Skip to main content ↑ ↓
Topic: USB IR toy V2.5 (Read 11316 times) previous topic - next topic

USB IR toy V2.5

Here is the IR toy v2 made on a standardized board. 3.1X5cm....
best regards FIlip.

Re: USB IR toy V2.5

Reply #1 –
I want to make sure that we're using dsm's recommended 0.5mm metric grid for positioning. ON this board I think the headers, USB jack, and RX/RX2/TX should probably all follow this rule to help make cases easier to build.

Any thoughts?
Got a question? Please ask in the forum for the fastest answers.

Re: USB IR toy V2.5

Reply #2 –
[font=Georgia:]USBIRtoy.v2.5.brd [r????~20Jan12] comments

I looked at the USBIRtoy.v2.5.brd [r????~20Jan12] board layout and have a few comments.

  • The overall dimensions of the board are 50.0 mm x 31.0 mm.
    These dimensions are approximately golden ratio to the nearest 1.0 mm.
    This allows the the major axis and minor axis board center-lines to be on a 0.5 mm grid. Nice.
  • The 3.2 mm mounting holes are symmetrically located 4.0 mm inward from the corners. Nice.
  • The radius of the corners appears to be 4.0 mm. Nice.
    ----------
  • Layer 39 tKeepout and Layer 40 bKeepout are not present.
    This makes it difficult to determine whether RX2 and TX are too close to the mounting holes.
  • The center of component RX2 is at location (1.778, 21.717).
    The center of the footprint (QSE15x) for this component is offset from the expected location.
  • The center of component RX is at location (2.286, 15.494). RX1 in the schematic.
  • The center of component TX is at location (2.286, 8.7884).
  • The center of the LED component is at location (22.352, 6.858).
  • The origin of the USB connector is at location (45.212, 15.494).
  • The center of the ICSP connector is at location (14.8336, 29.083).
  • The center of the UART connector is at location (23.7236, 27.813).
  • The center of the JP1 connector is at location (33.8836, 27.813).
  • The center of the push-button component is at location (19.2024, 14.7574).
The component and connector locations were determined after switching to a metric placement grid.
The current oddball location values suggest a "mil" grid was used for component and connector placement.
The Eagle "info" command was used to determine locations.
The origin of the USB connector probably should be at location (44.75, 15.5).
44.75 mm == 50.0 minus 5.25; 15.5 mm == major axis board center-line.
Since the ICSP, UART, and JP1 are 0.1" grid through-hole components, treat them as a group with the center origin located
between pins TX and IRTX. The center of this group should probably be near (25.0, 27.0). The current footprints for these connectors look different from each other on Layer 51 tDocu.
The legend on Layer 26 bNames will need to be relocated.
The Layer 49 Reference for these connectors needs to be updated.
The outlines of RX2, RX, and TX hang over the edge of the board.  This was not a problem when the board had no case, but now these components should be pulled back from the edge of the board.  TX especially needs some consideration because it is generally mounted at 90 degrees to the board (i.e. the footprint is misleading).
Incidentally there is a deep red acrylic intended for IR devices that would make a nice side window over the IR sensitive components for this design.

Thanks for your time.

dsm[/font:]

Re: USB IR toy V2.5

Reply #3 –
Thank you for the review.
Got a question? Please ask in the forum for the fastest answers.

Re: USB IR toy V2.5

Reply #4 –
thanks for the review, It really helps a lot.
My comments..
The usb cant be moved, to the edge, because the pads will stick out from the board.
I aligned the USB and the RX and PIC to the 15,5mm...

the TX cant be moved to the back because it wound be able to clear the 5mm standoff if its moved behind it...
I talked it over with Ian, and we agreed that if we do make sides to the boards, will just drill a hole for the TX led to fit, and stick out :) all the other components will fit inside the board..*same for RX...
Unfortunately the large IO header cant be moved down..that would mess up the gorund layer cohesion thats in place..but I moved it to the horizontal canter of the board(25mm)

Fixed the reference layers
best regards FIlip.

Re: USB IR toy V2.5

Reply #5 –
I think that's the wrong version :)

We weren't really sure what to do with the sensors at the front. Moving the LED back ran into the mounting hole/post, and it felt like we would have to move to a bigger footprint.
Got a question? Please ask in the forum for the fastest answers.

