Paul, Teensy in a commercial product?

Status
Not open for further replies.
Paul, what are your rules / recommendations / guidelines for the use of Teensy in a commercial product, please?

Thank you.
 
Hardware-wise, you have 3 options. #1: It's perfectly fine to use Teensy as a component in your product. This makes sense at low volume or for a first run for demos, eval units, etc. #2: If you're making your own PCB, but at modest volume, we have pre-programmed chips you can buy. Using that chip and similar parts (the same crystal frequency) gives your board compatibility with Teensy Loader, so you can get your own hardware working without extra much extra design work. This is often a smart intermediate step from low to high volume production. #3: For high volume production, there are many solutions available on the market to directly program the chip on your PCB. These require up-front costs and work, but they are the lowest cost per unit. For AVR (used on Teensy 2.0), Atmel's AVR ONE might work. For Freescale Kinetis (used on Teensy 3.0), P&E Micro's Cyclone programmer is the preferred solution. The .hex file you used on Teensy can usually work when used with those tools.

Software-wise, most of the code PJRC publishes is MIT license to allow usage in close-source products, or a modified MIT license which requires special attribution only if you're making a development board (eg, a Teensy competitor) and publishing the source code. Obviously that doesn't apply if you're making a closed-source product. Some Arduino libraries are GPL which does not allow publishing as closed source. Ultimately, you should check the license on every piece of code that gets built into your application. Also, as an obvious disclaimer, this message is not "legal advice". For that, you'd need an attorney.

Busienss-wise, if you're planning to sell physical a product, I'd highly recommend focusing on customers and sales. Obvious as that sounds, I've seen many great ideas fail because the founder couldn't make the transition from focusing on engineering to sales and marketing. A big red flag is repeatedly redesigning the product, especially if the goal is cost reduction, before making many sales. When you have a lot of units sold, which might be Kickstarter, or a single large client, or the result of slowly increasing sales, then it can make sense to redesign. But if you don't have customers lined up, all the best engineering effort in the world won't add anything other than more red ink.
 
Paul, thank you very much for taking the time to reply. Excellent advice.

Despite having 2 technology degrees and now being a programmer I've historically worked in retail and agree with you completely. My initial obstacles are more around hardware (rather than software or marketing the product - although not to say I'll be any good at any of it). It's been a long time since I did any electronics (20 years in fact, since my first degree) but I'm up for the challenge. I'll probably kill a few Teensys off at first, but I'll get there.

Thanks again!
 
I am interested in this as well. I currently have 6 custom prototype boards with a socket for Teensy 3.1; they work great. I understand the intermediate volume step of buying pre-programmed MINI54 chips and using the same software tool chain. That sounds like a good way to get everything onto one board without adding too much more complexity. For the next step up in production, you suggest buying MK20s and programming them with the same hex file that comes out of teensyduino. You mentioned the cyclone programmer, but would the USB Multilink Universal work as well? I ask because it is quite a bit cheaper and it doesn't seem like I would need all the features of the cyclone programmer. Along with that, it seems like I might need this as well?

On my pcb I would also break out the following pins to the header design the programmer used right?

MK20 Pin 23 (PTA1)
MK20 Pin 25 (PTA3)
MK20 Pin 25 (PTA3)
MK20 Pin 22 (PTA0)
MK20 Pin 24 (PTA2)
MK20 Pin 38 (PTB3)
MK20 Pin 37 (PTB2)
MK20 Pin 34 (Reset)

Thanks for your support. We were using Sparkfun Micros initially, but they were terrible to work with. The Teensy 3.1s have been great.

Scott
 
I really can't comment on which P&E Micro product would be best for your needs.

Given the P&E's prices, I'd image they ought to be willing to answer your questions!
 
P&E's web page is a splendid example of someone with too much product knowledge talking over the heads of prospective customers.
I read it 3 times trying to figure out what it is and of what value it would be to me. At those prices, this is important to figure out. I don't have nor want any of Freescale's Tower development platforms.
 
Status
Not open for further replies.
Back
Top