1541 DIAGNOSTICS Some symptoms and solutions for a sick disk drive. latest updates or corrections: 2-21-2014 WHEN THE DRIVE IS WORKING PROPERLY To properly diagnose a potential problem, you have to know exactly how the drive should respond when it's working correctly. Before you start, if the drive has sat unused for a long time, you might want to check a few things before you power it up. On an ALPS (push-down door) mechanism, the head assembly should move back and forth (drive off) easily with your finger. If it is hard to move, the rails should be cleaned. ALPS drives seem to suffer more from sticky rails than Newtronics mechanisms. The stepper assembly in a Newtronics (twist-down door) mechanism is normally harder to move with a finger, and it is less likely to suffer from sticky rails. Clean the head while you have the drive open. Here's how, after removing the drive top cover (4 screws in the case bottom) and the metal shield (two screws) over the PC board. To clean the head, use a Q-tip cotton swab moistened with alcohol to gently wipe the head surface after raising the pressure pad out of the way. If the head clog is very stubborn, you can use a pencil eraser (a very fine but gentle abrasive) to rub the head clean. I'll include a closeup photo, before and after cleaning. Make sure the head is dry before inserting a disk. Never spray anything into a disk drive but compressed air can be used to blow the dust out. Cleaning the stepper rails will be described later. DRIVE POWER UP: Green power LED comes on and stays on, red activity LED comes on and spindle (which turns the disk) motor runs for about two seconds, then red LED goes out and spindle motor stops. There is no stepper (which moves the head assembly) motor activity at this time. COMPUTER POWER UP (OR RESET): Drive red LED should come on and spindle motor should start, then LED goes off and motor stops within two seconds. READ DIRECTORY: Insert a known good disk and type: LOAD"$",8 and hit the RETURN key. The disk should spin and the stepper should move the head to track 18 and read the directory. The screen will show: SEARCHING FOR $. If it finds it, the screen will display READY. Then you can type LIST to see the contents of the disk. Note: some program disks will not have a directory you can list. If the disk read fails for any reason (drive door open, unformatted disk, bad chips in the drive, clogged R/W head, etc.), the drive stepper will gently "burp" four times (moves slightly back and forth), then the red LED will flash and an error message: FILE NOT FOUND will appear on the screen. If you read the disk error channel, it will display: 74, DRIVE NOT READY,00, 00. INITIALIZE: This command from the computer should move the head from wherever it was to track 18 (directory) and the disk should spin. The head will not move (but the spindle motor will turn) if it is already over track 18. If there is no disk in the drive, or you insert an unformatted disk, or if the drive door is open, INITIALIZE should cause the spindle motor to run and the head to seek track 18 (directory) anyway. When it tries and fails, it will pull the head back to track zero and "chatter" as it hits the head stop, then advance to where track 18 should have been. The red LED will flash because of the drive read error. No error message will be shown on the screen, but if you read the disk error channel, it will display: 21,READ ERROR,18,00. More on this later... FORMAT OR DISK "NEW": When you format a disk, the spindle motor will turn and the red light will come on. The drive will pull the head back to track zero and "chatter", then the stepper will advance to each track as it writes from track 1 to track 35. With a "stock" drive, the steps are about two seconds apart. When it finishes the format (about 1 minute 25 seconds on a stock drive), the head will return to track 18 (the directory). If the format fails, the red activity LED will flash, but there will be no error message on the screen. Reading the drive error channel will display: 21,READ ERROR,00,00. Format failures can be caused by a write protect (disk tab on), drive door open, bad disk, bad or clogged head, or bad chips in the drive, in roughly that order of likelyhood. The format will attempt to write to track 1, then do a read, and if that read fails, the format will terminate, and the head will not move from track 1. If it advances a few tracks and then stops or takes a long time to format, suspect a bad disk or a partial head clog. If a drive will read OK but fails to format a disk, check the head, UC1, UC2, UA1, and UD2. Swap out drive mechanics to verify the head is bad. It may test good with an ohmmeter and read disks OK, but if defective internally, may fail to format a disk. Note that all wires of the head should measure continuity (low resistance) to each other. If any line is open (pin #5 seems to be most common to open up in the Newtronics drive), the head is bad. Disconnect the plug from the drive to do the resistance tests and make sure you get the plug back on the connector the same way it came off. To do the resistance check, it's helpful to use straight pins to touch the connectors on the side of the plug shell. The twist-door Newtronics mechanisms seem to suffer a very high failure rate from an open R/W head, unlike the push-down ALPS ones. I don't know why but I