Image Map Image Map
Page 1 of 2 12 LastLast
Results 1 to 10 of 20

Thread: Tektronix 4054 troubleshooting

  1. #1

    Default Tektronix 4054 troubleshooting

    I opened my Tektronix 4054 to begin troubleshooting last night.

    It last worked in 2000 after I bought it, but it had not been used for many years - and after moving to my new house 10 years ago, it simply popped the main fuse.

    I followed my 1983 4052/54 4052A/54A Technical Data manual to remove the cover and begin troubleshooting, hoping that there would be lots of similarities to the 4052 troubleshooting that I'm still doing - but waiting to receive a replacement Tek custom display IC from an EBAY auction.

    It looks completely different - other than the ALU/MAS/I-O boards.

    And I believe the I-O board is slightly different than the 4052 I-O board, because I have the 4054A upgrade kit which replaces two of the boards including the I-O board, but haven't installed that kit yet.

    Here is a 2000 photo of both my 4054 and 4052 running my port of the Byte Magazine Hidden Line BASIC program:


    Just to the left of the 4052 keyboard you can see a Kraft PC Joystick I successfully rewired to use with the 4052/4054 joystick input.


  2. #2


    I opened my 4054 to start troubleshooting why it immediately blows the main fuse - the power supplies and display boards are completely different than the 4052. And the ALU, MAS and I/O boards are on a tilt out platform above the huge CRT. The third picture shows the display circuits are now on three boards with one extra slot for the Dynamic Graphics board.

    The first photo showed my 4054 left the factory with the 4-slot ROM back pack with serial (which also included a RAM board), the memory upgrade to 64KB, and "Refresh Graphics Opt 30". However - when I bought it, it did not have the Dynamic Graphics board Opt 30, nor the 4-slot backpack, and it has a Reconditioned sticker on the back

    According to the seller, he bought it from Caltech, but it wasn't working.

    As I recall, I bought it in 2000 and all I remember doing to get it working was replacing a couple of DRAMs.


    This photo shows that the ALU/MAS/I-O boards are on a tray that folds over the back of the 4054.

    The power supplies and display board are completely different than the 4052 that I've been repairing.


  3. #3


    This photo is a top view showing the power supplies and display card cage.

    The Display Graphics Opt 30 card would have plugged into the card cage between the two boards that are ribbon cabled together.



  4. #4


    I made progress today - found a bad 420uF 200VDC capacitor in the 5V power supply.

    Looking at the 4054 Service Manual pdf on bitsavers - I picked the following steps to debug the power supplies:

    1 - Chassis mounted Bridge Rectifiers: These are right below the power input module. I photographed the cables to the bridge rectifiers first - so I could ensure the wires removed were replaced correctly. Then I removed the smaller two wires from each rectifier (DC outputs) and tested that rectifier with my Fluke multimeter (any digital multimeter should work too). Put the DMM in ohms mode - or in my case I have a diode check capability - put the positive lead on one of the AC inputs and the negative lead on the 'top' positive DC output. You should get a reading. For meters with diode testing they give a voltage across the diode (slightly under 0.6V for these silicon rectifier diode bridges). Move the positive probe to the other AC input and the negative on the top will read the other diode forward voltage. If both of these show good diodes, test the other two by moving the positive lead to the lower negative DC output terminal and the negative to the two AC inputs (big wires to the transformer) and they should also test good - or you need to replace that bridge rectifier. I found a web site with those Tektronix rectifiers, but mine were all OK. Now replace the DC output wires on the correct terminals of each of the rectifiers.

    2 - Low Voltage Power Supply: This board is hard to get to below the neck of the tube. So remove the 5V Power Supply (photograph all the cable locations including the ribbon cable to make sure you can replace it correctly). There are three cables to remove, then remove the screw on the bracket on top of the 5V supply, then loosen the two thumbscrews below the 5V power supply heatsink on the outside of the case. Now pull the 5V power supply out of the chassis from the rear. Now photograph the big cable to the Low Voltage Power Supply noting that the black shrink wrap is around a couple of the wires on the end under the main fan. Unplug that cable - I temporarily put the end out the hole where the 5V Power Supply was removed so I had easy access to the pins to test the Low Voltage Power Supply. There are two more cables to that board that need to be unplugged elsewhere. First, look for the tiny two wire cable running from the bridge rectifier with two connections on each AC terminal. You will find the tiny connector P5004 near the tape drive - unplug that connector to remove the 20V AC from the I/O board. Second cable is also to the I/O board - P103 to J103. Unplug this ribbon cable from the I/O board (easy to find with the top board assembly rotated back). At this time, you can ohm test the fuses - mine were all OK on the Low Voltage Board. Now I was ready to test the Low Voltage Power Supply outputs - but first I grabbed my Variac transformer and connected it in series to the 4054 power input connector. Since all the regulators are linear on the Low Voltage power supply, I wanted to gradually increase the AC input to this power supply to reform those power supply capacitors. I set the meter for AC voltage and clipped the meter leads on the AC input of one of the three chassis mounted rectifiers. I watched the AC voltage as I slowly increased the Variac dial - over about five minutes (much shorter than times I saw on the web for this trick) to go from 0 to 100 percent on the Variac (my Variac dial goes above 100 percent - I don't recommend that for this capacitor reforming). The AC voltage gradually increased to 20VAC - and the main fuse did not pop. Now I turned off the 4054 power switch and left the Variac on 100 percent. I then carefully clipped a logic analyzer spring clip to each of the Low Voltage Power Supply connector pins on P32 - WITH THE POWER SWITCH OFF. Then I would measure the DC voltage on the DMM when I turned the power switch on, then turned the power switch back off, and waited a minute for the voltages to bleed off and moved the clip to the next pin and repeated test. All the Low Voltage Outputs looked good. I watched the voltage on the +490V pin climb - above +400V DC and continue climbing and turned off the power switch. Now it was time to ohm test the cable to the Primary Interconnect board for shorts on the pins that connect to the output voltages (obviously the ones connected to Ground show shorts - as this board is bolted to the chassis). At this point, I reconnected all the cables and plugged in the 5V power supply and recabled it - turned on power and the main fuse immediately popped.

