PDA

View Full Version : Teletek Systemaster



BubbaShakers
November 18th, 2015, 12:48 PM
I recently acquired a complete system on fleabay consisting of the following:



Integrand chassis
Teletek Systemaster (the only board in the system)
Morrow Designs dual 8" floppy (appear to be Shugart 851's based on internet reaserch)
Televideo 925
Several dozen 8" floppies, including several labeled as Teletek system disks.


I originally acquired this setup to get a vintage terminal and drives for my imsai, but after examining the system further (which appears to have been well kept and used at least until '80s/early '90s judging by the disk labels), I'd like to try and get the system going as configured. The terminal is definitely working as I have it hooked up through a usb serial adapter to my iMac currently. As for the system, PS voltages appear to be good, but I'm not getting very far trying to bring it up (I can't get anything from the terminal). I'd be very surprised if the drives work at this stage, so what I'd like to do is verify the systemaster itself is functioning (e.g. get into a system monitor) before worrying about the drives. It's unclear from reading the manual if this is possible, as apparently the system tries to boot from a floppy before the serial port is active.

Another issue is that it's unclear which serial port drives the terminal. I found two versions of the systemaster manual online -- one claims SIOA should be connected to the terminal, the other claims SIOB. What I have tried is the following.


Power up drives and terminal -> Drives spin continuously (assuming this is normal - not familiar with 8" drives).
Power on system -> At this point the rightmost drive led is lit solid (assume these means it is the boot drive and "selected", implying the processor and/or the disk controller are active).
Hit return on terminal repeatedly to determine baud rate (unclear if init happens from bootloader in ROM. Trying it at this point so I can see any errors) -> No response
Insert "System Disk" and close door ->At this point I here a click and drive door is locked, but no other indication of anything happening (e.g. head seeks). On occasion the led will go out momentarily after the click.
Hit return on terminal repeatedly -> Under no circumstances can I get anything out of the terminal no matter how many times I hit enter, regardless I try these steps while hooked up to to SIOA or B. Note that serial and floppy cables are keyed (as well as marked with magic marker by previous owner) so I don't think cable reversal is an issue.


Does anyone have a systemaster? Is there a way to interact with the system at all without a working boot disk? Should you at least get error messages of some kind through serial if there is a problem with the drives/disk? If not, is there a way to "break out" of the floppy boot and get to a monitor?

Thanks,
-Jason

new_castle_j
November 18th, 2015, 08:00 PM
I've got a Teletek Systemaster and some manuals and resources. I haven't tried bringing it up though, maybe I should pull it out and we can work on this project in parallel. Do you have any pictures of the board or at least can you post the version of the onboard ROM that is installed?

BubbaShakers
November 19th, 2015, 05:53 AM
Here is a scan of the card. The backside of the PCB is labeled "Rev 2" and (c) 1982-- don't know if that has any bearing on manual revisions (The two I found online are Rev 2 and 3). The ROM is a 2716 and is labeled STMC00.

One of my standard troubleshooting techniques one old equipment is to reseat chips. When I tried this on the disk controller chip, I found the system would no longer select the drive on startup. Further investigation revealed that some of the pins on the D765AC were not making contact in the socket, and wouldn't even after cleaning with deoxit and reseating several times. I was able to remedy this by sandwiching a machined 40-pin socket (which has thick, straight pins) between the original socket and the D765AC. The system is now behaves as it did when I obtained it (drive select on power up, no other activity).

Thanks,
-Jason

http://www.vintage-computer.com/vcforum/attachment.php?attachmentid=27904&stc=1

new_castle_j
November 19th, 2015, 08:36 AM
The board I have is still in the box from Teletek, unused - still in the plastic heat sealed bag. It has the same EPROM, the enclosed manual says SIO B is the proper one to use for terminal I/O (closest to the floppy connector). The EPROM is meant to boot Teletek's customized CP/M 2.2 revision C00. Do you have that disk? If not, one is enclosed in the box with my board. There is no onboard Monitor program on the Systemaster C00 prom and after dumping the contents of the ROM I also don't see any embedded ASCII indicating a sign on message or anything else. I have a second ROM (2716) for the Systemaster labeled 3D29, it's contents appears to have some embedded ASCII referring to a "system confidence check". I've attached the ROM contents to this post.

I have not tried booting my board yet, but am willing to dig out a chassis and give it a whirl.

