In the old days, when we had a diskette like you showed at the end of the video, we would take an exacto knife and cut carefully along one end pull out and clean the whole diskette, then put in back into it diskette jacket and use Scotch tape to seal the cut end. This works much better than just trying to clean the diskette through the drivehead hole! If you're worried about leaving fingerprints on the diskette, try washing your hands prior and handle the diskette by its edges, or wear lint-free cotton archival gloves.
Lots of useful hints there. May have already been mentioned below, but most of the drives have two speed adjust rings printed on the flywheel, one for 50Hz, one for 60Hz, same as some turntables. Track 0 is always where it starts reading from, for the disk info and the BAM, all referenced from counting the number of the tracks from track 0. Older drives just run the head back until it hits the stop, it's better for the mechanics of moving the head if it can be stopped by the optical sensor before it hits the stop, so in general it would be a good idea to enable it.
+Dave Curran Thanks! I thought the BAM info was track 18? The other strange thing is that alignment software says my head stop is track 1 which suggests it cannot get to track 0, yet it reads disks fine. All very confusing lol.
+GadgetUK164 Yes, sorry, I was trying to say it started counting from there to get all the disk info and the BAM etc., all referenced to the number of tracks since 0.
Most drives with a manual speed adjustment pot also have the "strobe" markings on the drive spindle. There's usually two sets of circular marks - the inner one is for 50hz and the outer one is 60hz. You shine a fluorescent light on it, get the drive to spin, and adjust the pot until the marks appear stationary. (It's the interference or "wagon wheel" effect that causes the marks to drift if they're not at a multiple of the frequency - in this case, 50 x 60 = 300.)
Only the 1541C uses the optical headstop, and the ROM some of them shipped with is designed to make use of it (many 1541C's shipped with the original 1541 ROMs). The 1541 and 1541-II ROM's do not have code to use a optical headstop. Further, the 1541-II has the VIA input that the optical sensor would connect to grounded, so a 1541C ROM in a 1541-II will think it's always (or never, I can't remember which) at track 1. Unfortunately, the optical sensor models were short lived as the code in the ROM to check the optical sensor caused problems with copy protection, and made it impossible to use aftermarket parallel fast loaders as the sensor occupies one bit of the spare parallel port on the VIA. The stepper motor needs two steps to move one track, so by ignoring the optical sensor you can get to track 0.5 by manually stepping the head in software, but never track 0 as the head stop prevents it. The head is, naturally, one track wide, so using half tracks isn't useful for anything but rudimentary copy protection. Of more interest is aftermarket ROM's that let you go beyond track 35 (depending on the disk you can physically move the head to track 42, but typically track 39/40 is the highest usable track). A properly aligned 1541C with a 1541C ROM was great for formatting a bulk amount of disks, it has no head rattle thanks to the optical sensor and will stay in alignment significantly longer than a head rattling unit. Though, no one uses the slow slow ROM routines for formatting these days, so that point is moot.
Another excellent video. I thought I would chime in about the parallel mod. I have a parallel modded drive I use with a ZoomFloppy and a PC to enable the reading and writing of copy protected disks. The parallel mod is crazy fast, my setup can write an entire side of a disk in mere seconds compared to normal. While I don't have any experience with using a parallel drive on a C64 (havent built the user por adapter yet) as long as it was more or less compatible it should absolutely fly. The 1541 drive mechanism itself isn't slow, the slow part is how the had to slow the serial protocol down in software to make it work with early bugged CIA chips as I recall. I use JiffyDOS on both my 64 and 1541 drive however and that solution is plenty fast enough for me...
+aphexteknol Thanks for the awesome replies there =D I have had a look at that ZoomFloppy board - it looks cool! Don't seem to be any in stock =/ Thanks for this information, it's sounds like a worthwhile mod.
+GadgetUK164 No problem, I love all your videos, especially the Commodore ones since those are my passion. Anytime I can help out a little with some of the stuff I've learned, I'm more than happy to share my experiences. I of course dabble in other games consoles up to about the 16 bit era, and even a little bit of Amiga as well.. I'm gonna have to find a way to contact you a little easier. YT comments can be a bit of a pain.
+aphexteknol Thanks =D Yes, TH-cam comments can be a pain but it's the only way I can manage so many messages tbh. It's taken me an hour and a half this morning just to reply to the comments of the last few days =/
Great video once again! Correct me if im wrong, but dont the instructuons for the alignment disk instruct you to flip the disk and run the head from track 19 or higher down to the 16.5 half track, then from 13 up to 16.5 to make sure the head can reliably hit the half track? I havent used my Free Spirit disk in a while but from memory thats what I thought. Maybe you did that but I missed it in the video. Regards
+Andy Mason Ah, you might be right! I didn't do that, or read the instructions properly lol. I will have another go and let you know the results. It's possible it might read the half tracks better on the other side.
The particals that can stick to these heads (and VCR heads) often needs something like PlastX to remove it. Another common technique for VCR heads is to use a business card with IPA on it. Hold the card against the head and rotate it.
I've still got my original 1541 which is the first one after it died, but will hopefully get around to seeing if I can fix it when I get more time. It powers-up but it doesn't do anything when trying to load
+GadgetUK164 I didn't have a disk drive until someone was selling one with a C64C in 1998 when my original C64C died or so I naively thought it did then I later learned it was just the internal fuse which is a telltale sign when there's no power to the tape deck and no sound, but it was too late because I'd already thrown it away. It taught me a lesson and that's why I've still got my original 1541, but I'm not technically-minded though what happened is I was loading a game back then (can't remember what it was) then whilst it was loading all the electrics went off. Once I got the electric back on I tried my 1541 and like it still does now powers-up with its green light and you can hear the motor initialising, but after you've typed: LOAD"*" it just hangs. Obviously if it's just a case of changing or replacing socketed chips I'll be able to do it but if it's a case of involving soldering then I'm no good at that so will have to see if I can get someone online to fix it for me though postage will cost a lot.
