+DrVenkman Posted July 4, 2014 Share Posted July 4, 2014 Testing with a known-good power supply causes the drive activity light to flash on and off. With the cover off, I can see the motor spinning up and stopping repeatedly, but the read/write head never moves at all. Googling lead mostly to other AA forum threads from years past and it seems like it might be a bad 6532 chip. Best sells them for $8. Clearly it's worth it if that's the problem. Anyone have anything else to suggest before I go ahead and order a new chip next week? Thanks in advance. Quote Link to comment Share on other sites More sharing options...
ricortes Posted July 5, 2014 Share Posted July 5, 2014 SOP is to resolder the diodes in the 12V supply. It's a good idea to check for puffs in the capacitors at the same time. What the manufacturer started doing in later models was mounting the diodes off the board an inch or so to reduce the thermal cycling but it didn't really work very well, better then nothing. If you try to replace the diodes with something a little more robust, you will find the PC board holes are too small to take anything but the types they came with. Quote Link to comment Share on other sites More sharing options...
+DrVenkman Posted July 5, 2014 Author Share Posted July 5, 2014 SOP is to resolder the diodes in the 12V supply. It's a good idea to check for puffs in the capacitors at the same time. What the manufacturer started doing in later models was mounting the diodes off the board an inch or so to reduce the thermal cycling but it didn't really work very well, better then nothing. If you try to replace the diodes with something a little more robust, you will find the PC board holes are too small to take anything but the types they came with. Why should I suspect the power supply? Honest question, not being snarky. The drive spins up just fine. The read/write head doesn't seek however. The caps are fine too - no smoke, no evidence of thermal cycling, no cracked solder. Prior threads (mostly reposts of old usenet posts) seemed to indicate the controller chip. Quote Link to comment Share on other sites More sharing options...
guus.assmann Posted July 5, 2014 Share Posted July 5, 2014 Hello, The motor starting and stopping is an error indication of the drive units OS. It means it didn't get through a number of intial tests. Can be the 6532, but also (E)Prom checksum, 6810 or FDC. See sourcecode. I'dd start with removing all ic's and reinserting them. Can you move the stepper motor by hand? It can also be an indication that the head will not step or the Zero position sensor doesn't change. (No stepping) 1490 START CLD ›F014 A2FF 1500 LDX #$FF reset stack ›F016 9A 1510 TXS ›F017 A93C 1520 LDA #$3C PORT A bit 2 - 5 output ›F019 8D8102 1530 STA DDRA ›F01C A938 1540 LDA #$38 set bits 3-5 ›F01E 8D8002 1550 STA DRA ›F021 AD8002 1560 LDA DRA read register ›F024 293C 1570 AND #$3C ›F026 C938 1580 CMP #$38 check for written value ›F028 D073 1590 BNE FAIL no, FATAL ERROR ›F02A A93D 1600 LDA #$3D PORT B bit 0, 2-5 output ›F02C 8D8302 1610 STA DDRB ›F02F A93D 1620 LDA #$3D set all bits ›F031 8D8202 1630 STA DRB ›F034 AD8202 1640 LDA DRB read register ›F037 293D 1650 AND #$3D ›F039 C93D 1660 CMP #$3D check for written value ›F03B D060 1670 BNE FAIL no, FATAL ERROR ›F03D A9D0 1680 LDA #$D0 ›F03F 8D0004 1690 STA FCNTRL force interrupt floppy controller ›F042 A215 1700 LDX #$15 ›F044 CA 1710 DEL1 DEX wait a little while ›F045 D0FD 1720 BNE DEL1 ›F047 AD0004 1730 LDA FCNTRL controller ready? ›F04A 2901 1740 AND #1 ›F04C D04F 1750 BNE FAIL no, FATAL ERROR ›F04E A955 1760 LDA #$55 test sector & track register ›F050 8D0104 1770 STA TRKREG ›F053 8D0204 1780 STA SEKREG ›F056 A21E 1790 LDX #$1E ›F058 CA 1800 DEL2 DEX some respite ›F059 D0FD 1810 BNE DEL2 ›F05B 4D0104 1820 EOR TRKREG ok? ›F05E D03D 1830 BNE FAIL no, FATAL ERROR ›F060 A955 1840 LDA #$55 ›F062 4D0204 1850 EOR SEKREG ›F065 D036 1860 BNE FAIL also for track register ›F067 A948 1870 LDA #$48 ›F069 8D0004 1880 STA FCNTRL HEAD LOAD & STEP IN ›F06C A228 1890 LDX #$28 delay ›F06E 2091F1 1900 JSR DELAY1 wait X * 71 cycles (?) ›F071 AD0004 1910 LDA FCNTRL busy? ›F074 2901 1920 AND #1 ›F076 F025 1930 BEQ FAIL no, something is rotten ›F078 A228 1940 LDX #$28 ›F07A 2091F1 