Jump to content
IGNORED

PlusCart(+) + SD


Andrew Davie

Recommended Posts

1 hour ago, Al_Nafuur said:

My first 5 protype boards had all PINs, but the "production" boards have omitted some of the unnecessary PINs:

 

I may have got one or two wrong, as I was working from an image for speed purposes. I marked the unused pins with a large green dot.

It wouldn't be too hard to modify, and I'm not so against this idea after all. However, I'd like to keep the "window" for the jumper, as I'm always trying in dim light I prefer to find the correct pins for adding/removing the jumper. With a hole in the board to guide, it should be easier.

I can see three groupings (all 2x3) where the pins could be dropped off the board.

unused.jpg

Link to comment
Share on other sites

Thank you, Now I was able to open the design.

You have used the curved tracks to get a retro style, correct? 

 

May I suggest you to reduce the track witdths to 16 or 12 mils, so you will have more clearance between the soldering pads and the tracks and hence less chance of shortings close to the solder joints?

 

Another suggestion is to move the tracks to the outer side of the pin header (for instance like below).

image.png.167cf9811165d2bdeadb72f4220899bf.png

image.png.c46499f13b5d6974d0776148226f4f70.png

 

 

When you are routing the board it is easier to get the tracks aligned if you use a minimum grid of 12.5 mils.

 

Another tip is when you are moving/positioning the components on the board, you can get them at precise coordinates using the "Info" button, as it allows you to change various parameters of either component or track or text being inspected (very useful for precise positioning of components and to keep tracks perfectly horizontal / vertical by copying the coordinates of one side and pasting on the other.

image.png.4408b8d4e39f3d495fe9132944ca6f75.png

 

 

Do you have a mechanical drawing of your SD card module or have you reverse engineered its dimensions ?  Have you considered using more than one type of SD card module ?

 

 

 

Edited by Danjovic
  • Like 1
Link to comment
Share on other sites

2 minutes ago, Danjovic said:

You have used the curved tracks to get a retro style, correct? 

Thanks for the review. I used what was pleasing to my eye. I very much dislike sharp corners/turns.

 

2 minutes ago, Danjovic said:

May I suggest you to reduce the track witdths to 16 or 12 mils, so you will have more clearance between the soldering pads and the tracks and hence less chance of shortings close to the solder joints?

I also very much like thick tracks! I've not run across issues with the spacing I use; I've made another project with the same design decisions. I guess since it's a shared project anyone can change it... I do like it as it is.

 

2 minutes ago, Danjovic said:

 

Another suggestion is to move the tracks to the outer side of the pin header (for instance like below).

Not sure I understand the whys and wherefores of that?

 

 

2 minutes ago, Danjovic said:

When you are routing the board it is easier to get the tracks aligned if you use a minimum grid of 12.5 mils.

If I need to, I align tracks on 0.635 mm, but really for the most part it's freeform based on the pin positions and eyeballing the spacing. You may notice the spacing on the curves is (reasonably) consistent.

 

2 minutes ago, Danjovic said:

 

Another tip is when you are moving/positioning the components on the board, you can get them at precise coordinates using the "Info" button, as it allows you to change various parameters of either component or track or text being inspected (very useful for precise positioning of components and to keep tracks perfectly horizontal / vertical by copying the coordinates of one side and pasting on the other.

Yes, this is what I do - although there may be a few slightly out. I switch to info when I really want to align things properly.

 

2 minutes ago, Danjovic said:

Do you have a mechanical drawing of your SD card module or have you reverse engineered its dimensions ?  Have you considered using more than one type of SD card module ?

I don't have a drawing or library file, so I dug out the calipers and did some rough measurements, and then created a library object which is good-enough. It's possible there may be other suitable modules, but the constraint is where the SD reader can sit (not many places) and the slot for inseting the card must be flush with the outside edge of the 3D shell in which it all sits.

 

 

Link to comment
Share on other sites

 

27 minutes ago, Andrew Davie said:
Quote

May I suggest you to reduce the track witdths to 16 or 12 mils, so you will have more clearance between the soldering pads and the tracks and hence less chance of shortings close to the solder joints?

I also very much like thick tracks! I've not run across issues with the spacing I use; I've made another project with the same design decisions. I guess since it's a shared project anyone can change it... I do like it as it is.

Ah ok. I have asked with the intention of changing it and post it back here. Nevertheless whenever you need to change all the track widths at the same time you can mark the whole board and then use change->witdth and then right click to change the whole group.

 

30 minutes ago, Andrew Davie said:
Quote

 

Another suggestion is to move the tracks to the outer side of the pin header (for instance like below).

Not sure I understand the whys and wherefores of that?

