RFID NFC Tap and Go Ring for Credit Card Payment

The idea behind this project is that if you have an RFID credit card, you can tell your bank that you need a replacement card, then once you have that second card, you can dissolve it in acetone, take out the RFID chip, solder a new antenna to it, glue that assembly to an inner frame, then put that assembly in a mold made from the full-size ring, pour resin around it, then have a ring that will let you pay at any establishment that accepts RFID credit cards (you'll be able to use your ring at pretty much anywhere that accepts google/apple pay) without you needing to take out your wallet.

To find out if your card has RFID contactless payment options, look for this symbol on the front or back: https://imgur.com/rQDpAac

If it doesn't, you can also call your bank and see if they have them available. My card did not have RFID but they were able to send me one that did when I called and asked.

I'd suggest reading through everything in this instructable before starting in on this.

This part is pretty straightforward. If you have a card with contactless payment options, call your bank and tell them it's wearing out. They should send you a second one and they should be very clear that your card number will not change. If it's not clear, double check with them. The reason that we want two duplicate cards is since not all stores accept tap to pay, you can make one into a ring and keep a regular card in your wallet for the usual chip or swipe payments. I am also told that in Europe, some banks issue "minicards" that are meant to go in your phone case or on a keychain and are an RFID-only duplicate of your main card, which is exactly what we want for this.

I made this in response to things like https://mclear.com/ which is an active version of this, but not available in America.

If your card doesn't have RFID currently, call your bank and ask them if they have a version that does.

Some card issuers call them different things. Tap and go, contactless payment, paypass, etc... Most banks should know what you're talking about if you ask about contactless payment or tap and pay.

Either way, you should get a second card. Note that not all banks have RFID enabled cards yet.

There are two approaches to this that I've found. Both involve soaking the card in acetone until you can get to the RFID microchip. The chip is encased in an epoxy blob that seems to be impervious to acetone. The backing for the chip that has the traces on it also seems to not mind acetone. The thing I've found that you have to watch out for is once you dissolve the plastic, is that the chip and backing themselves are not strong at all. Be careful to not bend them, the backing board will break and then you'll be waiting for another replacement card in the mail.

Originally, I scraped off the back and front plastic layers on the credit card until I got to the plastic "core." In my case, my card was a different color than the actual middle plastic core. I threw the whole thing in acetone for about an hour and a half, and was able to pull out and measure the antenna at 30.0 inches.

On subsequent attempts, since I knew the antenna length, I just cut out the small chip area on the credit card. With less mass for the acetone to dissolve, it went faster, probably a little over an hour, and I also didn't have to worry about covering a large container so the acetone wouldn't evaporate. Once a few people make this and weigh in that their cards are the same as far as chip location and antenna length, this step should go a bit faster.

Make sure you don't scrape too close to the metal contacts, and don't scrape too deeply into the back. Once you hit plastic, you're good. The acetone will do the rest of the work.

The hardest part of this step is probably going to be finding a high-resolution printer. I did mine on a resin printer. FDM printers won't be able to replicate the tiny arms on the inner frame that help center it in the mold.

All files are located at: https://www.thingiverse.com/thing:3550573 be sure to scale both files according to your ring size. The original files are for my right ring finger, which is size 10 and 3/4. If your ring size isn't 10 and 3/4, scale it using this chart before you print it: http://www.mazaldiamond.com/blog-en/wp-content/up...

Since the inside diameter above is shown as 0.806 inches for the original 10.75 size, (You're looking at the radius in the last picture above, so we're doubling it.) then if your ring size is 9 and 1/2, looking at the chart you'll see that's a diameter of 0.762 inches. 0.806 divided by 0.762 is 0.9454ish, so just scale the models to 94.5% of their current size before you print them. Note that bigger ring sizes mean more room for more resin around the RFID chip, so I wouldn't suggest making this a pinky ring. Middle finger or thumb is probably the safest bet if you have small hands. I could also be talked into making a larger profile ring STL if you need it, but that would probably be less comfortable to wear.

Random note that the absolute best way I've found to remove delicate resin prints from the supports is a Dremel, a cutting wheel, and going very slowly. Make sure when you're cutting you move the cutting wheel up and down so that it doesn't bind as you're cutting and you should be fine. It's better to cut the support and then grind/sand it down flat against the part than to try and save a minute or two and end up breaking the part. Focus on getting it out of the supports completely first and only then sanding things flat where the supports were.

Note: The second to last picture above is a different version layout from a newer card from my bank than the four earlier pics. Just to give an idea of some ways that I filed down different chips that worked.

The idea here is to make the chip backing board the least wide that you can since that's the smallest dimension in the ring that you have to worry about. From what I've seen there are usually larger traces running around the chip. Make sure you don't file into these, or any traces that run from these to the outside two pads. These two outermost pads are where the antenna attaches.