1950 JSR DELAY1 wait another time ›F07D AD0004 1960 LDA FCNTRL ›F080 2901 1970 AND #1 now it must by ok ›F082 D019 1980 BNE FAIL otherwise it's wrong ›F084 A9F0 1990 LDA #$F0 get checksum of PROM ›F086 8501 2000 STA SEKBUF+1 start address ›F088 A900 2010 LDA #0 ›F08A 8500 2020 STA SEKBUF ›F08C 18 2030 CLC ›F08D A8 2040 TAY ›F08E 7100 2050 PCHECK ADC (SEKBUF),Y add PROM bytes ›F090 C8 2060 INY ›F091 D0FB 2070 BNE PCHECK ›F093 E601 2080 INC SEKBUF+1 next page ›F095 D0F7 2090 BNE PCHECK ›F097 0900 2100 ORA #0 result must be zero ›F099 8500 2110 STA SEKBUF ›F09B F001 2120 BEQ TSTOK if so, everything is fine 2130 * 2140 * fatal error, break› 2150 * ›F09D 00 2160 FAIL BRK hardware defect› 2170 * 2180 * hardware test ok, do RAM test› 2190 * ›F09E A200 2200 TSTOK LDX #0 write to RAM ›F0A0 8A 2210 RAMTST TXA ›F0A1 9500 2220 STA SEKBUF,X ›F0A3 E8 2230 INX ›F0A4 D0FA 2240 BNE RAMTST ›F0A6 8A 2250 RAMTST1 TXA read it ›F0A7 D500 2260 CMP SEKBUF,X and compare ›F0A9 D0F2 2270 BNE FAIL ›F0AB CA 2280 DEX ›F0AC D0F8 2290 BNE RAMTST1 ›F0AE 8A 2300 TXA ›F0AF 9500 2310 RAMCLR STA SEKBUF,X write another time ›F0B1 E8 2320 INX to zero out ›F0B2 D0FB 2330 BNE RAMCLR› 2340 * 2350 * for now, init motor› 2360 * ›F0B4 A203 2370 LDX #3 step motor phase 3 ›F0B6 869F 2380 STX PHASE ›F0B8 AD03F0 2390 LDA PHASE4 bitpattern of phase ›F0BB 2D8202 2400 AND DRB to PORT B ›F0BE 8D8202 2410 STA DRB ›F0C1 2080F1 2420 JSR TMOTON motor on?, otherwise start motor ›F0C4 A900 2430 LDA #0 track 0 ›F0C6 858D 2440 STA TRACK ›F0C8 20B4F1 2450 JSR SIREST 10 steps in, then restore ›F0CB AD0004 2460 LDA FCNTRL controller status ›F0CE 858F 2470 STA CSTAT to RAM ›F0D0 2980 2480 AND #$80 drive ready? ›F0D2 D003 2490 BNE DRDY yes ›F0D4 2055F1 2500 JSR IDENT otherwise, init drive, density etc› (Not 100% sure of the stepping though ) BR/ Guus 1 Quote Link to comment Share on other sites More sharing options...
+DrVenkman Posted July 5, 2014 Author Share Posted July 5, 2014 Hello, The motor starting and stopping is an error indication of the drive units OS. It means it didn't get through a number of intial tests. Can be the 6532, but also (E)Prom checksum, 6810 or FDC. See sourcecode. I'dd start with removing all ic's and reinserting them. Can you move the stepper motor by hand? It can also be an indication that the head will not step or the Zero position sensor doesn't change. (No stepping) Thanks for that, Guus! I will disassemble it again tomorrow and remove/reseat the IC's. Quote Link to comment Share on other sites More sharing options...
ricortes Posted July 6, 2014 Share Posted July 6, 2014 Long story? I'm not real clear on details after all these years so as best as I can remember. "The drive spins up just fine." That wasn't in the first post. Not that it makes a huge difference. Circa 1990, whenever Atari was liquidating their warehouses, a local surplus shop bought a pallet or two of broken 1050s. Once again, kind of foggy, I think they were selling the drives for $7.50 each. I offered to anyone looking for a drive I would buy one and ship it with a money back guarantee for $20. I ending up fixing and shipping dozens. Of course I wasn't trying to make money, just promoting Atari and trying not to lose money on the deal. Somewhere around 70-90% had the 12V supply problem I mention. My personal drives that I had bought new developed the same problem. Second problem that was most notable was track zero sensor failed but that is easy to diagnose. The rest of the faults were rare and random enough to not be memorable: Bad 6532 here, fried LM3086 there, that kind of stuff. Like I said, the manufacturer found it enough of a problem that towards the end of the manufacture, they bent the leads of the diodes into a shape that looked like an FM bow tie antenna to get some distance from the board and improve cooling.<it didn't stop the problem> It's one of those things that should be fixed, even if it isn't broken yet. As far as the symptoms, the drive activity light will still come up/blink, then the 12V supply drops below what is needed to function and the drive craps out. It may boot when cold then flake out when warm. The thermal cycling causes the solder joint to crack. 1 Quote Link to comment Share on other sites More sharing options...