I think Erik Klein, this site's founder, has some Teletek resources, maybe he will see this post and can contribute as well. ;)
http://vintage-computer.com/teleteks100.shtml

27905

new_castle_j
November 19th, 2015, 03:43 PM
OK, I took my Systemaster board out of it's sealed bag an have just booted it with the C00 ROM and boot disk. This board was factory fresh from Teletek 33 years ago and never used, everything worked as described in the manual. Here's how it worked for me. I am using SIO B (closest to the floppy controller) and my terminal is set to 9600 8N1

Power On
Insert Disk and close door, red light comes on and drive clunks, red light goes off
Hit return on terminal keyboard (3 times in my case)
Red light on floppy comes on and drive clunks, I am then presented with the CP/M boot message

Next, I tried the ROM labeled 3D29:

Power On
Insert Disk and close door
Message displays on screen: "[02Z80 memory test is successful.
Red light on floppy comes on and drive clunks, no further output to the terminal (I'm guessing this ROM doesn't like the disk and therefore doesn't boot)

Pictures attached

27919 27918


I'm curious now, what other Teletek disks do you have? This board would run CP/M 2.2 CP/M 3 MP/M TurboDOS and Micro Mike's - which is an interesting story in itself.

BubbaShakers
November 19th, 2015, 06:14 PM
I have several disks labeled "Teletek System Disk, Mbasic, SSDD" - all with handwritten labels - none of the disks are originals. The oldest is labeled 1982, the latest from 87. The other disks are unlabeled (some appear to have been recycled from a Decmate), application disks (wordstar and dbase II -- again none original), and several data disks.

This morning I attempted to power up the system again (with the drive case cover removed so I could see if there is any head movement) and once again the disk will not select - on power up both drive leds light dimly and the doors will not lock. If I hit reset on the front panel the lights go off while reset is held down and come on again when it is released. No other signs of life. My board clearly has issues above and beyond the drives. Based on yesterdays results I suspect there are issues with bad sockets that I have made worse by reseating chips.

When you booted CP/M, how long did it take between the time you closed the door and the light went off? On mine, I have had the light go off on a couple of occasions when inserting a system disk, but it seems to do so immediately after the door clicks shut -- doesn't seem to be enough time to have read the disk. And it stays off for a second or two before starting back up again and remaining lit.

I'm curious about the 3D29 ROM. Was this included in the package with the systemaster? Also, what happens if you power on the system with the 3D29 ROM and hit enter a few times with no disk in the drive (and/or just close the drive door with no disk inserted)?

Thanks,
-Jason

BubbaShakers
November 20th, 2015, 07:27 AM
This morning I powered up the system, checked voltages again and noticed my +18V supply was dead, and two fuses (for the +and -18V) on the power supply were blown. Replaced the fuses, powered the system up and checked voltages without any cards. All voltages were present but a bit high (which seems to be normal for a system with no cards present). Powered off, inserted card, powered up and checked +18V supply. Dead. Fuses blown. Removed card and tested continuity on the card with multimeter.

There appears to be a short on the card between pins 2 and 50, and also betweens pin 51 and 50 (which I don't understand since . It seems the ps fuses blow immediately on powerup with this card inserted, so there are no fireworks or unduly hot components on the board (which will make tracking down the short a chore). My guess is one of the tantalum caps, so I will likely start by replacing any that show a short across their pins....

new_castle_j
November 20th, 2015, 07:38 AM
The 3D29 ROM came off of a spare SysteMaster board sitting in a box on the shelf. I believe that it is the boot ROM for TurboDOS. I was reading over some other Teletek materials and recall the mention of a "D" version ROM and that it was for TurboDOS. If I power the system on with the 3D29 ROM and hit enter any number of times without a disk in the drive it does not appear to change the system's behavior, the message "memory test is successful" gets displayed and system halts.

When I booted CP/M, it only took 1 second after closing the drive door until the light went off. In my estimation, the ROM did a quick check of the boot track, then paused to wait for the user to hit enter at the terminal to determine the baud rate. Once baud rate was established, the it automatically proceeded with the rest of the boot. To be clear, the drive select light on mine does not turn on until the drive door is closed, then the head loads just momentarily, and the light goes off.

I would take drive B off the ribbon cable for testing. Do you have the ability to write Dunfield disk images or burn ROMs?

Marty
November 20th, 2015, 07:42 AM
Hi All;

Bubba, "" My guess is one of the tantalum caps, so I will likely start by replacing any that show a short across their pins.... ""

They will all show a short, unless You can carefully take each one out, one at a time and then check it (That cap) for shorts..

THANK YOU Marty

BubbaShakers
November 20th, 2015, 07:49 AM
The only 8" drives I own are the ones now under testing. I do have a Rom burner and a spare 2716, so if I can get past the short issues on the board it may be worth burning the 3D29 rom and trying it just to see if I can get serial output and verify the memory. My worry is that the board may be thoroughly fried at this point...

