Shop OBEX P1 Docs P2 Docs Learn Events
Prop2 Analog Test Chip Arrived! — Parallax Forums

Prop2 Analog Test Chip Arrived!

cgraceycgracey Posts: 14,261
in Propeller 2
We've got 100 of these:

Prop2_Test_v1_packaged.jpg

I've been so busy struggling with the ROM code that I haven't made a board for these, yet. I'll make a board that will plug into the Prop123, so that these chips can be exercised as if they were part of the core logic.
Prop2_Test_v1_packaged.jpg
622 x 593 - 53K
«134567…13

Comments

  • cgraceycgracey Posts: 14,261
    Here's how the bonding went:

    OnSemi_Prop2_Test_v1.png

    And here is the pinout, as if it had gone into the originally intended 80-pin package:

    Prop2_Analog_Test_V1.png
    OnSemi_Prop2_Test_v1.png
    1645 x 2559 - 2M
    Prop2_Analog_Test_V1.png
    854 x 852 - 24K
  • Peter JakackiPeter Jakacki Posts: 10,193
    Looking good, some partial P2 silicon is always good. Which analog functions are you testing?
  • cgraceycgracey Posts: 14,261
    Peter Jakacki wrote: »
    Looking good, some partial P2 silicon is always good. Which analog functions are you testing?

    Everything. All elements of the pad frame are in there. So, there are two I/O pins (4 fuses in each), the clock pads (RCFAST/RCSLOW/XTAL/PLL), and the RESn and TESn pads.
  • ozpropdevozpropdev Posts: 2,793
    Fabulous! :)
  • TubularTubular Posts: 4,722
    Fantastic. Nice that its QFP rather than that strange double row qfn, too.

    Good luck with the testing.
  • Heater.Heater. Posts: 21,230
    Yea ha! Real silicon.

    Starts biting finger nails, nervously waiting for test results...
  • Cluso99Cluso99 Posts: 18,069
    Woohoo ;)

    Fingers and toes crossed that all functions as expected!!!
  • Cluso99Cluso99 Posts: 18,069
    Chip,
    You might be able to buy a PLCC128 breakout board off the shelf to start some testing while you layout and wait for your own pcbs.

    BTW what is the pitch of these chips?
  • samuellsamuell Posts: 554
    Great! That's good news. It seems that the P2 will be launched next year, if all goes well.
  • RaymanRayman Posts: 14,934
    Good luck! QFP is a lot easier...
  • PublisonPublison Posts: 12,366
    Cluso99 wrote: »
    BTW what is the pitch of these chips?

    Looks like it's .5mm. Plenty of breakout boards for that footprint.
  • cgraceycgracey Posts: 14,261
    It's a 20x20mm package with 0.5mm lead pitch.
  • Ken GraceyKen Gracey Posts: 7,405
    If you want this board done quickly, feel free to ask Michael or David to just get it done for you, Chip. We're all interested in the results of your first look at this test chip.

    Ken Gracey
  • TubularTubular Posts: 4,722
    Publison wrote: »
    Cluso99 wrote: »
    BTW what is the pitch of these chips?

    Looks like it's .5mm. Plenty of breakout boards for that footprint.

    I had a quick look but didn't find much in square 128 pin 0.5mm. There were rectangular 128 pin qfp package breakouts (RE470), however I suspect the 144 pin breakouts would work as these are also designed for 20x20mm

    http://uk.rs-online.com/web/p/surface-mount-smt-to-through-hole-adapter-boards/3675219/
  • potatoheadpotatohead Posts: 10,261
    Awesome!

    Hey, will we be able to test DAC input?
  • cgraceycgracey Posts: 14,261
    potatohead wrote: »
    Awesome!

    Hey, will we be able to test DAC input?

    It has two instances of the I/O pad. So, we've got both DAC and ADC.
  • potatoheadpotatohead Posts: 10,261
    edited 2016-10-23 04:00
    :D Cool! And of course, I had a brain fart. ADC, it is.

  • Bob Lawrence (VE1RLL)Bob Lawrence (VE1RLL) Posts: 1,720
    Excellent!! looks great.
  • samuellsamuell Posts: 554
    Hi Chip,

    Which foundry did you use? I expected to be ON Semi, and on a different package as well. Also, are those 180nm testchips?

    Kind regards, Samuel Lourenço
  • cgraceycgracey Posts: 14,261
    samuell wrote: »
    Hi Chip,

    Which foundry did you use? I expected to be ON Semi, and on a different package as well. Also, are those 180nm testchips?

    Kind regards, Samuel Lourenço

    Yes, OnSemi made these and they use the ONC18 process.
  • BaggersBaggers Posts: 3,019
    Fingers crossed it all goes well :D looking good Chip!
  • BaggersBaggers Posts: 3,019
    Wow, my first week off work after a long year of critical path stuff, and Prop2 is on a chip! so here's hoping it goes well with the tests.
  • SeairthSeairth Posts: 2,474
    Baggers wrote: »
    Wow, my first week off work after a long year of critical path stuff, and Prop2 is on a chip! so here's hoping it goes well with the tests.

    Not quite. This is only the ring (analog, smart pins, etc), no cogs or hub memory.
  • RaymanRayman Posts: 14,934
    Can you do grayscale VGA with this?
    Seems like maybe it could...
  • jmgjmg Posts: 15,186
    Rayman wrote: »
    Can you do grayscale VGA with this?
    Seems like maybe it could...
    It certainly should be possible to confirm the DAC Drive, and Slew rates on changes (bandwidth), and create some test patterns, and test things like Sync-On-Green.

    Limited Colour VGA and Limited Colour Component Video should be possible with 2 DACs.

    The final change-rate will be dictated by how quickly the test interface can update the DACS, so it might not manage full VGA pixel-change-rates.
  • RaymanRayman Posts: 14,934
    Actually, I guess you need 3 I/O for VGA: Vsync, Hsync, Luma...
  • ElectrodudeElectrodude Posts: 1,666
    I think I remember someone here saying there's some trick to send VGA with only two I/O, by XORing the hsync and vsync together or something like that.

    Could you instead just use two chips to get 4 I/O?
  • jmgjmg Posts: 15,186
    Rayman wrote: »
    Actually, I guess you need 3 I/O for VGA: Vsync, Hsync, Luma...

    Strictly, yes, but whatever is driving the test chip, I figured could easily manage H,V Syncs.

  • TubularTubular Posts: 4,722
    Well there are already signals including sync driving the THS8136 dac's on the P123 boards.

    You could just duplicate/divert the signals to the P2 test silicon and see how it looks, side by side.
  • RaymanRayman Posts: 14,934
    True. That should work.
Sign In or Register to comment.