• Please review our updated Terms and Rules here

TM-848 8" Floppy Drive Error Codes

MicrocomputerSolutions said:
I don't remember ever seeing sensors wired that way before. Normally, sensors run in parallel, isolated from each other. That makes troubleshooting much easier. Ever try to find the burnt out Xmas light on a string that's wired in series? I would not have thought to test the sensor off the drive. When a sensor doesn't work on the drive, there's no reason to think it would work off the drive (I've never removed a sensor and replaced it with the replacement sensor not fixing the problem). Normally the wires from the sensors run directly to a connector on the logic board, and to logic on the logic board, not from emitter to emitter (on a another sensor) that wired in series.
Click to expand...

If they're the same type sensor, it cuts down on a resistor and unnecessary power drain, so it makes sense.

Put it another way, if an LED in an opto draws 20 ma. at 2V two of them in series are still going draw a total of 20 ma. at 4V., but run independently, they'll draw 40 ma. In the former case, the series resistor need only dissipate 1/5 of the power, while in the second case, you're wasting more power in the resistor than in the LED.

Aside from power consumption and parts count, why else would the designer want to do this? There's a not-so-obvious answer as well.

Consider that, with 8" floppies, one attaches a "write enable" tab to the bottom edge of the jacket, unlike the 5.25" "write protect" tab. So, with a drive smart enough to do power-on diagnostics, how do you test this sensor, not knowing if there's a floppy already in the drive? You can't really--and a failed WP opto sensor will result in writes being allowed even on write-protected floppies--a very serious matter.

However, if the WP emitter is placed in series with the TK00 emitter, a failure of either will result in a failed power-up test and you'll be prevented from using a drive that could well overwriting your valuable write-protected data. Of course, it's always possible for the PT in the sensor to fail, but that's less likely than having the emitter fail.

I suspect this is why 5.25" floppies went to a "write protect" scheme, where a failed sensor would result in an unwritable floppy, but otherwise not jeopardize existing data.

Perhaps too clever by half, but not unusual--and it makes perfect sense.
 
Last edited:
Chuck(G) said:
If they're the same type sensor, it cuts down on a resistor and unnecessary power drain, so it makes sense.

Put it another way, if an LED in an opto draws 20 ma. at 2V two of them in series are still going draw a total of 20 ma. at 4V., but run independently, they'll draw 40 ma. In the former case, the series resistor need only dissipate 1/5 of the power, while in the second case, you're wasting more power in the resistor than in the LED.

Aside from power consumption and parts count, why else would the designer want to do this? There's a not-so-obvious answer as well.

Consider that, with 8" floppies, one attaches a "write enable" tab to the bottom edge of the jacket, unlike the 5.25" "write protect" tab. So, with a drive smart enough to do power-on diagnostics, how do you test this sensor, not knowing if there's a floppy already in the drive? You can't really--and a failed WP opto sensor will result in writes being allowed even on write-protected floppies--a very serious matter.

However, if the WP emitter is placed in series with the TK00 emitter, a failure of either will result in a failed power-up test and you'll be prevented from using a drive that could well overwriting your valuable write-protected data. Of course, it's always possible for the PT in the sensor to fail, but that's less likely than having the emitter fail.

I suspect this is why 5.25" floppies went to a "write protect" scheme, where a failed sensor would result in an unwritable floppy, but otherwise not jeopardize existing data.

Perhaps too clever by half, but not unusual--and it makes perfect sense.
Click to expand...


Different Drive Manufacturers used different methods to detect whether a disk was inserted in the drive. Some drives use a switch or sensor to detect when a disk is present. On other drives Index Pulse being present is used as a indication of a disk being inserted.

I have not had extensive experience working on the Tandon 848E drives. There simply aren't that many of them floating around. Most of the drives in circulation are the older 848-1 and 848-2 with the belt drive ac spindle motors.

Having been a victim of a malfunctioning Write Protect Circuit several times in the past (which resulted in damaged Alignment Disks), I've found that the best defense against a rogue Write Circuit with malfunctioning Write Protect is the use of a Drive Exerciser with a rigid test and observation procedure, before inserting an Alignment Disk into the drive. I don't ever remember seeing Write Protect Sensor Failure that causes a Write Malfunction. But I have seen several cases of Write Circuit malfunction on the logic board that enabled Disk Write all the time, or when it was not supposed to happen. Using a Drive Exerciser and a blank disk to observe the operation of the drive before inserting an Alignment Disk is the best defense against accidentally writing on the Alignment Disk. The Drive Exerciser should allow you to observe the presence of Index Pulse, Write Protect, TK 00, and Read Data while the heads are loads and allow switching heads to see the output of both heads (if double-sided). The presence of Read Data shows that the drive is not writing as a drive cannot both read and write at the same time.

Perhaps if we had seen the introduction of more intelligent drives (like the Tandon 848E), we would have seen more doubled up sensors. But at the collapse of 8" floppy production, most drives were not intelligent, and sensors were singles to make diagnostics easier.
 
That's all well and good, but you never know when an un-monitored sensor will go out, which is what makes the 8" system so treacherous. The 848E's sensing is good, but even on a "dumb" drive, series-connected optos makes sense--if the WP sensor goes out, so will the index sensor--and you won't be able to write to the drive, assuming that you pay attention to things such as the READY/ or INDEX/ signal
 
Last edited:
Chuck(G) said:
That's a Honeywell part (HOAxxxx), but my catalogs don't carry it. Absent finding an old operating one, you may want to consider something like an HOA1887:

RF321930-01.jpg
Click to expand...

Anyone know where I can get one of these Honeywell sensors? The HOA1887. They were discontinued in 2018 and I can’t find any from the usual suppliers.

I just used my last spare Write Protect sensor and I know I need at least 2 more for upcoming drive refurbs. Thanks !
 
I'm not much of a hardware guy so I'd appreciate some advice here.

I have an SMS MDX System (essentially a PDP 11/23) with that manufacture's "Flinchester" controller (floppy+winchester) and a Tandon TM848E drive; I can't get it to boot with my old RX02 RT-11 system disks. The HD is toast, and the floppy drive seeks once but isn't happy. LEDs on the controller indicate "no drives ready". I've tried swapping out the floppy for a TM848-2 and get either that same error or sometimes a "floppy 0 suspect" error (reading the controller's diagnostic LEDs). Trying a third floppy drive produces the same - except it squeals a bit - and now I'm out of things to swap in. I've jumpered everything as per the system manual... so I'm thinking it is a bad controller. Still, the hardware passes its self-test; just the drive test fails - "no drives ready". But I've never see a DRV? prompt - which makes me wonder; I've been booting via reset switch or @773000G.

Suggestions most appreciated!
 
Chuck(G) said:
That's a Honeywell part (HOAxxxx), but my catalogs don't carry it. Absent finding an old operating one, you may want to consider something like an HOA1887:

RF321930-01.jpg
Click to expand...
Hey Chuck. I needed a track 0 sensor for my Tandon 848E and this pointed me in the right direction. This part is no longer available but Digikey recommended part number OPB830W55Z. https://www.digikey.com/en/products/detail/tt-electronics-optek-technology/OPB830W55Z/1015250

I can report it works perfectly and is an exact replacement.

Hopefully this helps others going forward.

Quite interesting I can buy an exact replacement for a 40 plus year old drive.
 
You must log in or register to reply here.
Back
Top