• Please review our updated Terms and Rules here

Floppy disk controllers, 3 1/2" drives, or alternatives?

J

JNZ

Experienced Member
Joined
Sep 4, 2015
Messages
167
Location
Utah
I'm having no end of trouble trying to get both of my Siemens FDD 100-8 8" floppy drives to consistently work on my S-100 system. It has a Jade Double D rev C floppy disk controller which seems to work fine, but one of the drives seems to have a problem with the read logic, and the other is now intermittently failing to read. More debugging is required, but for a month or two now I haven't been able to do much else with the system.

I'd like to run CP/M on my system, using my other cards (Jade Big-Z Z80A, Jade 64K RAM, CompuPro Interface 4 SIO). I have an image of the existing ROM and images of the CP/M disk that actually runs on the machine.

So, is there an S-100 card that'll let me use a more modern and reliable solution, such as a 3 1/2" floppy drive, or even a 5 1/4?" CompactFlash or similar would also work.
 
The Jade controller documentation suggests that 5.25" drives should work. Doesn't say anything about 3.5" drives. I would be surprised if 3.5" drives can't be adapted to work but I never used that controller.

I missed all of the previous thread but the Jade controller documentation also has sheets listing fixes for both Big-Z and Siemens FDD-100 drives so check those out.

Start here: http://retrotechnology.com/herbs_stuff/drive.html#threefive and follow the links and get a good idea of how others adapted 3.5" drives in place of 8" drives.
http://majzel.blogspot.com/2009/04/converting-from-8-to-35-inch-floppy.html
 
Mostly, it's a cabling issue to get a 3.5 HD or 5.25 HD to work in place of a soft-sectored 8" drive. Depending on the software and/or controller, if the "/READY" signal is needed, you may have to tie that line active for 3.5" HD drives, but it's otherwise an exercise in wire-weaving.
 
So speaking for Cromemco systems they had a 16FDC and then 64FDC S100 controller with on card ribbon cable outs for 8 inch and 5.25 inch drives.

This enables cromemco users to cable onto the 5.25 inch diskette drive port a pseudo 3.5 inch disk drive (that fully emulates an 8 inch diskette drive)

or my later modification which is to use the HxC drive emulator.

Here is my blog post for Hxc although of course you can search our blog for anything using syntax under google search_term site:majzel.blogspot.com

https://majzel.blogspot.com/2014/04/the-hxc-solid-state-diskette-emulator.html


regards marcus b
 
Thanks, guys. This is exactly the sort of information I'm looking for.

It sounds like getting a 5 1/4" drive to work would be easier, though the Jade controller manual seems to imply that I'll need to change about four resistors. I'm not sure if any cabling changes are required.

There's always the chance of tearing out a via when desoldering such old equipment, so I'm somewhat hesitant to try this, especially since I'd have to do it again to get the original drives to work. I also face the issue of how to create the boot disk.

The Majzel.blogpost.com post looks promising, though I don't yet understand the purpose of the modifications to the cabling or the floppy drive.
 
There's a N8VEM/s100computers.com IDE board which is designed for use with either real hard drives or CompactFlash cards in an adapter module. You won't be able to use the CP/M images meant for the Jade controller.
 
glitch said:
There's a N8VEM/s100computers.com IDE board which is designed for use with either real hard drives or CompactFlash cards in an adapter module. You won't be able to use the CP/M images meant for the Jade controller.
Click to expand...

What modifications to CP/M are necessary to get it to work with that controller? My system also contains a CompuPro 4 serial/parallel board. Is it likely CP/M was modified to work with this card as well, or is it more compatible "out of the box" with various serial boards?
 
I suggest that you bite the bullet and either have your 8" drives serviced, and buy a pair of more reliable brand drives that have already been serviced.

You need to maintain a way to get new programs onto your computer, and save/store programs and data that you have created. Moving to a flash drive or hard drive takes care of storage, but doesn't solve the problem of getting new stuff onto the computer from other Users/computers. The most common method of interchange between CPM computers is 8" single-sided, single-density floppy disks.

