# @Abse @0hmx @Seve  please review:

Channel: #designreview
Source: https://discord.com/channels/1233487248129921135/1545409289034924072
Started: 2026-09-04T12:38:24.134000+00:00
Last activity: 2026-09-04T18:58:28.783000+00:00

## shibo — 2026-09-04T12:24:46.822000+00:00

<@398467566588264449> <@400684445054795796> <@757706909351411845>  please review: https://tscircuit.com/ShiboSoftwareDev/mspm0g3507-usb-c-dev-board#pcb

## Anas — 2026-09-04T12:38:26.703000+00:00

Try to use a `schematicsheet`/`schematiSection` to make the schematic readable

## shibo — 2026-09-04T13:05:06.503000+00:00

done

## 0hmx — 2026-09-04T14:36:43.224000+00:00

<@778624824875941908> if you are ordering the board lmk to i will order mine too

## shibo — 2026-09-04T14:37:21.128000+00:00

i would order just want a review from <@757706909351411845>

## Seve — 2026-09-04T14:37:56.847000+00:00

We need a pedometer, anything else im not paying attention to

## Seve — 2026-09-04T14:38:09.426000+00:00

Not today anyway

## Seve — 2026-09-04T14:38:24.617000+00:00

They said specifically in the vid they needed different ti chips integrated together

## Seve — 2026-09-04T14:38:35.701000+00:00

Anyway maybe it is good to have

## Seve — 2026-09-04T14:41:29.013000+00:00

Btw the top bottom board margins make it difficult to use with a bradboard

## Seve — 2026-09-04T14:45:24.996000+00:00

it looks like a nice board (other than the margins) but i don't think it creates the demo they're looking for rn

## Seve — 2026-09-04T15:08:03.593000+00:00

i think if you adjust the margins so it can fit into a breadboard you should try to order today

## Seve — 2026-09-04T15:08:11.128000+00:00

it is a good backup if all the pedometers fail

## shibo — 2026-09-04T15:42:04.806000+00:00

[attachment]
Attachment: Screenshot_2026-09-04_at_5.41.56_PM.png — https://community.tscircuit.com/media/1545458940731596880

## shibo — 2026-09-04T15:42:18.593000+00:00

[attachment]
Attachment: Screenshot_2026-09-04_at_5.42.14_PM.png — https://community.tscircuit.com/media/1545458998801727648

## Seve — 2026-09-04T15:48:41.748000+00:00

i'm surprised at that change lol

## Seve — 2026-09-04T15:49:05.283000+00:00

the pinheaders need to be very close to the edge of the board so they don't block the wires that you put in adjacent to them

## Seve — 2026-09-04T15:49:08.658000+00:00

i should have sent a pic

## Seve — 2026-09-04T15:49:29.174000+00:00

[attachment]
Attachment: IMG_5514.jpg — https://community.tscircuit.com/media/1545460804663386132

## Seve — 2026-09-04T15:49:59.934000+00:00

do you see how the board edge makes it difficult to insert a breadboard wire without bending it? In fact the width in general makes it a bit hard to wire tbh

## Seve — 2026-09-04T15:50:11.494000+00:00

your thin version is much better in this way

## Seve — 2026-09-04T15:50:20.412000+00:00

but i don't think it needs to be that long

## shibo — 2026-09-04T15:54:03.371000+00:00

oh, i see

## shibo — 2026-09-04T16:50:06.587000+00:00

<@757706909351411845> this is not what i intended but look what i got 😂
Attachment: Screenshot_2026-09-04_at_6.49.52_PM.png — https://community.tscircuit.com/media/1545476061029208064

## Seve — 2026-09-04T17:15:13.938000+00:00

Wild

## shibo — 2026-09-04T17:19:09.017000+00:00

<@757706909351411845> i have put too many components that it would have to be very long for this specific width

## Seve — 2026-09-04T17:52:59.156000+00:00

haha idk man

## Seve — 2026-09-04T17:53:05.973000+00:00

that board is super empty

## Seve — 2026-09-04T17:53:16.989000+00:00

but also i didn't say to change your width rofl

## Seve — 2026-09-04T17:53:19.479000+00:00

self-imposed

## shibo — 2026-09-04T17:55:33.488000+00:00

it needs to be a specific width to fit into the breadboard right?