If you melted the card down whole, you'll have noticed that the ends of the antenna that runs through the entire card will contact these pads. In my case, it was simply pressure holding the antenna to the contacts. If you have any questions or if your chip looks different, feel free to send me high-resolution pictures of your chip and I'll do my best to advise.

This is why you don't cut too close to the chip area on the card before you put it in acetone if you're not melting it down whole, as you could cut through these traces, and this is why I recommend filing it down gently to get it smaller, is to make sure you don't disturb any of these delicate traces as you work to make this assembly smaller to fit into what you're going to cast. The last picture is an example of what can happen if you aren't gentle enough.

Your antenna should measure 15.0" or half of what the full antenna in your old card measured, from the first point where it makes contact with the chip, to the second point, both times as soon as it makes contact with the pads on the chip board. If these two points measure differently than 30.0" in your card, please leave a comment letting me know, but I assume these antennas are all the same size since they're all the same frequency.

For the antenna, I used: https://smile.amazon.com/gp/product/B01IJ2JJG6/ but there are probably better options out there (See discussion in the comments.)

This is one of the more annoying steps for a few reasons. In order to attach the antenna to the chip, you have to first strip the magnet wire, which is a feat in and of itself. There are many guides online, but the best way I've found is to very briefly touch a bic lighter's flame to the end of the magnet wire, and it will burn off the enamel and leave a carbon deposit all over the end of the wire. If you then take some 400 - 1000 grit sandpaper, pinch it together and use it to repeatedly sand off the carbon deposits until you see copper, it seems to work well.

Then you need to do this to both ends, and solder one end to the chip's outermost pads. Note that the pictures above are of different chips, to show different possibilities you might run into. The reason that you're soldering only one end of the new antenna to the chip is that you can pinch the other end to the opposite contact pad and then test it out on a sale to make sure that everything is still working before you start gluing things together. The easiest way that I've found to do this is to look for vending machines that accept NFC. You'll want to make a few loops of the antenna as shown in the last picture, three or four loops seem to work well, then pinch the stripped non-soldered end to the other pad while holding it up to the reader.

If it works, get yourself a drink as a reward for getting this far. If it doesn't, you may have filed too far into a trace or had something else happen. Request another card from your bank and try it again. You can also try this part before filing the chip's board down as well so that if something does happen, you can figure out where you went wrong.

As for soldering one end, solder doesn't seem terribly willing to flow onto magnet wire, so go for a liberal use of flux. In my case, I ended up just making a large blob of solder on the pad and then managing to get the magnet wire end into the middle. Less so having the solder flowing onto it, and more so just having the wire attached by the fact that it's in the middle of the blob. Less than ideal, but it worked well.

Take your chip and antenna, still with only one end soldered to the chip, and wind the antenna around the inner frame, until you get to a point where you can't make another loop around the frame. At that point, solder the other end of the magnet wire antenna to the other pad on the chip's board, and place any excess antenna back into the inner frame channel.

I would again test the whole assembly at a vending machine at this point, and then if it works, glue it all together and test once more. If it works at this point, you've made it through the hardest parts successfully. Congratulations.

You're now ready to mold and cast the ring.

Take your 3D printed full ring that you scaled and wet-sand it with 1000 grit sandpaper for about 15 minutes. You shouldn't see any major ridges from printing at this point, and it should be cloudy all over. If it's not, sand a little longer.

If you intend to make a clear ring, I'd suggest wet-sanding it for another 45 minutes, then wet-sanding with 1500 grit sandpaper for about an hour, then wet-sanding with 2000 grit for about an hour. Painful, but if you need to make more rings down the road (and you will, when your credit card expires, or if someone gets a hold of your card number and you need to reissue a new one) then time spent here is time that you don't have to spend on each cast you make in the future. Much better to get it perfectly smooth now and not have to spend an hour sanding each ring that you cast in the future.

Take a chunk of mold-making clay and cut it into a roughly 2" x 2" x 0.75" shape. Push your ring that you've sanded to a high polish into that, roughly halfway, as shown in the first picture.

Then push the clay up to the ring on the inside and outside as flat as you can, and make a box around it so that you can pour the silicone over top of it. Some divots will make sure that you can always line your two halves of your mold up well down the road, and a channel around the outside will help prevent leakage when casting.

For the mold, I used the silicone (yellow) and hardner (blue) shown in the third and fourth pictures. I'm quite sure there are other options, but these worked very well.

I then mixed these together, and threw them in a vacuum chamber for a few minutes to get all the bubbles out as shown in the fifth picture.

I was told to pour the mold starting at one corner, and just let it slowly cover the ring. Once this was done, I let it harden for 24h, then carefully peeled and cleaned the clay off the other side without removing the ring from the mold or disturbing it in any way, then sprayed mold release over everything, then poured the other half in the same original box made of acrylic pieces and hot glue (or whatever is handy.)

