Image Map Image Map
Page 2 of 5 FirstFirst 12345 LastLast
Results 11 to 20 of 41

Thread: Tektronix 4052/4054 Multifunction modules available.

  1. #11
    Join Date
    Jul 2012
    Location
    Switzerland
    Posts
    149

    Default

    Here they are...

    TekMFM1.jpg

    ..and another pic with empty PCB and 3D printed casing...

    TekMFM2.jpg

    They are functional, for the RTC part and at least the first 4 ROM images. My eprommer doesn't want to program the 1M Eproms, broken pindriver or something, have to check. I also have not yet received the components ordered, that is why there is a DS12887 RTC module instead of the DS12885 RTC chip ( a DS12887 is a DS12885, potted up with a surprisingly small battery and a 32KHz chrystal). Serial not yet tested, also due to lack of components...

  2. #12

    Default

    Looks great! Can't wait!

    I've received EPROMs and blank PAL and hope to upgrade my DataIO to program big EPROMs too

  3. #13
    Join Date
    Jul 2012
    Location
    Switzerland
    Posts
    149

    Default

    Looks like you will have to wait some more.....

    The DS12885's I received after 5 weeks wait were fake ( some odd memory chip inside...)
    The other Chinese outfit send me 74ls90 instead of 74ls590......

    Ready to give up on Chinese part vendors and just sell the boards !

    Jos

  4. #14

    Default

    Jos,

    I believe that may be the best route. I think I can find all the parts in your list in the first post.

    Monty

  5. #15

    Default

    Saw this 24k 4051 memory board on EBay. https://www.ebay.com/itm/48pcs-TEKTR...Y/353117861256

    Not really related to multifunction boards. But at least a good spare board for 4051 owners. I probably would have bought it but shipping from US is to expensive.

  6. #16

    Default

    In debugging Jos Dreesen's 4052 Multi-Function ROM Pack, I encountered a System Error message this morning, which results in clearing the program from memory.

    I remembered a Tekniques article showed a small program to decode that System Error message - the article is in Tekniques Vol 6 Number 3 on bitsavers.
    http://www.bitsavers.org/.../Tekniqu..._Fall_1982.pdf



    This article uses the 4050 BASIC "POINTER" command to accept one character from the keyboard without requiring a RETURN.

    The interesting thing about the error code I got was it pointed to the seventh ROM Pack in the ROM Expander in the right slot - and I have the MFM ROM Pack plugged in the left slot, so we have an issue right now with the bank switch decoder on the MFM that emulates the bank switch decoder in the 4050E01 ROM Expander.

    Here is the System Error message I got this morning with the MFM ROM Pack plugged into my 4052 in the left slot:



    At the top of the screen I had typed one command:

    CALL "R12LVL"

    This is the command to a Graphics Enhancement R12 ROM Pack to PRINT the version of the firmware.

    The SYSTEM ERROR message at the bottom of the screen indicates that ROM Pack 7 in an eight slot ROM Expander "BE" plugged into the right ROM Pack slot on the back of the 4052 had an error.

    Since I have the MFM ROM Pack which emulates all eight ROM slots in a 4050E01 ROM Expander, and contained the Graphics Enhancement ROM in slot seven, I would have expected the system error message to report "B7" as the location of the ROM Pack with the error.

    I am debugging issues with ROM decoding in Jos Dreesen's new MFM ROM Pack, so this message indicates the 4052 'discovered' the Graphics Enhancement ROM was in slot BE, and likely had an illegal opcode trying to access it for the CALL "R12LVL".

    4052/4054 ROM Pack "discovery" performed immediately by the system BASIC ROM on power up is to write to a ROM bank switch in the system the bank address for a single ROM slot (table in article above) and then fetch two bytes from hex address 0x00010. Every 4052/4054 ROM Pack will respond with "4052" for those two bytes, and if there is no ROM Pack installed in that slot address, the system will ignore any other values and continue discovery with the next possible ROM slot.

    The 4050 ROM Expander (and Jos' MFM ROM Pack) emulate the system bank switch, adding decodes for 8 additional slots within one of the four major slot addresses.

    Each 'ordinary' ROM Pack gets a slot specific "BS" signal from the system bank switch to use to decode the ROM or I/O based on the additional A15 address signal. A15 is high for all I/O and low for all ROMs in a ROM Pack.