BubbaShakers
November 20th, 2015, 08:50 AM
Hi All;

Bubba, "" My guess is one of the tantalum caps, so I will likely start by replacing any that show a short across their pins.... ""

They will all show a short, unless You can carefully take each one out, one at a time and then check it (That cap) for shorts..

THANK YOU Marty

Most show high resistance while in circuit, but 2 showed dead shorts - both were connected to the +18V line on leads near pin 2. I removed them both, tested again and sure enough both were shorted out of the circuit. And there is no longer a short between pins +18V (2) and GND (50) on the card with these caps removed. How critical are the values? The removed caps are 1uF @ 35V. I don't have these handy, but do have 2.2uF @ 35V on hand...

Thanks,
-Jason

MicrocomputerSolutions
November 20th, 2015, 09:26 AM
Most show high resistance while in circuit, but 2 showed dead shorts - both were connected to the +18V line on leads near pin 2. I removed them both, tested again and sure enough both were shorted out of the circuit. And there is no longer a short between pins +18V (2) and GND (50) on the card with these caps removed. How critical are the values? The removed caps are 1uF @ 35V. I don't have these handy, but do have 2.2uF @ 35V on hand...

Thanks,
-Jason

The 2.2uf caps should be fine as long as they fit on the board where the old caps were removed from. As long as the rated voltage meets or exceeds the original part/s, and the rated capacitance is not lower, you should be okay. Dead tantalums usually look all burned up. Check the circuit board on both sides for copper trace damage, and solder shorts to other traces.

Integrand Research used to be in Central CA. I went there many times in the 80s to pick up orders when I was one of their Dealers. They used to sell a good, reliable product as reasonable prices. I was sad to find out that they shutdown a few years ago when I went looking for them to see if I could get some new S-100 Enclosures manufactured. The Company shutdown years ago, and another company is still making some of their PC based products in a different location. Integrand Research was unique in that they used to manufacture all major components themselves, unlike most other manufacturers. They bent, painted, and assembled the sheet metal themselves. They designed, manufacturered (including making the transformer plates, winding, and coating the transformer windings) the power supply transformers. And, they etched, and assembled the circuit boards (like motherboards) for S-100 enclosures. I sold a lot of their drive enclosures (more than their S-100 enclosures) and never had a single warranty claim for any of their products that I sold.

MicrocomputerSolutions
November 20th, 2015, 09:37 AM
The only 8" drives I own are the ones now under testing. I do have a Rom burner and a spare 2716, so if I can get past the short issues on the board it may be worth burning the 3D29 rom and trying it just to see if I can get serial output and verify the memory. My worry is that the board may be thoroughly fried at this point...

Hopefully replacing the burned caps will restore the processor board.

If you take the top off the drive enclosure (with the ac plug removed from the wall), you can reach the shaft of the stepper motor and rotate it to move the head carriage in towards the spindle. When you power the system up, insert a disk, and close the door, the drive select light should come on, the door lock will activate, and the head carriage should step out to Track 00, if the processor board, and drive: 0 is functioning. With 8" drives, do not turn the power off with disk/s inserted in the drives, otherwise you can glitch the disk/s when the power goes off. Always open the drive doors, and pull the disk/s partially out. If the drive select light is on, and the door is locked, trying pressing and holding the Reset Button on the system enclosure while pressing the door release on the drives.

BubbaShakers
November 20th, 2015, 06:04 PM
It is a bit strange I could find no sign of physical damage on the board or any of the caps. The shorted caps did not even overheat at all when the second set of fuses blew during testing (these were 1A, the originals were 3A). I have had tantalums below before on other systems - they usually annouce their failure pretty dramatically.


So after replacing the caps and a new set of 3A fuses I am now back where I started (drive A appears to select at bootup and will lock if a disk is inserted, but nothing else happens). I tried manually moving the stepper motor to the spindle before powering up the drive. When the system is powered up, the head does step out to track 0 but nothing else happens. I discovered the hold-reset-to-eject trick early on, but did power down the system (not the drives) a couple of times during testing before I saw your note. In any event I have at least 7 system disks, all display the same behavior so assuming these really are boot disks I'm inclined not to think it's an issue with the media.


It will be Monday before I can look at this system again, but next I may try removing the other drive (the drive that seems to be active on startup is first on the ribbon cable - the other one is on the end -- wasn't sure if there are termination issues / rejumpering required if removing the other drive).


