-
Notifications
You must be signed in to change notification settings - Fork 0
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Rev_G Wishlist #11
Comments
Thanks, Chip! |
Thanks Chip!
|
Yes to Bluetooth Audio
|
Nice to Have
Perhaps going to a PCB with more layers so we can pack everything into a smaller package. |
|
@eyuan-creare , Rev F already as the qwiic connector! (Spelling corrected. So hard to remember. Thanks Chris!) |
@eyuan-creare The quiic does not have a port in the current case design. Might be a good idea to post a design mod with a quiic port for folks to print if they want to? |
Good idea. Are the pads for a right-angle QUIIC at the edge of the board? |
Before we get too far with this, the correct name is Qwiic: "https://www.sparkfun.com/qwiic". |
Thanks for the qwiic reply, Chris! |
I knew you or Chip would correct me, so this was the most efficient path to the correct spelling ;) |
Large flat section on case for easy labeling with large font lettering |
Improve RTC coin cell battery life. (Re: #12 (comment)) |
Maybe a rechargeable coin cell? Shorter run time, but rechargeable. Hopefully find one with UN38.3 certs. |
We may have had someone damage their Tympan by applying 5V to the audio input jack (the pink jack). Perhaps we should add over-voltage on this connection? |
|
(2 to 3) Physical push buttons on the board |
@eyuan-creare , I know that you're a fan of HEX files. More of a fan than me, for sure. In my experience, the process is never as easy as one hopes, at least from a new-user perspective. When I've offered it to people, these are people who have just figured out how to compile our Examples using the Arduino IDE. That was hard for them, but now they feel a tiny bit of success. When I offer an "easier" solution that involves completely changing the process, it's given everyone pause. It seems like it's too abstract for inexperienced users who only know the comforting blue/green of the Arduino IDE. Perhaps if the HEX files used a comfortable web-based uploader or something, that would be different. But, no, you have to use the Teensy uploader (Teensy.exe). Ah, the Teensy Uploader...what a weird too-small piece of software. Also, it's impossible to find on one's computer! Here it is on mine:
Be aware that, "AppData" defaults to being hidden/protected on the default Windows installation. Yikes! Even if get over that hump, you've got to drill down 8 more directories, 3 of which are dependent upon your particular installation ("Local", "Arduino15", and "1.59.0"). That's a lot of opportunity for failure! As a completely separate criticism of HEX, I find HEX files difficult to manage. How does one keep them up-to-date with underlying source code? Currently, it has to be done manually...manually ask it to produce the HEX, manually copy it to the directory with its associated *.ino, then manually delete the build directory. Ugh. that's a recipe for being out-of-sync, for sure. Maybe we could set up an automated process, but maintaining automated processes has been difficult in the past. KXM and I tried a travis-based production release system years ago...the system didn't last but a few months before some dependency changed and broke the whole system. I didn't have the skills to fix it and those with the skills didn't really have time. Bummer! This is me being overly-negative about HEX. Compiling the examples oneself has its own problems. I know that you've had more positive experiences with HEX. Tell me more! |
Good point. I think there are two things that I am considering:
In terms of uploading hex files, we could start by including a hex whenever we release and test a new software feature (not necessarily keep all the hexes up to date with the latest rev). Just a snapshot in time when things had worked. Then when new hardware is released, upload them when we happen to test and verify a script is stable, for those sketches we commonly use. Perhaps not the time yet to put this into play. |
@eyuan-creare , you've painted a good picture, especially in drawing a difference between researchers vs developers. Continuing with this idea, I might suggest keeping the collection of hexs in their own repo. This repo being aimed at researchers / regular users rather than developers. The docs for the new repo (readme, for example) could be aimed at this kind of user..."how to use these files to easily discover the Tympan". I like that. A clear purpose, with words that help people realize that this is what they want. Additionally, using a separate repo for the hexs addresses my concern of the hexs being hard to maintain, it addressed the challenge of keeping them matching the latest Tympan_Library code base. If the hexs are in a separate repo, it isn't necessarily expected that the hexs will always perfectly aligned with the developer repo (ie, Tympan_Library). In other words, the hex repo could have its own releases, with notes as to what standard the hexs were made to. Having a hex-only repo gives up more options for keeping hexs for the diff versions of Tympan, too, if we desired. What do you think about the hexs being in their own "For Researchers" repo? From your viewpoint, does this achieve your vision better or worse? |
@chipaudette |
BAyotte has been loading Hex files directly on a Tympan Rev F, following the instructions in "https://github.com/Tympan/Docs/wiki/Program-the-Tympan-from-a-Single-Hex-File#uploading-a-hex-file-to-the-tympan-1". That involves digging into the Arduino app data directory that @chipaudette mentioned. I'd be in favor of adding instructions for how to use the standalone Teensy loader. |
Oh man! One can download the Teensy Loader directly from the web?!? That totally addresses my concern! Thanks @biomurph! |
There is an active tutorial on pjrc that uses the stand alone be loader to flash micro python... It would need some basic testing. |
The battery charger/Tympan power circuit needs some attention. Charging the battery should not take priority over powering the Tympan main functionality. From @chipaudette:
|
From @cab-creare-com
Agreed. As @biomurph has already made his own version of the Adafruit nRF52840 Express, we could use its charging / power supply as a model. https://cdn-learn.adafruit.com/assets/assets/000/068/545/original/circuitpython_nRF52840_Schematic_REV-D.png?1546364754 |
For the meantime, how about a software version of UVLO? When Tympan's analog battery voltage measurement is low, shut itself off. If reconfiguring power management, might as well consider USB-C fast charging. |
@chipaudette Yes, we can use that topology to ensure the VBUS will power the system even when charging the battery. |
It's worth reviewing recommended circuits for USB & rechargeable battery powered devices. The Adafruit nRF52840 Express design is not ideal:
|
We currently have no specific plans for making the next revision of Tympan. But, if we were to make another revision someday, what hardware changes would we like to have?
@biomurph, @eyuan-creare, @cab-creare-com, @mpu-creare, etc...if you have thoughts on useful hardware mods for the Tympan, please add a post here to this issue. Once I understand the request, I will then collect them into the list below.
Here are some changes for Rev_G where someone has expressed interest...
ELECTRONICS:
FIRMWARE:
HARDWARE:
This list is not a list of action items. It is a list of potential changes that we will consider if we ever decide to make a Rev G.
The text was updated successfully, but these errors were encountered: