Develop a QCA7000 board?

[catphish]

I tested this circuit today. The MOSFET failed after a few charging cycles
Definitely needs protection between D2 and Q1.
I use the same MOSFET circuit for 12V GPIO inputs too. Those also failed. They require the same protection.

These are my updated (hopefully much safer) designs for 12V general purpose input, and CP input/output.

[uhi22]

I just rotate the components in the JLC GUI when I place the order, but I guess it's better to do it correctly in the source files!

Yes, in case I try different options, it's annoying to correct the rotation again and again in the web tool. And also if somebody else just wants to order the same design without deep knowledge, it's much more convenient to have the correct rotation automatically.

Added a PCB version history, to make transparent which version has which features. Will prepare two version 3 boards https://github.com/uhi22/foccci#v3-2023-08-17 to ship to johu and muehlpower, we will discuss the details per PM.

[uhi22]

These are my updated (hopefully much safer) designs for 12V general purpose input, and CP input/output.

This looks good. I use an other way, hopefully also safe:

[uhi22]

What are the conditions to place the openinverter logo on the board? Is this available in KiCad compatible format?

[johu]

Check if this one imports

Conditions? I'm happy if you put it on there

[xvyDFatih]

I have codico red carrier board. I have used stm32f4 to communicate with qca chip. I have used https://github.com/uhi22/ccs32clara. Can I make a ccs charge controller with this code. I know that I have to implement cp duty cycle and pp reading. How can I test stm32f4 and red carrier board as a pev side.
Also I have white carrier board EI. I have ran Seventax FreeV2G code. I have connected CP and PP pins to red carrier board. But I havent set connection between these modules. I have raspberry pi, 3 red carrier board, white carrier board EI. CAn you help me about testing ccs charge controller with these board? How can I test these boards

SLAC matching succesfull. Im using white carrierd board EI with freeV2G as evse side. stm32 and red carrier board as pev side.
/* our link-local IPv6 address. Todo: do we need to calculate this from the MAC? Or just use a "random"? /
/ For the moment, just use the address from the Win10 notebook, and change the last byte from 0x0e to 0x1e.
You wrote these comments in code (ipv6.c). Idont understand here. Can you explain
Also after slac matching succesfull. [SDP] initiating SDP request. I dont pass the [SDP] initiating SDP request. How can I do sdp

thanks

I never did this before, and was quite insecure whether I have a chance to do this. But after watching some videos I understood it is easier then expected, when knowing the tricks. Flux is the key. And pre-heating the board to 100°C. And then hot air, and the part will float by itself to the correct position. Really amazing to watch. Ok, my first attempt went wrong, but nothing broken, just again flux, preheating, hot air, tip from the side, and it exactly jumped in the correct position. The next time will be faster I guess.

Ahh, and the 7005 is no real QFN, it is QFN with "wettable edges", this means we can inspect the solder connections. The real QFN is harder I guess, because you see nothing.



Yes, exactly. The SPI, the interrupt pin (which maybe is not necessary) and the reset, and some test points to be able to program the Flash.

Something is wrong with this ugly board: No smoke up to now. It just talks to the ccs32berta. This went a little bit too smooth.

image.png

[uhi22]

I have used https://github.com/uhi22/ccs32clara. Can I make a ccs charge controller with this code.

This is the plan. At the moment, the clara is heavy work in progress. The communication on PLC, some basic CAN communication and contactor driver is working, but other things like status LEDs, connector lock is still to be developed.

How can I test stm32f4 and red carrier board as a pev side.

I recommend to use pyPLC as EVSE to test the PEV side. This makes it possible to record pcap files of the ethernet communication, which are a big help to debug potential issues.

Also I have white carrier board EI. I have ran Seventax FreeV2G code. ... CAn you help me about testing ccs charge controller with these board? How can I test these boards

Sorry, I do not any experience with these boards.

/* our link-local IPv6 address. Todo: do we need to calculate this from the MAC? Or just use a "random"? /
/ For the moment, just use the address from the Win10 notebook, and change the last byte from 0x0e to 0x1e.
You wrote these comments in code (ipv6.c). Idont understand here. Can you explain

I think you can ignore this comment. The question was, whether the IP address of the PEV needs to be calculated with a certain formula out of
the MAC address. My current understanding is: No, it does not matter. So I just took the IP address which my notebook uses, and changed one byte. The only necessary thing is, that the IP addresses of the PEV and EVSE should be different. Theres a high chance, that this is fulfilled.

SLAC matching succesfull. Im using white carrierd board EI with freeV2G as evse side. stm32 and red carrier board as pev side.
Also after slac matching succesfull. [SDP] initiating SDP request. I dont pass the [SDP] initiating SDP request. How can I do sdp

The clara software sends SDP requests. The EVSE should respond to these requests. If this does not work, it would be helpful to have a log file of each side, and in best case a ethernet trace to see which side is sending which data, to understand, where the flow stops.

