Thanks to Old_Grey and vanepp.
I downloaded adafruit library and there are indeed some hats. But they all are ready and not blank.
I am not yet professional enough to create with that my own blank PCB-shape.
I contacted steelgoose and hope for a positive Reply.
In the meantime I wait for the postwoman to bring my very first PCBs created with Fritzing 
)
Ole
↧
Blank RPi HAT PCB
↧
Blank RPi HAT PCB
(post withdrawn by author, will be automatically deleted in 24 hours unless flagged)
↧
↧
First PCB for an own wordclock - any obvious mistakes?
While not wrong in the sense they will work fine, the vias can be eliminated with some better routing. Here I eliminated the pair on your resistor by doing the routing a bit better. It may be worthwhile to do the others as well.
Lange PCB 3.fzz (21.8 KB)
Peter
↧
SeeedStudio Relay shield V2.x
Hi, finally got it done. The schematic svg was messed up big time. Not sure what happened but I think it's good now. Take a look and let me know if there are any more problems.
SeeedStudio Relay Shield 2.1.fzpz (56.5 KB)
Thanks,
Jeffery
↧
SeeedStudio Relay shield V2.x
Still some problems. Schematic is I think at 90dpi (because fritzing aligns it to the grid correctly) but it loads in Inkscape .9.2.1 as too small (because it thinks px scale is 96dpi I expect). This is happening because the view box height / width are in px. The solution is to (assuming it loads in your editor at the proper scale) change the view box height and width parameters to in or mm. To do that edit->select all then change the tool bar to in or mm and copy the height and width parameters (in in or mm) from the tool bar in to the height and width parameters adding a trailing in or mm to the value in the first element of the drawing using xml editor. When you press set it will rescale the drawing to the real world coords which avoids the px confusion. For reasons I don't understand your terminalId definitions are being ignored (that is why the schematic lines in the sketch below terminate in the middle of the pin). That said the correct terminalId values appear in both the fzp file and the schematic svg. Ah! just saw it, your terminal ids are mostly paths (which don't work as terminalIds. The couple that work are polygons which do work although I prefer a rectangle with height and width at .01 in. I don't think fritzing can calculate the center point of a path to know where the center should be so it defaults to the center of the pin (which isn't what you want).
PCB has a couple of problems: the holes for the pads are .035 (IC size) they should probably be .038 (.1 in header post size). The board outline (which I think is one of the polygons) has a fill value set. It probably wants fill:none and a stroke width (although I don't know how to do that on a polygon) to draw an outline around the board. As it stands the entire silkscreen layer is one color (the fill from the polygon) and the lettering is wiped out. Isn't parts creation fun 
Below is a test sketch that shows the pin alignment problems in schematic (and an easy way to test that the pins all have correct terminalId defiintions.)
test.fzz (75.4 KB)
and this is the output from the gerbv gerber viewer of the top silkscreen layer
you can see the correct outlines of the connectors I added, but all the text on the board is washed out by the board outlinel
Peter
↧
↧
Homemade LED Driver Causing a Ruckus
Thanks for your inputs, much appreciated,
Not that I know of @Old_Grey, had a quick squizz at the data sheet could not find anything.
@vanepp I placed the capacitors as suggested to no avail. I think it does not work because they don't have time to charge up or they are not big enought!
What I did discover is that if i moved the LED outputs from pins 2 and 16 to pins 0 and 16... it works!
Holding down the reset button i took the volltage from pins 0, 16 and 2(the problem pin) to ground and got:
3.1v pin 0
3.2v pin 16
3.3v pin 2
would this have anything to do with it? I would think not as @vanepp said the voltage does not matter it is the current that is amplified by a tip122g darlingotn.
SO... maybe it is the type of pin (pin 2) that is causing it? if you look at the pinout diagram for thr huzzah feather it labels pin 2 as TXD1, would this have anything to do with it?
.
.
OR
maybe it is has something to do this bit of info from the adafrui website this, found here.
