Gregory's Serverunfortunately we don't have a dump of the protected mask ROM, but it seems to be somewhat similar, and the 6502 puts *every* address it accesses on the bus, so you can see it accessing special function registers and the stack, allowing you to infer quite a bit about what is going on.
|
119 comments
 well that was epic! i've completely reverse engineered this older revision. details in the repo (https://github.com/schlae/HardDriveInfo) another day, another board revision! this one has minor electrical changes from the previous one, mainly adding control over the spindle motor *power level* from the microcontroller. that way they were able to lower the average power consumption. https://github.com/schlae/HardDriveInfo/tree/main/st251/21020 this is the control board for the ST-225 hard drive. it's an earlier revision. so far it's not too dissimilar from the ST-251 but the stepper motor driver circuit is quite different. and just like that...the ST225 logic board is now in KiCad. check it out: https://github.com/schlae/HardDriveInfo @tubetime @tubetime I've never wanted to play a DOOM map this badly in my life. turns out there is another variation of the ST-225 logic board. looks like a cost reduction. OK, this one is done! amazingly, it also helped me solve a mystery on the ST-251. on this later drive, the board has a custom chip that captures data from the current track and outputs a "processed" version of it to the MCU through the PA2 GPIO pin. how does it process it? i didn't know until today. so the 20527 board in the ST-225 uses a discrete circuit for doing the same task. it has two 74123 one-shot pulse generators connected in series with each other and going out to the MCU (on PA7 because this board uses a variant). the key observation is that the pin state depends on the frequency of the signal coming in. for DC, it is normally low. for frequencies *higher* than about 2.6MHz, it also stays low. but for a particular window of frequencies (1.4 to 2.6MHz) the pin goes *high*. why is this useful? it turns out that these drives have a 1.75MHz signal recorded onto specific tracks, letting the MCU (which controls the stepper motor) know if it has gone outside of the data area. so basically, you step around, and if you see this pin go high (and stay high) then you know you're out of bounds. there is a special signal recorded at track -2. it is the same 1.75MHz but it has short bursts of twice the frequency (2F). this lets the MCU know that (1) it is at track -2, and (2) it lets it synchronize the hall effect sensors. you get multiple pulses per revolution, so you need a way to sync up the divider chain so you get an index signal that is repeatable. so from the ST-225 to the ST-251, this function got swallowed up in the drive interface chip. but at least now i know how it works. someone's dumped the MOS version of this microcontroller using a clever circuit and a test mode. i don't know if it will work with the NCR version of the chip, however. let's see what happens when i throw the switch. it's putting 10V bursts on the reset pin, so i hope nothing gets damaged. no luck, sadly. it's supposed to be clocking in a program over PORTC, but there appears to be output contention (see how the voltage levels on ch4 don't always make it to a logic high)  i'm not giving up. maybe it needs more clocks before bringing reset high. the datasheet says "at least 8" oh that's interesting, the port C output is pulsing sometimes (and the data line doesn't get contention). so the code that the exploit loads into RAM pulses the port C (all pins) with an LDA #$FF ; STA $82; KDA #$00; STA $82. the LDA is 2 cycles and the STA is 3 cycles. but the pin is high for 10 clock cycles... OH! it has a clock divider on the input. so it's actually 5 CPU cycles. i think this exploit actually is running, but not consistently. looks like the key finding is that i needed 2 more sets of clock cycles before bringing RESET up to 10V. finished the disassembly process, more or less. you can read the commented source code here: https://github.com/schlae/HardDriveInfo/blob/main/st225/firmware/80007-001_commented.txt and i learned some interesting things along the way. the drives have this weird set of resistors and driver chips tied to the stepper motor phases. it's for microstepping while in recovery mode! the idea is that you assert a control pin that puts the drive in recovery mode, and it will try to read a track 15 times while shifting the head side to side very slightly. this is apparently only implemented in the ST-01 disk controller. essentially all of the drives i've inspected do not actually have these components soldered down, so it must have been an uncommonly used feature that was removed to save cost. another fun one is the empty space for an EPROM and a latch chip. these weren't used to store drive firmware (as i thought before) but they store a giant lookup table of microstep values *for each individual track* so you could presumably trim individual drives to avoid bad areas on a platter. of course it controls the head position regardless of the physical platter, so it wouldn't really be helpful unless the drive only had a single platter in it. i've never seen an actual chip soldered in here and it was completely removed in later board revisions, so i guess they never fully implemented it.  @tubetime you got impressive knowledge from analysing these drives. It's unheard of. This will be so useful for anyone wanting to fix these drives, thank you so much for the full #retrocomputing community. time to reverse engineer the platters themselves. what secrets can we find? with this setup I can control the stepper from the Arduino without the drive's microcontroller interfering. then I can use the Saleae to digitize the data. confirmed that there are two index tracks, one at track -2 (relative to user data starting on track 0) and another on track 616 (user data ending on track 614). the D2 square wave is directly from the hall effect sensor, 2 cycles per revolution. the D1 pulses are this signal divided by 2. MCU read data is the index signal as seen by the MCU after being processed with a one-shot.  @tubetime Have you figured out yet how it syncs up the divider chain? @developing_agent yes it pulses a GPIO pin and that clears the divider flip flop. @tubetime I mean more like knowing when to do that (which 2F region should be the index pulse), or is that arbitrary and doesn't matter? @developing_agent the 2F region is once per revolution, so i think it doesn't matter as long as it is consistent. my guess is the edge between the 2F and 1.75MHz regions. @tubetime If there's only one 2F region, what do you get multiple pulses of per revolution? (and need a divider chain for?) @developing_agent you get multiple pulses per revolution off the spindle motor's hall effect sensor. you need a way to divide this down (usually with a chain of T flip flops) to get a pulse that occurs once per revolution. the disk controller uses this to place the sectors in their correct locations. it needs to be in the same spot every time otherwise the controller will get confused. @tubetime Ah, so the divider chain is driven from the spindle hall effect sensor, and the single 2F region sets which hall pulse is the zero index. @tubetime The ST-412 has an NCR 6500/1. This was also used in the Amiga keyboard controller. There's a schematic floating around from the Amiga ppl for a circuit which dumps out a 6500/1's internal ROM. I still have the 6500/1 off my dead ST-412, but haven't gotten around to making the dumper yet. Wonder if a similar circuit could be used for R6518... @cr1901 i found this: https://e4aws.silverdr.com/hacks/6500_1/ but it seems to be for the Commodore version of the 6500/1 @cr1901 also i think the st-412 (at least one version) also uses an external EPROM which has been dumped and is on bitsavers. @tubetime whatever happened to the capsoff people who used to dump arcade maskroms/maskrom microcontrollers? are they still around anywhere? |
here's where things get really interesting. i've pulled in the imagery of the early ST-251 board and overlaid my Kicad traces from the newer revision.
it's almost like a visual diff--now i can see exactly what changes were made between board versions, which means i can back out a schematic as well!