[olegiv]

I am very sorry - why is Serial Flash Memory installed for QCA7000?

[uhi22]

Because the QCA does not have internal memory for its firmware.

[olegiv]

Because the QCA does not have internal memory for its firmware.

Thanks, while I don't understand how to program and what firmware to use - I will read further

[uhi22]

The setup to program the SPI flash is described here: https://github.com/uhi22/foccci#program ... -spi-flash

[olegiv]

The setup to program the SPI flash is described here: https://github.com/uhi22/foccci#program ... -spi-flash

Thanks again - as I understand it - if RESET QCA is set to GND, the corresponding QCA pins go to impedance state and it is possible to connect to these pins externally. I will try it. I'm still redesigning the PCB for other connectors

[xvyDFatih]

Thank you very much indeed for making such a study and sharing it as an open source

johu: quote redacted

[johu]

I was asked via PM (please don't do that) about the BeagleBone

Read here: https://openinverter.org/wiki/CCS_EVCC_using_AR7420

[asavage]

(This is an animated gif, but is not show the animation inline?)

The forum software generates a GIF thumbnail, but animation does not survive this process.

If the thumbnail is selected, the animation is active in the resulting popup.

If you want the animation active in the post without user selecting it . . . there's a ridiculous workaround. Basically, after an image is in your "gallery", you can refer to it in other posts later:



But, in order to have the image in your "gallery", it has to be a finished attachment, which means you have to submit the post first, then edit it to refer to the external "gallery". And you end up with the "unused" attachment being displayed at the end. It's awful, but cosmetic. I realize I have not adequately described this, sorry.

[uhi22]

Great! Not an obvious and simple solution, but at least it works

Some progress here: To ensure that the units which I plan to ship will work as intended, I'm creating a little test bench. The main point is solder-less contacting, without the need of the Deutsch connector. I use standard quare pins, stuck in wood, and a knee lever to get sidewards pressure on these pins. Adding some switches to simulate temperatures, button, PP and so on. Some LEDs to show the state of the outputs, and a small display (https://github.com/uhi22/TFT-with-CAN) to show some measurements from CAN. Also will connect the CP to my EVSE-playground (which uses homeplug modem and pyPLC).

[asavage]

That's odd, I had originally just inserted a link.

I took a look at this today. The phpBB forum software evaluates any URL we post and postprocesses links to certain sites (like Amazon), to instead embed media rather than the actual link we post. This only occurs if you paste a bare link, without wrapping the link in [ url ]...[ /url ] bbCode tags (which I assume most people don't).

You posted something like

Code: Select all

https://www.amazon.com/IWISS-Crimper-Impression-Deutsch-Contacts/dp/B07V1C6N2V

which in a sane world would translate to this in HTML

Code: Select all

<a href="https://www.amazon.com/IWISS-Crimper-Impression-Deutsch-Contacts/dp/B07V1C6N2V" target="_blank">https://www.amazon.com/IWISS-Crimper-Impression-Deutsch-Contacts/dp/B07V1C6N2V</a>

(interpreted: https://www.amazon.com/IWISS-Crimper-Impression-Deutsch-Contacts/dp/B07V1C6N2V)

But, because the forum's code recognized that the URL pointed to one of its favored corporations, it instead translated that to a media embed for Amazon

Code: Select all

<span data-s9e-mediaembed="amazon" style=[blah blah] style="background:url(https://www.amazon.com/favicon.ico) [blah blah] src="//ws-na.assoc-amazon.com/widgets/cm?l=as1&amp;f=ifr&amp;o=1&amp;t=&amp;asins=B07V1C6N2V"></iframe></span></span>

The link that's in the final display HTML contains a pull from Amazon (//ws-na.assoc-amazon.com) and on my end the combination of FireFox's Enhanced Tracking Protection feature -- which I have enabled -- and a FF add-on, Ad-Block Ultimate, conspired to not display anything on my end.

That's why I saw nothing, even though you'd dropped a URL in your post.

I almost always post URLs using the bbCode link tags, rather that a bare URL, because I've used several forums that like to take what I wrote and mangle it, just as in this example. I hate when what is displayed isn't what I wrote.

It's not your problem that I have various website/corporation tracking methods disabled on my browser, and I don't expect anyone to change how they post, I'm only providing the explanation why what you think you posted may not align with what people are seeing.

[I've hosted my own phpBB forum since 2007. They sneak these kind of "features" in every update.]

[asavage]

The next version will have the TE DEUTSCH DTM13 connector.

