Parallax P2 Eval Board

14567810»

Comments

  • cgraceycgracey Posts: 11,738
    edited 2018-11-16 - 00:24:49
    jmg wrote: »
    Ken Gracey wrote: »
    Publison wrote: »
    Will the PCB be Open Source?

    Our PCB files have always been open source long before an organization existed to validate the purpose.

    We will put out PCB files after we prove out the functionality of the design. You will be able to do what you want with them, free of restrictions.

    I think most vendors publish their eval info, because it is in everyone's interest to have good reference designs.
    That was one reason I tried to get Chip to improve the Switcher design, to something more 2018+ and elegant.

    We can do that next round.

    Now, I know you probably stated it several times, but what would be the best 3A switcher for our use here?
  • jmgjmg Posts: 14,008
    edited 2018-11-16 - 01:46:55
    cgracey wrote: »
    We can do that next round.

    Now, I know you probably stated it several times, but what would be then best 3A switcher for our use here?


    That's a bit of a moving target, as in that space things seem to be moving rapidly, and prices are falling, as well as integrated inductors are on offer.
    Fully integrated modules (all caps included) are still too expensive.

    If you are price focused, & ok with external inductor, and are OK with Asian vendors, then the
    Fitipower Integrated Tech FR9889SPCTR HSOP-8 LCSC Part #:C81337 4.5~23V 3A 340kHz PGood 120/100mOhms STK 5115 500+ $ 0.1843 UVLO typ 4.2v
    & that is quite a good price point.

    Or, Asian vendor with integrated inductor is MUN12AD03-SH, 4.5~16V, 3A, PG, 3.5*3.5*1.7mm, prices 25c/2k5 lcsc, or 40c/10+ yoycart.


    If you prefer more mainstream suppliers, and like integrated inductor, the MPM3632C 3A 4.5~18V PG, is QFN-20 (3mmx5mmx1.6mm) 2500+ $1.77 Stocked at DK, Mouser,
    TI have some similar, but supply looks uncertain atm, and I think MPS are a US company, traded as NASDAQ: MPWR

    MPM3632C has low loss fets, which means less switcher loss (compare with FR9889, which also claims 3A rating).

    HS switch on resistance HSRDS(ON) VBST-SW = 3.3V 36 mΩ
    LS switch on resistance LSRDS(ON) VCC = 3.3V 18 mΩ
    Inductor DC resistance LDCR 25 mΩ
    VIN UVLO rising INUVVth 3.2 3.6 3.9 V



    Emerging parts to watch, are those with i2c, and there is also a trend to buck-boost (4 FETS) allow very wide supply ranges.

    I would also suggest a chat with the OnSemi Power supply product manager, as they may have something coming ?

    If I was doing an Eval board, as a banner product, more compact design always 'looks better' and you do not want to spook users with a large BOM in the power supplies.
  • Jmg, agreed. The power supplies should not cause things to look complicated.
  • My impression is buck-boost function is less about wide input voltage and more about operating around a battery voltage that crosses the regulated voltage as it discharges. Particularly handy for single cell LiCo with their 4.2 to 2.5 volt range.

    "We suspect that ALMA will allow us to observe this rare form of CO in many other discs.
    By doing that, we can more accurately measure their mass, and determine whether
    scientists have systematically been underestimating how much matter they contain."
  • jmgjmg Posts: 14,008
    evanh wrote: »
    My impression is buck-boost function is less about wide input voltage and more about operating around a battery voltage that crosses the regulated voltage as it discharges. Particularly handy for single cell LiCo with their 4.2 to 2.5 volt range.
    Err, yes, note that anything that allows the Vin range to increase to below Vout, has naturally widened the supply range.
  • Wide is mostly from 10 to 70 volts or more. Filtered vehicle/industrial/telco/large-battery supplies.

    "We suspect that ALMA will allow us to observe this rare form of CO in many other discs.
    By doing that, we can more accurately measure their mass, and determine whether
    scientists have systematically been underestimating how much matter they contain."
  • Wow!.... I haven't checked for a while.. and now it's real close.... I want one ! ..... please....
  • 10gigbill wrote: »
    Wow!.... I haven't checked for a while.. and now it's real close.... I want one ! ..... please....
    So you're willing to drop a $10G bill on this? Then you can probably have whatever you want! :smile:

  • VonSzarvasVonSzarvas Posts: 1,654
    edited 2018-11-16 - 12:19:09
    @W9GFO 3D Models… here you go.

