MWCT1000 calibration errors with FreeMASTER

[droberson]

Hello,

We are attempting to calibrate the MWCT1000 wireless charging transmitter using "WCT1000 A11 Reference Design Calibration User Guide.pdf" and the firmware provided by NXP. Below are some technical details:

FreeMASTER version: 3.2.2.2

MWCT1000 Software: WCT1000_A11_V3.3

FreeMASTER interface: UART (19200 baud)

Board: WCT-5W1COILTX

We are attempting to calibrate using the reference board (WCT-5W1COILTX) to rule out any issues with our hardware and are receiving the error in the attached screenshot. I am performing the following steps to run calibration:

1. Set LOW_POWER_MODE_SUPPORTED to FALSE

2. Set FMSTR_USE_SCI to 1

3. Set FMSTR_USE_JTAG and FMSTR_USE_JTAG_TXFIX both to 0

4. Program the board

5. Open the .pmp file in the project folder.

6. Set the interface to UART (19200 baud) in the Project->Options dialog

7. Click the "GO!" icon and observe a successful connection. I can now monitor the RunTimeParams and observe the RaW Currents graph receiving data.

8. Select "Wireless Charging" at the top of the Project Tree.

9. Scroll down to FOD Calibration and complete steps 2 and 3 and receive green check marks.

10. Set the coil frequency to 200 MHz and receive a green check mark.

11. Click "Read" and observe a red triangle with the message "Board is not responding, timeout error: 2.1s"

Why is this timeout occurring? Our devices will not work without FOD calibration. I found a similar thread with the same issue here: https://community.nxp.com/t5/FreeMASTER/Freemaster-error-when-calibrating-MWCT1000/m-p/1501369

However, that thread's solution of selecting "Refresh" does not work in that the "Input power" field is never updated after clicking "Read" in the FOD Calibration section. The timeout occurs before this field is updated and the field remains at a preset constant value.

I also created NXP support case with ID: 00568891 a few days ago but I have not received a reply.

Thanks,

Derek

[droberson]

For anyone else that comes across this thread, I managed to get FreeMASTER communicating with the MWCT1000 A11 board using the USB-TAP interface by following the the steps in the thread linked below.

https://community.nxp.com/t5/FreeMASTER/FreeMASTER-3-1-and-USB-TAP-not-working-0x84b30004/m-p/147707...

Summary: The timeout errors in my initial post above still occur even when using the USB-TAP interface. I do not recommend using the MWCT1000 even though it is still active on the NXP website as of 10/20/2023.

Derek

[aaronwang]

From the description, the running steps are all OK, the Freemaster can be connected, and some buttons get correct green marks. In usual, the FOD calibration should work, there is no any other tricks here. I suppose the failure causes can be:

1) The Freemaster connection is unstable, please try to re-plug the connecter and try again;

2) There may be some compatibility issues between latest Freemaster version and old Freemaster pmp file, the WCT1000 is an old project, using Freemaster V2.0 or V2.5, please find old Freemaster version tools for try;

3) The Tx may get some hardware errors, for example, when setting the frequency to 200KHz, the board gets stuck and collapses, it needs some debugging works.

4) Etc.

[droberson]

Hey Aaron,

Do you have download links to the older versions of FreeMASTER you recommended? I can't seem to find them on the NXP website or by Googling.

Regarding "The Tx may get some hardware errors, for example, when setting the frequency to 200KHz, the board gets stuck and collapses, it needs some debugging works." Not sure I follow this. I am using the NXP eval board with no load and the NXP software, why would I need to debug anything? I would expect it to work out of the box per the calibration guide. Does there need to be a load for FOD calibration to work?

Edit: I found a link to FreeMASTER version 2.5 by modifying the link to the latest version. Version 2.5 has the same issue. Unfortunately, this trick doesn't work for version 2.0, so I still need a link to FreeMASTER v2.0.

Version 2.5 for reference: https://www.nxp.com/webapp/sps/download/license.jsp?colCode=FMASTERSW25&appType=file1&DOWNLOAD_ID=nu...

Thanks!

Derek

[aaronwang]

Hi, Derek

Frankly speaking, WCT1000 A11 wireless charger is a very old product, the development enviroment is Codewarrior 10.7 and Freemaster V2.0+, these tools have been updated to new versions in our website, like CW11.2 and Freemaster V3.2.  I searched the website and found no old versions remained.

If the A11 board on your hand is a new complete kit, the software inside should be with golden calibration parameters, no need to redo the FOD calibration, it should work with all Qi standard BPP Rxs. You can basically try the charging function with standard Qi receivers and check out if there is any HW errors with the board.

I am not sure about the specific cause about the issue you meet, and can't give the exact answer without field researching. Just from my point of view, the possibles reasons can be belows: development tools complibility, hardware error and connection unstability.

Some more suggstions for you,

1) After any software changes, should re-complile the entire project for downloading;

2) Make sure the elf file imported into the Freemaster project is the right one running in the board;

3) Try other freemaster functions, like variables monitoring, to check if the Freemaster connection is OK or not, and make sure the elf file matached.(If wrong elf used, the varaibles' values will be in chaos)

4) Try JTAG connection way for the calibration steps.

