Wiring Forum
http://wiring.org.co/cgi-bin/yabb/YaBB.pl
Wiring Hardware >> How to get it? >> Wiring 2.0 Hardware
http://wiring.org.co/cgi-bin/yabb/YaBB.pl?num=1288809181

Message started by b on 11/03/10 at 18:33:01

Title: Wiring 2.0 Hardware
Post by b on 11/03/10 at 18:33:01
Hi!

My name is Brett Hagman and I will be working on the next hardware design for Wiring.  I have discussed with Hernando all of the ideas of developing a cheaper, smaller Wiring board.

I wanted to start a discussion on what things people would like to see on the hardware board.

So, this thread is an open discussion for the next Wiring board design.

Here are a few things I would like to add to the new design:

1) MCU: I'd like to make several options (maybe different boards)
[list bull-redarrow]
  • ATmega1281, ATmega2561, ATmega640, ATmega1280, ATmega2560 (all pin compatible - 64 pin package)
  • ATmega644P/ATmega1284P (44 pin package)


    2) Make the FT232RL optional - besides cost savings, there are power savings, and you can utilize the pins for the hardware serial for other purposes (alternate hardware capabilities).  Regardless of the FT232RL, a header break-out for an FTDI USB to serial cable (or something similar).

    3) Board size - I'd like to go for 3 inches by 2 inches.

    4) Power regulator/external power - minimum of 1 amp regulated 5 volts - debating whether we should use a buck/boost regulator, or just a plain ol' 7805.

    5) Use a switch for the USB/External power selection.

    6) Footprints for both USB B and USB mini-B 5-pin jacks.

    Please feel free to add to this thread, any comments/suggestions for the next hardware design.

    b

  • Title: Re: Wiring 2.0 Hardware
    Post by Zap on 11/04/10 at 06:06:53
    Yes I like all the suggestions, as for
    2: yes make it optional, I use my own max232 cct with auto reset for uploading and most standalone projects don't need the extra cost of USB.
    4: They are many switchmode regs now available in the To-220 (with HS) footprint, so the user could either depending on cost and efficiency required. (just allow for the slightly bigger PCB footprint)
    5: Use a three pin header with jumper strap.

    Zap

    Title: Re: Wiring 2.0 Hardware
    Post by roypardi on 11/04/10 at 15:02:31
    Hi, I'm not that hardware savvy so don't have much to contribute. I've used the 1st gen Wiring boards for many projects and they have served very well.

    2.) FTDI cable. I use the Boarduino with an FTDI cable and it works fine. If this saves on cost/frees up pins that would be a plus.

    5.) The existing jumper is fine (how often does one need to change this, etc.). My issue is with the fuses - I seem to have blown the fuse on both my 1st gen Wiring boards so I can't use USB power at all. Is there a design issue with the fuse?

    My particular usability concerns:

    a. Is the overloaded pin functionality due to the chip or the board design? (e.g. pins 2/3 are hardware serial and interrupts). If it is board design then eliminating this overloading would add greater functionality (e.g. accessing all 8 interrupt pins while also using hardware serial).

    b. use screw terminals instead of pins ...  ;)  I know a change is not likely but after many projects, the hardest, most error prone aspect of using the Wiring board is cabling up - making cables w/ a hand crimper, header blocks slipping off, etc.

    Title: Re: Wiring 2.0 Hardware
    Post by b on 11/09/10 at 22:49:43
    Unfortunately the "overloaded pin functionality" (Atmel calls them "Alternate Pin Functions" - bug/feature... your call) is part of the chip design.  Not much we can do about that.  ([sarcasm]Yeah, I'm particularly fond of the INTx interrupts on the hardware serial pins[/sarcasm]).

    RE: screw terminals - it's already in the works to make the headers holes spaced from the edge of the board to accommodate 0.1" spaced screw-downs.