speculate that Newtronics (Mitsumi) didn't seal the heads well enough and moisture seeps in and corrodes the wiring. Storing a drive in a humid environment is therefore to be avoided. An open head is not repairable. I usually just swap out the entire mechanism. NOTE: The two mechanisms are interchangeable. Each have four connectors that match the controller board. Just make sure all are inserted as the originals. The headers are not keyed and it's possible to get the plugs inserted backwards! I put a line from a magic marker on the pin 1 end of each plug so I'm sure to get them all back correctly. I did once substitute an ALPS head element into a Newtronics assembly just to see if it would work... it did. However, that required the epoxy to be chipped away from both, without breaking them, and shims were needed to raise the height of the new head to match the old. I wouldn't do it again. It was a lot of work! For write-protect problems, check UC1, UC2, UA1 and, of course, the sensor. Look for a stuck write protect tab that has fallen off a disk. If the drive is dusty inside, there could be some dirt inside the sensor. The spacing is tight, so a quick shot of "canned air" will take care of it. Note: if cleaning the head with solvent doesn't help, try rubbing it with a pencil eraser. It is a mild abrasive and will not hurt it. Try the read again. As mentioned above, it is sometimes helpful to read the disk drive error channel when the drive red light is flashing. Here is a small BASIC program to do that. It reads the channel, displays the error message, and turns the red activity LED off. 10 OPEN 15,8,15 20 INPUT#15,EN,EM$,ET,ES 30 PRINT EN,EM$,ET,ES 40 CLOSE 15 This program and all of the possible drive error messages are listed in the back of the disk drive operators manual. Note that JiffyDOS provides an easy way to check the error channel... just press the @ key, then hit RETURN. DRIVE DEAD... OR NEARLY Lets take it from the top. Does the drive start up properly when turned on? If the power light (green LED) doesn't come on, or is dim or flickers, you probably have a power supply problem... the 5 volt line is bad. That usually results in a spindle motor that runs continuously with red LED off. Check the bridge rectifier (CR3 in early version drives with PCB# 1540050, and CR1 in later version drives with PCB# 251830) and the 5 volt regulator VR2. Note: if the regulated 12 volt supply fails, the motors will not run at all. That's a rare failure but I've seen a few shorted tantalum caps (C15, 10uF 25V) pull that line down. At power up (without the computer connected) if the red activity LED stays on and the motor runs continuously, it means that the drive failed to complete its startup sequence. The most common causes are a bad DOS ROM UB4 (901229-xx) or failing 5V bridge rectifier, but other bad chips can produce those symptoms. The easiest place to check for correct voltages in and out of the power supply regulators is at the diodes CR2 and CR4 located near the two rectifiers. The anodes of those diodes are connected to the outputs of the +5V and +12V regulators. The cathodes (designated by a line or stripe at one end) show the unregulated source from the rectifiers that feed the regulators... about 10V for the 5V source and 20V for the 12V source. With drive startup problems, some chips to check are: UC4 (6502 MPU) and UC2 (6522 VIA). The smaller "glue logic" chips are pretty rugged, but do sometimes fail. Check UA1 (74LS14) and UD2 (7407)... they have also been known to cause those symptoms. DRIVE POWERS UP OK, BUT WILL NOT LOAD THE DIRECTORY OR PROGRAMS When the computer is turned on, the reset signal from the computer should cause the drive (and other peripherals like the printer) to reset. The red LED and spindle motor should come on and go off within a few seconds. If that doesn't happen, try a substitute serial cable. If that's OK, suspect the interface chips in the computer or VIA chip in the drive. If the computer resets other peripherals, it's probably OK. Note that a drive may stay in reset (red LED on and spindle turning) if connected to a computer that is turned off. That's normal. Disconnect or turn on the computer to check. If the computer can't "see" the drive on the serial bus, (serial cable unplugged or drive turned off, for example) you will immediately get an error message: "DEVICE NOT PRESENT" when you try a LOAD command, and the red LED will start flashing. The default (factory setting) for a 1541 is device #8. If your drive is hardware modified as device 9 for example, and you try to read the directory with LOAD"$",8 you will get that error message but LOAD"$",9 will work. If the VIA (UC3) 6522 interface chip in the drive is bad, the drive will likewise be "invisible" to the computer and you'll get "DEVICE NOT PRESENT". If you get "SEARCHING FOR" and nothing else happens, check ICs UB1 (7406) and UA1 (74LS14). These two chips carry data to and from the VIA chip. When one of them fails, if you try to load the directory or a program, the computer will display that error message until you turn it off or reset it. 74,DRIVE NOT READY,00 00 from the drive error channel indicates the computer can "see" the drive on the serial bus, but 1. there is no disk in it, 2. the disk is not formatted, 3. the drive door is not closed, 4. the read/write head is completely clogged or 5. the drive has an electrical problem. With any of these problems, the drive head can't find -any- data. The drive will respond by flashing its red activity LED and may step the head back and forth slightly looking for data. Note that this takes only a second or two before the drive "gives up" and the spindle stops. A partially clogged R/W head may allow the drive to see data but still not read it properly. Other similar false reads would include a corrupted disk or trying to load the directory of the reverse side of a 1571 formatted disk. In any case, if the drive can see data but can't read it properly, it takes some time "hunting" before it gives up trying... more time than if it doesn't see any data. That's an important clue. You may hear the head assembly "chatter" as it bangs against the head stop seeking track zero... a normal process if disk errors are encountered. DRIVE SUDDENLY WILL NOT READ DISKS OR LOAD PROGRAMS One quirk of the 1541 is the "drive lost" symptom. Normally, the drive will "park" the head over the directory track (18). If the head, for some reason, stops past the directory track, an INITIALIZE command from the computer will return it to track zero and it should then work normally. Note: turning the drive off and back on again will -not- reset it if that's the problem! Some disk errors can do that to a drive and make it look "dead", as can exiting incorrectly from some programs by just turning off the computer. So, if the computer can access the drive, but you can't load even the directory of a known good disk, try the INITIALIZE command (with or without a disk inserted), then try reading A disk again. To INITIALIZE the drive: OPEN15,8,15 PRINT#15,"I" CLOSE15 As an alternative to Initializing, you could try formatting a disk. That will also return the head to track zero. Lastly, if you insert the original CBM transit card shipped with the drive (drive turned off) it will push the head back to track zero. Inserting a disk will not do it. The transit card has a tab on the front (the longer of the two tabs if there are two) that moves the head back. Don't have your transit card? With the top cover off and metal shield (if your drive has one) removed, you can push the head assembly back with your finger. The drive must be turned off, of course, or the head assembly may not move. The transit card is preferred to Initialize or Format as you don't have to turn off the computer (just the drive) so you don't lose a program in memory. One important note to make here is to avoid turning the drive on or off with a disk inserted. Whether or not the disk is write protected, it's a risk and sooner or later it will corrupt a disk and make it unreadable! PROGRAMS FAIL TO LOAD COMPLETELY OR COMPUTER LOCKS UP If your computer setup or components have been moved recently, take note: drives or cables too close to a TV or monitor can sometimes pick up "noise" interference from the high voltage circuits inside the monitor which can garble the data. Move the drive and cables at least a foot away from the monitor and try it again. If that helps, move the drive to the other side of the monitor and keep the cables as far away as possible. If you have re-initialized the drive and it still doesn't work (can't read a disk), it may be out of alignment. Keep in mind that actual alignment problems are not as common as once thought. Try formatting a disk and see if it can read the (empty) directory of that disk. If it can't, clean the head and try it again. If it can, but can't read other disks, then misalignment is a good possibility. There is one other thing you should check first: see if the head assembly rails are sticky, especially on a drive that has sat unused for a long time. With power off, the head assembly should slide back and forth easily and the tension band should be taut. It's kept in place with a spring but old grease can cause it to stick too far forward. If the rails seem sticky (experience helps to know the difference between good and bad), the rails can be cleaned with a drop of WD-40 on a Q-tip cotton swab. WD-40 is not a lubricant but it works well as a solvent to clean up the rails. Spray a small amount of solvent on a Q-tip (cotton swab) and then wipe the rails, work the head assembly back and forth, clean the rails again, and repeat as necessary until no more residue is found. I would avoid putting oil on the rails. It will work for a time, but it eventually picks up dirt and the rails will get sticky again. I run them "dry" or one can use a small amount of graphite based or silicone lube. Avoid the use of sprays directly into the drive. The spray goes everywhere and can contaminate the next disk you insert in the drive. SOME ODD FAILURES If you smell or see smoke from your drive, of course turn it off quickly. OK, where did it come from? Look at all the tiny electrolytic capacitors (known in the trade as Tantalum) to see if any are burned. When new, they are orange or dark blue in color and shaped like a teardrop. Those capacitors have a history of catastrophic failure and can pop like a firecracker or just make a lot of smoke when they fail. I've seen them short out and drag the +12V line down so the motors didn't work. The regulator IC's have an internal protective feature called "fold-back current limiting, so they will not immediately burn out from an