↧
SeeedStudio Relay shield V2.x
Wow, I use Illustrator, I tried Inkscape and It seemed less intuitive than Illustrator. I couldn't even get any layers to show up when I opened the SVG file. I may try it again.
Do you think inkscape is better than illustrator for these SVG files?
Seems they both have problems from what I've read though not sure how to change the view port in Illustrator unless Illustrator calls it art board then it is already in mm so I'm at a loss on that one.
I'll read over your comments carefully and see what I can do to fix the issues. I did use the eagle2fritzing program and did very little to change the PCB layout as I don't use it but I'll see what I can do.
Thanks,
Jeffery
↧
Homemade LED Driver Causing a Ruckus
I think it is more likely that bypassing isn't your problem. If it was the bypass caps, it should also happen after boot not just at power up. As we agreed earlier the bypass caps in the power supply and the cpu should be enough
.
This seems the likely problem, for whatever reason pin 2 objects to having the transistor on it at boot and hangs the boot process somehow. If you don't need it to be on that specific pin for some reason moving it would be a good solution. I still don't see what it would object about having a transistor on a pin that can drive a led, but it seems a likely cause.
That is fairly normal and not a problem. The transistor base junction will always be two diode drops (about 1.2 to 1.4 volts) because it is a darlington transistor (a normal transistor is .6 to .7 volts at the base). So as long as you have more than 1.4 volts present and a sufficiently small series resistor to provide current the transistor will turn on without problem.
It seems this is probably the problem (although I still don't see why). There isn't much information either on that page or in the full data sheet (it may depend on the firmware that is loaded) to say what "detect boot mode" means. I'd expect that if you pull it down something boot related does or doesn't happen but it isn't clear what. Pin 16 may also be a problem as they say it should be connected to reset (although ti also appears to be a gpio pin) as it causes wakeup from deep sleep. If you aren't using the spi ports you may be batter to use pins 12 13 or 14 and avoid any of the boot related pins. I think what is happening is that the drive needed for the transistors is affecting the level of on the pin at boot up and causing the cpu to drop in to an unexpected mode (which is poorly documented which is unusual for Adafruit!). That would explain why it works when the leds are connected after boot occurs.
Peter
↧
Tree of Ténéré 175,000 LEDs
thanks for sharing it with us
i like it!!!
do you have a link / information on the 'making off' & used technics side of the story?
↧
↧
SeeedStudio Relay shield V2.x
Inkscape's main feature is that it is open source and thus free , I use it for that reason and because a lot of our folks use it but which ever one you are comfortable with should be fine though. I don't know how to set the parameters in Illustrator, maybe someone else will though. Here is the xml that you want to change (from editing your schematic xml file) in case that tells you what to change:
<?xml version="1.0" encoding="utf-8"?>
<svg version="1.1" id="Layer_1" xmlns:xml="http://www.w3.org/XML/1998/namespace"
xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" x="0px" y="0px" width="108px" height="165.6px"
viewBox="0 0 108 165.6" enable-background="new 0 0 108 165.6" xml:space="preserve">
what you want to do is replace the width="108px" and height="165.6px" with the equivelent in in or mm. Then the scale is set without fritzing needing to guess at how many px per inch is being used. It isn't vital because Fritzing appears to be guessing correctly as the part is properly scaled, only Inkscape is having problems. It looks like you have an older illustrator that is using 72dpi as px as changing the height / width in Inkscape to height 2.3in width 1.5in (calculated by dividing the px values by 72) looks to fix up the scaling in Inkscape. The main issue is that terminals defined as a path don't seem to work, so replacing them with either a polygon (such as connector12terminal) or a rectangle will fix that. Setting fill to none with a stroke width set to make an outline around which ever polygon is making the board shape should fix the silkscreen problem (not that I know how to do that in Inkscape let alone Illustrator though ) one of our more experienced board makers may know how to do that though. In Inkscape the hole size is set by stroke width and hole diameter. The current values (in Inkscape) are diameter .065in increasing that to .068 will make the hole .038, but it (at least in Inkscape) will also move the x y coords a bit and misalign the board. I usually use the xml editor to figure out what the new radius should be and manually change that (which keeps the x/y coords constant in Inkscape).
edit: A bit of experimenting in the pcb svg indicates that setting fill=none in the polygons and path in group silkscreen XMLID_1_ g6 and g10 and path13 fixes up the silkscreen problem.
Peter
↧
Blank RPi HAT PCB
Thank you, Old_Grey !!!
This is great support of newbies like me. I really appriciate your work and support.
And yes, this is what I was looking for. Wohooow!!
And on top my exploratioy spirit is awake now. I will try to better understand of how to do these shapes by myself.
Inkscape will be the next step for me I guess.
Ole
PS: my first PCBs arrived today from AISLER. It is a bit like Christmas (at least the first time )
↧
Tree of Ténéré 175,000 LEDs
Unfortunately like many opensource projects it seems to be fragmented all over the place. The second video I posted has a general overview of the design process and the link at the bottom of my original post to their home page has some information as well. The software you can find on github. The one place I found had the most information is their crowd funding page.
Some great pictures of the construction and the hand sculpted bark.
↧
Blank RPi HAT PCB
Make sure you throughly check it with a gerber viewer - Gerbv is free - before you order stuff or you could be wasting $.
↧
↧
Homemade LED Driver Causing a Ruckus
It is bizzare! might place it in the adafruit forum to see what they think.
Thanks for the help!
↧
Homemade LED Driver Causing a Ruckus
That's probably a good bet. I expect someone there can tell you what the pin does during boot (I expect it is documented somewhere that we haven't found yet). I'd guess that the transistor is dragging the level of the pin down and making the micro think the pin is low.
Peter
↧
Help for ESP8266
Registrering a proper ECG of a moving person is challenging. You mostly get motion artifact and wandering baseline https://www.aclsmedicaltraining.com/blog/guide-to-understanding-ecg-artifact/ .
Artifacts can be reduced using a proper filtration. Have a look at AD8232 datasheet https://cdn.sparkfun.com/datasheets/Sensors/Biometric/AD8232.pdf page 24 and 25. It requires a bit tinkering to change SMD-resistors and capasitors on the breadbord. The original breadboard is supposed and suitable to record a resting ECG. It means that when recording, the only moving muscle is your heart.
A heart rate can also be measured by pulse sensor or ready made heart rate monitor https://www.adafruit.com/product/1077.
↧
Very positive experience with Fritzing
I have been using Fritzing for a couple of months, now. And I've taken several boards all the way to fab.
I have blogged about my experiences here:
↧
↧
Split internal elements in an array
Did anyone figure out how to split internal elements for parts that are in one part.
I want to use the op-amp symbol in a twin op-amp LM358, but I want to split it so that the 2 internal elements are in different places in the SCH.
If you look at the pic there is an single op-amp that is split with 1 element above and 1 below the MEGA2560 IC.
There are also lots of resistor arrays that I want to split.
↧
Split internal elements in an array
Yep, you are looking for schematic subparts in the part file format document. I think you need to edit the fzp file I don't think (but of course may also be wrong!) that parts editor will do it. There are a bunch of posts from 6 months or so ago about the problems doing this. Buses for instance won't work with it. A number of parts in core use it which will give you an example to work from the adtl082 is an dual op amp that does this for instance. The 4001 is a digital IC with subparts and there a bunch more in the same series. I did it (not all that successfully due to no buses and performance problems with many subparts) for brad's project board so I can provide advise if needed. Its not all that hard just as usual somewhat obscure and poorly documented.
Peter
↧
Split internal elements in an array
I installed EagleCAD yesterday, and boy is FZ is far behind. I still like FZ more, because to me Eagle looks very messy and complicated. FZ just look cleaner.
I was going to redo the Eagle Arduino MEGA2560 SCH to FZ - yes I'm insane - and keep it sort of similar, but without those features it will be vastly different.
It's already becoming a PITA because I'm practically making new part for everything, so maybe I should make the thing totally different.
↧