You could add a format changing package (driver programs and floppy controller board), like the one once marketed by New Generation Systems. Designed for used on CPM-80 2.2 Systems, it composed of a floppy controller by Sun Microsystems, and software that would allow the User to read over 100 different 8" floppy formats, and another 100 5.25" floppy formats. Compupro (CDOS, and CCPM-86 for 5.25" and 8"), Morrow Designs (CPM-80 2.2 5.25"), and Kaypro (CPM-80 2.2 5.25"), also supplied format changing software with their computers at one time or another.

All floppy drives need to be serviced if you expect them to continue to read and write disks that can be rear and written on by other computers. Head alignment and index timing drift over time, and need to be adjusted (like the wheel alignment on your car).

Single-sided drives will need to be serviced more often because the felt head load pads wear out. Double-sided drives don't have a head load pad that wears out, but the upper head can go out of alignment (relative to the lower head). Some brands of floppy disk drives have more of a problem with the upper head assemblies going out of adjustment than others, and upper head alignment (if it drifts relative to the lower head) is not supposed to be adjustable, except by the factory).

Personally, I think the the most reliable 8" floppy drive would be a band stepper driven double-sided drive, being run as a single-sided drive. No upper head to go out of alignment or compromise head alignment and index timing, no head load pad to wear out, and no worm screw threads to wear-out. To get a drive like this, you would look for a Shugart full or 1/2 height double-side drive, Mitsubishi full or 1/2 height double-sided drive, NEC 1/2 height double-sided drive, or a Qume 1/2 height double-sided drive, with a bad upper head, and disconnect the upper head index pulse sensor so the drive will act as a single-sided drive.

The NEC FD-1165 1/2 height drive is an excellent drive, that has a tendency to lose the head alignment on the upper head. Double-side drives with a bad head should be relatively cheap, when you come across them.

Your Siemens FD-100-8 drives are not bad drives, they just aren't as reliable as Shugart SA-800/801 single-side full height drives. I don't have much in spare parts laying around here. but I can service/align your drives, and I do have the Siemens FD-100-8 Service Manual. In the past I have sold, and serviced hundreds of Siemens FD-100-8 and FD-200-8 floppy drives. That means I can service/repair the logic boards if necessary. The cost of servicing your drives would be $25 each (single-sided drives) for an inspection without repair (when I get a drive in, I plug it in to see what wrong, and if I can't fix it (or uneconomical to repair), the charge is $25). If the drive needs a straight alignment the service charge would be $100 per drive. A drive that needs repair and parts (other than logic board, heads, motors) would cost $150. A bad logic board is an additional $100 for a repair. Return shipping is a separate charge.

I can also sell you some used 8" floppy drives (with a 6-month warranty). I usually have Shugart SA801 full height single-sided and SA851 full height double-sided drives, as well as Mitsubishi 1/2 height double-side drives service and ready for delivery.

Prices normally range between $200-$300 for single-side drives and $300-$400 for double-sided drives, depending on what I paid for the drives as cores. Good used 8" drive are getting scare. I found that most of the drives for sale on eBay are out of alignment, and many are broken junk.

The copy of CPM running on you computer has been altered for use of the Compupro I/O board. I prefer to stay with all Compupro boards, and Operating Systems because Compupro supplied versions of Operating Systems are all supplied with commented source with support with all of the Compupro boards that were being made at the time that the Operating System shipped from Compupro.

Replacing any board in your system with a different board, or adding additional board/s is going to require reassembly of the Operating System. You can replace your 8" single -sided drives with other 8" single-sided drives (or maybe double-sided drives) without reassembling the Operating System (if your Operating System was generated to work with double-sided drives).
 