    Anyone have any comments on the board size?  GET YOUR RULERS OUT! :)

    b

    Title: Re: Wiring 2.0 Hardware
    Post by mowcius on 11/10/10 at 20:26:31
    Just to comment, had a quick chat to Brett. Someone mis-read  :P USB mini B is definitely staying (thankfully) so +1 for both sets of pads.
    I like the USB B as it's definitely much less easy to damage. I am always careful with things but the USB B ports, they feel as if you can plug them in and out with no worry. USB mini B is a bit more delicate and somehow I seem to have broken one (but I think it may have been a slightly duff socket to begin with).

    Board size sounds good. - Arduino is... well we won't go into the random arduino board sizing but it's roughly that.
    Should give a good amount of space for all the pins and other components.

    I would probably stick to simply a 7805 but pads to allow other regulators/boost converters as people always seem to like using 9V batteries - with a boost converter they could go with something more sensible!
    If a nice regulator/boost combo can be purchased without too much in the way of price increase then I'd go with that. Not sure how expensive they are as a combo.


    Quote:
    b. use screw terminals instead of pins ...

    Not convinced by this, Perhaps a different 'ruggedised' board with these and a bit of protection for things. I dislike screw terminals most of the time. I only use them for things like motors and larger stuff. Tiny multi strand wires are terrible in screw terminals.

    Mowcius

    Title: Re: Wiring 2.0 Hardware
    Post by b on 11/11/10 at 03:20:00
    RE: Regulator - the biggest problems with using a buck/boost regulator are the cost and the real estate taken up by the other components for the switcher (i.e. inductor, mostly).  I'm leaning towards an LDO linear regulator.

    RE: Screw-downs/screw-terminals - This will only be an option.  i.e. the screw-down terminals can be added by the end-user.


    Back on the subject of regulators, if we don't have the FT242RL on board, we will need a 3V3 regulator.  My thought is to just have a 3V3 regulator, regardless.  A 250mA LDO should do the trick.

    Also, I was thinking of this:

    1) Standard 2.1mm Jack, with protection diode.
    2) 2 pin header (JST?) for battery connection (i.e. 6V or better) bypassing the protection diode.

    This way you can run the board by supplying 6V via the battery connector to a LM1084 (or similar) and you can get 2A output.

    Though, a buck/boost would be nice to put on there.  Then you could also run the board from 4.5V (or even 3.0V!) if you want.  Kind of a tough call.  I think the linear regulator will be a better choice in the long run.  (Can make an external boost regulator board to apply to the 5V input maybe?)

    Thoughts?

    b

    Title: Re: Wiring 2.0 Hardware
    Post by mowcius on 11/11/10 at 09:30:38

    Quote:
    Regulator - the biggest problems with using a buck/boost regulator are the cost and the real estate taken up by the other components for the switcher (i.e. inductor, mostly).  I'm leaning towards an LDO linear regulator.

    Yeah but as presumably it would be onboard when you buy it you could make it some tiny surface mount components.

    Doing it all surface mount, you could fit it all into the space of a typical cheapo 7805.


    Quote:
    we will need a 3V3 regulator.  My thought is to just have a 3V3 regulator, regardless.  A 250mA LDO should do the trick.

    250mA sounds fine. Again it can be something tiny.

    Not sure I'd go with 2A, seems a bit ott, I would have thought 1A would be plenty although I may be wrong, what's the price difference?

    Mowcius

    Title: Re: Wiring 2.0 Hardware
    Post by b on 11/11/10 at 16:38:55

    mowcius wrote on 11/11/10 at 09:30:38:
    Yeah but as presumably it would be onboard when you buy it you could make it some tiny surface mount components.

    Doing it all surface mount, you could fit it all into the space of a typical cheapo 7805.

    ... well, not for running at 5V @ 2A+.  The coil is the problem.  This point is moot, though, since we aren't that hard-up for space.

    After looking into it some more, I think the cost is actually going to be the prohibiting factor (~$5 per switcher @ 100Qty, and that doesn't include the other components - ~$1 for 2-6A rated coil).  If anyone has a good reference design/line on parts that could sway me to go with the switcher, let me know.

