• Please review our updated Terms and Rules here

4P Floppy Frustration

NF6X

NF6X

Veteran Member
Joined
Sep 9, 2013
Messages
1,554
Location
Riverside, CA, USA
I'm working on a model 4P that I just acquired from eBay. If you've seen my recent disjointed ramblings elsewhere, then you may recall that I have no bootable floppies for it yet, and rather than taking the easy (practical, sane...) way out, I'm planning to try to hack together a scheme for writing downloaded floppy images onto real disks on the 4P, using its RS-232 boot mode to download some sort of disk writing program that I'll need to develop. Similar schemes exist for the Color Computer and Apple II, using the Wired/DriveWire and ADTpro tools, respectively.

I'm digging around inside the 4P to take care of preliminary tasks like washing mouse pee out of the case, cleaning up drive heads and card edge contacts, and so forth. I found that drive 0 is no good. Its spindle bearing is shot, and the drive belt is nowhere to be found. I got the bright idea of replacing the original pair of drives with a pair of nice Teac FD-55B drives that I just got, originally planning to use them with one of my Color Computers. They look like they're either unused or refurbished, since they're very clean and they came in what looked like OEM packaging. The detail that makes me think they might be refurbished is that their serial numbers are very close together, yet they have different spindle motor assemblies in them. Same markings on the drive motor PCBs, but different PCB and motor designs. Weird... I wonder if they slapped in motors from different alternate suppliers on a single production run?

Anyway, I had them all bolted in to the drive mounting frame, and then I discovered that they don't fit in the case because their faceplates stick out about 1/8" farther than the faceplates on the original Tandon drives. Nice.

So, what do y'all think I should do next? Punch new holes in the drive mounting frame so that those nice Teac drives will fit in the case? That gives me beltless double-sided drives in the 4P, but it'll make purists and preservationists cringe due to the new mounting holes and different front-panel appearance. I'm not sure how much that bothers me, but it clearly bothers me a little bit since I'm here asking for input. :) Or shall I fix the bad Tandon drive? The spindle bearing may not be a problem if it's a standard size of ball bearing, but I gather that belts for these things can be hard to find nowadays. Or maybe I should look for another good used Tandon drive to replace it?

For now I just swapped the two drives so that drive 0 is the hopefully-good one.

Even though it's a bit frustrating when things don't work or don't fit, I'm having lots of fun playing with this old stuff!
 
Well, you could take a round file and simply elongate the mounting holes a bit. That doesn't have the same ugliness factor as drilling new holes and yet still allows you to mount the old Tandon drives, should you so desire.

Or you could look around for some other FH drives (e.g. MPI, Qume, Shugart, etc.)
 
This is a 4P, so it needs half height drives. I need to measure, but I think that the faceplate thickness is enough that the new screw locations won't overlap the existing holes. If I carefully punch new holes with my Whitney punch then they'll probably be neater than if I try filing the existing holes into slots. Your mileage may vary if you're not as sloppy with a file as I am! :)

I always had the impression that Teac drives like the FD55B were pretty good drives, but I don't know if there's any factual basis to that impression. It seems to me that a direct drive design ought to be more reliable than a belt driven one, since there's no belt to stretch, slip or break.

Well, duh! I just realized that a pair of drives I got for adding second drives to my CoCos are both Tandon drives with the same bezels. One is already installed in a CoCo drive housing, and the other ended up to be bad with an apparent stepper motor or stepper driver problem (head only steps in one direction). I'll try swapping parts between those two bad Tandon drives to see if I can get one good drive out of them.
 
Yah, 4P... I've put Teac FD55s in them many times, you gots to re-drill the stupid holes.

I tend to drill them oversized and put large washers on the screws so you have a little bit of wiggle space to get a nice flush fit.

Don't waste your time with the Tandon TM50s in your 4P, but at the same time, thoroughly test the Teacs with your 4P before going to all the trouble of drilling the holes.

Ian.
 
Well, I think I've made a good drive by taking the motor driver board out of the one with the bad bearing and missing belt, and installing it in the one that would only step in one direction. Reading and writing still need to be verified, but I can spin it up and move the head with my Kryoflux interface. If nothing else, the Kryoflux is handy for checking spindle RPM and maximum track number.

If these Tandon drives give me any more problems, then I'll probably take Ian's advice and put some better (?) drives in the 4P.

Thanks!
 
Be advised that the TEAC 55B drives floppy edge connector is upside down in comparison to the rest of the industry. Meaning pin 1 is on opposite side.
Just FYI
 
I say... elongate the holes!