That's a lot of very good information. At this point I'm going to make it my goal to use 5 1/4" drives with the Versafloppy II and a modified version of CP/M, as it fits in better with my idea of making a custom enclosure and a custom frontpanel for this computer. However, I'd still like to get these drives working (or find suitable replacements), which is why I'm not modifying my "Double D" disk controller instead of buying the VF II.

There are a number of issues with my drives that I already know about, which would increase the service cost right off the bat:

* The screw-on front panel for one drive is completely broken

* The latching mechanisms for both drives don't work

* The lead screw on one drive is stiff (but less stiff after I lubricated it with machine oil)

* One drive is missing a felt pad. The other felt pad is probably worn

* One drive was missing the head load plastic assembly that presses down on the disk. I designed and 3D printed a replacement (this thing), but the plastic isn't as tough as the original and likely won't last as long

* One drive doesn't read at all, and I don't yet know how reliably the other one writes

Replacements might make more sense. The current enclosure may have been designed with these drives in mind, however. Are the outer diameters of all 8" drives standardized?
 
If the front panel is damaged, the index pulse timing will be affected.

If the front door latches do not work, index pulse timing will be affected.

If index pulse timing is affected, the head load mechanism may not work correctly

If the head load pad is missing or worn-out, head output will be low, drive may be intermittent.

I have a drawer full of used head load pads. If I service your drives, I will replace the head load pads.

The height and width of 8" rackmount floppy drives are standardized. The depth is not standardized. Most rackmount 8" drives enclosures can be used with most any rackmount floppy drives.
 
Last edited:
MicrocomputerSolutions said:
If the front panel is damaged, the index pulse timing will be affected.

If the front door latches do not work, index pulse timing will be affected.

If index pulse timing is affected, lead load mechanism may not work correctly
Click to expand...

What is it about the front panel and latch that affect the index pulse timing? Isn't that a function of the rotational velocity of the magnetic media? If the media makes good contact with the spindle, it'd seem to me that the disk would rotate at the proper speed. The way I close the drive is by lowering the top mechanism and just wedging something in there to keep it closed.
 
The distance between the index pulse sender, and the index pulse sensor is critical.

A properly mounted faceplate and door latch will reproduce the same distance between the index sender and the index sensor every time the door is closed and latched, so the index pulse timing and index pulse length can be adjusted properly to spec.

Closing the door and wedging something to keep the top carrier grip/clamp the disk to rotate it is not going to be the same every time, it is not repeatable. How can you tell whether you have closed the door the same distance as it was when the index was adjusted?

There are specs for when the pulse should appear, and how long it lasts.

The distance between the sender, and the sensor affects the distance that the light travels (and so when the light is detected by the sensor). Rotational speed affect the detectable light pulse. Most drive index pulse sensors have a mask with a slit in it to control the amount of light from the sender that can enter the sensor. Most drives have a pot on the index pulse sensor circuit to adjust the length of the pulse, AFTER you adjust the timing of the pulse, which is done after inspecting, and adjusting the front face panel, and the door latch.

Floppy Drive Alignment can be a simple procedure if you have the correct equipment, know what the proper steps are, the correct order of the steps are, you know how to correct defects on the physical drive and electrical problems that crop up in the sensors and on the logic board. But there is a reason that drive servicing costs what it costs. The equipment is a significant investment (I have over $2000 invest in my drive testing equipment and alignment disks (the disks wear out or get damaged and must be replaced periodically), and it does take years of experience to make the procedure look easy, while servicing an individual drive properly, combined with testing and necessary repair could take hours or days. There is a reason that I have 15-20 drive logic boards sitting on my bench waiting for repair, and a reason that I charge $100 to repair or exchange a defective logic board.
 
Last edited:
Very interesting, thanks. I didn't consider the distance between those two sensors. I'd like to suggest that the distance between the sensors can't be important due to the time-of-light propagation, because the timing difference between an emitter 1mm away and 2mm away is such a miniscule change that it would require scientific equipment to detect it. However, the further away the light source is from the detector, the more defined the shadow cast by the index hole will be when it's projected. Perhaps that would account for the necessity of calibrating to its distance?

Anyway, the part of the drive I'm talking about being broken isn't the internal structure that holds those sensors on this drive, but rather the front plate, which on this drive lets the top armature open and close within a narrow path. If it opens too much the springs pop out and the eject mechanism gets set improperly. On these drives there aren't any electronic components on that armature, so wedging it closed seems to do the job.
 
I've been working on floppy drives since the 70s. On hundreds, or thousands of 5.25" and 8" floppy drives. I can tell you that the distance makes a difference. You aren't talking about a variance in distance of 1mm, it could be 5mm.

When I'm working on a drive, I can pull the door closed to the hard stop, and release it to where the latch holds it, and I can see the timing difference on my test gear. The Service Manuals for drives tells you where to set the door latch at the beginning of the alignment and testing procedure. The door latch is set, BEFORE the first alignment check, which is the index timing and pulse length adjustment.

I run each drive on a computer,at the beginning of the service procedure, and can see a difference between holding the door against the stop AND letting it sit on the latch. I can see when the index is not set right. A drive may able to read and write disks, but not read and write on disks may on another drive. Or it may be able to read a disk from another drive, but when you write with it, the drive the disk was originally written on might not be able to read or write it any more.

You may not think that how the door is held closed makes a difference, BUT you're also the one with two drives that you admit are not working properly.

The way I see it, you don't have two working drives. You need two drives for parts donors to fix your drives. That means it may be easier to look for two good drives, rather than two parts donors to fix what you have. Even if you located two parts donors, and put the parts on your two broken drives, you would still need to get your drives aligned because the index pulse timing on both drives is going to need adjustment.

The other choice would be to use this opportunity to change drive brands to something more reliable. You could take a chance on two matching drives from eBay Sellers, and plan on selling off your broken Siemens drives when you get the replacement drives running. The odds of getting two drives that are running and in alignment off eBay are not good, but if you can get solid drives that need alignment and maybe a little TLC, at least after you get them service, you will know what you have and can go on from there.

I think you could get two untested single-sided drives of a good brand for about $200 total shipped, if you shop off eBay. When you get them there are ways to do an informal test to see if they will generate interchangeable disks. From there, you can decide whether you want to get the drives serviced. Hopefully you can sell your broken drives for at least $25 each.

If you would like, you can shop, then message me, and I'll be happy to take a look at the eBay ad/s and tell you what I think.

I can also sell you some good drives if you like. You'll probably pay me more up front than you'd pay eBay Seller/s for drives in unknown condition, but you would be getting drives of known quality, with a 6-month warranty that you know are working.

The choice is up to you.
 
MicrocomputerSolutions said:
I can tell you that the distance makes a difference. You aren't talking about a variance in distance of 1mm, it could be 5mm.
Click to expand...
Even with 5mm the timing difference would be in the range of 1*10^-11 seconds or thereabouts.. I don't doubt you seen an effect, but that's probably due to light dispersion or some other factor (shadow? as has been mentioned) that *does* vary noticably with that kind of distance.
 
OP,

Here's an example of what you might try looking for on eBay.

http://www.ebay.com/itm/Siemens-FDD...039936?hash=item542a2ff940:g:UZIAAOSwoydWjaBY

It's a Siemens FD-200-8 8" double-side drive for $65 (shipping looks to be a little excessive, but you could try offering the Seller less), plus shipping. It looks pretty clean in the photos. Unlike most double-sided drives, it uses the same ac power connector as single-sided drives, so it will plug right into your drive enclosure without any electrical modifications.

Seller says there a sticker on it that says "scratches disks". The solution could be as simple as cleaning the heads. Head cleaning on a double-sided drive is pretty simple. Take a cotton swab, douse it in alcohol, gentle lift the upper head ONLY HIGH ENOUGH to expose the heads and swab both the lower and upper heads until clean.

While buying any eBay used floppy drive is a risk, you almost can't lose. If the drive is dead, you can salvage the front face plate and door latch parts to fix one of your broken Siemens single-sided drives. If the lower head is bad, you can configure the Upper drive to run as a single-side drive. If the upper head is bad, you can disconnect the double-side index sender or sensor, and run it as a regular single-sided drive.
 
MicrocomputerSolutions said:
OP,

Here's an example of what you might try looking for on eBay.

http://www.ebay.com/itm/Siemens-FDD...039936?hash=item542a2ff940:g:UZIAAOSwoydWjaBY

It's a Siemens FD-200-8 8" double-side drive for $65 (shipping looks to be a little excessive, but you could try offering the Seller less), plus shipping. It looks pretty clean in the photos. Unlike most double-sided drives, it uses the same ac power connector as single-sided drives, so it will plug right into your drive enclosure without any electrical modifications.

Seller says there a sticker on it that says "scratches disks". The solution could be as simple as cleaning the heads. Head cleaning on a double-sided drive is pretty simple. Take a cotton swab, douse it in alcohol, gentle lift the upper head ONLY HIGH ENOUGH to expose the heads and swab both the lower and upper heads until clean.

While buying any eBay used floppy drive is a risk, you almost can't lose. If the drive is dead, you can salvage the front face plate and door latch parts to fix one of your broken Siemens single-sided drives. If the lower head is bad, you can configure the Upper drive to run as a single-side drive. If the upper head is bad, you can disconnect the double-side index sender or sensor, and run it as a regular single-sided drive.
Click to expand...

Thanks, that looks like a pretty good listing! That particular one sold, but I see here there's another just like it...though sadly another instance of the single-sided version.

Are there other issues that could cause scratching of the disk other than dirty heads? On the single-sided ones I bought some felt pads that seem to be suitable replacements for the little felt things that push down on the media.

I might not need a replacement latch, as I'm waiting for a 3D printed one I designed to arrive (https://i.imgur.com/5otsAPb.png) but I could certainly use the front panel. (Of course, the replacement latch wouldn't be as original as a real one.)
 
JNZ said:
Thanks, that looks like a pretty good listing! That particular one sold, but I see here there's another just like it...though sadly another instance of the single-sided version.

Are there other issues that could cause scratching of the disk other than dirty heads? On the single-sided ones I bought some felt pads that seem to be suitable replacements for the little felt things that push down on the media.

I might not need a replacement latch, as I'm waiting for a 3D printed one I designed to arrive (https://i.imgur.com/5otsAPb.png) but I could certainly use the front panel. (Of course, the replacement latch wouldn't be as original as a real one.)
Click to expand...

If the disks themselves are failing, the oxide layer can co.r off creating the appearance of scratches on the disk.

The material that falls off then ends up on the heads, and may scratch the next disk you use (even if it is okay) or cause read errors. At that point you need to clean the heads.

I tried reading some old 5.25 disks from when I was a kid, that my mom had stored in an un-conditoned shed (saw freezing temps in winter and 90F in summer) and even in cleaned, working drives that work fine with other disks I had bought new, most of the disks I tried from that box fall apart and foul the heads.
 
That drive was available for a while. Just bad luck that someone finally picked it up (or maybe someone read my last post and grabbed it up).

The single sided drive that's going for the same price doesn't look to be as original. Someone jammed a different power connector onto the back of it, and it looks like it's had a much harder life.

Oops, looks like it's gone too.
 
I learned from that experience and grabbed the single-sided one! Especially after I saw that some joker wanted $325 for a similar drive.

I looked at the power connector for a bit, and it seems like they didn't modify the existing connector on the PCB, just attached some connectors to it, so I'm thinking all I have to do is pull it off.

Also, if that drive looks like it had a hard life to you, you should see mine! I think these things were used constantly for years, as this machine actively ran a BBS during its lifetime.

Example of the busted-up faceplates:
 
You must log in or register to reply here.
Back
Top