Author Topic: uChip: open source, DIP16, CortexM0+ (Arduino Compatible) LIVE on Kickstarter!  (Read 1221 times)

0 Members and 1 Guest are viewing this topic.

Offline nwvlab

  • Regular Contributor
  • *
  • Posts: 64
  • Country: it
    • next-hack.com
Hi There!

I would like to show you my new project, uChip, which will be live on Kickstarter on March 10th! (preview link: https://kck.st/2SABZPr ).

EDIT: Now the campaign is live!   ;)  https://www.kickstarter.com/projects/1186620431/uchip-arduino-zero-compatible-in-a-narrow-dip-16-p

uChip is a small, narrow-body DIP sized, Cortex M0+ dev board, Arduino Zero Compatible.

It features a 0.3” spacing between pin rows, therefore you can mount it on a breadboard or you can even use a standard 16-pin IC socket! Of course, you can also solder it on your PCB or 0.1” prototyping board if you want!

uChip has an ATSAMD21E18, with 256kB of Flash and 32kB of RAM, and it features also integrated high efficiency boost and buck converters. This reduces component count and allows:
-   to power USB devices (when uChip operates as USB host) with 5V @ 500mA, even when the external power supply is @ 3.3V.
-   to power uChip (when operates as USB device), and the external circuitry, with 3.3V or 5V @ up to 1A (the voltage can be selected by software).
uChip can be programmed with the integrated USB micro AB connector, or by connecting any SWD programmed/debugger to the SWD pins (which can act as GPIO).
Most of the sketches for Arduino Zero will run unmodified on uChip, and you can of course use your favorite IDE.
We have also developed several applications/examples, you can find in the link provided!

After the campaign, uChip will be released as open source!

I hope to see your comments and criticism about uChip, and I hope I haven’t been too spammy with this post!

Cheers!

Nicola
« Last Edit: March 10, 2019, 07:21:22 pm by nwvlab »
 

Offline BroMarduk

  • Supporter
  • ****
  • Posts: 47
  • Country: us
I'll most likely pledge for one (or more) once available.

The logic IC emulation has me intrigued (when I saw the pin out diagram I noticed the VCC and ground pins placement!).   Do you have a full list of emulate-able logic ICs beside the 74HC191 and 74HC138?
 
The following users thanked this post: nwvlab

Offline dunkemhigh

  • Super Contributor
  • ***
  • Posts: 1275
That's an intriguing project. I think I would be unable to resist a pledge once it goes live...
 
The following users thanked this post: nwvlab

Offline blueskull

  • Supporter
  • ****
  • Posts: 11272
  • Country: cn
  • Power Electronics Guy
Two recommendations:

1. You don't need the RCD reset circuit. The MCU comes with its own POR circuitry.
2. You can replace the USB port with a through hole one (SMT signal pads, THT mechanical clips) for better mechanical strength. Your connector is too close to PCB edge, so external force can actually peel off the copper under the pads. A through hole mechanical clip is much harder to be damaged by brute force.
 
The following users thanked this post: nwvlab

Offline nwvlab

  • Regular Contributor
  • *
  • Posts: 64
  • Country: it
    • next-hack.com
Thank you for your feedback!

Quote
The logic IC emulation has me intrigued (when I saw the pin out diagram I noticed the VCC and ground pins placement!).   Do you have a full list of emulate-able logic ICs beside the 74HC191 and 74HC138?

Actually any 3.3 V logic IC can be emulated, provided it has the same VCC/GND position. We just made some examples with some old ICs we got from the 90's , and we later found that the 74HC191 was the most effective for a breadboard demo :) (some other ICs we tested were: 74HC109, 74HC153 and 74HC175, but they were " boring" - and "difficult" without adding a lot of other components  - to show :) )

Quote
1. You don't need the RCD reset circuit. The MCU comes with its own POR circuitry.

We are surprised that you spotted the RCD reset circuitry :) Good sight!

However It's a bit different than a simple RCD reset circuitry: we originally wanted the CPU to be reset in bootloader mode if the button was held for at least 1 s (while a quick push was a simple reset). While we found that this works fine, we found later it was rather annoying waiting for 1s. The (Arduino) standard "quick double click => bootloader" is much more practical, so we will remove this feature in the final release. In this way we save also one pin on the MCU.

The RC will remain though, as suggested by the datasheet, section 38.4.

Quote
2. You can replace the USB port with a through hole one (SMT signal pads, THT mechanical clips) for better mechanical strength. Your connector is too close to PCB edge, so external force can actually peel off the copper under the pads. A through hole mechanical clip is much harder to be damaged by brute force.
 
You're right, the PTH version was indeed our first choice, because it grants a much better strenght. However there are a lot of components on the bottom side: we will try to optimize and see if we can make room for a TH version. Still, the the copper area is quite large (there are large mounting pads even under the connector).

cheers!

 
The following users thanked this post: blueskull, BroMarduk

Offline chickenHeadKnob

  • Frequent Contributor
  • **
  • Posts: 733
  • Country: ca
  • doofus programus semi-retiredae

Quote
2. You can replace the USB port with a through hole one (SMT signal pads, THT mechanical clips) for better mechanical strength. Your connector is too close to PCB edge, so external force can actually peel off the copper under the pads. A through hole mechanical clip is much harder to be damaged by brute force.
 