To reduce the chance of a solder blob to cause a short circuit in a place where is hard to reach, either to visually inspect or to remove the blob.

 

 

36 minutes ago, Andrew Davie said:

 

Quote

Do you have a mechanical drawing of your SD card module or have you reverse engineered its dimensions ?  Have you considered using more than one type of SD card module ?

I don't have a drawing or library file, so I dug out the calipers and did some rough measurements, and then created a library object which is good-enough. It's possible there may be other suitable modules, but the constraint is where the SD reader can sit (not many places) and the slot for inseting the card must be flush with the outside edge of the 3D shell in which it all sits.

Ok, in this case I strongly suggest you to print the image of the board to check if the dimensions are all ok before ordering the prototype board.

 

A have edited the SD card library, adding the footprint and the symbol with the component and assigning the pins between them. The library (eagle 7.0) is attached

 

image.png.29664eb311b2776b0c0ae80024966951.png

 

I am also working on some minor things on the board (misalignments of the cutting lines), I will post it back here, just in case :)

MICROSD_7X.lbr

  • Like 1
Link to comment
Share on other sites

5 minutes ago, Danjovic said:

Ok, in this case I strongly suggest you to print the image of the board to check if the dimensions are all ok before ordering the prototype board.

The version I based this off is the same measurement in the essential dimensions. I checked with the measurement tool to make sure I got the basics right. I'm confident it's OK. As to the MicroSD object, it's only the pin spacing that's really critical for the first version. But I didn't think about printing, so I'll do that and see how good the positioning of the MIcroSD is relative to the cartridge shell edge, and how close the board is to the original in the parts that matter. Tomorrow.

 

5 minutes ago, Danjovic said:

A have edited the SD card library, adding the footprint and the symbol with the component and assigning the pins between them. The library (eagle 7.0) is attached

OK, the library was a WIP. The MICROSD is/was unused. It's the MICROSD2 that I use. And it already has a footprint.

Thanks for your update - I'll look at it tomorrow. I had to follow a tutorial to get anything at all, so it's really a first-effort hack.

 

5 minutes ago, Danjovic said:

I am also working on some minor things on the board (misalignments of the cutting lines), I will post it back here, just in case :)

Yes, the dimensions lines are a bit of a mess. I had the original outline, and I had to cut/paste/move/delete/add stuff, and what I guess happened was that because I was in freeform mode (effectively no grid), unless you get the endpoints exactly exactly exactly right on top of each other (even though they look on top at high magnification, they aren't necessarily so).... well, then it's not a connected outline, and so you don't get the cutout.

 

So I ended up drawing a second outline over the top of the first one, and then adjusted it a bit here and there, deleting bits of the first one when I could see 'em, etc.  I just wanted/needed to get something that worked out the door today, and the dimensions layer really does need a whole rework - especially now I understand why it wasn't working as expected!  In short, if you want points to join, then use the info edit coordinates to make sure they're coincident, or at least use a grid. Otherwise... no cutout!  The black/dark dark grey shading difference between the board and void, as defined by the dimension outline, isn't nearly obvious enough on my display.

 

 

 

 

 

Link to comment
Share on other sites

20 minutes ago, Danjovic said:

Ok, in this case I strongly suggest you to print the image of the board to check if the dimensions are all ok before ordering the prototype board.

 

6 minutes ago, Andrew Davie said:

The version I based this off is the same measurement in the essential dimensions. I checked with the measurement tool to make sure I got the basics right. I'm confident it's OK.

I have noticed at times that printing to paper is not always accurate (printer scales it a bit)

 

I have used KiCAD a few times, and it can output an STL of your board, so you can 3D print it.  I don't know if Eagle can do that since I've not used it. 

 

KiCAD can import in Eagle files, too. 

 

So if there is any doubt about the board size, shape, dimensions, etc, then 3D print the board. 

 

Link to comment
Share on other sites

2 minutes ago, Andrew Davie said:

OK, the library was a WIP. The MICROSD is/was unused. It's the MICROSD2 that I use. And it already has a footprint.

Ops, totally missed that.

 

Well then, here follows the board with some modifications. Feel free to ignore whichever changes, lol!

PCB

  • Attached the board drawing. Moved the whole board so the headers now match the 0.1" grid.
  • Reduced the tracks to 16 mils (can be increased back to 24)
  • Moved the straight lines to out of the middle of the rows of the esp8266 2x4 header
  • Rerouted the wires close to the SD header to remove one Via
  • Rerouted the wires of signal D4 to remove another via
  • Straightened the dimension lines (and set it width to zero)
  • Changed the "orphans" attribute of ground polygons to OFF.

image.png.dbff247cd132ca7f7938d29e513e594a.png

 

Schematic

  • Added a frame around the components
  • Rearranged the headers to be in the same disposition as the board

image.thumb.png.696afc671b19e1663d4dd1adc6e549a1.png

 

PlusCartUnified_7X 2021.03.14_DJV.zip

Link to comment
Share on other sites

28 minutes ago, fdr4prez said:

I have noticed at times that printing to paper is not always accurate (printer scales it a bit)

Indeed, some printers (and operating systems) can mess with the scale. It is always good to verify the scale first.

Once I had a printer that worked OK under windows XP and 7, but when i moved to windows 10 it shortened the scale of the prints, no matter what I have tried to troubleshoot.  In that case the workaround was to print to postcript and windows 10 did not messed with the sacle of postcript files :)

 

  • Like 1
Link to comment
Share on other sites

44 minutes ago, Albert said:

@Al_Nafuur, I've made the PlusCart User club public, which means everyone can see it and post without joining.  If you'd like, I can make it "open", which means it's still visible, but people will have to join before they can post.  Please let me know which you prefer.

 

 ..Al

I would prefer "open".

Link to comment
Share on other sites

Here's an update.

 

I started with @Danjovic's version as a base, and first fixed up the couple of issues I spotted there. I always pass the design through the self-checking ERC and DRC tools, and there were two issues. a) VCC missing junction, and b) I can't remember ;).  The VCC one came about because of the addition in the design sheet of the VCC junction, and because I didn't really want to spend time fixing it, just removing that junction object fixed that problem. It can be fixed later; it's just a visual thing for the design. Oh yes, now I remember - the other was frame doesn't have a value, and that was a simple addition of a value to the frame. Well, yes. In my programming, I always treat warnings as errors, and I see no reason that shouldn't apply to hardware, too.

 

So, onto the changes - I was OK with most of what I saw, and appreciated the removal of a couple of the vias. I hate those with a passion, so good to see them go. Vias are the through-holes from one side to the other side and they generally show a lack of skills/foresight in routing traces. IMHO. I'm a bit OCD as you will see later. Maybe you've noticed already.

 

So, firstly I printed a STL of the board, thanks to the suggestion earlier in the thread. I exported the design as a STEP file, and then used an online converter ( https://anyconv.com/step-to-stl-converter/ ) to convert that into a STL file for 3D printing. That worked a treat, and it highlighted a couple of issues.  First, there's the size comparison; the paper version was woefully incorrect, and the STL was perfectly-sized. My first 3D print didn't really show on the print itself where the MicroSD was mounted, so I put in the two mounting holes as on the SD board. These may or may not be useful in the final version, but they're super for getting things exact in the testing.  Initially I found that the board was about 3 mm shy of the edge of the cartridge shell, so my calculations were wrong after all. I've modified that, and will print a new shell and dummy-board overnight but I'm pretty confident it will be good to go.

 

IMG20210214231556.thumb.jpg.789c28a3645afecfd4e401aade3bcd6d.jpg

 

The above picture shows the penultimate version, with the SD card reader about 1 mm shy of the correct position. Now fixed.

 

IMG20210214231625.thumb.jpg.b7838d894f60814eca72b183a0371b3c.jpg

 

I modified the PlusCart 3D case to have a slot, and this time I could actually plug-in my dummy board and measure exactly where the slot needed to go. It's going to look pretty good, I think. The above are the slightly not right version. Will show pics of the final version tomorrow.

 

As to the changes in the board; I modified the MICROSD library to include the holes - but also I accurately modelled the dimensions and position of the card holder on the board. It's not symmetrical, so there's that. Also, the pin that I had the square hole trace around was GND and this turns out to be incorrect; on the module board itself, the square one is 3V3, so I corrected that!  Now, it's a really odd way of doing things, but I'm now mounting the reader underneath the PlusCart board, with the components pointing up towards it. This is because of the geometry of where the lip of the cart shell is, and what seemed to make the most sense.  You can see in the pictures, I think. Those white things are just little printer filament 'pegs' that I've put through the mounting holes for this test. In the real version they can be screws if you like, but really the soldered pins should be more than enough to hold everything in place.  I may or may not have to mount the ESP-01S a fraction of a mm higher; it will sit OK on the reader, I hope.

 

I updated the PlusCart shell to have a slot in hopefully the right place. You can also see the little "lip" I put on it. That hides the fact that the two halves sit flat against each other - if it wasn't perfect you could see light through the join. The lip now extends into the other half, so you don't see the light anymore. An unexpected side effect is that the cartridge now (lightly) holds together without any pins at all, just the lip, so that's a plus.  I've included the two STL files in case anyone wants to print this during development.

