Tech Questions for the Cartesian Co Team

Do you have a question for the Cartesian Co team? If you can't find your question in the FAQ's on this page or already in this thread, we would love to help you out.

Fire away!

Hey Windraver,

We will be selling our ink packaged inside of ready-made inkjet cartridges which could be mounted onto a rep-rap with some clever engineering. The more frustrating part however will be making the electrical connection to the cartridges and creating a board that drives it. If you are really dedicated and willing to sort out your own gantry system (which is a bit more difficult than FDM printing due to the inertia you're moving and the fact that it needs to be as stable as a rock) then it might be a good idea to grab a set of our electronics to tack onto it. This will also work with a pre-existing printer you have but you may have to perform some heavy modifications to achieve any real accuracy.

As of this point, we don't have any immediate plans to support this kind of development (aside from being open source and letting you do it yourself) and we don't have the electronics as a stand-alone purchase item (although it will be by the time we ship the Argentum).

Feel free to ask us more about the idea or make some suggestions as to what you would like us to produce.

Thanks for your interest in the project

Hey Andrew,

1. Unfortunately neither the mechanical or electrical systems are planned to support more than two cartridges at a time. I actually pushed very hard to incorporate this for some potential upgrade ideas as simple as printing silk screens in conventional ink as well as more ambitious things such as an insulator. In the end, the size of the cartridges meant that allowing for more than 2 would *severely* limit the print area in all cases. On the electronics side, we decided it was best to leave the other i/o ports free for open experimentation and expansions considering that we couldn't foresee the mechanical ability to guide more than two cartridges in this model. It makes me sad too that we can't have a spare slot in this model but it was definitely for the best.

2. We had only briefly considered using sodium silicates previously but now that you bring it up again, there's something it could be useful for in the future.
One issue that we are having with the printers at the moment is how to automatically remove the water that is left behind after Silver has precipitated out from the two chemical solutions. The largest difficulty is that the water will contain minute amounts of dissolved dehydroascorbic acid and Nitric Acid which prevents us from steaming the water off (or some such similar) without leaving them behind. Removing the contaminants helps to improve conductivity a reasonable amount and so removal of the excess water seems to require it remaining in liquid form. The sodium silicate could potentially be tuned in concentration to react with the Nitric acid and/or dehydroascorbic acid to produce a gel that would soak in the moisture (with other chemicals dissolved) and allow for a small vacuum or perhaps even a wiper/brush to remove it as a solid without affecting the Silver print.

The downsides to that though are:
- if I'm not mistaken silica gel has to be baked which could potentially be done with a heated build plate but this risks ruining the inkjet nozzles and takes time.
- I'm not sure exactly how the baking process would go given that the silicic acid would contain so much water (necessary to be 'inkjetable') but this might be fine.
- brushing the silica gel against the silver traces while they are still fresh could cause some problems

In terms of the silicone polymers from alcohol - it could be awesome because of the adhesive properties and we'd really have to try it to see how well everything would go. However, my concerns would be that it would probably require a 4th cartridge for the alcohol and I have no idea on the temperature resistance (for soldering) that the polymers we could create would have. We would absolutely love to explore it in the future (or even better you or anyone else that ends up with a printer) but for now we'll back bench it to focus on making sure this model does the basic stuff as well as possible. We don't want ambition to ruin what we already know works.

If you do any more research or tinkering on the topic please keep us and everyone else posted!!

Thanks for your ideas and sorry if I prattled on too long.

To be honest, it wasn't more than a fleeting thought in the torrent of other ideas for an insulating ink - we haven't given it much thought let alone tested some printing with it.

Yeah I actually did a little bit more reading after my last post and it seems that a lot of the polymers are quite stable up to around 250C. That's usually the temperature around we solder with our SMD heat guns so it definitely seems doable. The next hurdle to tackle with it would be making the surface it creates non-porous or at least closed cell as well as able to be printed on top of again.

The ability to print on a surface is dependent on how wettable it is in relation to the conductive chemical solutions and the strength of the bond that is made by the silver particles. I wouldn't have a clue about the micro structure that would form as the silicone polymer chains grew or potentially how to alter this but the structure we'd be looking for in a multilayering insulative ink may be difficult to attain. The problem is that if you generate a porous, open celled structure, there is nothing to stop the subsequent layers of silver salt and acid shorting through to that below. On the other hand a very smooth surface will likely act hydrophobic and the traces will not form evenly as well as lending little strength to the connection that the first crystals of silver form. An example to illustrate this is that paper is a great substrate for printing to because the first layers wick into the fiber and form a good hold, however the resolution is reduced slightly from something such as FR4 due to the small amount of bleeding away from the original artwork.

Yeah, we've had a few people excited about integration with a system such as pwdr - it would really be awesome to see how the inks react with Gypsum or other powders (another reason why I wanted to have a third spare cartridge bay). Who knows what the future holds, I just wish I had more hours in each day to build everything I want haha. But I guess that's the whole beauty of open source, someone else might get to testing something before we get the chance.

Hey khdevel,

So, my preface would be that I don't think a circuit printed onto fabric would sustain a pass through of a washing machine. That being said, a large amount of our ink is just water and we now regularly wash our printed circuits under the tap after printing to clear it of contaminants - literally turn tap on, put circuit under, dry with paper towel.

Unfortunately we've been doing this with circuits made on paper, FR4 and other materials that act as pretty good substrates. Having circuits on fabric is never going to be quite as robust in the foreseeable future, it's not at a point where you would be able to crumple the circuit (on fabric) and expect it to work 100% afterwards.

If you're keen to try printing onto t-shirts, keep a few things in mind such as:

• Use a very fine/tight weave. Our results where we've had repeatable printing onto fabrics have been using micro-fibres.
• You're going to have to attach components with conductive glue. Conductive glue is safe and easy to use but it's not as great at holding components as solder and will not allow you to attach fine pitch components.
• The circuit won't be indestructable, simple circuits can flex quite a lot (like the wristband shown in our KS) but unless you're using some additional chemicals to make strong and flexible bonds to the fabric (we're working on it), you can't just ball it up and throw it on your clothes pile.

As for the effect of the sun, we're yet to notice any UV damage on any of the circuits we've made and we have no reason to believe that it would occur. I'll admit, we haven't done any field testing specific to that though.

Hey Dave
You are definitely on to something with double sided printing. Once you have resolved alignment by using some sort of guide on the print area, there is really nothing stopping you from connecting them together (auto alignment is already in the works). To be honest we haven't really experimented enough with vias, but I'd be extremely interested to see how you go. My one concern would be the adhesion of the silver to the sides of the hole, it varies quite a bit with surface finish so with rough sides it just might work.

One alternative is to print on paper and use staples as vias... maybe not so much with FR4
Rob

Dave, we just tried it, and it worked! This is just a very preliminary result, but using a predrilled via we were able to through plate it by printing into the hole.

We'll post more on this soon. It's very exciting!

Interesting idea. We do intend on doing some kind of solution for multilayer boards, we just have to try some competing methods to see which is best. Whichever works, we'll attempt to release as a backwards compatible upgrade.

When we first started prototyping the Argentum (feels like forever ago now) we bought a 5W (I think) IR diode laser with the intention of using it to sinter the printed silver, but that turned out not to be necessary, so the laser went back on the shelf. When we have the time, we'll probably break it back out and give it a try for making holes.

Some good questions, I'll see what I can do about replying to all of them:

What kind of fabric? We have used microfiber cloth, like the type you use for cleaning glasses. It worked the best for us because the fibers are weaved so close together, which makes it easier for the silver to connect. Other fabrics may very well work, we've barely scratched the surface of fabric printing.

Will it survive the washing machine? I don't know, but I don't think so. At least, not with the same level of conductivity as it started with. But that's just a guess because we haven't tried, I've been surprised by this kind of thing before. But I think if the washing machine is a problem, there's probably something that that you can coat the circuit in after printing to protect it. This is something I hope everyone will help us research because it's very exciting.

How long do the prints last? I think it depends on what you print it on, but there's nothing we know of that makes the prints break down. We have an 18 month old (our very first print) paper circuit that still works fine, so I'm not sure what the maximum lifetime is.

How many times can you bend it? It would seems infinitely, or at least a lot! We've never had a problem with the traces breaking under bending, with the exception of printing on paper and then creasing over a trace, that breaks it. But if you look closely at a creased bit of paper, you can see that the paper itself has cracked in little bits, so it's not the trace necessarily. The main problem with bending is that components aren't flexible and if you try to bend a circuit too much near one it will break the connections and possibly pop the component off the board. The solution to this is to use smaller components, or make the board less flexible around them by gluing something rigid behind them.

What's the resolution on fabric? Depends on the fabric. The best resolution we've achieved is this which was a footprint for a SOIC chip printed on micro fibre.

What's the mean time between failures (MTBF) of the printer? Unfortunately, I don't know the exact answer to this one. Before finalising the design for the printer we did a stress test on it by getting it to print continuously for 3 days, and it didn't show any signs of wear. The only maintenance required on the printer is to occasionally clean dust off the rods and relubricate them. Aside from that, everything else should keep working for a very long time, with replacements parts available cheaply and easily from us.