Cascade Kobayashi @cascade.moe · 15/05/2025Completed soldering of my CS4398-REV2 board, looks not bad after cleansing. 070
Cascade Kobayashi @cascade.moe · 29/04/2025@raysky-rt.bsky.social さんのイラストです！ 是自己 OC （或者說是自設）的插畫。 An illustration of my OC by RaySky. 0102
Cascade Kobayashi @cascade.moe · 07/04/2025some one screenshotted me when I was streaming having difficulties with BGA packages.jpg 有人在我直播時截圖了 被 BGA 封裝為難.jpg 050
Cascade Kobayashi @cascade.moe · 04/04/2025They are actually net colors~ You can assign colors to nets or net classes in this tab. I would prefer assigning different protocols & power rails with different colors. 100
Cascade Kobayashi @cascade.moe · 03/04/2025WIP Isn't it strange to use a STM32H7 + HS PHY, with a old DAC chip (CS4398) which is almost as old as me...? But it's interesting that, CS4398 actually lived 20 years and it's still not discontinued. 020
Cascade Kobayashi @cascade.moe · 02/04/2025WIP Price of STM32 microcontrollers are rising, the interface chip is twice more expensive than that DAC chip... Maybe I will use CS43131 later, although it's not challenging as it integrated an HP-AMP which can simplify the design. 140
Cascade Kobayashi @cascade.moe · 31/03/20253rd order low-pass filter. A simple RC filter followed by a Sallen-key filter. The latter filter is compensated to achieve flat group delay in pass band. 三階低通濾波器，一個簡單 RC 濾波器之後跟著一個 Sallen-key filter. 後者經過補償，可以在通帶中實現平緩的群延遲。 050
Cascade Kobayashi @cascade.moe · 29/03/2025DAWN PRO They may have better performance, but I can hardly identify the difference between playback devices with my ears. Test result is slightly better than on board audio, but I suppose the result is limited by the recording device & testing environment, which are quite bad. 070
Cascade Kobayashi @cascade.moe · 29/03/2025Tested my CS4398 DAC (Rev 1), its performace isn't as worse as I expected... Although THD+N is really bad (0.02% @ -5dB, 0.01% or 0.008% @ -10dB) 041
Cascade Kobayashi @cascade.moe · 27/03/2025From the experience on last CS4398 DAC project, I decided to make a new CS4398, with STM32H7 as interface and some second hand OP1612 apmlifiers. H750 is quite cheap now. It's very fast, plus bunches of peripherals that I may or may not use. 040
Cascade Kobayashi @cascade.moe · 20/03/2025沒活了，找點東西臨摹。這次是小塗真紀。 是聽講座的時候開始摸的，不過沒有怎麼優化線條，所以大概顯得很亂。 原作 Pixiv ID: 126483504 040
Cascade Kobayashi @cascade.moe · 17/03/2025Nia-chan / にあちゃん Original Character of ゆーね (x.com/ayanepuna) さん 上課的時候摸的，不過我太菜了，所以不知道身體應該怎麼畫。 然後是發現了正臉其實很難畫這一事實。 1100
Cascade Kobayashi @cascade.moe · 11/03/2025Good news: My fever reduced Bad news 1: I still have a sore throat Bad news 2: I broke my thermometer and I get some dangerous, liquid mercury 060
Cascade Kobayashi @cascade.moe · 07/03/2025Not good if there is few reference images... This time it's (probably) not shota, but this doesn't matter. (I draw unisex characters more like shota) 1100
Cascade Kobayashi @cascade.moe · 04/03/2025colored sketch without lines * Actually feel betther than outlined because I didn't draw the outline smoothly 060
Cascade Kobayashi @cascade.moe · 27/02/2025rkgk 應該不難猜出是什麼角色……！ 還需要很多練習。臨摹參考了 BOOTH 上 Rusk 的圖像，不過沒有畫上 hoodie 上的圖案。 It should be easy to guess the character...! I still need lots of practice. Tracing of image of Rusk on BOOTH, although I didn't draw patterns on the hoodie. 071
Cascade Kobayashi @cascade.moe · 21/02/2025I Added a 1 pole LPF, this reduced "ΔΣ-like noise" and overshoot. Beautiful 12kHz Triangle & Square wave. The sample rate is set to 192kHz. They become worse at 96kHz sample rate, but it's reasonable because higher harmonics (>4th) will be filtered by DAC internal filter. 040
Cascade Kobayashi @cascade.moe · 21/02/2025So, I also need a LPF to filter out these "Delta Sigma" like noises... CS4398 Datasheet specified a 2 pole LPF. Now I know why this exists in their datasheet... I kept the pads on my board so I can still achieve a 1 pole LPF, but I don't know whether it is enough... 040
Cascade Kobayashi @cascade.moe · 20/02/2025I managed to get the DAC working, although the headphone amplifier is off... There is a huge LPF on my computer onboard audio output. The waveform will completely become a sinewave after 12kHz. My CS4398 DAC peforms better at 96kHz. Pics are the waveforms of 4kHz square wave. 040
Cascade Kobayashi @cascade.moe · 17/02/2025USB 3.1 Hub Recognized! Working well with external supply. Also possible to work on host supply but need to connect external supply first (there are some issues with power path management). 070
Cascade Kobayashi @cascade.moe · 17/02/2025USB 3.0 Hub I fixed the power supply part and now USB 2.0 is working. I will test USB 3.0 later. 030
Cascade Kobayashi @cascade.moe · 16/02/2025Finished my CS4398-DAC Board after 4 hours of work. Haven't tested power supplies yet but STM32 should be working normally. 071
Cascade Kobayashi @cascade.moe · 05/02/2025Here is a diagram of the trace connections (forgive my poor drawing since I'm not home rn...!) 120
Cascade Kobayashi @cascade.moe · 05/02/2025Should be the final mass product version of my touch sensor. The board shape is optimized for direct surface mounting on glass. 150
Cascade Kobayashi @cascade.moe · 01/02/2025An copy that I drawn & scanned a few years ago... Original work: www.pixiv.net/en/artworks/... 151
Cascade Kobayashi @cascade.moe · 29/01/2025Classic match, they must be using a customized firmware. Maybe CT 7601 is actually running in slave mode? And I didn't get what does "USB surface IC" mean. Maybe it's a typo for "interface". 130
Cascade Kobayashi @cascade.moe · 09/01/2025Yet another try on USB SuperSpeed Hub... I still can't forget about high speed PCB design. This time I tried another IC and it should be able to work without burning a firmware into it. 190
Cascade Kobayashi @cascade.moe · 02/01/2025It's working! Although some pins are not soldered properly. I will fix it later. 280
Cascade Kobayashi @cascade.moe · 31/12/2024I bought some X7R 47nF MLCC because I noticed that 47nF MLCC I bought 2 weeks ago are Y5V caps. 040
Cascade Kobayashi @cascade.moe · 24/12/2024After some adjustments on this board (adding resistors on gate), I successfully blew this board. The high side MOSFET burned... I also noticed ringing on SW / Gate when it's connected with high load. So there is actually some issue in my design. Maybe the gate is too hard to drive... 070
Cascade Kobayashi @cascade.moe · 23/12/2024The inductor (as heatsink of MOSFETs) is really hot when running at 6.6A, I don't know it's okay or not... 🥵 There isn't much overshoot and ringing on SW and MOSFET gate, so other designs are still okay. Maybe I should try new designs with large areas of copper... 170
Cascade Kobayashi @cascade.moe · 22/12/2024Another trip to Macau on yesterday morning. This time we also visited Cotai~ 160