One other thing I noticed is that according to the manual, the serial port requires active hardware handshake lines. The tvi925 has an internal switch that allows DCR, DCD and DTR lines to be individually enabled. The terminal had DTR enabled but the others disabled. I tried enabling the other two and now I will sometimes see an @ symbol or ERR on the terminal when the system is powered up. The terminal is set to 19200 baud, which is supposedly supported by the SIO ports on the systemaster.

Thanks,
-Jason

BubbaShakers
November 23rd, 2015, 10:26 AM
Had some time to try some additional things this morning:



Tried removing Drive B from the ribbon cable (and reseating the cable on A) -> no change (Drive a selects, no other activity)
Tried burning 3D29 ROM and booting system with it-> Both drive LEDs light dimly. Drive does not select, lock, or attempt to read with disk inserted. No serial activity.


The next thing I tried (after replacing the original ROM and hooking drive B back up) was gently cleaning the drive heads on drive A with a qtip and isopropyl alcohol. Now, when I insert any disk labeled as a system disk, I can see the drive head start to advance to the next track, then the drive head immediately disengages, the drive door unlocks and the drive led either dims or remains lit (about 90% time it dims). This all happens in < 1 second after closing the drive door. If I insert a blank disk or some other disk not marked as a system disk, the same thing happens, except I never see the head try to advance.

It's unclear whether new behavior is a sign a progress (it appears to be trying to read the disk at least, whereas before it would ignore it), or whether my cleaning attempt has buggered the heads.

Marty
November 23rd, 2015, 10:44 AM
Hi All;

Jason, two things..

One)..
"" The terminal is set to 19200 baud, which is supposedly supported by the SIO ports on the systemaster. ""
I wouldn't go that High, until You have it working at a lower Baud rate..
9600 Baud is about tops for these old systems, some had 19200, But, I would NOT assume so for Now..
Two)..
"" whether my cleaning attempt has buggered the heads. ""
Actually, it's probably just the opposite, it is Now trying to Read that Track, I would suggest doing another cleaning..
And try again..

THANK YOU Marty

new_castle_j
November 23rd, 2015, 03:00 PM
I would continue to do some testing with the 3D29 ROM since it will output a message to the terminal at 9600 8N1 without any prior user input. Is the terminal known to be good?

BubbaShakers
November 24th, 2015, 12:55 PM
After having spent more time with this it's clear the board still has some serious issues:



Tried setting the terminal down to 9600 and even 1200 baud (always 8N1).
Tried enabling/disabling hardware handshake lines on the terminal (via dip switches).
On the off chance there are issues with the handshake lines in either the terminal or the cable, tried trying the RTS, CTS and DTR lines on SIOB together as described in the manual for cases where the terminal does not provide them (note that there is apparently a software utility that can configure the serial port post-boot, and when I received the terminal the DTR line was disabled).
Tried replacing the 89 and 88 serial driver IC's (along with several others that appeared to behave intermittently when reseated -- see below) as I read elsewhere on this forum these tend to get fried by static.
Tried the above with both ROMs.
There were 2 serial cables provided with this system. Tried both, and in both ports.



Under no circumstances can I get anything on the terminal other than a single '@' sign, or very rarely an 'ERR' when the system is initially powered up (never on reset) - and only when hw handshake lines are active. The terminal definitely works as I can use it as a serial console for my OS X box via an FTDI USB serial adapter (getting that whole setup working was an interesting project itself).

During my testing it's become clear the drive behavior is erratic, and seems to be tied to IC issues on the card. When I try the 3D29 ROM I consistently never get any drive activity, and when I switched back to the C00 ROM this morning after testing a bit with the 3D29 ROM, the drives would no longer try to read as they did yesterday (no matter how many times I tried to reseat the ROM, drive cables, etc). Sometimes both drive LED's light solid and the drives will not try to read disks when inserted. Sometimes both drives will remain dimly lit (and won't lock the doors or try to read disks when inserted). Other times Drive A will engage (but not try to read as near as I can tell). When this happens drive B's LED may be dim, off, or flickering at random. All of these different behaviors are brought about by reseating and/or replacing chips on the card. E.g. I can remove a chip and reinsert it (e.g. U54) -- checking each pin for continuity before and after the reseat -- yet get different drive behavior before and after the reaset (even after cleaning the pins with deoxit).


I'm guessing there may be one or more marginal solder joints or an intermittent short somewhere on the board thats getting flexed when chips are removed or inserted.


In any event I've decided to throw in the towel on the systemaster before I am tempted to do physical violence to the machine in frustration. Thanks for everyone's assistance.


-Jason

new_castle_j
November 24th, 2015, 04:25 PM
Can I send you one of my spare Systemasters? It's all there except for the voltage regulator since this one came out of a regulated buss. Just pay shipping and it's yours.