[droberson]

Hey Aaron,

I am using the "A11 board" as a test to rule out any potential issues with our custom designed board. We have integrated the MWCT1000 on a custom board hence the need to run calibration there. Other FreeMASTER functions work fine such as real-time current measurements and plotting, etc, so I don't think the UART interface is unstable. I was unable to get the JTAG interface to work with FreeMASTER last year due to a bug in the FreeMASTER software reported by NXP, but I may give that another go.

Worst case, we may just disable foreign object detection (FOD) using the FOD_ENABLE define.

Question: Is there any risk in disabling foreign object detection (FOD)? Will this result in the device attempting to charge foreign objects that could lead to bodily injury, fire, etc?

I didn't think that meaningful power transfer would occur until the wireless charging transmitter negotiated with the wireless charging receiver?

Thanks,

Derek

[aaronwang]

Hi, Derek

By now I can't figure out the root cause about your problem, so if you strongly want to fix this Freemaster issue, I suggest you to contact our marketing guys and send us your hardware for further debugging.

About the FOD, I am pretty sure that the calibration data in our A11 reference code is the golden one which passing Qi certification, there should be no such big power loss error if the Tx board is OK. So instead of disabling the FOD, I suggest you to loosen the FOD thresholds to avoid some false trigger cases, the FOD thresholds can be set in Freemaster GUI as NVM data.

If disabling FOD in a BPP wireless charger, the FO can't be detected in time, no matter it is placed with Rx together or inserted during charging, which can lead to overheat and burning in worst case.

The Tx can only start to transfer power after successfully negotiate with Rx, but it has no relationship with FOD. If the FO exists during BPP Tx and Rx negotation, it can still success to enter charging state, another case is that FO can be insert into a charging system. 

FYI.

[droberson]

Hey Aaron,

I have a few follow-up questions to your comments.

1. Are you saying that a foreign object (FO) can be energized when placed between the TX and RX device?

2. Can a foreign object (FO) be energized if placed on the TX coil with no RX device present?

3. Since we have integrated the MWCT1000 reference design into our custom board, my understanding is that we must run the calibration procedure. Are you saying that the A11 "golden" calibration data will work on a custom designed board if it matches the reference design from the A11 board?

To clarify, I am testing on the A11 board to verify the design of our custom board. If we can't run calibration on the A11 board, then we definitely won't be able to run it on the custom board.

Thanks,

Derek

[aaronwang]

Hi, Derek

For your questions,

1) If the FOD function disabled and the FO(metal object) can't be detected and protected, the FO between in the magnetic field area can be heated and get hot, like a induction cooker;

2) Without Rx, the Tx is in standby state, there are some analog and digital pings running with fixed intervals, the energy is weak, the FO should be safe;

3) If you want to bring up a customer's board, you'd better to do full calibration steps to get best FOD performance. The golden calibartion data of NXP A11 can only be used for some basical tests, can't be directly used for customer's formal products;

I understand your situation now, maybe you can ask our marketing guys for help, or directly contact our local FAEs for field debugging.

Frankly speaking, A11 is an old Qi V1.2 design, it can't be registed and certified in Qi lab now.(latest supported version is Qi 1.3.3). And since the software is a little bit old, there may be some compatibility issues with new Rxs. I suggest customers to develop the products based on our latest reference designs.

[droberson]

Hey Aaron,

Thanks for your clarification on FOD. In regard to your comment " I suggest customers to develop the products based on our latest reference designs", we chose the MWCT1000 last year because it was the only part available due to the global supply chain shortage and because it was recommended by NXP due to this. NXP reassured this part was not EOL (End of life) when we asked if it was still supported. Their response was "the MWCT1000 is an active part, adopted by many customers and there is no current plan for end of life... The majority of the customers using and designing w/MWCT parts are auto related so we design to and support long term these products. The reason that we are on this part with your business case is because of the inventory that you saw and that’s what drove the MWCT1000 design you are working on."

The part is still active according to the NXP website: https://www.nxp.com/part/MWCT1000CFM#/

Given the above, I am somewhat confused and would expect this product to still receive support and compatibility updates with FreeMASTER as needed.

Can you provide contact details for our local FAE's, or know who I can ask to provide them? We are located in North Carolina, USA.

Thanks again for your help!

Derek

[aaronwang]

Hi, Derek

I'm not saying that the A11 design has problems, it is a success project with lots of design wins. I just mean it is a little out of date, so I gave the suggestion. Anyway, A11 can still be used, if the design can meet your requirements.

I post the contact information of our marketing and local AE as below. You can ask them for help.

Marketing:  Vaclav Halbich <vaclav.halbich@nxp.com>

AE: Jaehee Park <jaehee.park@nxp.com>

[droberson]

Hey Aaron,

I have reached out to the contacts you mentioned, thanks for providing those!

Last question: Did you try to run FOD calibration on the MWCT1000 A11 development board using UART and FreeMASTER?

Thanks,

Derek

[aaronwang]

Hi, Derek

WCT1000 A11 was released by us, I definitely tried this connection methold and confirmed no problem. But as I said in previous answers, A11 is an old project, I havn't touched it in several years, so I am not sure it is still OK with new development tools. Right now we don't have any A11 reference demo board on hand, and can't do any tests. Please contact our marketing guys for any request. Thanks!