    The 2 amps I was referring to in the last post is the maximum that the 1084 regulator would provide @ 6V input.  The 1084 can provide up to 5A regulated (@ above 6.5V).

    So I am thinking this:

    Just go with the "good ol'" 7805 @ 1A (nice and cheap).  The 1084 is still more expensive (almost $3 @ 100Qty), nice LDO and all, but the benefits don't outweigh the costs.

    Thoughts?

    b

    Title: Re: Wiring 2.0 Hardware
    Post by b on 12/09/10 at 19:21:47
    So, decisions are below:

    New board name: Wiring S (ATmega324P/644P/1284P based)
    5.0V Regulator: 7805 or similar, min 2V drop-out, min 500 mA
    3.3V Regulator: LP2992 or similar, min 200 mA
    FT232RL USB optional
    FTDI USB to Serial cable compatible header as option to USB
    USB B Connector only

    If you have anything to add, do it soon.

    b



    Title: Re: Wiring 2.0 Hardware
    Post by mowcius on 12/09/10 at 19:49:57

    Quote:
    USB B Connector only

    Can't you add the pads for the mini B connector then?

    I'd have thought you might as well as it aint gonna cost any more :D

    Title: Re: Wiring 2.0 Hardware
    Post by b on 12/09/10 at 20:42:14
    2 thoughts:

    1) As you've mentioned before, and I have to agree, the mini B connector is not nearly as durable as the standard B connector.

    2) Unless there is an overwhelming amount of support for the mini B, there is no reason to have an option for the mini B on the board, since producing a different board with only a different connector just makes production more complex.

    Convince me otherwise :)

    b

    Title: Re: Wiring 2.0 Hardware
    Post by mowcius on 12/09/10 at 21:54:33
    As I had forgotten about the 'other' header I am happy with the USB B only owing to the complex nature of trying to fit both in.

    I personally don't like mini USB B so much as I have managed to break one. Never broken a USB B!

    Mowcius

    Title: Re: Wiring 2.0 Hardware
    Post by b on 12/10/10 at 23:33:38
    Ok, schematic complete.  Part placement complete.

    Please have a look, review and yell at me if you think something should be changed.


    Wiring-S-V10-Schematic by Rogue Robotics, on Flickr


    Wiring-S-V10-FTDI-Option-Various-Headers-2 by Rogue Robotics, on Flickr


    Wiring-S-V10-FTDI-Option-Various-Headers-1 by Rogue Robotics, on Flickr


    Wiring-S-V10-Cable-Option-1 by Rogue Robotics, on Flickr


    Wiring-S-V10-FTDI-Option-1 by Rogue Robotics, on Flickr

    b

    Title: Re: Wiring 2.0 Hardware
    Post by mowcius on 12/11/10 at 09:27:59
    For once, I can't think of anything wrong with that :D

    I'm trying to decide if I like that reset switch or not though :p

    Title: Re: Wiring 2.0 Hardware
    Post by mga on 12/11/10 at 23:08:11
    Is the idea to choose from female, male, screw headers? I notice you use all three in the 3D example. My vote would go for the male. There is convenience when having females to just put the cable in but later it can become a mess when you want to "go pro" with the project.

    Title: Re: Wiring 2.0 Hardware
    Post by mga on 12/11/10 at 23:09:27
    And, in some situations having the (male) headers below the board would be nice (as in the Mini).

    Title: Re: Wiring 2.0 Hardware
    Post by b on 12/12/10 at 02:52:28
    @mga: I've discussed this with a few people, and the idea for the headers is to have an option.  That is, when you order the board, you can get 1) no headers, 2) male header pins, 3) female header sockets, or 4) 0.1" screw-downs.  None are mounted in production - the option connectors are simply shipped with the board and must be soldered in.

    RE: mowcius' concern for the reset switch - that switch is 1/2 the size of the typical Omron switch and we have used it on a few boards.

    b

    Title: Re: Wiring 2.0 Hardware
    Post by mga on 12/12/10 at 02:53:57
    Shipping your desired headers sounds like a good idea.