I have a single 3.5" FDD in my 4p, and a 2.5" IDE HD in the other spot (by way of 5.25 drive adapter with faceplate painted flat black) on the outer location. (the avatar was pre-mod!)

Do whatever floats your boat... but if you want to keep it original, I still have my HH SSDD drives and would be happy/willing to part with them to whomever.
I can only verify that they have booted TRSDOS before they were removed, and have been protected from mice. :p

IOW, fix it up the way YOU want it! :D
 
After using a lot of different HH 360K drives, I settled on the FD55B (actually, the 55BR) as being absolutely solid. One of the best designed "360K" drives made.
 
I wish the 55B drives were more common on eBay. And cheaper, too!

I got my RS-232 bootloader working this afternoon, and I made another posting here about it. Now comes the much harder task of writing the Z80 code for formatting and writing floppies without a DOS to lean on. Since the LSDOS code is published, I'll give that code a good looking-at.
 
Sorry I don't have use of the 5.25s...

Id be happy to build you a CP/M system disk if I could.

Im pretty sure I have an image of the system in file form... tho I would have to check to see if it is the floppy or IDE version.
Attaching the relocator program I built would allow you to pull it down and then move CP/M to the correct spot...

This might be something I should look into... setting up a CP/M configuration that will boot from serial... :cool:
 
Last edited:
That would be convenient, but if I had a way to boot my 4P off a disk then I'd never finish the disk transfer tool. ;)
 
Hell! You're in Riverside! :eek:

If I had a connected 5.25, I could build them and bring them over to you!
Ive been making MANY trips to Loma Linda (chemo) it is not much further!

... but you wouldn't want that anyway... :rolleyes:
That's okay! :D

Oh... have you run the memory test to see if it is all good?
 
How do I run the memory test? I remember seeing something about that, but I didn't find it last time I went looking for the keys to press.

Chemo? Egad! I hope it does the trick!
 
IIRC it is the PERIOD you hold down...
(at least I know something!) :p

When I first got mine I managed to get a TRSDOS disk to run, but a CP/M disk would always hang up...
I found the info about the test but ignored what it showed me... basically if you start it and numbers start flowing down the screen that means there are bad memory locations...

Come to find out that I had bad memory in the area that CP/M was trying to load into! Once I replaced them (moved to bank 2) with new, it came right up!

Meh... they want to do that 'stem cell bone marrow transplant' thing now... At least I'll get to see (parts of) Seattle! (or San Antonio...)
:D
 
Thanks! Yes, it's the period... and I'm seeing errors! Looks like I have a stuck two's bit in a block starting at 8401h, and problems in multiple bits in a couple other spots. I'll try reseating the RAM chips to see if that helps. If not, IIRC, the RAMs are just 4164 chips with Tandy numbers on the top, so I can look for any old 4164 chips? If I find 16 of them, then I'll upgrade to 128k while I'm at it.

That stem cell stuff is interesting. I'm looking forward to the day I can get a prosthetic brain.
 
My replacements weren't 4164s but an equivalent... will have to open it up to remember... perhaps someone will chime in w/ the 'new' number.
Yep... might as well get 16, but I recall some 'tricks' need to be done if it is not a 'gate array' model...

Hopefully a simple reseating will do the trick!
 
Well, reseating the RAMs didn't fix the problem, so I'll hunt for some new RAM chips.

Edited to add: Actually, I should play chip-swap with the existing RAM chips first to make sure I have one or more bad RAM chips but no other problems.
 
Last edited:
You might first want to try moving them around... see if the errors move too... if they don't you may have an addressing or trace problem.
just sayin' :D
 
Yup, I realized I should so that after I posted. Since I was getting lots of errors in bit 1, I swapped it with bit 6. I haven't studied the schematic to see which chip is bit 1 yet, but I guessed that the chips were probably lined up in bit order and then swapped the second one from each end of the row. Most of the errors moved to bit 6, so I definitely have a problem with that chip.

That doesn't rule out other board problems yet, because I see errors come up in other bits sometimes. Sadly, I don't have any 4164 chips in my junk box at the moment. I shall hang my head in shame.

BTW, I found this old usenet posting from 1988 which is informative:

http://www.megalextoria.com/usenet-archive/news072f1/b92/comp/sys/tandy/00000753.html
 
That will likely change once new 150ns chips (I think I got 120s...) are put in!
PRETTY SURE the lowest # chip is bit 1...

At least we can hope so! :D
 
You must log in or register to reply here.
Back
Top