bdale, the small radius corner arcs are easy: they start at a 90 degree angle at the end of the vertical line si I'd do this: start with the vertical line draw the 2 90 degree arcs connecting to the vertical line then draw one large radius arc for the whole upper section and another for the whole lower section (you can use the propedit to set the exact radius, and you can use the vertical size specified in the drawing to tune the placement) Really stupid question... are the R345.40 arcs assumed to be referenced center to board? if you did everything as above, then you will have the small, almost-90-degree arc crossing the large arc and that's when you can use the trim action to cut off the excess part easily this will guarantee the endpoint of the two arcs will be at the intersection Igor2: given a complex board outline, is there a quick way to create a plane that has copper to a given distance from the board edge everywhere? bdale, you can copy the outline to copper nd mark the objects noexport we even have an user script that can copy the outline to every (copper) layer and update it quickly if you change the outline noexport: http://repo.hu/cgi-bin/pool.cgi?project=pcb-rnd&cmd=show&node=noexport user script: http://repo.hu/cgi-bin/edakrill.cgi?cmd=show&krill=igor2/script/pcb-rnd/on_every_layer.krill 2021.04.25 - for the LiPo charger LEDs, it might be nice to have them shine out through a hole in the end adjacent to the connector, so that if you're plugging it in it's easy to see them. If so, a right-angle SMT dual LED like the Digikey 350-2072-1-ND (about price parity with our current dual LED) could be a good choice. 2021.04.28 - included 40 PCC-SMP-CLIP on a Newark order so I'll have hold-down clips for the connectors I acquired on eBay - it turns out the board outline for the 1553BBK and 1553CBK are the same, the latter is just thicker. The 1553BYLBK is a yellow with black edges variant of the thinner one, and seems like the ideal choice for this project, looks very "speed yellow" fast car'ish. - adding BTLE module requires connecting: P1_5(CONNECTED) bt_p1_5 P2_2(SW_BTN) bt_sw_btn P2_3(WAKEUP) bt_wake_up P3_1(RSSI_IND) bt_p3_1 P3_7(CMD) bt_p3_7 RST_N bt_rst_n RXD TXD CTS RTS 2021.05.11 - the MAX6675 turns out to be expensive .. $12.68 q1. The MX31855KASA+T looks to be pin compatible and only $5.31 q1. The biggest difference appears to be that the 6675 is 12-bit ADC and 16 bits of data per reading, while the 31855 is 14-bit and sends 32 bits per reading. Let's try them! 2021.06.08 - samples of possible tire pressure sensors made by Fujikura arrived from Kim Solomon at ServoFlo. What came are 4 pieces of APB20R-001MG, and 2 pieces of AH30R-050KG-X1. The APB2 series are the parts Kim suggested for my application, have a barbed port, and these are 1MPa parts. However, right now they only have them in 5V. Should have more 3.3V parts eventually, but sent these so I can thy them out with an Arduino or something. The big problem I see is that the barb comes off orthogonal to the PCB surface, so using them would make for a "thick" case need. The AH3 series are quite similar, but smaller, and have the barbless port coming off parallel to the PCB surface. The parts available to send as samples are 5V and 500 kPa. Unfortunately, the AH3 line is down at Fujikura due to a fire at the ASIC supplier, plans call for the AH3 series to be available again in mid-2022. In both cases, these are 6-lead DIP'ish through-hole parts, with only 3 pins active (power, gnd, analog output), and are "Gauge/Positive" parts. The AH3 line is down at Fujikura due to a fire at the ASIC supplier, plans call for the AH3 series to be available again in mid-2022. If the APB2 series seems like an acceptable fit otherwise, then it could just be a crazy mechanical hack would allow the part to be mounted so the barb is parallel to the board surface somehow? Maybe a small carrier board? Much study required. 2022.10.06 - sent Kim at Servoflo a query about AH31R-001MG availability (3.3V, 0-150psi) This pressure sensor has an analog output of 0.3 to 3.3V from 0-150psi, so with a 12-bit ADC running 0 to 3.3V, the psi per q-level of the ADC is about 0.05, so we should "easily" be able to record pressure to the nearest tenth of a psi. Seems sufficient. Interestingly, the sensor also has a threshold comparator. For the price of 2(3?) resistors, I could set a threshold that I could sense the state of. That might allow knowing when a pressure is applied to make it possible to record a pressure without having to push a button on the UI? Maybe set it for something like 20 psi? When that triggers, average some number of ADC input readings, record it and beep? Vth needs to be 0.66V for 20psi, 2022.10.13 - after thinking about it, trying to machine the provided plastic endplate for the enclosure seems like a recipe for disaster. It's small, it's plastic, holding it while trying to CNC route it seems really hard. It's probably a lot easier to just design a replacement piece and 3d print it?