Re: USB IR toy V2.5

Reply #6 –
[font=Georgia:]more comments
[quote author="arakis"]
The usb cant be moved, to the edge, because the pads will stick out from the board.[/quote]
The Kobiconn 154-15320-E Type B 5-pin mini-USB connector data sheet [1] recommends an origin that is 5.25 mm from edge of the board. Since the USB connector is mounted on the right side, 50.0 mm minus 5.25 mm == 44.75 mm for the USB connector "X" coordinate to achieve this placement. The USB pads will not stick out from the board. This placement allows the connector housing to hang over the edge of the board slightly (0.65 mm) in order to maximize engagement margins with the mating mini-USB plug given typical enclosure wall thicknesses. Note that the proposed SOB wall thicknesses and wall-to-board margins make this USB housing overhang even more critical.

Thanks for your time.

dsm

[1] Search for "Kobiconn" in the "Sick of Beige" thread for more details.[/font:]

Re: USB IR toy V2.5

Reply #7 –
[font=Georgia:]more comments

You may have loaded the wrong "brd" file for the 23Jan12 version of the USBIRtoy layout.

The idea behind moving the TX LED further from the edge is to allow all of the IR components to fit inside the case.
This would reduce the number of custom-cut SOB case sides or alternatively to use the same deep red filter (which would improve the signal-to-noise ratio for RX and RX2). Moving the IR LED back from the edge of the board in the "X" direction should not affect the spacing of the IR LED to the standoff in the "Y" direction. Your overall intent for mounting TX would be clearer if you used a right-angle LED footprint for TX.

A similar argument applies to moving RX and RX2 back from the edge of the board (without the right-angle footprint issue).

Thanks for your time.

dsm[/font:]

Re: USB IR toy V2.5

Reply #8 –
[font=Georgia:]more comments

[quote author="arakis"]
Unfortunately the large IO header cant be moved down..that would mess up the gorund layer cohesion thats in place[/quote]
I'm not quite sure what that means. I was looking at the USBIRtoy design from the viewpoint of creating a "slide-in" enclosure or a "clamshell" enclosure and realized that the proximity of the IO connector to the edge of the board is a problem for either case variant. Shifting the PIC (which doesn't interact with the enclosure and doesn't need to be on the board center-line) toward the bottom edge of the board would free up space to shift the IO connector (which does interact with the enclosure) away from the top edge of the board.

Thanks for your time.

dsm[/font:]

Re: USB IR toy V2.5

Reply #9 –
sorry about the files...here are the real ones
best regards FIlip.

Re: USB IR toy V2.5

Reply #10 –
You are off course right about the USB, i have no clue why when I inputted your coordinates, the usb went off the board, but now its ok and at the specified coordinates...I would like not to move the IC from center cause the boards will be sold without cases as well, and IMHO the IO header is not the primary purpose/or selling point of the IR toy..

About moving the LED..well if you move it in the x coordinates, you are closing the gap to the hole since its off center...if you go past the hole center line, the right angled LED will not be able to clear a 5mm standoff...At least that the way it looks when I add a 5mm circle to the board at the same center as the hole..I made a quick sketch to better illustrate the problem..
best regards FIlip.

Re: USB IR toy V2.5

Reply #11 –
[font=Georgia:]more comments

I realize that the TX, RX, RX2 area of the board is fairly tight, but at the same time I believe moving all of the optical components behind the edge of the board would be a good thing. I'll take a look at the USBIRtoy.v2.5.brd [23Jan12] layout tomorrow and see whether anything occurs to me.

You might want to take a look at the "Standoff size and side height" thread.  In particular, Scheme B (described in the thread) would remove almost all board standoff obstructions from the top of the board (by using swaged standoffs).
Scheme B also lends itself to thinner cases and simpler mechanical design.

Another possibility would be to just up the size of the board from 50 mm by 31 mm to 60 mm by 37 mm.
The incremental board cost would fairly small and the layout would be more relaxed.

Just for reference, part of the reason I'm interested in the USBIRtoy design is because I designed some IR remote controls for a medical product.  To help with product development, we had all sorts of little IR protocol measurement widgets to analyze and reproduce IR signals.