Bryan Posted July 6, 2014 Share Posted July 6, 2014 Is this your only 1050? If possible I'd swap some chips with another 1050 (or maybe a 6532 from a 2600?) before ordering anything. I've seen bad 6532's and bad FDC's fairly frequently. A bad CPU usually results in a completely dead drive. Quote Link to comment Share on other sites More sharing options...
+DrVenkman Posted July 6, 2014 Author Share Posted July 6, 2014 Is this your only 1050? If possible I'd swap some chips with another 1050 (or maybe a 6532 from a 2600?) before ordering anything. I've seen bad 6532's and bad FDC's fairly frequently. A bad CPU usually results in a completely dead drive. I have two others that work; I'm trying to avoid turning this one into a parts-donor if I can avoid it. So, today I disassembled it, pulled the socketed IC's, cleaned the contacts and reseated them. A couple (especially the 6532!) were seriously difficult to get out. Anyway, no change: it powers up, drive spins up and the activity light illuminates, the drive spins down and the light goes out. The read/write head never moves. Interestingly, it continues to do the same thing even with the drive latch is open. Ideas for chips to buy/replace? I'm probably gonna put in an order with Best for some miscellaneous bits and bobs in a couple weeks anyway. Quote Link to comment Share on other sites More sharing options...
ricortes Posted July 6, 2014 Share Posted July 6, 2014 (edited) Does it matter where the R/W head is initially positioned? Specifically if with the power off you move the R/W head to the inner tracks, does it ever step back out to track 0 when power is applied? Edited July 6, 2014 by ricortes Quote Link to comment Share on other sites More sharing options...
Bryan Posted July 6, 2014 Share Posted July 6, 2014 I have two others that work; I'm trying to avoid turning this one into a parts-donor if I can avoid it. So, today I disassembled it, pulled the socketed IC's, cleaned the contacts and reseated them. A couple (especially the 6532!) were seriously difficult to get out. Anyway, no change: it powers up, drive spins up and the activity light illuminates, the drive spins down and the light goes out. The read/write head never moves. Interestingly, it continues to do the same thing even with the drive latch is open. Ideas for chips to buy/replace? I'm probably gonna put in an order with Best for some miscellaneous bits and bobs in a couple weeks anyway. Yeah, the IC's can get stuck in the sockets after a while. Just work them out gently. I'd still try swapping the FDC and the 6532 with a working drive (if you haven't already). The door latch doesn't mean anything if the drive is in error mode. It will simply ignore everything and turn the motor on and off. Quote Link to comment Share on other sites More sharing options...
+DrVenkman Posted July 7, 2014 Author Share Posted July 7, 2014 I'd still try swapping the FDC and the 6532 with a working drive (if you haven't already). Yeah, I might do that later in the week before I order any new parts, just to try and narrow down what might be wrong with it. I hate to disassemble one of my working examples, simply for fear of damaging something, lol! At least one of my working 1050's is a Tandon but I don't know about the other - I've never taken it out of the case. I might start by just swapping mechanisms between the working and non-working Tandons to make sure that one that isn't currently operable doesn't have anything ELSE wrong with it, mechanically. Does that make sense? Quote Link to comment Share on other sites More sharing options...
+DrVenkman Posted July 10, 2014 Author Share Posted July 10, 2014 Follow-up: Tonight I swapped the mechanisms from both drives (they are both Tandon mechanisms). The mechanism from the previously-dead drive now works like a charm - reads, writes and formats disks just fine. Conversely, the other drive mechanism (which I already know to be good) doesn't work inserted into the other drive. Now I just need to figure out which chips to replace from that non-functional body to have three functional drives. I'd really like to have a spare, or one I can use as swap-bait for something else sometime. Quote Link to comment Share on other sites More sharing options...
Bryan Posted July 11, 2014 Share Posted July 11, 2014 First swap the 6532 and if that doesn't do it, try the WD2793/97. I bet it works after swapping one of those, but if not try the rest. One thing to remember, some 1050's have a mask ROM and some have an EPROM. There are white jumpers that are configured differently for each type, so if they're different don't try to swap them. Quote Link to comment Share on other sites More sharing options...