 

 

438533431_ScreenShot2021-02-14at11_46_45pm.thumb.png.ac3fd4eec34ef18be7d47fc5e8ec5d39.png

 

I restored orphans on the ground plane. @Al_Nafuur likes 'em, and I like 'em too. Besides I went to all the trouble to find how to specifically turn 'em on, so I guess they stay for a while. It's a style preference here.

 

I removed one block of 2x3 header which weren't used. Removing others would be a much more complex job.

 

Now here's what it's like to have OCD... I mentioned earlier that I didn't like the VIAs - those through-holes. Well you can't always eliminate them, but I noticed that they were distracting from the logo. So, I spent quite some time getting things re-spaced so that the vias are aligned carefully with the letters of the UNOFIED so that they look as if it were meant to be. Funny to me, really, because I'm 100% certain that that was useless/pointless work because the name's gonna change anyway. Welcome to my life.

 

via.png.b3890434c1ef205033b44849bbbedcc0.png

 

See how they align along the bar of the "A", and the "I" with a dot on top, and the "N" and "C". I just need people to notice that. That didn't happen by chance. My OCD did that.

 

263905327_ScreenShot2021-02-15at12_17_01am.thumb.png.f85bd65549c252718d3dd4af65eacbaa.png

 

The PlusCartUnified zip file contains all the Eagle files (including library), along with 7.X variants of the same.

 

 

STL.zip PlusCartUnified 2021.02.14.zip

  • Like 6
Link to comment
Share on other sites

3 hours ago, Andrew Davie said:

So, firstly I printed a STL of the board, thanks to the suggestion earlier in the thread. I exported the design as a STEP file, and then used an online converter ( https://anyconv.com/step-to-stl-converter/ ) to convert that into a STL file for 3D printing. That worked a treat, and it highlighted a couple of issues.  First, there's the size comparison; the paper version was woefully incorrect, and the STL was perfectly-sized. My first 3D print didn't really show on the print itself where the MicroSD was mounted, so I put in the two mounting holes as on the SD board. These may or may not be useful in the final version, but they're super for getting things exact in the testing.  Initially I found that the board was about 3 mm shy of the edge of the cartridge shell, so my calculations were wrong after all. I've modified that, and will print a new shell and dummy-board overnight but I'm pretty confident it will be good to go.

 

IMG20210214231556.thumb.jpg.789c28a3645afecfd4e401aade3bcd6d.jpg

just curious - did your STL not get the holes?  They look like little dimples - or is just a limitation of the 3D printing due to the small diameter of the holes?

 

When I take your BRD file into KiCAD and export STEP and then covert to STL I can see the holes

 

image.png.ff0ca4c340d50491c0f25997b59074f6.png

 

I don't have a 3D printer, but it is next on my list of things to get.

 

 

Link to comment
Share on other sites

6 hours ago, Andrew Davie said:

Now here's what it's like to have OCD...

Yeah, I know the feeling, lol!!!

 

One question, wouldn't it be good to grow the board some millimeters to have more supporting for the SD card module?

image.png.6c922da51fd9c2e5e9f929f504e5030b.png

 

6 hours ago, Andrew Davie said:

I removed one block of 2x3 header which weren't used. Removing others would be a much more complex job.

Which other header do you want to remove? Can I help you on doing that?

 

 

Link to comment
Share on other sites

3 hours ago, fdr4prez said:

just curious - did your STL not get the holes?  They look like little dimples - or is just a limitation of the 3D printing due to the small diameter of the holes?

 

When I take your BRD file into KiCAD and export STEP and then covert to STL I can see the holes

 

The ever-so-slight inaccuracy of a filament printer is enough to hide holes this size.

They are printed as shown, but the filament just needs to be half the radius "too big" on doing the tiny circumference passes that they're not seen well. In any case, they are superfluous for the most part, in this test.

 

  • Thanks 1
Link to comment
Share on other sites

20 minutes ago, Danjovic said:

Yeah, I know the feeling, lol!!!

 

One question, wouldn't it be good to grow the board some millimeters to have more supporting for the SD card module?

 

 

Which other header do you want to remove? Can I help you on doing that?

 

 

I am trying to keep the board growth to a minimum, because actual production on these things is usually priced on board area.

Particularly if you're using the $5 prototype print service. I'm not sure we're still inside that, but I think the support is sufficient now.

There's not a particular header that I want to remove -- more that @Al_Nafuur has suggested having unused pins means more soldering (2x for each pin), and so these unused pins can be removed from the board. So, any of those really. Personally I prefer them all there, but if someone wants to remove them I don't see the harm, I guess.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...