BQ51003: Issues to initiate stable power transfer with certain Qi transmitters

Leonhard Renk

Part Number: BQ51003

Tool/software:

Dear TI support team,

I am working on a design including the BQ51003 as a wireless QI receiver. It consists of the RX coil (760308101214 from Wuerth Elektronik), two ferrite sheets of 0.3mm thickness (354006 from Wuerth Elektronik), a PCB (see schematic attached) and a LIR2050 battery.

I meanwhile managed to stabilize this design in terms of the occasional interruptions that I observed earlier during my investigations: BQ51003: Waveform at FOD and ILIM pins - Power management forum - Power management - TI E2E support forums In particular I have added two circular ferrite sheets of 23mm diameter and 0.3mm thickness between the RX coil and the PCB to improve the shielding of the PCB. On top of that, I have replaced R6 in the schematic by a fixed 10kR resistor to disable the TS functionality since there are currently two resistors missing in the design in order for it to work as intended.

With these changes the battery is now reliably charging with 4 different Qi charging pads that I have available for testing. At the same time I am still struggling to establish a stable power transfer with two other transmitters (NORDMÄRKE from Ikea and 21310 from Trust). My assumption is that this a result of increased distance or worse coupling between the TX and RX coil for these two charging pads compared to the others.

Please see below two waveforms of the RECT pin when I place the RX coil directly on the Ikea charging pad. At first it seems like the power transfer is successfully established for appr. 6 seconds before the RECT pin eventually drops low. The second screenshot is a zoom of the first screenshot when the RECT voltage drops low.

When I remove the housing of the Ikea charging pad and place the RX directly on top of the TX coil I end up with a stable power transfer without the voltage drop as seen above. Please see the picture below for a comparison of the coil dimensions.

I have already adapted the resonance circuitry for the additional ferrite sheets with the following values:

LS measured as 28.11uH
LS' measured as 31.18uH
C1 calculated as 81.24nF --> 47nF + 33nF chosen for testing
C2 calculated as 911.38pF --> 910pF chosen for testing

Apart from that I have also tried different values for the COMM capacitors (10nF, 22nF, 33nF) without any significant improvement.

Do you have any suggestions for changes to my design that could help me to overcome this problem?

I am very limited in terms of the available space inside the receiver housing, so choosing a larger coil is not really an option. Nevertheless, I could give another RX coil a try that I have available here. It's the 760308101303 from Wuerth Elektronik with 47uH and 26mm in diameter. Would that make sense to improve the coupling factor between the TX and RX?

Any kind of help is highly appreciated. Thank you in advance!

Best regards

Leonhard

3 months ago

0 Bill Johns 3 months ago

TI__Guru 68886 points

Hi Leonhard

The AD pin (G4) could be the problem -- It is floating. Needs to be connected to ground with 10k or 1uF capacitor.

0 Leonhard Renk 3 months ago in reply to Bill Johns

Prodigy 30 points

Hi Bill
Thank you for the quick reply. I appreciate it.

You are totally right about the missing ground connection of the AD pin. I am going to implement it in the next revision of the PCB. With the current PCB design it is unfortunately not possible to manually create that connection, so I can not test the impact of it.

What's the reasoning behind your assumption that this could lead to the problem I am facing right now? Is it based on the assumption that the voltage at the AD pin could reach 3.6V which would then trigger the wireless charging to shut down? Have you observed something like that before? Isn't the same shutdown behaviour also to be expected then for the operation with all the other TX charging pads that I am testing with?

When connecting the AD pin to ground in the next PCB revision, is it really necessary to use a 10k resistor or 1uF capacitor for this purpose? According to the datasheet a direct connection to ground is recommended for my case with no wired adapter input available.

Thank you and have a nice day!

0 Bill Johns 3 months ago in reply to Leonhard Renk

TI__Guru 68886 points

Hi Leonhard

Floating AD pin needs to be grounded to prevent random shutdowns. Things to look for to determine if it is causing shutdown is a comm packet just before the TX shuts down. We see this in the scope capture, the TX turns off just after a packet. The packet is an End Power Transfer (EPT).

I have seen this problem before.

0 Leonhard Renk 3 months ago in reply to Bill Johns

Prodigy 30 points

Hi Bill

I will then go for a redesign of the PCB and connect the AD pin to ground. Hopefully this will solve the problem.

Is it sufficient to directly connect the AD pin to ground, or should I include a 10k resistor or 1uF capacitor as mentioned in your initial reply?

0 Bill Johns 3 months ago in reply to Leonhard Renk

TI__Guru 68886 points

Hi Leonhard

Yes, OK to ground AD pin.

The 10k pull down gives you a little more fixability, but not a must.

1uF would be used with a circuit where the voltage is applied to the AD Pin and low leakage current is desired.

0 Leonhard Renk 3 months ago in reply to Bill Johns

Prodigy 30 points

Hi Bill

Thank you a lot for your support! I will go with the 10k resistor from AD to ground then and continue with a redesign of the board. Hopefully this will solve the observed shutdowns.

Have a great weekend!