    Title: Re: Wiring 2.0 Hardware
    Post by mowcius on 12/12/10 at 19:14:25

    Quote:
    My vote would go for the male. There is convenience when having females to just put the cable in but later it can become a mess when you want to "go pro" with the project.

    The idea is that you can use any of the three. The boards will require you to solder the headers/screw terminals. Perhaps pre-soldered sockets/screw terminals will be an option for the starting-out hobbyist who wants to flash a few LEDs but hasn't got onto soldering yet (or is not too good at it yet).

    Might I also suggest an onboard LED. I know that wiring is often used mainly by people a bit more into it and beginners seem to go more towards the larger community of ard*ino but it's nice to have one. Perhaps give this a pad and solder jumper so it can be completely disconnected (so pullup resistors work correctly) or left enabled.

    Mowcius


    Title: Re: Wiring 2.0 Hardware
    Post by b on 12/12/10 at 20:26:19
    * points to the LED that is right next to the MCU (it is labeled "D15" - may want to rename that?).

    I picked D15 on this board because D15 is on PORTC, TOSC2 - which is used for connecting an external crystal as an independent clock source for Timer/Counter 2.  People will rarely use this.

    Using D15 as an input will still work.

    Using D15 for "high current" (i.e. > 10 mA) applications will suffer, but only slightly.  In that situation, the user can de-solder the resistor for the LED to disable it.

    b

    Title: Re: Wiring 2.0 Hardware
    Post by roypardi on 12/12/10 at 20:37:41
    I agree @ the LED. It is often a useful diagnostic/sanity check.

    While I'm here... any considerations of auto-reset a la arduino? I ask because I am working on a project now using both a 1st gen Wiring board and an arduino and being able to upload without pressing reset is helpful. I've also done some embedded pieces using arduino where access to the board was difficult so auto reset enabled me to easily program it 'in situ'

    There are no doubt pros + cons on this.

    Title: Re: Wiring 2.0 Hardware
    Post by mowcius on 12/12/10 at 21:56:28

    Quote:
    b points to the LED that is right next to the MCU (it is labeled "D15" - may want to rename that?).

    Heh, didn't notice that.

    Perhaps 'LED' is a nice label to have :D

    Title: Re: Wiring 2.0 Hardware
    Post by cosmogod on 12/17/10 at 11:03:53

    I did love to see the following specs:
    1)Atmega - (your choice)
    2)32KB RAM
    3)32KB EEEPROM
    4)* bit Analog-Digital Converter
    5)Input Capture and Output compare port (for sensors)
    6) DC/Servo motor driver + (PWM)
    7)LCD module with backlight(for debugging and selecting   
       programs)
    8) Switch (Start & Stop)
    9) Knob/joystick/buttons for Analog to digital projects(also   
        to use to select programs
    10)LED/Speaker for user friendly indications.
    11)RS-232/I2C/whatever you like for downloading and 
         interacting with the software.
    12)Rechargeable Circuit.

    and   low cost!!

    Well, I am a bot tinkerer. Which includes toying around with AI & stuff like Display and gadgets. And This is what I fell should be a really great board (for technostuff)  that'll last for a century.

    Cosmogod.


    Title: Re: Wiring 2.0 Hardware
    Post by b on 12/22/10 at 18:03:23
    Here's the latest layout for the Wiring S.


    Wiring-S-V10-PartPlacement-RC2-1 by Rogue Robotics, on Flickr

    Comments welcome.

    b

    Title: Re: Wiring 2.0 Hardware
    Post by b on 12/30/10 at 02:57:34
    Routing is complete.


    Wiring-S-V10-Routed-RC1-1-front by Rogue Robotics, on Flickr


    Wiring-S-V10-Routed-RC1-1-back by Rogue Robotics, on Flickr

    After some thought, I wonder whether a polarity diode on the barrel connector would be a good idea for protection.  I don't think that it would be a good idea at the raw input connector (VIN) though, since we'd like to keep that voltage as close as possible to the minimum required at the regulators.