overload like that. If the +12V line is lower than normal, suspect a shorted C15, a 10uF 25V tantalum on the front of the board. Another odd failure I've seen is the spindle motor turning all the time, but the drive still works perfectly. The cause is a bad UD2, IC number 7407 or 7417 (those are interchangeable parts). When just part of that IC fails, it keeps the spindle motor running, but with no other symptoms... strange as it seems. You have to listen closely to hear it, especially with no disk inserted. DRIVE ALIGNMENT Drive misalignment is something that doesn't usually happen all at once. It's normally a gradual process that begins with occasional disk errors (bad disk or intentional errors from copy protection) while loading (red LED flashing), failure to work with some programs, or excessive head chatter (the drive getting "lost" and having to go back to track zero to "find" it's place again.) Drives are forced out of alignment mostly while hot from extended use. Heavily copy protected programs or disk errors can cause the head to chatter against the track zero stop repeatedly. If the alignment is far enough off, you will get "FILE NOT FOUND" errors and red LED flashing with all disks, and the drive may try several times before "giving up". Note that misalignment will cause read errors with known good disks but such a drive will probably still be able to format a disk and read it back again. A disk formatted on a misaligned drive will not read properly on a correctly aligned drive. To properly realign a drive, you need special software. I use "1541/1571 Drive Alignment" by Free Spirit Software. The flipside of the program disk is the alignment disk, and as such, should not be copied. A copy is only as good as the drive that made it. The program provides a menu screen that indicates what track you're on, drive speed, etc. You make adjustments to the drive while watching the screen. The instructions even tell you how to load the program when nothing else will load. You can -check- the alignment of the drive without taking it apart, but of course realignment requires disassembly. I always final-check my drives with several different program disks to verify alignment... any commercial program disk can be used to do that. Drive speed can drift over time, but it's rather rare to find it off far enough to cause problems. Spindle speed (Note: some drives have no adjustment) is reset with a small screwdriver adjustable control on a small PC board near the spindle motor. On older belt-driven spindle motors, the belt may be slipping. On all drives, the spindle collar (clamper) can get sticky and a tiny bit of lube helps. Don't overdo lubrication. Excess oil will be thrown off and could get on the disks. Make sure the latch clamps the disk properly. Without a disk, move the lever down and see if the spring presses the collar against the spindle to clamp it securely. You can bend the tab down -slightly- (Newtronics drives only) so it makes more firm contact if necessary. A slipping or stalling disk will produce random read and write errors, a problem that's very hard to track down. Some disks may work better than others. The DD ones without a hub ring seem to slip more easily. HD disks should never be used in a 1541. Bad (sticky) old grease on an ALPS drive can actually stall the spindle motor. If that is suspected, or the clamper is noisy, you can try a drop of oil in the center of the spindle clamper and run the drive to let the oil work its way in. If that doesn't help, you wiil have to remove the metal bar that the hub lock is mounted on (two screws in the back) and take apart the clamper assembly by removing the C clip. It has half a dozen assorted washers, a brass bushing and a spring along with the plastic hub. Don't allow any of the spring-loaded parts to fly off when you remove the C clip. These parts need to be cleaned with solvent and put back together in the same order as before. A bit of moly lube or light oil finishes the job. Make sure the hub clamper is centered on the spindle (drive door closed) before you fully tighten the its screws. If the Newtronics (twist door) mechanism stepper seems sticky after cleaning the rails, you can try adding a few drops of oil to its shaft by tipping the drive on its rear and applying the oil to the space between the stepper collar and the metal base. The oil will run down and collect on the shaft (without disassembly) and wick it's way into the bearing. The same thing can be done to quiet a noisy spindle, but I've had to take a few of those apart to replace a really noisy bearing. Don't overdo the oiling, and keep a rag handy to clean up the excess. I made a survey of all the repair jobs I did on about 40 1541's. It might be helpful to know the number of bad chips I replaced to give you some idea of the statistics of the most common failures: 6502 CPU 6 6522 VIA 10 most often UC3 901229-05 DOS ROM 2 5 325302-01 DOS ROM 1 1 325572-01 MTR CTRL 1 7406 LOGIC 4 7407 (7417) LOGIC 4 74LS14 LOGIC 1 311 COMPARITOR 1 BRIDGE RECTIFIER 4 either 5V or 12V source can drop out TANTALUM CAPACITOR C15 3 can burn up or short and drop 12V power BAD NEWTRONICS HEAD 16 ALPS head is a rare failure Ray Carlsen CET CARLSEN ELECTRONICS... a leader in trailing-edge technology. Questions and comments are welcome, especially if you spot a mistake here. Thanks!