I have, but unless I’m configuring it wrongly (quite possible since I don’t have test / proper gerbers) the setup costs are insane, making 300 quantity cost the same as 10!!
1 Like
Yeah, you’re gonna pay $120 no matter what, but you’ll get a few hundred boards. You’d pay ~$80 for a few hundred Fr-4 boards so it’s not a huge jump. The barrier to entry is just higher. The tolerances are like 8x more precise by default though so you can make smaller boards than you ever would be able to otherwise.
Ultimately I don’t think the cost is that bad. 10 years ago you’d have paid $500 for 100 flex PCBs, and that was in back then money.
1 Like
From PCBWay yes, but not from OSHPark if I understand their website correctly. Their price is $10 per spare inch for 3 copies… which means for a 6.35mm x 14mm flex pcb (their recommended minimum width is 6.35mm not 2mm) three copies should cost $1.30 (I only want it for myself, obviously amal might want the much larger quantity)
If you find out more about their MOQ let me know. OSHPark is great, but I was under the impression you would need to order more than 1 square inch of board space per customer.
There’s also the downside that you have to wait for them to fill up a panel before they’ll order, which for something less common like a small batch flex PCB could be a bit longer than you would like.
I work in synthetic biology designing and studying autoluminescent plants and animals. I came across an interesting article about wavelengths of light and how green is not an ideal florescent protein for animal tissues, as blue and green light is readily absorbed by human tissues “…signals from enzymes with spectral peaks in the blue/green wavelengths (∼475to515nm)(∼475to515nm) were strongly absorbed with a pattern suggestive of a significant influence of hemoglobin. Enzymes with spectral peaks in the orange/red wavelengths (∼590to650nm)(∼590to650nm) were less absorbed, with greater penetration through tissues”
https://www.spiedigitallibrary.org/journals/journal-of-biomedical-optics/volume-10/issue-04/041210/Emission-spectra-of-bioluminescent-reporters-and-interaction-with-mammalian-tissue/10.1117/1.2032388.full?SSO=1
So I’m definitely not saying it won’t work with green light but I hypothesize that the green light will appear much more dim beneath the skin if you ever plan to implant, and this is really cool I would just recommend red LED’s for maximum brightness and visibility once implanted 
1 Like
It seems like that article mainly focuses on light produced quite deep within the tissue (refers to centimeters of tissue), rather than just barely subdermally. For existing LED implants (including the xSIID), anecdotally it’s been found that green is actually the brightest, followed by blue, then red, and amber as the least bright. Amal mentions this in this forum post, and comparing my own blue implant to pictures of those with red implants, it seems to hold true.
1 Like
@darthdomo Well hey I’ve never heard that but it certainly makes sense! You’re definitely right this is talking about multi-layered tissues from deep to superficial and I was wondering that myself, as the article largely accredits this phenomenon to hemoglobin, which there’s very little hemoglobin in the fascia tissues. So I did wonder how this effect would pan out in the strictly superficial tissues with little tissue to disperse the light, but that’s very interesting I’d have to try this out perhaps and test it! If anyone gets this implanted by chance I’d love to see the results 
1 Like
You most certainly can’t do that from other people’s videos. You need to meet the other implantee in the flesh, put your hand in the same light and same position as theirs and then compare - and even that is subjective.
Ultimately, it’s just a matter of perception. Maybe this or that light is dimmer, but if the eye sees it better, then it looks brighter.
2 Likes
I wonder how this translates to eyeborg, I wonder if it’s possible to email images and him upload them as camera signal
… then I suppose you are back to square one with different cameras lol
Not a full derail, as yes my xSIID is quite bright, I guess I can’t say brighter than others because aforementioned visual biases, but it is definitely not being absorbed quickly, as I can see it illuminate .394 inch of the surrounding flesh when powered
Dick 
Color, brightness, luminosity… are things that are surprisingly difficult to measure.
I learned that years ago when I worked for a small arms supplier: our customer complained that the pistol holsters we had supplied didn’t conform to the desert brown they had requested. Since we had no real acceptance criteria other than “it looks similar to the reference model”, I was tasked to find a metrology supplier and a procedure to measure the color objectively.
That sent me down the rabbit hole - DEEP down. I ended up visiting all kinds of scientific labs, fabric manufacturers, certification agencies… and spent north of 50k euros on test equipment and third-party colorimetry, to end up with not very many conclusive results that wouldn’t have ended up making the cost of the holsters spiral out of control.
In the end, we produced a 200 page report on the subject to the customer, and we mutually agreed that one guy comparing the production to a reference model in decent daylight without too much alcohol in his system was probably gonna be as good as it’ll ever get
1 Like
Here’s a 45 minute video you can send them too if you want to blow their minds. As you no doubt would have come across, lighting makes it hard to determine the true colour of something, but the reverse is also true - different objects make it hard to get lights to look consistent.
Here’s a great video with demos of how two colours that look the same can be entirely different in the world of theatre lighting.
1 Like
@Satur9…as I’ve just found it, for info this is one of the NXP NTAG I2C demo boards. It consisted of a second board with uC etc. Put that somewhere safe years ago. Think NXP currently do some flexi thing with the device mounted.
Might be an older version of https://www.newark.com/nxp/om5569-nt322em/explorer-kit-energy-harvest-nfc/dp/45AC7433?
It’s a nice little kit. They have some interesting antenna designs on their dev boards. Each concentric trace in the antenna is going the opposite direction because of some clever via routing.
Unfortunately I got my hands on one of these kits after I designed my business card based on the NTAG I2C Plus. It worked out though.
1 Like
Clever, but to what effect? Is there a significant point?
Oh almost certainly 
the RF engineers at NXP, with $100k+ in analysis equipment and simulation software, wouldn’t do all that extra work for no reason.
Now what the point is would be a bit more difficult to nail down. It could have to do with eliminating parasitic capacitance between the traces to allow more accurate estimation of the inductance value while designing. If you review this test board I made you’ll see that the coil design where I split half the turns on the top of the board and half on the bottom was spot on to the calculated Inductance using NXP’s formulas.
It makes sense that they would need that accuracy for the field detector board, because there’s no tuning capacitor, the LED is the the tuning cap (and a small one at that). Doing it this way could also present a lower impedance so that more power could be delivered from the reader, which would also make sense for lighting an LED.
Looks as though you’re pretty well sorted. Mine were the early version chip but I’ve since changed them…well I was forced into it when I bricked them 
Can’t say that I’ve noticed but probably more show than go.
You’re probably aware of this tool…but just in case not. http://www.circuits.dk/calculator_planar_coil_inductor.htm
1 Like