    A different and also unique ROM slot signal "BSX" is decoded by the system bank switch to address a (possible) bank switch decoder which is located inside the 4050E01 ROM Expander.
    This BSX signal is then used in the 4050E01 ROM Expander, and Jos' MFM ROM Pack to decode the three LSB bank switch register data bits which further decode to select one of eight ROM Packs in the Expander or MFM.

    After initial ROM Pack "discovery", the system ROM will go back to the discovered ROM Pack locations and read ROM CALL names followed by ROM CALL entry addresses. The system BASIC ROM will add the ROM bank switch address information with the base address of these ROM CALL names for use when this call is made in a program or from the immediate command line.

    The 4052/4054 computers supported up to four ROM slots, each slot capable of supporting an eight-slot 4050E01 ROM Expander for up to 32 ROM Packs!

    Each 4052/4054 ROM Pack could contain up to 16KB of ROM and was typically constructed with up to four 2732 4Kx8 EPROMs.

    The 4052/4054 ROM Pack architecture could also support ROMs plus I/O adapters such as RS-232, Real-Time-Clock, A/D converter, D/A converter, even memory expansion.

    Jos' Multi-function ROM Pack includes not only up to eight ROM slots, but the hardware for the Tektronix RS-232 Printer Interface, and the hardware for the TransEra 741-RTC Real-Time-Clock module!

    The 4052/4054 ROM Pack data bus was only 8-bits wide, but the 4052/4054 computers had a 16-bit bit-slice processor architecture which always fetched 16-bits at a time from ROM or RAM, and the system board handled even or odd starting addresses for program or data and 8-bit I/O devices properly.

    The 4052/4054 has 128KB of address space with 64KB of BASIC ROM (actually 8KB more ROM plus I/O register decode is at the top of RAM space) and 64KB of RAM (minus the 8KB of ROM plus I/O space for 52KB of usable RAM).
    The ROM Packs share a 16KB bank switch space at the bottom of the ROM space. The System ROM contains 16KB of code including RS-232 and other CALLs in the low 16KB space that is bank switched out to support a CALL to an option ROM.

    This bank switch works as the 4052/4054 is single-threaded code, and after an optional ROM Pack CALL is made, the system may or may not have to change the bank switch to run the next step in the program.
    Last edited by nikola-wan; September 20th, 2020 at 04:25 AM.

  7. #17

    Default

    Hi Monty,

    Good work! It keeps you quite busy I have added a 74LS138 into Jos's
    module to use the serial port as the primary port. This is because I don't
    like to cut traces inside my machine... But I still have to test that serial port.
    Then I got the message from Jos that he had updated the GAL program.
    So now I'm waiting for my new programmer to update that chip...

    Very nice to read the progress on the module! Next week I get the tapes
    and some other documentation for the 4052.

    Regards, Roland
    WTB: Case for Altair 8800 ...... Rolands Github projects

  8. #18

    Default

    Quote Originally Posted by Roland Huisman View Post
    Hi Monty,

    Good work! It keeps you quite busy I have added a 74LS138 into Jos's
    module to use the serial port as the primary port. This is because I don't
    like to cut traces inside my machine... But I still have to test that serial port.
    Then I got the message from Jos that he had updated the GAL program.
    So now I'm waiting for my new programmer to update that chip...

    Very nice to read the progress on the module! Next week I get the tapes
    and some other documentation for the 4052.

    Regards, Roland
    Roland,

    I believe I'm dealing with a GAL timing problem on my board.

    I've hooked my HP16500B Logic Analyzer to my MFM board - and it looks like the clock to the 74LS173 is less than 10nsec from the end of the data bus buffer enable. If I had a 25nsec GAL, I might get an additional 10nsec of setup time on the very critical writes to the bank switch register.

    Jos sent an AMI 22V10-AP25, but I can't reprogram it on my Data I/O LogicPak or TL866ii-Pro.

    I purchased Atmel 22V10C-15, which I can program with my TL866ii-Pro inexpensive programmer.

    Best I can do with the Atmel 22V10C installed on my MFM is see a couple of quick option rom calls like CALL "750SPU" to print the version of the TransEra Special Utilities or CALL "R12LVL" which prints the version of the Graphics Enhancement R12 ROM.
    Anything else fails and locks up my 4054A or 4052.

    Also, I have found the slot position matters. On my 4054A I can only get the above results from the right slot - MFM in the left slot locks up with all front lights on
    On my 4052 it is the opposite - MFM in left slot locks up the 4052 with all lights on and right slot works (to the limit of the commands mentioned)

    I was getting a bit farther on the AMI GAL, but may have damaged it trying to reprogram it It is now erased but not programmable.

    You should have better results if you can program the GAL you received with the MFM from Jos.

    I have ordered 3 NOS Lattice 22V10D-25 parts from EBAY, but I've already been burned with three EBAY Lattice 22V10 which cannot be programmed by my programmer - I suspect clone 22V10 ICs that are not programmer compatible

    I also found that Digikey - a US Electronics distributor has about 1000 Atmel 22V10C-30, so I ordered five of them, should get them by Friday to try on my board since my programmer can only program Atmel or Lattice 22V10 GALs.

    I haven't been able to test the RS-232 or TransEra 741-RTC as I'm waiting for the ROMs to work first

    What programmer have you ordered?

    I have made progress on my program that calls the Diagnostic ROM pack CRC - I now can get Jos' Diagnostic ROM pack CALL "CRC" to list the CRCs for ROMs plugged into my 4050E01.

    When I get my new (SLOW) GALs, my program should be able to list those same ROM CRCs in Jos' MFM! Currently, that program shows a CRC equal to no ROM found

    Monty
    Last edited by nikola-wan; September 20th, 2020 at 03:17 PM.

  9. #19
    Join Date
    Jul 2012
    Location
    Switzerland
    Posts
    149

    Default

    Roland,

    The GAL Jedec file ver. C, with checksum 7553, that is on my FTP site is the current one. I am using that on my non-A 4052 and do not see issues. It is however the same version that Monty is having severe problems with. It seems non-A and -A machines have sufficiently different timing to cause issues.

    Monty,
    do try the "slow" option that you can populate on my MFM board. It might very well be the solution. I will also try to come up with a solution to defuse the timing issues.

    All signals BSX-R, BSX-L, BS-R and BS-L are outputs of the same 74s138 decoder. And they are the only difference between right slot and left slot.....The timing must be very critical indeed if that causes issues.

  10. #20

    Default

    Quote Originally Posted by jdreesen View Post
    Roland,

    The GAL Jedec file ver. C, with checksum 7553, that is on my FTP site is the current one. I am using that on my non-A 4052 and do not see issues. It is however the same version that Monty is having severe problems with. It seems non-A and -A machines have sufficiently different timing to cause issues.

    Monty,
    do try the "slow" option that you can populate on my MFM board. It might very well be the solution. I will also try to come up with a solution to defuse the timing issues.

    All signals BSX-R, BSX-L, BS-R and BS-L are outputs of the same 74s138 decoder. And they are the only difference between right slot and left slot.....The timing must be very critical indeed if that causes issues.
    Jos,

    My 4054 is the only one that is A-series, my 4052 is non-A, which is why I suspect my faster GAL timing may be the issue: -15 vs -25 on the GAL you sent me that I cannot reprogram on my Data I/O LogicPak nor TL866ii-Pro, but must have been able to erase while trying to program it

    I have tried my Atmel 22V10C-15 and same EPROM as you sent ST 27C1001-15 on both your original board and my assembled board on both my 4052 and 4054A in both slots with same results

    On my 4054A I can do the CALL "CLIST" and see the TransEra 750SPU, Character & Symbol, and Graphics Enhancement ROMs, but most commands lock up the machine or throw the System Error message.

    Since I can get the MFM to boot in the right slot in the 4054A - I can plug in your Diagnostic ROM pack in the left slot and run my CRC test program on the MFM 'slots' but they all show up with 5000 (all FFs) on this test.

    When I plug my 4050E01 into the 4054A right slot with the Diagnostic ROM pack in the left slot - and insert those actual ROM packs in the same slots as the MFM images, my CRC test program gives the correct checksums for all those ROM packs and they work fine - same thing in the 4052.

    Monty

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •