LDC2112: LDC2112 coil questions

sorry since picture didn't show, I write again 

1. Do conductor target need placed at INn side or COM side or both are ok?

2. Or conductor target need placed at induced magnetic direction side like below yellow mark arrow?

3. Below is our coil layout, is this type ok?

Thank you~~

BR/Poki

Hi Poki,

Thanks for your post. We will have Kristin get back to you next week as we are in training early next week. 

For your question 1, the conductive target needs to be where the user is expected to push the target. 

Hi Arjun:

Thanks for feedback

And please taka a look detail at above green area highlight

Is our pattern(left) ok? Or need follow design guide pattern(right) ?

Thank you~~

BR/Poki

5852.ITB_FPC_0518.pdfITB_FPC_FEMM_v1p28_0518.pdf

Hi Arjun:

Sorry add more questions

1. Any criteria between coil and GND distance? For our design, the coil and GND distance is 6mil, is it ok?

2. Any criteria for INn length? For our design, from golden finger to coil is 20mm, will it too far?

3. Attached below is our SCH and design parameter, could you help check if it ok?

Thank you~~

Hi Kristin:

Thanks for feedback

1. Ok I will try to increase coil turns

2. OK I will let sournd GND as much as possible far away from coil

3. Could you help me to run this FEMM by our parameter and feedback me the result? Since we have run once but can't understand the report(just see two coil chart)

    Or can we use the excel "LDC_Tools-ext49", the sheet "SkinDepth", enter the target material , sensor frequency(calculated by sheet "Spiral_Inductor_Designer") and material thickness

    to check if target is at least 3 skin depths thick?

Thank you~~

BR/Poki 

Sorry add one question

Since coil top/right side has more trace, I can just let GND faw away at left/down side, it help?

Or the GND sournd coil need symmetrical?

Thank you~~

Hi Kristin:

If we increase the turn in the center of the coil, the Coil Fill Ratio will under 0.2 (spec. is 0.2~0.8)

So we can ignore this spec. or still need larger 0.2?

Thank you~~

BR/Poki

Hi Poki,

I can verify the FEMM simulations, and I can provide feedback by the end of the week.

Yes, it would help to reduce the ground plane on the left side. The GND does not need to be symmetrical around the coil, and the less GND that is around the coil, the better. This is true for all copper pours near the coil, not just GND.

You can ignore the coil fill ratio spec. Most applications should stay within the recommended range, but button applications are a special case that benefit from more inner turns. You should maximize the number of inner turns.

Best Regards,

Hi Kristin:

Thanks feedback

1. You said "It would also be good to verify the COM capacitor value with the FEMM output data. The simplest way to do this is to copy the outputs of the FEMM simulations and input them into the LDC2114 section of the LDC Tools excel calculator." you mean the "LDC_tool-ex49.xlsx" excel file and the sheet "LDC2114_Config_tool"? Or any other one?

2. So we only need consider FEMM result for L/Rp/Q/sensor frequency with target right?

3. Since FEMM simulation need a lot of time, could you help create a LC sensor calculations file that use 430 stainless steel to me? So that we can speed up

Thank you~~

BR/Poki

ITB_FPC_LDC_Tools-ext49-skin depth-0521-1500.pdf

ITB_FPC_FEMM_v1p28_0521-1500.pdf

Hi Kristin

Sorry add more questions

Since we need use stainless430 as target material, and the coil dimension need around or smaller than 11.75*7.95

With FEMM simulation, we found only thin trace width with large gap can pass all criteria (W2.5/G24), and only 4 turns (please see attached)

Could we design real layout with attached parameters? Becuase larger gap and few turns seems unreasonable

If can't, could you help how to design with stainless430 and coil dimension around or smaller than 11.75*7.95??

Also, where can I find the Ccom value with FEMM simulation?

Thank you~~

BR/Poki

Hi Kristin:

We have another FEMM simulation with stainless430 (W2.5/G15), and only 4 turns, dimension is 4*11.75 (please see attached)

Could you help check if it's ok?

Thank you~~

BR/Poki

ITB_FPC_FEMM_v1p28_0521-1500(Poki).pdfITB_FPC_LDC_Tools-ext49-skin depth-0521-1500(Poki).pdf

Hi Poki,

I'd recommend using more turns than 4. In the latest simulation you've also decreased the size of the coil, which is a good idea, but I think you should shorten the longer dimension, not the shorter dimension (if possible). The length of the shortest side of the coil will determine the sensitivity and the sensing distance. You could try a 4mm x 8mm or 6mm x 8mm coil with more turns to see if this would work. 

Unfortunately the FEMM simulation results don't recalculate the Ccom value for you, but you can copy the FEMM outputs to the Quick Sensor Rp/Rs/Q Calculator at the bottom of the LDC2114_Config_tool.

Best Regards,

ITB_LDC_summary stanless 430-summary-0524.pdfHi Kristin:

I increase coil turns but it need decrease spacing between trace

And the Q factor will higher than 30(please see attached)

Seems it can't be done with such dimension but more turns

What will be if all parameter in spec. but  turns is less and spacing between trace is bigger?

And what will be if more turns but Q higher than 30?

Thank you~~

BR/Poki

Hi Kristin:

Since FEMM tool only provide two Aluminum(Aluminum, 6061-T6 and Aluminum, 1100)

Could we use Aluminum, 6061-T6 to simulate other Aluminum material(GM55 H38; TM60 HSM(AL6063); AL5052 H32,  as attached)?

Or could you set those  Aluminum material into FEMM tool to let us simulate?

thank you~~

BR/Poki

AL Comparsion-GM55TM6050526061.pdf

Hi Poki,

Thanks for your patience while I was out of office for part of this week. 

If all of the parameters are within spec, but the trace spacing is very wide, you might see a reduction in sensitivity. Typically we recommend setting the trace spacing to the minimum possible on your board.

If the sensor Q is higher than 30, the LDC2112 may not be able to drive the sensor. The Q value should not exceed 30.

I agree that it seems the current dimensions can't meet the above constraints. You could try reducing the coil size a little more. Alternatively, using an aluminum target will solve the issue without needing to change the coil design, and you can likely add more turns.

I would definitely recommend using the AL6063 over the GM55 H38. It’s almost twice as conductive, which will improve the target response and reduce resistive losses in the eddy currents. There is a way to add custom materials to FEMM for the simulations. You can follow steps (d) and (e) on pages 9-10 of this FEMM app note, which shows how to create FEMM simulations from scratch. You should only follow sections i-iv of step (e). Then you can specify the custom material in the excel FEMM tool. Alternatively, the AL6063 conductivity is in between AL1100 and AL6061. If the simulations look good for both of those materials, you can assume that the AL6063 will yield a valid sensor as well.

Best Regards,

Hi Kristin:

About INn trace length, I know you have feedback my question as below

20mm should be fine. The coils can be remotely located from the device

Since we will do a EVB that need extend INn trace like attached picture

So the INn trace will be 8mm(on MB trace length)+GF(zif connector)+32.8mm extended FPC+GF(zif connector)+20mm FPC

Can we add this extended FPC? Is this INn trace length OK?

If not, how long will be ok?

Thank you~~

Poki