    There are lots of export options. If any of these work, could you say which is most useful ?

    https://drive.google.com/drive/folders/1t-KQwLfhzz7xI-AghJN2dIrHyn9Px4C8



    @jmg About the psu… Maybe I can offer a few too many words from a background perspective...

    We needed a stout and flexible PSU for this EVAL board. Something that would allow us to figure out the best power requirements for a future board. Something that meets the needs of the task, limited build qty and target market of this product..

    The circuit used makes more sense if you consider that Parallax already builds the same supply in other products. It’s a known entity from which the necessary testing can be performed more efficiently. The same buck also shares a lower current same-footprint part, which we can swap in for testing at lower energy switching.

    The extra part footprints seen in the 3D images around the PSU enable a lot of tuning and experimentation at different supply voltages and conditions. Not all will be populated by default. Such flexibility is often not possible when switchers have internal parts with pre-determined design assumptions.

    The target market for P2-EVAL is closely aligned developers and contributors, mainly from the forum. It seems that this board will have sold out on release already, just to members here that have expressed interest! So it’s not expected to go onto general sale exactly in the current form. P2-EVAL is not meant to be a reference design just yet. It will deliberately be offered as an "Evaluation Board" and have a part number suffix "-ES" to denote that the P2 onboard is an Engineering Sample. Not production quality or reference design.

    This board includes a lot of extra features for experimentation and evaluation. I see the mission of the EVAL board is to get P2 as fast as possible into the hands of experimenters, and to provide very many flexible powering and interfacing options. For all this, the power supply we have is ideal, IMO. And with the header combinations, voltage levels can be adjusted, and external supplies can be attached.


    All that said, I agree with some comments here. Once P2 moves to mass production build qty, and maybe something like a P2-EDGE, FLiP or QuickStart style board is released, then a simplified and optimised PSU would be appropriate to develop and tune for a reference design. The lessons and experience of P2-EVAL will confirm for us what those optimal power requirements are!


    I really hope the people that get these boards will not be limited in what they want to achieve. We didn’t want to go through endless design cycles tuning this product. These should be in your hands as fast as possible!


    Curiosity and experimentation is fully supported with P2-EVAL !


    Edit: Just added a couple "dimensions" drawings to the 3D Model folder.
  • VonSzarvas wrote: »
    @W9GFO 3D Models… here you go.

    There are lots of export options. If any of these work, could you say which is most useful ?

    https://drive.google.com/drive/folders/1t-KQwLfhzz7xI-AghJN2dIrHyn9Px4C8



    @jmg About the psu… Maybe I can offer a few too many words from a background perspective...

    We needed a stout and flexible PSU for this EVAL board. Something that would allow us to figure out the best power requirements for a future board. Something that meets the needs of the task, limited build qty and target market of this product..

    The circuit used makes more sense if you consider that Parallax already builds the same supply in other products. It’s a known entity from which the necessary testing can be performed more efficiently. The same buck also shares a lower current same-footprint part, which we can swap in for testing at lower energy switching.

    The extra part footprints seen in the 3D images around the PSU enable a lot of tuning and experimentation at different supply voltages and conditions. Not all will be populated by default. Such flexibility is often not possible when switchers have internal parts with pre-determined design assumptions.

    The target market for P2-EVAL is closely aligned developers and contributors, mainly from the forum. It seems that this board will have sold out on release already, just to members here that have expressed interest! So it’s not expected to go onto general sale exactly in the current form. P2-EVAL is not meant to be a reference design just yet. It will deliberately be offered as an "Evaluation Board" and have a part number suffix "-ES" to denote that the P2 onboard is an Engineering Sample. Not production quality or reference design.

    This board includes a lot of extra features for experimentation and evaluation. I see the mission of the EVAL board is to get P2 as fast as possible into the hands of experimenters, and to provide very many flexible powering and interfacing options. For all this, the power supply we have is ideal, IMO. And with the header combinations, voltage levels can be adjusted, and external supplies can be attached.