Thanks for your time.

dsm[/font:]

Re: USB IR toy V2.5

Reply #12 –
[font=Georgia:]USBIRtoy.v2.5.brd [23Jan12] comments

I looked at the USBIRtoy.v2.5.brd [23Jan12] board layout and have a few comments.
  • The overall dimensions of the board are 50.0 mm x 31.0 mm.
        These dimensions are approximately golden ratio to the nearest 1.0 mm.
        This allows the the major axis and minor axis board center-lines to be on a 0.5 mm grid. Nice.
  • The 3.2 mm mounting holes are symmetrically located 4.0 mm inward from the corners. Nice.
  • The radius of the corners appears to be 4.0 mm. Nice.
  • The origin of the USB connector is at location (44.75, 15.5).
    Origin is 5.25 mm from edge of board as recommended by the connector manufacturer.
    Positioned on the major axis board center-line. Nice.
    ----------
  • Layer 39 tKeepout and Layer 40 bKeepout are not present.
        This makes it difficult to determine whether RX2 and TX are too close to the mounting holes.
    2.5 mm circle added to the lower left mounting hole on Layer 20 Dimension layer.
  • Layer 41 tRestrict and Layer 42 bRestrict are not present.
    Layer 43 vRestrict is present, but only around the crystal.
    For reasons discussed in the "Best Positioning Practices" forum thread,
    I recommend a 3.0 mm radius "clear" keepout around the mounting holes (including the GND fill).
    Note that 5.0 mm (flat-to-flat) hex standoffs are 5.774 mm (vertex-to-vertex).
    ----------
  • The center of component RX2 is at location (2.0, 21.717).
        The center of the footprint (QSE15x) for this component is offset from the expected location.
  • The center of component RX is at location (2.5, 15.5).
  • The center of component TX is at location (2.5, 8.7884).
    ----------
  • The center of the LED component is at location (22.352, 6.858).
  • The center of the ICSP connector is at location (14.84, 29.083).
  • The center of the UART connector is at location (23.73, 27.813).
  • The center of the JP1 connector is at location (33.89, 27.813).
  • The center of the push-button component is at location (19.2024, 14.7574)
    X-ray location since on the push-button is on the bottom of the board.
  • The center of the Q1 component is at location (33.02, 6.985).
    Included because the crystal is comparatively tall.
The Eagle "info" command was used to determine component and connector locations.
The current oddball location values suggest a "mil" grid was initially used for component and connector placement.
I recommend a 0.5 mm placement grid for all components that interact with the enclosure.
The vertical position of the TX, RX, and RX2 may need to be tweaked as mentioned below.
The other components might also benefit from being on a 0.5 mm placement grid.

Since the ICSP, UART, and JP1 are 0.1" grid through-hole components, treat them as a group with the center origin located
between pins TX and IRTX on the UART connector. The center of this group should probably be near (25.0, 27.0). The current footprints for these connectors look different from each other on Layer 51 tDocu. Shifting this group would also allow slide-in enclosures. The legend on Layer 26 bNames would need to be relocated. The Layer 49 Reference for these connectors would need to be updated.

The outlines of RX2, RX, and TX hang over the edge of the board. This was not a problem when the board had no case, but now these components should be pulled back from the edge of the board. TX especially needs some consideration because it is generally mounted at 90 degrees to the board (i.e. the footprint is misleading). As you point out, the layout of these components is very tight after the mounting holes are added. I investigated swapping RX2 and RX and moving RX and TX back from the edge of the board. The vertical position of the TX, RX, and RX2 may need to be tweaked slightly (move to 0.1 mm grid).

Both of the above suggestions would require moving IC1 slightly (vertically for the former, horizontally for the latter),
but there should be enough room.

Thanks for your time.

dsm[/font:]

Re: Best positioning practices

Reply #13 –
Hello everyone,

I have here the board for the USBIRToy v2.5 on the 60x37mm DP standard board.
As required most of the parts that interact with its enclosure are on the 0.5mm grid on its center-lines
The coordinates of those parts are as follows:
TX LED (6,11)
RX (4, 18.5) centered vertically on the board
RX2 (4, 26)
USB-mini B (54.5, 18.5) centered vertically on the board, opposite to RX component
JP1 header (38, 32)
UART header (27.5, 32)
ICSP header (18, 33)
I also include these, if they are somehow might affect on the enclosure for its height.
PIC (30, 18.5)
Crystal (38, 10)

Hole Restrict diameter for top and bottom is 6mm
Board edges are with 2mm component Keepout on top and bottom. TX LED, RX and USB mini B connector are the only parts that passes through the component keepout (am I right?)

board and image are attached for checking.

advices are welcome.

cheers,
vimark

Re: USB IR toy V2.5

Reply #14 –
Thanks Vimark.

We decided to go to the next bigger size. That means we have room to add crazy high-power transmitters in v3 without a lot of fuss or moving to a bigger PCB.

I am eager to hear dsm's thoughts, especially about the 2mm keepout on the edge - do you think that is enough/too much, etc?
Got a question? Please ask in the forum for the fastest answers.