I burned a few hours reading scattered PDFs, and I'm hoping I've ordered the correct enclosure/mating plugs/sockets/cavity plugs:

• Deutsch Connector, PCB 90°, dual receptacles (grey, black), DTM13-12PA-12PB-R008, https://www.mouser.com/ProductDetail/571-TM1312PA12PBR008 (b/o until Jan2024)
  https://www.newark.com/deutsch/dtm13-12pa-12pb-r008/wire-to-board-connector-dual-row/dp/26W9682
• Deutsch Enclosure for focci, EEC-325X4B-E016, translucent (I prefer over opaque), https://www.mouser.com/ProductDetail/571-EEC-325X4B-E016
• Deutsch Plug ‘A’ (grey), w/short (grey) cap, DTM06-12SA-E003, https://www.mouser.com/ProductDetail/571-DTM06-12SA-E003
• Deutsch Plug ‘B’ (black), w/short (grey) cap, DTM06-12SB-E003 (black cap would be -E005, but N/A), https://www.mouser.com/ProductDetail/571-DTM06-12SB-E003
• Deutsch WedgeLock for 12-way Plug, WM-12S, (2), https://www.mouser.com/ProductDetail/571-WM-12S
• Deutsch terminal sockets, 16-18 AWG (purple stripe), nickel plated, size 20, 0462-005-20141, https://www.digikey.com/en/products/detail/te-connectivity-deutsch-ict-connectors/0462-005-20141/7200802
• Deutsch terminal sockets, 20 AWG (no stripe), nickel plated, size 20, 0462-201-20141, https://www.digikey.com/en/products/detail/te-connectivity-deutsch-ict-connectors/0462-201-20141/7200930
• Deutsch cavity plug, size 20, 0413-204-2005, https://www.digikey.com/en/products/detail/te-connectivity-deutsch-ict-connectors/0413-204-2005/7387531

I'd ordered the terminal sockets and cavity plugs a month ago. They arrived yesterday; the order was delayed due to B/O of the cavity plugs. I ordered the enclosure/plugs/wedgelocks just now. I sure hope I got most of it correct.

[uhi22]

What is your plan about the J20 connector (the part which need to be soldered to the PCB) itself? The JLCPCB does not have it, so the PCBs will arrive without this plug. In your list above I do not see it.

[asavage]

Thanks, I missed that. I've edited the list above to include J20.

[ns0708]

The setup to program the SPI flash is described here: https://github.com/uhi22/foccci#program ... -spi-flash

Also thank you very much for doing such a research and sharing it as open source！！！

I have a question for your guidance：

Transmit power calibration
Directly burn bin without calibration, can the PLC signal be sent normally on the new PCB?
Whether special calibration operations are required. (CCM_FlashDump_SpiFlash_2MB_Ioniq_00_33_79.bin)
The transmitting power, no anomalies?

th！

[catphish]

I've just had a look at the FOCCCI board (actually mostly just the buck regulator), and I would strongly recommend making some layout improvements here for the next revision of the board. In particular, the current paths for the input and output capacitors are much too long.

1) The input capacitor (C65) should be as close as possible to the TPS54302 (U4).
2) The output loop U4->L13->C66->U4 should be as short as possible.
3) There should be no copper under the inductor L13.

Please see attached my layout of this SMPS as an example. Feel free to copy it, or if you prefer, I can do a pull request for this. It would likely be easier to use the smaller inductor I suggested.

Apologies for the criticism, and clearly it works, and I'm sure more tidying up is planned anyway, but I definitely think it's worth improving upon this.

[uhi22]

I appreciate this constructive feedback very much. Yes, would be great if you create a pull request. And yes, many details I did in a hurry, and are not optimal. There is definitety room for improvements. Currently I'm concentrating to finish a small batch of the V3 version, the boards run, software will grow.

[uhi22]

Transmit power calibration
Directly burn bin without calibration, can the PLC signal be sent normally on the new PCB?
Whether special calibration operations are required. (CCM_FlashDump_SpiFlash_2MB_Ioniq_00_33_79.bin)
The transmitting power, no anomalies?

I do not have the top secret documentation of the QCA, so I'm not even aware that a calibration is possible.
It is anyway total overkill, to use a physical layer which was developed for power line comm in big noisy and unpredictable building networks in the totally different use case. We have max 5meter cable, well defined setup, no coupling to something outside, only two participants, No other RF. So the high transmit power (several volts in the line) is not necessary, also not the microvolt senitivity of the receiver. In the boards of my Test setup, I usually add attenuation in both, transmit and receive path, a to avoid unnecessary radiation and increase immunity. A difference of some dBs doesnt matter. So: no, no calibration at all.

[olegiv]

Has anyone tried using QCA's JTAG?
Is it possible to use JTAG for flash programming?
So far I have decided to program via the free SPI U3 (STM32).

Pin QCA PLL_BYPASS "normally 10кОм to ground" - datasheet. The fact that this QCA pin isn't connected to ground will not affect the operation of the modem?
https://github.com/Millisman/QCA7000 - Pin QCA PLL_BYPASS through 10кОм is connected to ground.