Once you have your full mold made, cut a sprue out of the silicone and two small vent holes for bubbles to escape from as shown in the first two pictures. Put the inner assembly into it, with the chip facing 180 degrees from the sprue hole, then put together and wrap the full mold with two rubber bands.

Make sure they're not so tight as to deform the mold, just hold it together firmly. Then mix up your epoxy resin, THOROUGHLY. Make sure to mix for at least a few minutes, scraping any off the walls of the mixing container as you go. I used: https://smile.amazon.com/gp/product/B07BM9LHRB/ and it worked well. Get a syringe and pour the epoxy into the syringe. Inject it into the sprue until it comes out the vent holes.

Put the whole thing into a pressure pot for no less than 24h at 60psi, this is to collapse the bubbles into solution so that there are none in your final ring.

You should be able to pull it out of the mold/wear it at 24h, but be gentle with it. That epoxy takes a week or so to fully cure and strengthen. Until that point, be careful with it. It will scratch and potentially even snap easily until full cure. When you cut the sprue and vent hole pieces off the ring, don't cut terribly close to the ring. 1/4" or even 1/2" away is good. If you try to cut flush with the ring body or even too close, it can break off below the ring surface. Just cut 1/4" or more away and then file it down. It's not worth it to try for closer after all this effort.

Enjoy the fruits of your labor!

If you've made it this far, please leave a comment as to your experience so I can refine the instructions.

For most point of sale readers, you'll want to bend your knuckles as shown in the above picture and basically put your fingers on the screen of the reader with the ring parallel to the screen. I've found that some readers like it positioned just above the keypad. Most of them work best when it's in the direct middle of the screen, so start there and move it down a little if it doesn't work after a few seconds.If you move it away too fast, it will also fail. Give your ring a few (3 or 4) seconds before trying in a different area. You'll figure out quickly enough where it works at your favorite retailers that support it. I've also seen some options where you do a kind of "fist bump" and place the ring directly touching the reader, but I haven't tried that too often yet so I can't report as to how well it works.

Enjoy!

Won't hackers steal my credit card?

This is a very common question I get, that I tend to break down into a few parts. The first is that (AFAIK) with the new generation of RFID cards, the only current feasible attack is one in which you need to have a reader near the ring that's relaying the wireless data to a repeater that's near a point of sale terminal. Not impossible, but from what I know (and that brings me to my second point...)

Due to the difficulty, the fact that their reader would have to be within an inch of the ring, and the ease of skimming magnetic strips at gas stations or buying CC numbers by the hundreds off insecure website hacks, most people just don't bother with the first route. It's the same as when people ask me about my RFID car door reader that unlocks my car. "Couldn't they skim your chip and replay it?" Well yes they could, for a few hundred in equipment and a fair bit of time, or they could pick up a brick off the sidewalk for free and break the windows and achieve the same result. Honestly, I'm a lot more worried about the second one.

Will this work for other RFID/NFC/whatever chips?

Definitely. As long as you can get the main chip down to a width of less than your ring width, then you're good. I'd also be happy to modify the models a big if you're trying to fit a bigger chip in, feel free to contact me. What I found is that since the antenna is acting as both an inductor and an antenna, you generally want the fewest loops, but also want to keep the length of the antenna multiples of the original antenna's length.

For other RFID chips I would say measure the overall antenna length, then wind one made of magnet wire that's double that length, one that's the same length, and one that's half that length. Since most of us don't own antenna analyzers (Or am I the only one that's not a cool kid?) you can just test real-world range on a reader with those three and you'll very quickly find out what works.

I tried the original antenna size wound around the ring and found it worked, but you had to press the ring right up against the reader. I also tried 1/4 size and same thing. Half of the original 30 inches and you don't have to talk the ring off your finger to use it. It's a balancing act between picking up power and relaying back the data RF packets. You'll probably find similar results no matter what you're starting with. If your original antenna length is very short, you might even try 4x when you wind the new one on your ring, just to see if it works better.

Is a ring the only form factor this will work for? Can I make a bracelet or something like that the same way?

Absolutely! The best way to figure out if it will work is just to try.

The good news is that the larger you go, the longer the range should be, although that falls off at some distance (A few inches.) no matter what. I'd be curious if you could bend your wrist enough to have a bracelet work at a point of sale terminal without having to take off the bracelet.

The other good news, is that with a larger form factor, there's more room to work. For a bracelet what I would do is start by taking out the chip as outlined in the instructable, then looping it to about the size that it would be looped for on your wrist. You can easily solder new length antennas on to test, and just trying it is definitely the best way I've found to see if it works.

There's talk in the comments of magnet wire that will strip itself when soldered, and that's definitely going to be the easiest option for testing. Let me know what you find out, sounds like an awesome project!

Super long/single loops like you'd have to have for a necklace would not work very well. You need a few loops for the inductive properties to work, although you might be able to take off the necklace and loop it back on itself and have it work. A bracelet should work well, although you may find out you have to take it off. Max range no matter what will probably end up being about two inches.