    Please let me know what you think.

    b

    Title: Re: Wiring 2.0 Hardware
    Post by mowcius on 12/30/10 at 12:57:38
    I would personally would't really care - I have made a few barrel jack adaptors which connect to batteries. Although I don't use them very often (temporary connection to a 9v battery for testing) if I did use one and I had messed it up, a protection diode could be helpful.

    If you can make it relatively easy to get to and either THT or a large SMD package, people who are less likely to reverse polarity and need the lower dropout can replace it with a wire easily enough. Cost is negligible ;)
    But then again, those people are probably using VIn anyway.

    I can't say I have seen many 6v power adaptors. Most are 9v or 12v so I'd go for it.

    Mowcius

    Title: Re: Wiring 2.0 Hardware
    Post by mowcius on 01/03/11 at 01:03:40
    Just noticed how badly I managed to fail at writing that.
    I must apologise :)

    Title: Re: Wiring 2.0 Hardware
    Post by b on 01/21/11 at 02:50:24
    FYI: First run of PCBs expected in tomorrow.  I'll post some snaps as soon as I get them.

    b

    Title: Re: Wiring 2.0 Hardware
    Post by Alex on 01/24/11 at 21:33:39
    Hi All,

    Is the intent of "Option 2" to run the board on a 3.3V rail or is it just to have 3.3V handy for powering other parts?

    Title: Re: Wiring 2.0 Hardware
    Post by b on 01/24/11 at 21:39:47
    The intention is to have 3.3V power available for peripherals/components/shields, etc.

    b

    Title: Re: Wiring 2.0 Hardware
    Post by Alex on 01/24/11 at 22:20:38
    Cool, thank you.

    Title: Re: Wiring 2.0 Hardware
    Post by b on 02/03/11 at 08:07:57
    Sorry it took so long to post the image of the board.  Here it is:


    First Prototype Wiring S Boards by Rogue Robotics, on Flickr

    The board lot will be built up over the next few days, and sent off to beta testers.


    b

    Title: Re: Wiring 2.0 Hardware
    Post by Vaccumdiode2 on 02/05/11 at 16:14:31
    Hello,
               Looks like it is missing some ports.  What about port 4 and port 5?   Maybe i'm missing somthing?  Please let me know.

    Thank you,
    Vaccumdiode2 :-/

    Title: Re: Wiring 2.0 Hardware
    Post by b on 02/05/11 at 18:59:57
    The Wiring S board uses an ATmega644P Atmel controller.  This will make the board more cost effective.

    Some ATmega644P Specs:

    • 64KB of Flash
    • 4KB of SRAM
    • 2KB of EEPROM
    • 4 IO Ports
    • 8 ADC channels
    • 6 PWM channels
    • 3 hardware timers


    Hernando and I are working on converting the core to work with the new boards.  The core will continue to work with the old boards as well.

    The Wiring M board (in the works) will use the ATmega1281, with which you're more familiar.  It will have the same form factor, but have additional ports 4, 5 and 6.

    b

    Title: Re: Wiring 2.0 Hardware
    Post by Vaccumdiode2 on 02/07/11 at 19:01:18
    B,
            I had not understood that you were use a different chip.  I really like the ATmrga128 series of chips.  So I assumed that you were using it.
            Will the new board cost about $25 like the Arduinos? 
            The Board looks really nice.  the color and masking top noch!

    Thank you, 8-)

    Vaccumdiode2

    Title: Re: Wiring 2.0 Hardware
    Post by b on 02/07/11 at 19:19:32
    The idea is to have a more affordable board to use.

    I'm really happy with how they turned out.  We agreed that the black mask, which has been used on the older boards, should stay.

    b

    Title: Re: Wiring 2.0 Hardware
    Post by mowcius on 02/07/11 at 19:49:09
    The Uno is $30 and that is with a processor that isn't as good so I think asking even $30 for the new wiring board is a bit of a squeeze.

    Mind you, I don't know how much it's gonna be...


    Wiring Forum » Powered by YaBB 2.5 AE!
    YaBB Forum Software © 2000-2010. All Rights Reserved.