    All that said, I agree with some comments here. Once P2 moves to mass production build qty, and maybe something like a P2-EDGE, FLiP or QuickStart style board is released, then a simplified and optimised PSU would be appropriate to develop and tune for a reference design. The lessons and experience of P2-EVAL will confirm for us what those optimal power requirements are!


    I really hope the people that get these boards will not be limited in what they want to achieve. We didn’t want to go through endless design cycles tuning this product. These should be in your hands as fast as possible!


    Curiosity and experimentation is fully supported with P2-EVAL !


    Edit: Just added a couple "dimensions" drawings to the 3D Model folder.

    What do you guys use to open .STEP files. I can open .WRT with MS 3D Builder.
  • potatoheadpotatohead Posts: 9,807
    edited 2018-11-16 - 19:22:56
    Those generally need to be opened by a real cat system. The 3D Builder only works with polygons files. Step is actual solid model geometry. You could sign up for free addition of Autodesk Fusion 360 for example, and open those files. Someone could open them in SolidWorks and Export STL files as well.

    I just knock this out with Google Voice sorry for the typos.
    Do not taunt Happy Fun Ball! @opengeekorg ---> Be Excellent To One Another SKYPE = acuity_doug
    Parallax colors simplified: https://forums.parallax.com/discussion/123709/commented-graphics-demo-spin<br>
  • VonSzarvas wrote: »
    @W9GFO 3D Models… here you go.

    There are lots of export options. If any of these work, could you say which is most useful ?

    https://drive.google.com/drive/folders/1t-KQwLfhzz7xI-AghJN2dIrHyn9Px4C8
    Thanks Von, the .step files are most useful to me.
  • Publison wrote: »
    What do you guys use to open .STEP files. I can open .WRT with MS 3D Builder.

    I use Fusion 360.
  • RaymanRayman Posts: 9,760
    edited 2018-11-16 - 19:49:21
    Is there an IGES (.igs) or SolidWorks output?
    I also can import the step file. Maybe it's good enough for me...

    902 x 737 - 75K
    Prop Info and Apps: http://www.rayslogic.com/
  • Rayman wrote: »
    Is there an IGS or SolidWorks output?

    Those files are the only formats I can export from Diptrace.
  • VonSzarvas wrote: »
    Rayman wrote: »
    Is there an IGS or SolidWorks output?

    Those files are the only formats I can export from Diptrace.

    I can export .igs if needed.

    What is missing is the top silk layer.png (without dimensions) and the top/bottom layers. I don't need them but they would be nice in order for a render to look right.

  • Silk png. Could do that. 100% print scale suitable?
  • VonSzarvas wrote: »
    Silk png. Could do that. 100% print scale suitable?
    Yes, that would work. The png of the top and bottom layers would be great too.

    Also, I noticed that the tactile switch seems to be scaled in one direction, the button is oval instead of round.
  • VonSzarvas wrote: »
    Silk png. Could do that. 100% print scale suitable?

    Actually what would work best (for me) is another top down view (...09_44_05-3D Preview.png) but with the perspective turned off.
  • We have not yet opened the gate on this right?

    I just had a flash fear. Missed it!
    Do not taunt Happy Fun Ball! @opengeekorg ---> Be Excellent To One Another SKYPE = acuity_doug
    Parallax colors simplified: https://forums.parallax.com/discussion/123709/commented-graphics-demo-spin<br>
  • W9GFO wrote: »
    Actually what would work best (for me) is another top down view (...09_44_05-3D Preview.png) but with the perspective turned off.

    I've added some new images and a new .step file with a different (rounded) button.

    The Diptrace 3D renderer doesn't support orthogonal projections without perspective... is that possible to do with some 3rd party software, after importing the .step file ?



  • VonSzarvas wrote: »
    W9GFO wrote: »
    Actually what would work best (for me) is another top down view (...09_44_05-3D Preview.png) but with the perspective turned off.

    I've added some new images and a new .step file with a different (rounded) button.

    The Diptrace 3D renderer doesn't support orthogonal projections without perspective... is that possible to do with some 3rd party software, after importing the .step file ?
    What you have posted is more than enough. Thanks!

    I didn't realize that perspective could not be switched off in DipTrace. Yes, most CAD programs I have worked with will display in 3D without perspective, in fact it seems to be the default setting and it makes my brain hurt. Once in a while though it is useful.

Sign In or Register to comment.