIN",
"g_errn": "0",
"header.SessionID": "0102030405060708",
"header.Notification_isUsed": "0",
"header.Signature_isUsed": "0",
"DC_EVStatus.EVRESSSOC": "25",
"DC_EVStatus.EVReady": "1",
"debug": "Line9057Line9260Line364"
}
[15523ms] [EVSE] Received CableCheckReq. Extracting SoC parameters via DC
cable check is finished and failed
[15525ms] [EVSE] CableCheck failed. Will not answer the request.
[15526ms] [EVSE] from 4 entering 0
[17476ms] connection closed

I did a lot tests:

- Resistor between PP et PE => 1.5k ok
- PWM enabled using cppwn.py (thank you @uhi22) it show -11v on multimeter after enable the service
- CP on focccicape -> CP on cable -> -11v
- GND on focccicape -> PE on cable

(I'm trying to light bulb on my Tesla MY (I have the step-up converter for the next step, precharging)

Any insight that could help me?

[uhi22]

That is half-way good news. Congratulations that you got it running so far.
The error message comes from the situation, that during cable check we want to apply a voltage to the HV pins, and physically want to read back this voltage. This is to confirm that there is no shortcut between DC+ and DC- in the cable, plug or car side. Such a shortcut would damage a lot (welded contactors in the car, melting cables, tripped fuse). We had such a case, caused by installation error on car side.
So I see two options to make the cable check finishing:
A) Control the small HV supply during pre charge and add the voltage-sense board, so that a real cable check happens and succeeds.
B) Change the config/software to pretend a successful cable check.
For option B I need to check what we already have. Will give an update later.

[vbarrier]

Thank you for the reply.

The small HV supply is the famous dc boost from Amazon with The setup explaind by Johannes (a resistor 1M and a 450v 10uf). I power it up using P9_5 and OUT1 relay for 10 seconds.

There is a fuse on each DC line.

Should I start the precharging at cableCheck? I was thinking to trigger the boost at precharging phase only.

[uhi22]

Do you use a voltage measurement board, to give the beaglebone the possibility to know the physical voltage? If not, then turning on the boost converter during cable check does not make sense, because we can anyway not measure the voltage. So option B is the way to go.

[vbarrier]

Do you use a voltage measurement board, to give the beaglebone the possibility to know the physical voltage?

No I don't have that. I will look into but I would like a sccuess with option B before .

If not, then turning on the boost converter during cable check does not make sense, because we can anyway not measure the voltage.

That's what I thought

So option B is the way to go.

Ok!

[vbarrier]

I managed to put the lights on

I changed the powersupply value hardcoded from 0 to 300.

Also use mqtt to listen the precharge request and power up the dc boost.

[uhi22]

Added the feature "pretended cable check". It can be configured to be
"no", so doing a real HV cable check (option A above) or
"yes", which just does nothing but reports "cable is ok".

git commits https://github.com/uhi22/pyPLC/commit/2 ... ce29dd4021 and https://github.com/uhi22/pyPLC/commit/6 ... ce8c129f6d

So just run "git pull", restart the pyplc and let's see what it changes.

[vbarrier]

Added the feature "pretended cable check". It can be configured to be
"no", so doing a real HV cable check (option A above) or
"yes", which just does nothing but reports "cable is ok".

git commits https://github.com/uhi22/pyPLC/commit/2 ... ce29dd4021 and https://github.com/uhi22/pyPLC/commit/6 ... ce8c129f6d

So just run "git pull", restart the pyplc and let's see what it changes.

Thank you I will try that way tomorrow morning, I stopped on my light success