BubbaShakers
November 25th, 2015, 01:20 PM
Can I send you one of my spare Systemasters? It's all there except for the voltage regulator since this one came out of a regulated buss. Just pay shipping and it's yours.

If your are willing to part with one, certainly! PM me details.

I assume converting it back to use on a standard bus is just a matter of cutting a trace (or removing a jumper) and adding the voltage regulator for the 5V line?

Thanks again,
-Jason

BubbaShakers
December 3rd, 2015, 12:13 PM
Thanks to new_castle_j's generosity I now have this system up and running. It was necessary to remove 3 jumpers, add two transistors and a voltage regulator on the replacement board, which appears to have been factory built for a regulated bus. The replacement board appears to be a later revision than the old board as well.

After booting the system I was able format a SSDD disk in Drive B:, then backup the boot disk to it. The new backup boots perfectly in Drive A: and I can run apps from it, so it looks like the drives are working at least in SSDD mode. I attempted to format a couple of disks in DSDD mode but it refused to format side one (no disk activity and "Dive not ready" error). I need to look into this further as it's possible the disks I was using were not double sided (are there holes/notches to denote this on the media?).

Thanks for everyone who's helped in getting to this point!

-Jason


http://www.vintage-computer.com/vcforum/attachment.php?attachmentid=28142&stc=1http://www.vintage-computer.com/vcforum/attachment.php?attachmentid=28143&stc=1

simkiss
December 3rd, 2015, 04:07 PM
Congratulations, Jason, that is wonderful!

new_castle_j
December 3rd, 2015, 04:51 PM
I love it when a plan comes together!
Thanks for posting a follow up. Any other plans for this system? CP/M 3 perhaps?

MicrocomputerSolutions
December 7th, 2015, 11:55 PM
Double-sided floppies have the index hole in a different.

On a single sided floppy, the index hole is close to centered inline with the spindle mounting hole, and the center of the floppy on the side towards the door.

One a double sided floppy, the index hole is visibly to the left of center.

Double sided floppy drives have dual index sensors and dual index senders, so they can detect whether a floppy disk is double sided or single sided.

BubbaShakers
December 9th, 2015, 05:55 PM
Double-sided floppies have the index hole in a different.

On a single sided floppy, the index hole is close to centered inline with the spindle mounting hole, and the center of the floppy on the side towards the door.

One a double sided floppy, the index hole is visibly to the left of center.

Double sided floppy drives have dual index sensors and dual index senders, so they can detect whether a floppy disk is double sided or single sided.

I spent some time exploring the disks the came with the system. Only one is actually DSDD -- the rest are SSDD (though most are not labeled as such). Apparently this confused the previous owner as well, as a few of the disks have handwritten labels "will not format side 1 - system problem? (all of these are SSDD by the index holes)." Of the 80 or so disks that came with this system:



Roughly a 1/3 are DEC branded unused blanks.
Roughly a 1/3 appear to be recycled disks from a DECmate.
The remainder are essentially all various generational copies of the boot disk newcastle_j sent me, plus mbasic, and various BASIC programs and data files for managing a high school volleyball team. The system appears to have been primarily used for this purpose from 1982-1991. These disks all contain an incomplete and misconfigured version of wordstar 3 (wrong terminal), and some have non-functioning copies of supercalc as well.


As a next step I would like at some point try to get some of my CP/M software running on this machine (all of it downloaded from various archives and running in z80-pack or the AppliCard in my Apple //e). The only other systems I have that can handle floppies are Apple II's, which won't be of much use here. I am mulling trying something like the following:


Connect my Mac to this system via FTDI USB adapter + a terminal emulator like ZOC (which can actually emulate the TVI925).
Open file to be x-fered (probably start with Kermit-80 or something similar) with a hex editor and copy the hex dump contents to a new file.
Insert CR/LF's after each byte via BBedit or scripting foo.
Fire up DDT, and "type" the program into memory one byte at a time by pasting the entire hex dump file into the terminal session.
Save the memory image to disk.

I haven't played much with DDT, but it looks like I may be able to do something like the above. I use a similar process to load programs into my IMSAI (in that case I use a monitor in ROM which has an intel hex loader and x-modem routines). It may be a while until I get a chance to fiddle with this system further.

Thanks,
-Jason

Chuck(G)
December 9th, 2015, 06:15 PM
So, do you also have TurboDOS for this?

MicrocomputerSolutions
December 10th, 2015, 10:52 PM
The TVI 925 can emulate other terminals. Have you tried resetting the terminal personality to see if you can match it to the WS that you have?