Heyo,
Just keeping this thread updated on what's going on.
I've put 2 of these boards together now and they are both solidly reliable for use as hard disks. Speed seems limited to 600KB/s now, the 1.1MB/s reading might have been a fluke/glitch. But anyway, 600KB/s is still ridiculously fast.
I've tested the real-time clock is able to run and keep time via the backup battery which is a CR2016 coin cell (43 years worth of energy according to the datasheet). I would have used a CR2032 - because everything else does - but I wasn't super confident the holder would fit under the board, the space is only 4mm tall. A CR2025 might fit...
My main job at the moment is writing utilities to get and set the clock on the ST. Got stuck on writing DMA to the peripheral, the official docs are pretty light on info. They give an example only for read, not write. For now I'm only receiving zeros from the ST. But I'll figure it out eventually.
The whole project is on github now at https://github.com/smonson78/smonson-disk, warts n' all.
Can all registered uses please login, even just for a few minutes..
It helps build a picture where our "good traffic" is coming from..
Thanks :)
It helps build a picture where our "good traffic" is coming from..
Thanks :)
Smonson's untitled simple harddisk project
Re: Smonson's untitled simple harddisk project
(Spooky Hallowe'en noises)
Rising from the dead, I got around to building up the PCB for the last round of changes I did - I think it was around 2024 shortly after my last post - to see if I actually managed to fix the issues that I was hoping to fix. It seems to work pretty well now. Data transfer speed is about 1400KB/s which is probably now maxed out on the ST side.
Rising from the dead, I got around to building up the PCB for the last round of changes I did - I think it was around 2024 shortly after my last post - to see if I actually managed to fix the issues that I was hoping to fix. It seems to work pretty well now. Data transfer speed is about 1400KB/s which is probably now maxed out on the ST side.
- I switched over to an STM32C031 microcontroller because they can do SPI at 24MHz and are dirt cheap
- I solved a major issue with the power supply to the SD cards - they were drawing massive currents on startup and causing random glitches and resets (mainly when removing and inserting cards). To this end, a 4th regulator is dedicated just to the SD cards and supplies a low-ESR electrolytic through a current-limiting resistor to power them. This might be overkill, I'm not an engineer!
- I moved the RTC to a separate SPI bus to keep the SD card SPI bus as clean as possible
- I tried harder to adhere to a star topology for power supply
Re: Smonson's untitled simple harddisk project
Nice to see this project is still alive.
Re: Smonson's untitled simple harddisk project
1400KB/s is really great achievement,Smonson wrote: Sun Nov 01, 2026 6:20 am Data transfer speed is about 1400KB/s which is probably now maxed out on the ST side.
as far as I remember the ACSI limit is 2MB/s
Atari: V4sa TT030 Falcon030 Mega ST Mega STe 1040 STe 520 STm 800XL 65XE Jaguar Lynx I/II 7800 Portfolio / VME: ATW800/2 Mach64 Mach32 ET4000 CrazyDots PAMNet / SM146 SC1435 SM124 / Antonia2 ATSpeedC16 TF536 PAK68/3
Megafile30 Atari1050 DDDHDD XCA12 SF314 / AVGCART SUBCART SidecarTridge ACSI2STM FujiNet LynxMultiCard SkunkBoard CosmosEx SatanDisk UltraSatan USBFloppyDriveEmulator SDrive SIO2PC Eiffel
http://260ste.atari.org
Megafile30 Atari1050 DDDHDD XCA12 SF314 / AVGCART SUBCART SidecarTridge ACSI2STM FujiNet LynxMultiCard SkunkBoard CosmosEx SatanDisk UltraSatan USBFloppyDriveEmulator SDrive SIO2PC Eiffel
http://260ste.atari.org