Mitsumi drives do not have the speed calibration wheels on the spindle on the bottom like Alps drives do. I dont know about the difference between PAL and NTSC as far as the drives are concerned, but I know you can find an image of the speed calibration wheel, print it out, and stick it on the spindle of a mistumi drive and use that to set the speed correctly. I'm not sure I would trust free spirit enough to set the speed using the software. For what its worth, free spirit says my drives operate at 310 rpm, another drive alignment package (CSM) reports 300 rpm, so who knows? My gear is NTSC by the way...
+aphexteknol Thanks for the info =D I might just print off and stick that wheel on there - just to see how well it works vs the speed test software. I think 310 is normal with that software when running on a NTSC C64, and 320 for a PAL C64. It seems strange to me that they could be more accurate with the reported RPM. It certainly seems possible since you've got some other software that reports 300 rpm! What's the other software you use?
+GadgetUK164 I have a couple other 1541 drive alignment packages actually, I have CSM Software's 1541 drive alignment written by TN Simstad. It seems to be fairly obscure, only a couple references to be found online of it actually. Seems like a decent program, but I'm missing the manual so I can't interperet some of the results. I've tried copying some of these alignment packages with a kryoflux and a well aligned drive, but haven't gotten very repeatable results with the copies yet.
I just watched this video because I have a very similar drive at home (working). When I opened it up it turned out that mine is a special one. It has a 1541C white case, a 1541C type motherboard and even a 1541C ROM, but it's drive mechanism is different: it doesn't have a circuit board next to the motor and it doesn't have a track sensor. Also the model name printed at the back is 1541 not 1541C. I guess at mass production Commodore had different 3rd party floppy drive mechanisms in stock (mine is the old toaster style Alps drive but with white front plastic) and depending on which one had track sensor and which one doesn't they named the assembled unit as a simple 1541 or 1541C regardless of the case color or the motherboard version. Maybe that's why they did not cut the track sensor jumper by default on the board and why did they gave up on the whole track sensor concept later with the 1541-II.
9:59 - Anchor IPA? Libertine Black Ale? Another way to check if the drive works it to write 1-4k memory chunks to the disk and test read them in sequence and see if the drive reads the data properly.
I have a 1541 drive I am repairing where the drive spins WAY too fast, and i mean on the slowest vr1 setting its 500RPM. this was my own personal 1541 drive from the early 80s and i initially thought it was logic chips and dropped several in. 7406, 311, 592s, etc. but issue still there. On using diagnostic cart i finally noticed the rpm speed and think this is the reason for the 21 and 74 drive errors. Have you seen drives spin too fast and how did you fix it? I assume its on the lower board that drives the spindle motor but I am still troubleshooting. If I figure it out i will post it here but curious if you have seen this before. Thanks for your time.
It sounds like there's a problem with the speed control for sure. Adrians Digital Basement fixed an Apple drive a month or two back that had the same problem, and coincidentally had the same speed controller PCB. But, your board revision might be different.
The manual for the 1541C states that the format command should be as follows OPEN 15,8,15 "N0:DISK NAME,01":CLOSE 15 I think typing "CLOSE 1" was confusing it. The number after the close command needs to be the same as the number after the open command.
+TheDeeplyCynical I think I made a mistake in the video there then, but the problem remains - You can see an example of the exact same issue in the previous video, but instead of me typing in the command it was done automatically by the cart I booted, and the result was the disk formated with 65 blocks free. That's what happens if the CLOSE 1 (or CLOSE 15) is appended to the format command with this drive.
+GadgetUK164 Did you try it with the sensor disabled? I have exactly the same mitsumi drive as you and J3 is bridged and the format process starts with the head-thumping, i notice it didn't on your drive. The manual states that the thumping is normal so I believe that the sensor should be disabled by default on these models
+TheDeeplyCynical I've seen a few articles suggesting that some of the 1541C drives shipped with J3 open and mine didn't look like it had ever been soldered. I did test with the sensor disabled earlier and it worked just the same way regards formatting, but the bump test passed in the alignment disk test. I know they removed the bump in the 1571 so I think I will just leave the sensor enabled as its less likely to cause an alignment issue or motor failure in future. I might re-visit this in a future video.
On my drive you have to look very closely to see the join on J3, no blobs of solder just a very fine trace between the pads. The head-bang allignment issue was only present on early 1541 drives, they fixed the issue for the 1541C and later. I don't know what it is with Commodore, no two units they made was exactly the same!
Also, I wanted to tell you that copy of the free spirit alignment disk you bought, go ahead and ignore the half tracks while testing. After some research I've done, it appears these are merely copies of the original disk and the half tracks on the calibration side are missing. If you analyze the disk with a Kryoflux card, you'll see there is nothing there. This makes me believe these disks are just being duplicated using a Commodore drive and not a real duplication machine but I could be wrong. I've conferred with a few other people about this issue and they all reported that the half tracks on their copies they got off ebay do not work either on their drives so I know it isn't just us.
+aphexteknol Thanks =D I will re-try with the other side of the disk later. I meant to do it at weekend but didn't get chance. It looks like an old box and disk, so maybe its an original? I did wonder if it was a copy but after spending a lot of time examining everything (including the label) it does look original. Mind you, it could have been copied decades ago I suppose =/ I will let you know, and perhaps do a follow up video.
+aphexteknol I re-tested with the other side of the disk btw - same think, half tracks toggle between poor, acceptable and excellent. Not sure if there's supposed to be a 2nd disk or something. The disk looks like the original, but I guess it would be easy to fake a repro label.
wasn't there a bug on the 1541 serial bus routine that slowed the xfer rate down significantly? I guess changing to parallel mode, with an appropriate driver, would resolve this? I know Jiffy DOS did similar with the serial IEC routine.
I have an 1541-II which can't read/write, can't even format (it reports "21 Read Error" on first track). I suspected the head is bad but head resistances are within specs.
GadgetUK164 I've been using only 70% alcohol for head cleaning but it was already clean when it arrived. Don't have a tape head demagnetiser nor Datasette (if that's where I could get one). There are also other issues which point to electronic fault. For instance, head alignment test reports error instead of RPM and it always reports that the actual track being read is 1 even when the head physically moves to other tracks. Another 1541-II will arrive soon and I'll try to diagnose the problem by swapping electronics/mechanics and comparing measurements on both.
Somone told me that the parallel mode generates more heat on the drive chips, but don´t know if that´s really true. Would not be a problem for your drive with all the heat sinks installed. ;-)
Chase HQ on the C64 is very bad conversion, Spectrum version is brilliant which includes the awesome music that included beeper included with the intro music.
Hi! great video, thanks for upload! i have exacly the same issue with my 1541 Mitsumi drive, spended some days to fix it without sucsess..i have measured the head resistance and i got totally different results as Yours: red & yellow 15.26 kOhm red & black 15.26 kOhm red & blue 15.25 kOhm yellow & black 16 Ohm yellow & blue 27.4 Ohm Black & blue 13.7 Ohm according to the resistor on the bottom side of the head it's a 1.5k Ohm one (brown, green, red, silver) but it gives 10x higher resistance..How comes? Am i wrong? have already swapped it to 1.5kOhm one with no sucsess.. It seems that the demagnetising is my last hope, however, do not have that unit and i'm unable to do it :(
Comparing to what you should be getting (working on left, yours on the right hand side) These 3 I suspect have a problem relating to all 3 of them - maybe a short, but typically a short would result in a lower resistance not a higher resistance?!? Unless a common connection on them has gone high resistance. Have you checked the diode to see if that is OK? And just to check, you are taking the measurements when the head is disconnected from the logic board? What ever those first 3 windings have in common is the cause, maybe the diode or maybe one wire has some damage. red & yellow: 25.6 Ohm vs 15.26 kOhm red & black: 12.4 Ohm vs 15.26 kOhm red & blue: 23.8 Ohm vs 15.25 kOhm These 3 windings are fine and likely not a problem:- yellow & black: 13.8 Ohm vs 16 Ohm yellow & blue: 25.2 Ohm vs 27.4 Ohm black & blue: 12.0 Ohm vs 13.7 Ohm
Also double check you do have a mitsumi drive (same model as shown in the video). There's a chance the heads are not mitsumi or something - does the head look the same as in my video?
Yes, it's a Newtronics/Mitsumi drive i have (however, it has a white housing!) with exacly the same short board, same head type with sensor arm and speed adjustment board on top. I have spent another three hours with it, barrowed a working Newtronics 1541 drive from a chap, swapped the boards out - board works with no problems. Measured the diod and resistor under the drive head - same measurements, also doublechecked the leads - everything is good...except R/W coil: www.zimmers.net/anonftp/pub/cbm/schematics/drives/new/1541/service/Page_19.html so i am pretty sure the head is dead. Need a spare one, so if someone knows where can i obtain some - please let me know.. Funny thing, it seems that only Newtronics heads are dying, the quality of the heads used in Alps is much better....
I found this service manual for the 1541 at this site, it may help answer some questions about how they adjust spindle speed. personalpages.tds.net/~rcarlsen/cbm/1541/1541_Service_Manual%20(1985).pdf
+Crazy Monkey (Jame Pond) I did wonder that, but I am not sure if dare - once I move that there's a chance all track alignment will be out. I might do that at some point, once I've done a bit more research first.
If indeed it has slowed down a tiny bit, yes, trimmer cap related to crystal circuit, or the caps associated with the crystal, or the crystal itself. I cannot remember what speed these are supposed to run at, but on the C64 0.98 is normal for a PAL system. But the drives, I cannot remember what speed they run at.
thanks for answer in this 1541II - sadly trimmer can't adjust this so I'll replace C12, C31(C31 probably remove only), trimmer and crystal Crystal have 16MHZ and divide per 16 for 6502 is equal 1MHZ density signal is divide from 16 MHZ by - 13 for track 1-17 = 1.2307 MHz - 14 for track 18-24 = 1.1428 MHz - 15 for track 15-30 = 1.0666 MHz - 16 for track 31-35 = 1 MHz cheers
PROBLEM SOLVED After removed C31 work perfectly drive have 294 cycles more than C64 per every one C64 frame Of course cappacitors and adjustable cap. will be change as well. We tried repair the drive and can't adjust it by trimmer simply: C64 6510 going on 985 248 Hz 6502 in 1541 going on 1 000 000 Hz for simplicity divide by 100 and when C64 wasting 9852 cycles you need expect in drive past 10 000 cycles about 150 more than C64 This is very important feature for example Action Replay loader which after programming sync C641541 sending a few bytes without acknowledgement and when 6502 CPU in drive is slowing down it simply DOESN'T WORK and many other programs which using this method. I wrote 2 programs PAL/NTSC version first one measure a spin RPM and in this drive spin is 299 RPM less than 0.3% error so it is good BUT second program mesure how many cycles going in the drive CPU in one frame C64 C64 PAL case have a frame 312 scanlines * 63 cycles = 19656 cycles Program check how many cycles gone in this time in CPU drive from my experiency I know properly value is about 270-300 cycles more than C64 in this drive we can't adjust this by trimmer and this drive is ONLY about 50 cycles more than C64 - AR doesn't work once again my experiences is properly with a mathematic: 985248/1000000 = 0.985248 now 1/0.985248 = 1.0149728799246484133943941017896 AND 19656 * 1.0149728799246484133943941017896 = 19950 FINALLY: ------------------------------------------------------------------------- 19950 - 19656 = 294 CYCLES MORE IN DRIVE THAN C64 -------------------------------------------------------------------------
That's only on certain models (like the Alps). This model does not have the timing disk underneath. It's also better to use a frequency counter on this model, or on any model its better to use software to report the RPM. Doing it by eye using the timing disk on some drives is inaccurate when compared to other methods.
My understanding is the motor speed on the 1541 is not affected by the mains frequency. I believe the logic frequency comes from a crystal. The mains input drops the supply to 12v and 5v DC only, and there's no use of the original AC frequency in the 1541. On the C64 the Time of Day component of the system does hijack the mains input frequency in order to run at 50hz or 60hz. An NTSC C64 runs faster than a C64 - due to the clock generator on board, and I believe the CPU is different between NTSC and PAL - probably due to the number of scanlines, and as a result the VIC chip also runs faster on NTSC. I am told that a european 1541 runs at exactly the same rpm when powered in a 110v region (using a step up from 110v to 220v).
Hi, and thanks for this wonderfull video. I didn't read all comments of other people, having said that i have a few pointers and comments for your consideration: Check out a long-term repair experience collection by Ray Carlsen: personalpages.tds.net/~rcarlsen/cbm.html About the parallel cable. The parallel cable speeds up disk access quite a bit. Utilising the unused VIA it directly accesses info which is made available by uploaded floppy routines using the second CIA. This enables low-level access to the drive, allowing things like copying many copy-protections, or copy a whole 1541 disk (single side offcourse) in 10 seconds ;) As for compatibility, no worries there. The parallel cable itself doesn't impact compatibility (you can just let it disconnected from your C64 if you wish). A parallel speeder still uses the serial cable to give the drive commands, so you can't omit it. Floppy speeder-kernals like SpeedDOS, DolphinDOS, etc that made use of those parallel-cables did have compatibility issues with some (copyright protected) software. They came with their own kernal for the C64 and 1541 diskdrives. To bypass this you would use a kernal-select switch on the C64 and 1541 allowing you to use the original kernals. Offcourse many Copy programs using the parallel cable required the special kernals to be present. Here is some info on the drive geometry including half-tracks markus.brenner.de/binary/vcf3.pdf A lot of (especially) games did look for things like a speeder, cartridge, 2nd drive, ...., or simply could not correctly handle the extra connected hardware. In order to test a game/program it is wise to try it on a stock C64 and a stock 1541 first. From there you can try it with cartridges, and kernals, REU, etc. For example, if you configure Maverick to use a REU for copying without actually having a REU connected, the copier will copy garbage, because it copies parts of ROM/RAM instead of REU-RAM ;) Things like this can be quite confusing. It' s also wise to bare in mind that any hardware/software can have quirks/bugs/etc that only show themselves on non-stock configurations ;) Also check the text on Flippy-Disks. Flippy-disks were regular 1541 DS/DD disk which had an extra notch made by the user to allow it to be used/formatted on both sides. The drawback is that these flippy-disk-sides were not readable on a drive that relied on the index-hole-sensor for track 0 (as PC drives do). You can read more on the KryoFlux forum: forum.kryoflux.com/viewtopic.php?t=253 And finally, when powering up your C64 and 1541, wait for the 1541 till it stops turning, if you don't it might not respond to the C64 when it sends a command through the serial cable, locking up the serial bus ;) Have fun ;)
+MissingDaughters Thanks for watching and thanks for the awesome reply there! Lots of really useful information! =D I've had a look at Ray Carlsons site before - it's a good resource! Very interesting info on the parallel cable, I might have a go at that at some point. I will have a good read of those links on disk geometry and the kyroflux website =D
310 RPM ??? Ideal speed is 299 RPM or even lower if you calculate GCR bytes (5 bytes sync, info header, gap, sync of sector, sector indycator, sector data, checksum, gap - multiply by 21 times ) writes on the 1-17 tracks (16MHZ div by 13 for write tact per bit) you can obtain: between last one byte of last sector and first byte of sync first sector (trully 0 called) you got about 30 free bytes to write space (gap) under 300 RPM. Don't forgot if the crystal circuit wasting some edges and get a bit lower tact speed - when use VIA timer for measure motor speed this method gave incorrectly higher value of RPM. You can try (like in Amiga) making own double density write file system by set lower value of RPM to 148. In this case only this one drive will be able to read it but I think it is possible. The wasting edges problem you can recognize: Normal load work properly, Action Replay and few other fastloaders doesn't work. Cheers GadgetUK164
That software reports 310 on a PAL system even when its set to 300! It's well known! This is why software is not the best way to check it. A frequency counter is better.
@@GadgetUK164 It make sense if you know what you do and how to do. I measure motor speed by program: Drive code - 1. find first sector on the track and toggle state line on the serial port 2 go to 1 C64 code: By 1 minute (3000 frames in C64 PAL mode) count the changes on the $DD00 I think it's accurate enought I wonder why the program recommended 310 RPM as ideal speed. I suppose in the program is any inaccurate.
I found this service manual for the 1541 at this site, it may help answer some questions about how they adjust spindle speed. personalpages.tds.net/~rcarlsen/cbm/1541/1541_Service_Manual%20(1985).pdf
![BOWKYLION - วิงวอน (ex-change) [Official MV]](http://i.ytimg.com/vi/cLdnZWDaO-w/mqdefault.jpg)
In the old days, when we had a diskette like you showed at the end of the video, we would take an exacto knife and cut carefully along one end pull out and clean the whole diskette, then put in back into it diskette jacket and use Scotch tape to seal the cut end. This works much better than just trying to clean the diskette through the drivehead hole! If you're worried about leaving fingerprints on the diskette, try washing your hands prior and handle the diskette by its edges, or wear lint-free cotton archival gloves.
+ResistanceIsUsefull That's a brilliant tip - thanks =D
Lots of useful hints there. May have already been mentioned below, but most of the drives have two speed adjust rings printed on the flywheel, one for 50Hz, one for 60Hz, same as some turntables. Track 0 is always where it starts reading from, for the disk info and the BAM, all referenced from counting the number of the tracks from track 0. Older drives just run the head back until it hits the stop, it's better for the mechanics of moving the head if it can be stopped by the optical sensor before it hits the stop, so in general it would be a good idea to enable it.
+Dave Curran Thanks! I thought the BAM info was track 18? The other strange thing is that alignment software says my head stop is track 1 which suggests it cannot get to track 0, yet it reads disks fine. All very confusing lol.
+GadgetUK164 Yes, sorry, I was trying to say it started counting from there to get all the disk info and the BAM etc., all referenced to the number of tracks since 0.
Most drives with a manual speed adjustment pot also have the "strobe" markings on the drive spindle. There's usually two sets of circular marks - the inner one is for 50hz and the outer one is 60hz. You shine a fluorescent light on it, get the drive to spin, and adjust the pot until the marks appear stationary. (It's the interference or "wagon wheel" effect that causes the marks to drift if they're not at a multiple of the frequency - in this case, 50 x 60 = 300.)
+josephmmuller Thanks very much for clearing that up =D That makes perfect sense! I did have a look under my drive and I don't have the disk =/
Floppies and tapes are a pain. Luckily there's alternatives for nearly every system now!
+custardo I know lol - It's making me really appreciate the 1541 Ultimate!
Only the 1541C uses the optical headstop, and the ROM some of them shipped with is designed to make use of it (many 1541C's shipped with the original 1541 ROMs). The 1541 and 1541-II ROM's do not have code to use a optical headstop. Further, the 1541-II has the VIA input that the optical sensor would connect to grounded, so a 1541C ROM in a 1541-II will think it's always (or never, I can't remember which) at track 1.
Unfortunately, the optical sensor models were short lived as the code in the ROM to check the optical sensor caused problems with copy protection, and made it impossible to use aftermarket parallel fast loaders as the sensor occupies one bit of the spare parallel port on the VIA.
The stepper motor needs two steps to move one track, so by ignoring the optical sensor you can get to track 0.5 by manually stepping the head in software, but never track 0 as the head stop prevents it. The head is, naturally, one track wide, so using half tracks isn't useful for anything but rudimentary copy protection. Of more interest is aftermarket ROM's that let you go beyond track 35 (depending on the disk you can physically move the head to track 42, but typically track 39/40 is the highest usable track).
A properly aligned 1541C with a 1541C ROM was great for formatting a bulk amount of disks, it has no head rattle thanks to the optical sensor and will stay in alignment significantly longer than a head rattling unit. Though, no one uses the slow slow ROM routines for formatting these days, so that point is moot.
+Peter Owen Thanks for all the information there - very interesting! I've picked up a lot of new knowledge there, much appreciated =D
Another excellent video. I thought I would chime in about the parallel mod. I have a parallel modded drive I use with a ZoomFloppy and a PC to enable the reading and writing of copy protected disks. The parallel mod is crazy fast, my setup can write an entire side of a disk in mere seconds compared to normal. While I don't have any experience with using a parallel drive on a C64 (havent built the user por adapter yet) as long as it was more or less compatible it should absolutely fly. The 1541 drive mechanism itself isn't slow, the slow part is how the had to slow the serial protocol down in software to make it work with early bugged CIA chips as I recall. I use JiffyDOS on both my 64 and 1541 drive however and that solution is plenty fast enough for me...
+aphexteknol Thanks for the awesome replies there =D I have had a look at that ZoomFloppy board - it looks cool! Don't seem to be any in stock =/ Thanks for this information, it's sounds like a worthwhile mod.
+GadgetUK164 No problem, I love all your videos, especially the Commodore ones since those are my passion. Anytime I can help out a little with some of the stuff I've learned, I'm more than happy to share my experiences. I of course dabble in other games consoles up to about the 16 bit era, and even a little bit of Amiga as well.. I'm gonna have to find a way to contact you a little easier. YT comments can be a bit of a pain.
+aphexteknol Thanks =D Yes, TH-cam comments can be a pain but it's the only way I can manage so many messages tbh. It's taken me an hour and a half this morning just to reply to the comments of the last few days =/
Great video once again!
Correct me if im wrong, but dont the instructuons for the alignment disk instruct you to flip the disk and run the head from track 19 or higher down to the 16.5 half track, then from 13 up to 16.5 to make sure the head can reliably hit the half track? I havent used my Free Spirit disk in a while but from memory thats what I thought. Maybe you did that but I missed it in the video.
Regards
+Andy Mason Ah, you might be right! I didn't do that, or read the instructions properly lol. I will have another go and let you know the results. It's possible it might read the half tracks better on the other side.
+GadgetUK164 ahh lol.
Yeh the flip side of the disk is (allegedly) a special track layout for use with alignment.
Will give it a go at weekend! =D
I saw in a VCR repair video that IPA will not remove certain types of organic contamination. The guy recommended just a damp rag.
The particals that can stick to these heads (and VCR heads) often needs something like PlastX to remove it. Another common technique for VCR heads is to use a business card with IPA on it. Hold the card against the head and rotate it.
I've
still got my original 1541 which is the first one after it died, but
will hopefully get around to seeing if I can fix it when I get more
time. It powers-up but it doesn't do anything when trying to load
+Mark Thompson Let me know how you get on!
+GadgetUK164 I
didn't have a disk drive until someone was selling one with a C64C in
1998 when my original C64C died or so I naively thought it did then I
later learned it was just the internal fuse which is a telltale sign
when there's no power to the tape deck and no sound, but it was too late
because I'd already thrown it away. It taught me a lesson and that's
why I've still got my original 1541, but I'm not technically-minded though what happened is I was loading a game back then (can't remember what it was) then whilst it was loading all the electrics went off. Once I got the electric back on I tried my 1541 and like it still does now powers-up with its green light and you can hear the motor initialising, but after you've typed: LOAD"*" it just hangs. Obviously if it's just a case of changing or replacing socketed chips I'll be able to do it but if it's a case of involving soldering then I'm no good at that so will have to see if I can get someone online to fix it for me though postage will cost a lot.
Mitsumi drives do not have the speed calibration wheels on the spindle on the bottom like Alps drives do. I dont know about the difference between PAL and NTSC as far as the drives are concerned, but I know you can find an image of the speed calibration wheel, print it out, and stick it on the spindle of a mistumi drive and use that to set the speed correctly. I'm not sure I would trust free spirit enough to set the speed using the software. For what its worth, free spirit says my drives operate at 310 rpm, another drive alignment package (CSM) reports 300 rpm, so who knows? My gear is NTSC by the way...