You're right, the PTH version was indeed our first choice, because it grants a much better strenght. However there are a lot of components on the bottom side: we will try to optimize and see if we can make room for a TH version. Still, the the copper area is quite large (there are large mounting pads even under the connector).

cheers!

That advise from Blueskull  should be considered mandatory, not optional. Those smd only micro usb peel off if you look at them the wrong way, they should be banned from the marketplace.
 Consider that even the cheapest bluepill stm32f103 board has the through-hole tabbed kind, although you have to solder them yourself. The tabbed strain-relief type only cost pennies in shenzen.
 
The following users thanked this post: nwvlab, BroMarduk, OwO

Online OwO

  • Frequent Contributor
  • **
  • Posts: 340
  • Country: cn
  • Hewwo, I'm Gabwiel from OwOComm!
Overall nice design, but price is a little steep especially if you want to avoid people just sending the design off to a PCB fab and then making these themselves rather than buying from you. Since this is a fairly simple product with a big potential user base if this takes off I would expect Shenzhen cloners to be on it really quick, so be prepared to compete and be sure to optimize your costs to the bare minimum.
OwOComm - Intellectual Communism reinvented!
 
The following users thanked this post: nwvlab

Offline nwvlab

  • Regular Contributor
  • *
  • Posts: 64
  • Country: it
    • next-hack.com
Hi there!
Thanks again for your feedback!

Quote
That advise from Blueskull  should be considered mandatory, not optional.

Message received, working on that :)

Quote
but price is a little steep

tl;dr: believe us, we made our best to cut the price down without risking of being unable to fulfill the rewards !

Long answer:

There are many factors that forced us to make such price (let's talk about the first pledge, i.e. 20€).
- Part of the shipping fees is included (some countries have no additional shipping fees). Shipping is expensive! You get a much better deal by ordering more than one uChip.
- The goal corresponds to a very limited number of uChip. This is about 1k units. This might seem a very huge number, but actually it is not!
- We must order some % more, to account for lost packets (even if shipping has tracking number...) and defective units.
- Kickstarter fees.
- Variabilty. Just to give you two examples: 1) we are in Italy, so Kickstarter only allow us to run a campaign in EUR. But manufacturers and shippers are paid in USD! EU elections are on the way: we must not forget the effect of Brexit Referendum on the USD/POUND exchange rate... 2) if suddenly a component become out of stock, and there is one much more expensive 100% compatible, we cannot tell the backer to wait 20 weeks more, so we will try to upgrade that component.

Still, the cost for a single piece is the same of an Arduino Uno.

Quote
especially if you want to avoid people just sending the design off to a PCB fab and then making these themselves rather than buying from you.

tl;dr: making few uChip boards will cost you much more than the first pledge!

Long answer:

From our personal experience with uChip, a single individual would not save money from making his uChip at home. The cost of the components for a single unit already exceeds the price of a uChip of the first pledge. The PCB also is quite expensive (4-layer board), even if you consider the cheapest China pcb manufacturer on PCBShopper.com. Not to mention PCB assembly: if you don't want to pay a fortune, you must do it at home. Then you need the stencil to perform reflow soldering. And pick and placing all the 0402, UDFN and CSP devices is a really long hard nightmare! At the end, even if you make 10 units (to amortize PCB+stencil cost and to get a better price break on components), you won't get cheaper than ten times a single pledge (i.e. 200€)! And the 10x pledge is even cheaper!

Still, uChip will be opensource. We think instead that the single individual might want to make his improved version of uChip (and why not, sell it!).


Quote
I would expect Shenzhen cloners to be on it really quick

Yes, that's fine. uChip will be open source, and anyone can make it, and even sell it!
We also believe that Arduino has become so popular thanks to clone manufacturers too. They made it affordable for everyone. Still some people are buying original Arduino boards, which are much more expensive.

It would be our delight if uChip became as popular as Arduino. And if this happens because of clones, well, we thank them!


Cheers!

Offline sajattack

  • Contributor
  • Posts: 13
  • Country: ca
Will there ever be a DIP40 version?
 

Offline xaxaxa

  • Regular Contributor
  • *
  • Posts: 247
  • Country: ca
I just noticed this didn't reach campaign target. IMO the target was set too high, for a device like this PCBA becomes economical at about 200pcs, and at the price offered you would still have plenty of margin.

I saw the post where you mentioned finding commercial partners/investors, and my advice would be DON'T. A small project like this can make you plenty of profit for an individual/small team, but investors are far more cash hungry and will force the business in directions you will almost certainly disagree with. I am also an individual business owner who started my project on kickstarter, and to this day I don't accept outside investment. With one or two projects there can be enough profit to make a decent living, WITHOUT massive pressure to grow or having someone tell you how to run your business.

If I were in your position my next steps would be to manufacture a batch of 100-200 using your own funds, then sell these on a webstore/tindie that you will direct your kickstarter backers to. The kickstarter campaign did give you a valuable piece of information which is that there are people willing to buy this device at the price you chose, and that at 200pcs first batch the risk of not selling out is low. The other advice I would give is keep business/operating expenses at an absolute minimum, especially at the beginning. You should be able to end up with at least a 2x profit on the first batch, all expenses considered.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf