Making a linear slider control with the LDC1000

Hello,

The coil and target design is defined by available space in your application. The larger - the better precision you will get.

The material most commonly used is copper layer on PCB.

If you will tell us more about your application, we'll be able to provide more detailed recommendations.

Hi Evgeny,

The application is for a slider control, on a hand-held device for wireless control of a lens in the film industry.

As the device is hand-held, space is of course limited, particularly in the length of the slider which needs to be somewhere between 50 to 75 mm in length approximately.

I was hoping that I'd be able to have the target in a fixed position with the coil moving over the target, and again, the smaller the diameter of the coil, the better; I'd say ideally no more than 12mm diameter. It would also be good if the coil could be remote from the LDC1000, so that it could be at the end of two-core folded flat ribbon, rather like an inkjet printer head.

The coil could be copper tracks on a pcb, but I'm not clear what material I'd need for the target, or how to design the target shape as shown in your illustration of a similar application. Also, advice on the coil design for this configuration, and the optimum distance from the target would be appreciated.

The aim is to have a slider control sealed from the weather and with higher resolution than using a slider-pot and adc.

Thanks,

Geoff

Hi Geoff,

2-layer 12mm coil will work. 12mm dia, 4mil pitch/width will get you around 28 turns on each side. Spacing between the target and the coil will have to be ~1mm

What kind of precision are you aiming for?

Sliding coil is possible, but care should be taken about the wires that route coil to the LDC. Wires, changing its shape/routing may influence readings of LDC. LDC footprint is not that big, we could place it on the same coil board and run 5-core ribbon to it (power, gnd + comm).

I would suggest to print the target on another PCB - cheap and robust. I will suggest to make the shape as truncated triangle, with the thin part being 0.2 of the coil diameter, thick part - 0.8 coil diameter. The response will not be exactly linear, but you can do linearization in post-processing in MC

Thanks Evgeny, that's all very helpful information.

Mounting the LDC with the coil sounds like the best idea and connecting on the digital side to the micro with a flat ribbon, as you say.

I have a few other questions - 

I think 14-bit resolution from end to end would certainly be good enough, but what is the maximum resolution I could expect from a setup like this ?

Do you have any cad tools or ready-made coil patterns that could be imported into a pcb design package ?

As regards the target, would linearisation of the signal be done with an algorithm of some kind, and is there any information available on this, or perhaps fitting to a known curve with a look-up table would be best ?

With the truncated triangle, would the end points for the coil position be with the centre of the coil over the top and base of this triangle, or with the outer edge of the coil in these positions ?

Thanks again, Geoff