+aphexteknol Thanks for the info =D I might just print off and stick that wheel on there - just to see how well it works vs the speed test software. I think 310 is normal with that software when running on a NTSC C64, and 320 for a PAL C64. It seems strange to me that they could be more accurate with the reported RPM. It certainly seems possible since you've got some other software that reports 300 rpm! What's the other software you use?
+GadgetUK164 I have a couple other 1541 drive alignment packages actually, I have CSM Software's 1541 drive alignment written by TN Simstad. It seems to be fairly obscure, only a couple references to be found online of it actually. Seems like a decent program, but I'm missing the manual so I can't interperet some of the results. I've tried copying some of these alignment packages with a kryoflux and a well aligned drive, but haven't gotten very repeatable results with the copies yet.
I just watched this video because I have a very similar drive at home (working).
When I opened it up it turned out that mine is a special one.
It has a 1541C white case, a 1541C type motherboard and even a 1541C ROM, but it's drive mechanism is different: it doesn't have a circuit board next to the motor and it doesn't have a track sensor. Also the model name printed at the back is 1541 not 1541C.
I guess at mass production Commodore had different 3rd party floppy drive mechanisms in stock (mine is the old toaster style Alps drive but with white front plastic) and depending on which one had track sensor and which one doesn't they named the assembled unit as a simple 1541 or 1541C regardless of the case color or the motherboard version.
Maybe that's why they did not cut the track sensor jumper by default on the board and why did they gave up on the whole track sensor concept later with the 1541-II.
9:59 - Anchor IPA? Libertine Black Ale?
Another way to check if the drive works it to write 1-4k memory chunks to the disk and test read them in sequence and see if the drive reads the data properly.
Hehe - IPA = Isopropanol Alchohol. Good idea testing the drive that way!
Pretty cool stuff on the 1541
Sorry there's a couple of 'de-ja-vu' editing errors in this video =/
not to worry, appreciate you producing these vids
I have a 1541 drive I am repairing where the drive spins WAY too fast, and i mean on the slowest vr1 setting its 500RPM. this was my own personal 1541 drive from the early 80s and i initially thought it was logic chips and dropped several in. 7406, 311, 592s, etc. but issue still there. On using diagnostic cart i finally noticed the rpm speed and think this is the reason for the 21 and 74 drive errors.
Have you seen drives spin too fast and how did you fix it? I assume its on the lower board that drives the spindle motor but I am still troubleshooting. If I figure it out i will post it here but curious if you have seen this before. Thanks for your time.
It sounds like there's a problem with the speed control for sure. Adrians Digital Basement fixed an Apple drive a month or two back that had the same problem, and coincidentally had the same speed controller PCB. But, your board revision might be different.
The manual for the 1541C states that the format command should be as follows
OPEN 15,8,15 "N0:DISK NAME,01":CLOSE 15
I think typing "CLOSE 1" was confusing it. The number after the close command needs to be the same as the number after the open command.
+TheDeeplyCynical I think I made a mistake in the video there then, but the problem remains - You can see an example of the exact same issue in the previous video, but instead of me typing in the command it was done automatically by the cart I booted, and the result was the disk formated with 65 blocks free. That's what happens if the CLOSE 1 (or CLOSE 15) is appended to the format command with this drive.
+GadgetUK164 Did you try it with the sensor disabled?
I have exactly the same mitsumi drive as you and J3 is bridged and the format process starts with the head-thumping, i notice it didn't on your drive.
The manual states that the thumping is normal so I believe that the sensor should be disabled by default on these models
+TheDeeplyCynical I've seen a few articles suggesting that some of the 1541C drives shipped with J3 open and mine didn't look like it had ever been soldered. I did test with the sensor disabled earlier and it worked just the same way regards formatting, but the bump test passed in the alignment disk test. I know they removed the bump in the 1571 so I think I will just leave the sensor enabled as its less likely to cause an alignment issue or motor failure in future. I might re-visit this in a future video.
On my drive you have to look very closely to see the join on J3, no blobs of solder just a very fine trace between the pads. The head-bang allignment issue was only present on early 1541 drives, they fixed the issue for the 1541C and later.
I don't know what it is with Commodore, no two units they made was exactly the same!
I got one, going to replace the psu with switched. Power supply that is going to remove a lot of heat
That's a great idea! Let me know if you upload a video!
From the service manual: Newtronics short J6, Alps J6 is open. Probably wrong schematics, but still..
+bwack Yeah, I don't see a J6 on this drive =/
Also, I wanted to tell you that copy of the free spirit alignment disk you bought, go ahead and ignore the half tracks while testing. After some research I've done, it appears these are merely copies of the original disk and the half tracks on the calibration side are missing. If you analyze the disk with a Kryoflux card, you'll see there is nothing there. This makes me believe these disks are just being duplicated using a Commodore drive and not a real duplication machine but I could be wrong. I've conferred with a few other people about this issue and they all reported that the half tracks on their copies they got off ebay do not work either on their drives so I know it isn't just us.
+aphexteknol Thanks =D I will re-try with the other side of the disk later. I meant to do it at weekend but didn't get chance. It looks like an old box and disk, so maybe its an original? I did wonder if it was a copy but after spending a lot of time examining everything (including the label) it does look original. Mind you, it could have been copied decades ago I suppose =/ I will let you know, and perhaps do a follow up video.
+aphexteknol I re-tested with the other side of the disk btw - same think, half tracks toggle between poor, acceptable and excellent. Not sure if there's supposed to be a 2nd disk or something. The disk looks like the original, but I guess it would be easy to fake a repro label.
wasn't there a bug on the 1541 serial bus routine that slowed the xfer rate down significantly? I guess changing to parallel mode, with an appropriate driver, would resolve this? I know Jiffy DOS did similar with the serial IEC routine.
+stupossibleify Very interesting! That's highly possible, i wonder how much faster it is?
I have an 1541-II which can't read/write, can't even format (it reports "21 Read Error" on first track). I suspected the head is bad but head resistances are within specs.
Have you tried cleaning the head with something like PlastX? Or you can also dry running a tape head demagnetiser over the head.
GadgetUK164 I've been using only 70% alcohol for head cleaning but it was already clean when it arrived. Don't have a tape head demagnetiser nor Datasette (if that's where I could get one). There are also other issues which point to electronic fault. For instance, head alignment test reports error instead of RPM and it always reports that the actual track being read is 1 even when the head physically moves to other tracks. Another 1541-II will arrive soon and I'll try to diagnose the problem by swapping electronics/mechanics and comparing measurements on both.
Somone told me that the parallel mode generates more heat on the drive chips, but don´t know if that´s really true. Would not be a problem for your drive with all the heat sinks installed. ;-)
+VectrexRoli Thanks for the info there =D
Chase HQ on the C64 is very bad conversion, Spectrum version is brilliant which includes the awesome music that included beeper included with the intro music.
Totally agreed! The Spectrum version is far better to play!
I subscribed to Commodore Disk User!
Hi!
great video, thanks for upload!
i have exacly the same issue with my 1541 Mitsumi drive, spended some days to fix it without sucsess..i have measured the head resistance and i got totally different results as Yours:
red & yellow 15.26 kOhm
red & black 15.26 kOhm
red & blue 15.25 kOhm
yellow & black 16 Ohm
yellow & blue 27.4 Ohm
Black & blue 13.7 Ohm
according to the resistor on the bottom side of the head it's a 1.5k Ohm one (brown, green, red, silver) but it gives 10x higher resistance..How comes? Am i wrong? have already swapped it to 1.5kOhm one with no sucsess..
It seems that the demagnetising is my last hope, however, do not have that unit and i'm unable to do it :(
Comparing to what you should be getting (working on left, yours on the right hand side)
These 3 I suspect have a problem relating to all 3 of them - maybe a short, but typically a short would result in a lower resistance not a higher resistance?!? Unless a common connection on them has gone high resistance. Have you checked the diode to see if that is OK? And just to check, you are taking the measurements when the head is disconnected from the logic board? What ever those first 3 windings have in common is the cause, maybe the diode or maybe one wire has some damage.
red & yellow: 25.6 Ohm vs 15.26 kOhm
red & black: 12.4 Ohm vs 15.26 kOhm
red & blue: 23.8 Ohm vs 15.25 kOhm
These 3 windings are fine and likely not a problem:-
yellow & black: 13.8 Ohm vs 16 Ohm
yellow & blue: 25.2 Ohm vs 27.4 Ohm
black & blue: 12.0 Ohm vs 13.7 Ohm
Also double check you do have a mitsumi drive (same model as shown in the video). There's a chance the heads are not mitsumi or something - does the head look the same as in my video?
Yes, it's a Newtronics/Mitsumi drive i have (however, it has a white housing!) with exacly the same short board, same head type with sensor arm and speed adjustment board on top.
I have spent another three hours with it, barrowed a working Newtronics 1541 drive from a chap, swapped the boards out - board works with no problems. Measured the diod and resistor under the drive head - same measurements, also doublechecked the leads - everything is good...except R/W coil:
www.zimmers.net/anonftp/pub/cbm/schematics/drives/new/1541/service/Page_19.html
so i am pretty sure the head is dead. Need a spare one, so if someone knows where can i obtain some - please let me know..
Funny thing, it seems that only Newtronics heads are dying, the quality of the heads used in Alps is much better....
I found this service manual for the 1541 at this site, it may help answer some questions about how they adjust spindle speed. personalpages.tds.net/~rcarlsen/cbm/1541/1541_Service_Manual%20(1985).pdf
notice how quickly the disc formats, i.e. not limited by external xfer bottlenecks
+stupossibleify Yes, I think it's a brilliant design really.
can u move or adjust the sensor maybe back a bit?
+Crazy Monkey (Jame Pond) I did wonder that, but I am not sure if dare - once I move that there's a chance all track alignment will be out. I might do that at some point, once I've done a bit more research first.
Hey GadgetUK164 what is wrong when CPU 6502 slowing down instead 1MHZ to 0.98-0.99 MHZ
Trimmer ? Quartz ? Cappacitors ?
If indeed it has slowed down a tiny bit, yes, trimmer cap related to crystal circuit, or the caps associated with the crystal, or the crystal itself. I cannot remember what speed these are supposed to run at, but on the C64 0.98 is normal for a PAL system. But the drives, I cannot remember what speed they run at.
thanks for answer
in this 1541II - sadly trimmer can't adjust this so I'll replace C12, C31(C31 probably remove only), trimmer and crystal
Crystal have 16MHZ and divide per 16 for 6502 is equal 1MHZ
density signal is divide from 16 MHZ by
- 13 for track 1-17 = 1.2307 MHz
- 14 for track 18-24 = 1.1428 MHz
- 15 for track 15-30 = 1.0666 MHz
- 16 for track 31-35 = 1 MHz
cheers
PROBLEM SOLVED
After removed C31 work perfectly drive have 294 cycles more than C64 per
every one C64 frame
Of course cappacitors and adjustable cap. will be change as well.
We tried repair the drive and can't adjust it by trimmer
simply:
C64 6510 going on 985 248 Hz
6502 in 1541 going on 1 000 000 Hz
for simplicity divide by 100 and
when C64 wasting 9852 cycles
you need expect in drive past 10 000 cycles
about 150 more than C64
This is very important feature for example Action Replay loader which after
programming sync C641541 sending a few bytes without
acknowledgement
and when 6502 CPU in drive is slowing down it simply DOESN'T WORK and
many other programs which using this method.