## Seve — 2026-09-04T17:58:19.059000+00:00

it needs to match pico but my point has always been about the margin beyond the pinheaders

## Seve — 2026-09-04T17:58:32.954000+00:00

pico is your standard

## Seve — 2026-09-04T17:58:43.884000+00:00

and i never mentioned shrinking the width etc.

## Seve — 2026-09-04T17:58:58.226000+00:00

(never mentioned shrinking below pico)

## shibo — 2026-09-04T18:42:13.963000+00:00

<@757706909351411845> https://tscircuit.com/ShiboSoftwareDev/mspm0g3507-usb-c-dev-board#pcb pico width

## Seve — 2026-09-04T18:44:03.444000+00:00

lgtm ship it

## Seve — 2026-09-04T18:45:31.483000+00:00

it's not programmable i think- is there a programming jumper....

## shibo — 2026-09-04T18:46:29.691000+00:00

what is the programming jumper?

## Seve — 2026-09-04T18:46:47.153000+00:00

we just discussed in voice but 99% of the time you can't program over USB

## Seve — 2026-09-04T18:46:54.690000+00:00

especially because you're using a chip to facilitate USB

## Seve — 2026-09-04T18:47:12.602000+00:00

you have to break out SWD (SoftWare Debug) in a JST to program 99% of chips

## Seve — 2026-09-04T18:47:38.181000+00:00

you can also just ask ai "how do i program this" and if it suggests connecting to pin headers you know you messed up b/c then i have to solder 😁

## shibo — 2026-09-04T18:50:30.402000+00:00

the agent did tell me it's programmable with usb, here's it's response to your concern:

The reviewer raised a reasonable question, but his conclusion—“it’s not programmable”—is incorrect for this MCU and board.
TI explicitly documents that the MSPM0G3507:
- Ships with its ROM bootstrap loader enabled.
- Can program flash over UART.
- Uses PA10 as default BSL TX and PA11 as default BSL RX.
- Uses PA18 as an active-high BSL-invoke input.
- Enters BSL automatically when blank.
  MSPM0G3507 datasheet
Our board connects exactly those signals:
- CH340 TX → PA11/BSL RX
- PA10/BSL TX → CH340 RX
- BOOT button → PA18 high
- RESET button → NRST low
- PA18 has a normal-state pulldown
TI’s own programming documentation describes using PC software with a USB-to-UART bridge, so the reviewer’s statement that the bridge makes USB programming unlikely is backwards—the bridge is what enables it. TI BSL implementation guide
What is valid in his feedback:
- This is not native USB or Pico-style UF2.
- The CH340 cannot automatically toggle BOOT and RESET in this design; you must use the buttons manually.
- A dedicated keyed SWD connector would make debugging and recovery more convenient.
- The board exposes SWDIO, SWCLK, RESET, 3V3 and GND on the headers, but not as one convenient JST/Tag-Connect socket.
- USB BSL must be tested on the first physical board with the selected TI host software. If firmware later disables BSL or enables incompatible fast-boot settings, SWD becomes the recovery method.
So: no programming jumper is required. The BOOT button performs that function. His SWD suggestion is a reasonable convenience recommendation, but SWD—and specifically a JST connector—is not required to initially program this board over USB-C.

## shibo — 2026-09-04T18:51:53.348000+00:00

<@757706909351411845>

## Seve — 2026-09-04T18:53:55.059000+00:00

ok deal

## Seve — 2026-09-04T18:54:02.575000+00:00

let's order

## Seve — 2026-09-04T18:54:37.537000+00:00

actually

## Seve — 2026-09-04T18:54:41.771000+00:00

you should break out SWD lol

## Seve — 2026-09-04T18:55:02.631000+00:00

because of what the AI is saying, we won't be able to automatically reboot and reload etc.

## Seve — 2026-09-04T18:55:08.520000+00:00

which is a very nice feature with ai

## shibo — 2026-09-04T18:55:35.248000+00:00

ok

## shibo — 2026-09-04T18:56:35.532000+00:00

i ordered after you said let's order, and you can't cancel

## Seve — 2026-09-04T18:58:21.576000+00:00

ha ok

## Seve — 2026-09-04T18:58:28.783000+00:00

you can cancel but wtvr it's fine