+DrVenkman Posted July 12, 2014 Author Share Posted July 12, 2014 First swap the 6532 and if that doesn't do it, try the WD2793/97. I bet it works after swapping one of those, but if not try the rest. One thing to remember, some 1050's have a mask ROM and some have an EPROM. There are white jumpers that are configured differently for each type, so if they're different don't try to swap them. Thank you for those pointers. Can you explain or point me to a source that explains the difference between the EPROM and ROM versions of the chips? I just today downloaded a copy of the 1050 Field Service Manual but I haven't had time to read through it yet. Quote Link to comment Share on other sites More sharing options...
1050 Posted July 12, 2014 Share Posted July 12, 2014 SpartaDOS construction set has jumper info for 1050 maskable ROM vs standard eprom set up in the section on installing the US douber upgrade for the 1050. Available in this thread as text file, pdf in other threads - just don't have a handy link for that one though. http://atariage.com/forums/topic/226714-spartados-32d-manual-or-help/ Quote Link to comment Share on other sites More sharing options...
Bryan Posted July 12, 2014 Share Posted July 12, 2014 The EPROM will be a ceramic EPROM with a paper label. The mask ROM is a normal plastic DIP. Quote Link to comment Share on other sites More sharing options...
+DrVenkman Posted July 12, 2014 Author Share Posted July 12, 2014 1050 and Bryan, thank you for those tips. These are the IC's of the troublesome unit. It looks like it's the EEPROM version. Do you think I need to replace that chip as well? Or is it even worth it, if swapping the 6532 and/or WD2793/97 don't fix the problem? Quote Link to comment Share on other sites More sharing options...
re-atari Posted July 12, 2014 Share Posted July 12, 2014 1050 and Bryan, thank you for those tips. These are the IC's of the troublesome unit. It looks like it's the EEPROM version. Do you think I need to replace that chip as well? Or is it even worth it, if swapping the 6532 and/or WD2793/97 don't fix the problem? IMG_2557.JPG It indeed is an eprom equipped 1050, so it must be an early drive. It's unlikely, though, that the eprom is causing the error you encounter. A defective rom/eprom would result in other symptoms. I'd leave it in place, and start by swapping the socketed IC's, 6532 first, then the 2793. My bet is the 6532 is acting up. I have seen a 1050 with exactly the same error before, replacing the 6532 cured it. Still no idea why it went bust. re-atari Quote Link to comment Share on other sites More sharing options...
Bryan Posted July 12, 2014 Share Posted July 12, 2014 1050 and Bryan, thank you for those tips. These are the IC's of the troublesome unit. It looks like it's the EEPROM version. Do you think I need to replace that chip as well? Or is it even worth it, if swapping the 6532 and/or WD2793/97 don't fix the problem? IMG_2557.JPG I can't predict what the problem will be, I can only tell you what it's likely to be. It's not hard to verify the chips so I don't know why you're waiting. It's cycling the motor so the firmware program is running. This means the 6507 is probably okay (since it's rare for one to fail in a way that allows execution of code). The EPROM is probably also okay, although it could have some corruption. Worry about that after you've tried swapping the other chips which are known to be the problem in most cases. There's a chance other drive you open for parts will have the EPROM and make it easy for you anyway. Try the 6532 first. Quote Link to comment Share on other sites More sharing options...
+DrVenkman Posted July 12, 2014 Author Share Posted July 12, 2014 I can't predict what the problem will be, I can only tell you what it's likely to be. It's not hard to verify the chips so I don't know why you're waiting. It's cycling the motor so the firmware program is running. This means the 6507 is probably okay (since it's rare for one to fail in a way that allows execution of code). The EPROM is probably also okay, although it could have some corruption. Worry about that after you've tried swapping the other chips which are known to be the problem in most cases. There's a chance other drive you open for parts will have the EPROM and make it easy for you anyway. Try the 6532 first. Well, basically I'm waiting for a couple of reasons: first is my Inner Neurotic Nerd - I have two working units and I'm afraid to mess with them too terribly much at risk of angering the Capricious Gods of Obsolete Technology. One wrong slip of a screwdriver or chip-puller and bam! My drive will be smitten to the nether regions of hell. The second reason I'm waiting is that I've got an SIO2PC in work and want it in my hands and working before I start pulling chips out of the guts a working drive. Anyway, I really do appreciate you guys giving me this info. I've owned these drives for probably 8 - 10 years but never even tested them. Prior to that, the last one I owned was around 1986 or 1987, and I certainly never took it apart to poke around. I'm gonna order a bunch of bits and pieces from Best in the next week or two anyway, and adding in a couple of chips won't be a big deal - they may in fact help put me over the $20 minimum order. Once that SIO2PC get working, I'll swap some chips and figure out what's what. Quote Link to comment Share on other sites More sharing options...
sup8pdct Posted July 13, 2014 Share Posted July 13, 2014 Be careful with the 2793 or 2797. if using a 2797, i think one leg needs to hang out of the socket. Cannot remember which one tho. James Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.