Would it be possible to have an NFC antenna integrated into a membrane key panel, using the same construction techniques as the key switch circuitry?
I know you can make NFC antennas using flat-flex PCBs, but those are still using a copper conductive layer just like a regular PCB, so I would assume they perform much the same. However, membrane keypads are typically constructed using a conductive ink as the circuit layer. Would it be possible to make a decently-performing NFC antenna that way?
Also, would any kind of 'shielding' layer be necessary behind the membrane? I seem to recall reading somewhere that when flat-flex antennas are applied to the surface of other objects (e.g. a mobile phone battery), a ferrite layer is placed in-between, so that the object does not interfere with the antenna.
I don't really know much about NFC or RF myself. I was just recently having a discussion with an acquaintance about a device his company makes where they want to upgrade it with NFC capability, and it already has a membrane keypad front panel - the thought occurred to me that integrating it into that might be possible.
You'll have to figure out what the ESR is. Then you can do some estimates on inductance, number of turns, and what matching components are needed.
Silver ink isn't necessarily bad. Resistance will cost Q, which shouldn't really be a problem as the selectivity isn't a big problem as far as I know. Mainly it'll cost range. Maybe unless your circuit is emitting enough junk that it confuses the system, or that it's susceptible to the emitted field? How close is the nearest circuit, or metal panel?
Yes, ferrite can be used to keep fields away from such things. Available materials range from flexible tape to rigid plate or bar. They're not the cheapest option, so making do without may be worth looking at first? Distance is your best friend, and if distance is not available, there's this option.
Solid metal shielding just shorts out the field; this reduces range some, but also reduces inductance, so tuning needs to be done in situ.
It's not uncommon for example, to have circuitry in the middle of an antenna region, and the ground plane simply shields the circuitry from interference. It shorts out the field so reduces inductance relative to the antenna traces alone in free space, but it's fixed for the circuit board so you don't have to faff around tuning it.
Tim
I guess it would be pretty hard to design if you didn't have exact specifications on the silver conductive ink. I'm guessing your average membrane manufacturer isn't going to have terribly precise control of their printing (e.g. variance of layer thickness) to meet a tight ESR spec. I suppose it would be a lot easier if the membrane was constructed with copper layer like your regular polyimide FPC.
Does the length of 'tail' of connecting traces of an antenna coil circuit have much impact on the performance of the antenna? I'm imagining if an antenna integrated into a membrane keypad has its terminals routed out alongside the rest of the keypad signals, it could be quite long.
Nah, just more resistance, and inductance but less than from going around the loop.
I mean, it's at least not so bad that they can't put LEDs on the stuff, so there's that. On the order of ten ohms or less, would probably be acceptable; a hundred, no.
If you can't tell until the thing is fabbed... you can always bring out two or four pins (crossings might be difficult, or added cost?, so just doing single loops around the periphery might be preferable over a multi turn loop?), and if they don't work out, just ground them -- it's recommended to have this anyway, to collect ESD sparks that might creep in from the edges. Which, keep that in mind also if it works out, some ESD protection on the antenna.
Or heh, could always put an FPC underneath it anyway... almost no height cost, but does require an additional part.
Tim