BQ51050B Coil capacitors

Let me describe my problem bit more clearly.
I'm working on receiver that is build on BQ51050B and custom PCB. The thing is that it does not work at all.

I don't have oscilloscope so I cannot measure it properly. I can only see that there is voltage about max 3V between GND and BOOT (both 1 and 2).

That is why I have asked previous question - how to choose proper coil capacitors ?
I cannot say that I know how to, because the level of my knowledge is not good enough. Still.

Thank you,
Robert

Robert,

Please review the section on "Series and Parallel Resonant Capacitor Selection.  Also, I would suggest reviewing the bq51050BEVM0764 User Guide.  There are a number of helpful items there.  Key to the Cs capacitors is they must be 25V rated (or higher).

Debugging without an oscilloscope would be very difficult.

Look at the RECT pin to get more information on actual power transfer.  If RECT does not get above VUVLO, the system cannot power up.

During "ping" the RECT will rise and if high enough, will start communication.  If it's not enough, the ping will continue with a ping every 500ms or so.

Regards,

Dick

Hi Dick,
Thank you for your answer.
Cs capacitors are rated 25V.

The voltage on RECT is about 1,5-1,6V max.

Refering to e2e.ti.com/.../1758479 I was looking for sluc566 but without luck.

Robert,

I am checking, but the SLUC566 document is no longer available.  It was specific to the bq51221 and based on what I could find on this, it would not have helped with the issue you're seeing.

Are you starting with a known good transmitter?  Have you measured Ls' and calculated the required Cs value? For a start, you can estimate the Ls' to be 20% higher than the rated inductance.  I do not recommend doing this for your final product, but to get started, it is a decent range.  It could vary widely.

The coupling of your system is not good enough if RECT is that low.  Again, using a DC meter is not a good method since it averages the voltage across time.  An oscilloscope is really required for meaningful debug.

Regards,

Dick

Hi Dick,
This is what I asked about at the top - how to calculate Cs value.
Your answer provides another question - how to known what is the value of rated inductance ?

I will try to build simple oscilloscope and then get back with my findings.

Thank you,
Robert

Robert,

As noted in my first reply, the section in the bq51050B datasheet 10.2.1.2.1 titled Series and Parallel Resonant Capacitor Selection covers how to calculate Cs.

Cs (also referred to as C1 in some sections) is:

C1 = 1/((2*Pi*fS)^2 * Ls')

fS is defined as 100 kHz which is the resonant frequency of the RX and TX systems.

Ls' is the inductance measurement done on a specified ferrite shield.

The rated inductance should be in the datasheet of the inductor.

Regards,

Dick

Dick,
Thank you for pointing this. How about value for C2 ?

And what should be the (min) value on RECT to start ?

Thank you,
Robert

Am I calculating this properly - my coil have 10uH.

C1 = 1/(2*3,14*100)^2 * 10)

Robert,

C2 calculation is in the same location of the datasheet.

You will note that C2 is dependent on C1.  You must calculate C1 and choose the actual value of the capacitors chosen (not just the calculated value) for the C2 solution.

Your calculation shows "100" for the frequency and it should be 100,000.

For 10uH, the C1 (Cs) value would be 252.2nF.  But, I expect that 10uF is your Ls, not Ls'.  As noted earlier, a starting point would be 20% higher - although this is very dependent on the coil shielding.

For 12uH, the C1 value would be 211.1nF.  I would recommend using 3 capacitor to get this value.  Two choices would be:  68nF + 68nF + 68nF (204nF total) or 68nF + 68nF + 82nF (218nF total). 

Using 204nF as the value for C1, the C2 value would be 1.63nF.  Choosing 1nF and 0.68nF would be a reasonable solution.

The Electrical Characteristics table shows VUVLO to be 2.7V nominal.  See section 9.4.3.9 of the bq51050B datasheet for details.

Dick

Hi Dick,

I really appreciate you help!

I made simple oscilloscope and this are the results from RECT line:

First coil, I have no idea what is the inductance (waiting for multimeter to measure it).

Looks clean, but as I mentioned I have no idea about inductance so I cannot calculate capacitance.

Second coil. Inductance is known 68uH:

I did not yet switch Cs to match this.

In addition sometimes there is some kind of noise:

The question is what should the value of C1 and C2.

Am I right that C1 should be around 31nF and C2 arround 38pF ?

Robert,

These plots show me that RECT is not getting enough power to begin communication with the TX and to start power transfer.  The coupling is not high enough.  This could be the anything from the transmitter being used to the coil, to the spacing to the Cs (C1) values discussed above.

What transmitter are you using?  What is the physical size of the coil?  What is the size of the coil shield?  68uF is quite high compared to the coils typically used for wireless power transfer.  The series resistance is also quite important.

Can you send a schematic of your system?

Regards,

Dick

Hi Dick,

Please find bellow schematic of my solution.

I'm testing now two coils from Wurth Elektronik. I know that they are not designed for Qi but due to small size I would like to use them.

First, size Ø 10mm:

inductance 24.2uH, C1 - 87nF (22+22+47), C2 - this is where I'm not sure how much it should be - with 2nF I was able to run it but it was not stable

Second Ø 6mm:

inductance 7.2uH, C1 - 293nF (100+100+100), C2 - the same as previous.

If you could help me with choosing right capacitance it will be great.

Thank you,

Robert

Robert,

The calculations for the capacitors seem reasonable.

The issue now is the coil size.  If you are using a standard Qi Transmitter, you will have great difficulty getting this system to work.  The coupling between the coils is not enough to work consistently strictly due to the ratio between the coils.  The optimum ration is 1:1.  As the ratio get farther apart, the coupling decreases and it becomes quite difficult to get enough coupling for a satisfactory power transfer.

See the TIDA-00334 or TIDA-00415 for transmitter options that are not Qi compliant.

Further, with your charge current set to 77mA, you may not get consistent results across a wide sampling.  Setting the termination current to 7.7mA will give a wide variation as well.  For a very precise solution, you may consider the bq51013B receiver (drop in replacement for the bq51050B footprint) and the bq25100 low power charger.

Regards,

Dick

Robert,

C1 and C2 are based on the inductance of the coil. Without knowing the inductance, we cannot calculate C1 and C2.
Have you seen this article on "Designing a Qi coil"?
www.ti.com/.../slyt477

Regards,
Dick