I wrote 2 programs PAL/NTSC version
first one measure a spin RPM and in this drive spin is 299 RPM less than
0.3% error so it is good
BUT
second program mesure how many cycles going in the drive CPU in one frame
C64
C64 PAL case have a frame
312 scanlines * 63 cycles = 19656 cycles
Program check how many cycles gone in this time in CPU drive
from my experiency I know properly value is about 270-300 cycles more than
C64
in this drive we can't adjust this by trimmer and this drive is ONLY
about 50 cycles more than C64 - AR doesn't work
once again my experiences is properly with a mathematic:
985248/1000000 = 0.985248
now 1/0.985248 = 1.0149728799246484133943941017896
AND 19656 * 1.0149728799246484133943941017896 = 19950
FINALLY:
-------------------------------------------------------------------------
19950 - 19656 = 294 CYCLES MORE IN DRIVE THAN C64
-------------------------------------------------------------------------
You supposed to use an strobe light to get the speed
That's only on certain models (like the Alps). This model does not have the timing disk underneath. It's also better to use a frequency counter on this model, or on any model its better to use software to report the RPM. Doing it by eye using the timing disk on some drives is inaccurate when compared to other methods.
shit, shit and more shit!
Guys and Gals' Aww thanks.
+Amy Marie Haha =D And you're welcome =D
I'm no engineer, but can someone explain to me how the drive motor, which runs on direct current (DC) is effected by mains voltage (AC) frequency?
My understanding is the motor speed on the 1541 is not affected by the mains frequency. I believe the logic frequency comes from a crystal. The mains input drops the supply to 12v and 5v DC only, and there's no use of the original AC frequency in the 1541. On the C64 the Time of Day component of the system does hijack the mains input frequency in order to run at 50hz or 60hz. An NTSC C64 runs faster than a C64 - due to the clock generator on board, and I believe the CPU is different between NTSC and PAL - probably due to the number of scanlines, and as a result the VIC chip also runs faster on NTSC. I am told that a european 1541 runs at exactly the same rpm when powered in a 110v region (using a step up from 110v to 220v).
GadgetUK164
I hate when somebody gets me thinking about something so hard that it makes my brain hurt, lol I'm a coco user anyway, cheers and thanks
Hi, and thanks for this wonderfull video. I didn't read all comments of other people, having said that i have a few pointers and comments for your consideration:
Check out a long-term repair experience collection by Ray Carlsen: personalpages.tds.net/~rcarlsen/cbm.html
About the parallel cable. The parallel cable speeds up disk access quite a bit. Utilising the unused VIA it directly accesses info which is made available by uploaded floppy routines using the second CIA. This enables low-level access to the drive, allowing things like copying many copy-protections, or copy a whole 1541 disk (single side offcourse) in 10 seconds ;) As for compatibility, no worries there. The parallel cable itself doesn't impact compatibility (you can just let it disconnected from your C64 if you wish). A parallel speeder still uses the serial cable to give the drive commands, so you can't omit it. Floppy speeder-kernals like SpeedDOS, DolphinDOS, etc that made use of those parallel-cables did have compatibility issues with some (copyright protected) software. They came with their own kernal for the C64 and 1541 diskdrives. To bypass this you would use a kernal-select switch on the C64 and 1541 allowing you to use the original kernals. Offcourse many Copy programs using the parallel cable required the special kernals to be present.
Here is some info on the drive geometry including half-tracks markus.brenner.de/binary/vcf3.pdf
A lot of (especially) games did look for things like a speeder, cartridge, 2nd drive, ...., or simply could not correctly handle the extra connected hardware. In order to test a game/program it is wise to try it on a stock C64 and a stock 1541 first. From there you can try it with cartridges, and kernals, REU, etc. For example, if you configure Maverick to use a REU for copying without actually having a REU connected, the copier will copy garbage, because it copies parts of ROM/RAM instead of REU-RAM ;) Things like this can be quite confusing. It' s also wise to bare in mind that any hardware/software can have quirks/bugs/etc that only show themselves on non-stock configurations ;)
Also check the text on Flippy-Disks. Flippy-disks were regular 1541 DS/DD disk which had an extra notch made by the user to allow it to be used/formatted on both sides. The drawback is that these flippy-disk-sides were not readable on a drive that relied on the index-hole-sensor for track 0 (as PC drives do). You can read more on the KryoFlux forum: forum.kryoflux.com/viewtopic.php?t=253
And finally, when powering up your C64 and 1541, wait for the 1541 till it stops turning, if you don't it might not respond to the C64 when it sends a command through the serial cable, locking up the serial bus ;)
Have fun ;)
+MissingDaughters Thanks for watching and thanks for the awesome reply there! Lots of really useful information! =D I've had a look at Ray Carlsons site before - it's a good resource! Very interesting info on the parallel cable, I might have a go at that at some point. I will have a good read of those links on disk geometry and the kyroflux website =D
310 RPM ??? Ideal speed is 299 RPM or even lower if you calculate GCR bytes (5 bytes sync, info header, gap, sync of sector, sector indycator, sector data, checksum, gap - multiply by 21 times ) writes on the 1-17 tracks (16MHZ div by 13 for write tact per bit) you can obtain: between last one byte of last sector and first byte of sync first sector (trully 0 called) you got about 30 free bytes to write space (gap) under 300 RPM. Don't forgot if the crystal circuit wasting some edges and get a bit lower tact speed - when use VIA timer for measure motor speed this method gave incorrectly higher value of RPM.
You can try (like in Amiga) making own double density write file system by set lower value of RPM to 148. In this case only this one drive will be able to read it but I think it is possible.
The wasting edges problem you can recognize: Normal load work properly, Action Replay and few other fastloaders doesn't work.
Cheers GadgetUK164
That software reports 310 on a PAL system even when its set to 300! It's well known! This is why software is not the best way to check it. A frequency counter is better.
@@GadgetUK164 It make sense if you know what you do and how to do. I measure motor speed by program:
Drive code -
1. find first sector on the track and toggle state line on the serial port
2 go to 1
C64 code:
By 1 minute (3000 frames in C64 PAL mode) count the changes on the $DD00
I think it's accurate enought
I wonder why the program recommended 310 RPM as ideal speed. I suppose in the program is any inaccurate.
I found this service manual for the 1541 at this site, it may help answer some questions about how they adjust spindle speed. personalpages.tds.net/~rcarlsen/cbm/1541/1541_Service_Manual%20(1985).pdf
Thanks, very useful =D