    3 - 5V Power Supply: It was easy to remove that power supply. I then removed the screws holding the cover - and diode checked the diodes and capacitor check the capacitor on the small board called Primary Interconnect Board in the schematic with that name and Chassis Wiring and labeled 670-5942-00 in that schematic (not to be confused with the Primary Interconnect board with the card cage plugged in). I also checked the big diodes on the 5V Power Supply Board, but couldn't check the two biggest capacitors without removing that board. That removal took awhile - as you also need to remove the power supply heatsink and all the screws bolting the board rectifiers and power transistors to the heatsink and the small power resistor. Once I had that board removed - I quickly determined one capacitor had OK capacitance value - the other was shorted. I removed that capacitor with a fairly large tip on my temperature controlled soldering iron AND (highly recommended on Amazon) Chem-Wik 10-50L 0.100 inch rosin solder removal braid. To remove the solder from the capacitor pads - I clamped the board vertically in a vise, dipped the tip of the copper braid in a small container of Radio Shack rosin soldering flux, put the tip of the braid on the pad touching the capacitor lead and then applied heat with the soldering iron to the top of the braid. In seconds - you can see the solder get sucked into the braid with capillary action. Remove and clean the tip of the soldering iron and then cut off the small piece of braid covered with solder and repeat - several times for each pad until the hole is clear of solder. At that point - try to wiggle the capacitor off the board. If it doesn't move - which it didn't for me, then wick more solder from the holes and try again. For these Sprague capacitors - there are four ground tabs and one positive capacitor tab to remove. I tried a little persuasion by levering the edge of the capacitor from the board while heating the pads next to that point and got the capacitor moving away from the board - and then finally off. I retested that capacitor and it was a dead short. A quick search for that Sprague part number turned up an ebay auction for three of them for $0.99 - and I snapped them up. They should arrive next week, I'll put one in and see if that fixes the 5V Power Supply problem.

    Last edited by nikola-wan; May 5th, 2018 at 04:41 PM.

  5. #5


    Here are some photos from today's repair:

    First the desolder braid I used:
    Chem-Wik Rosin Solder Removal Braid.jpg

    Then the front side of the 5V Power Supply board before removing the board and capacitor - marked bad:
    5V Power Supply board.jpg

    And finally the back of the power supply board before removing the bad capacitor.
    Note - you don't need to remove the aluminum block that the stud rectifiers are bolted to.
    Back of Power Supply Board.jpg


  6. #6

    Default My Tektronix 4054 is working - sort of

    I found a second 5V power supply cap was bad - replaced it and now I get 4054 screens - sort of.

    I need to run calibration on it.

    First screen shows running ROM CRC test with Jos Dreesen's 4052/4054 Diagnostic ROM Pack. I tried increasing the character intensity - all the way up, but getting character dropout.

    ROM CRCs match 4054 V4.1 according to the Tektronix CRC list that Jos supplied with the Diagnostic ROM Pack.

    Memory size is reported on each test - DFFF is 64KB memory, just like my 4052.

    Next screen is long RAM test - it passed


    Last screen is ALU test - it passed too


    Last edited by nikola-wan; May 20th, 2018 at 07:01 PM.

  7. #7

    Default 4054 Analog issue on screen characters

    I typed in a 4052 calibration program that fills the screen with "8" characters by simply pressing User Definable Key 1:

    4 PRINT "8";
    5 RUN

    Here is the output on my 4054 - looks like a bit of ripple (hum bars) while printing each line


    Now I need to input a simple program loop to draw various size rectangles - and see if lines are affected - or only characters.


  8. #8

    Default 4054 lines are also affected by the issue

    I entered this simple program to RDRAW a series of nested squares:

    100 MOVE 0,0
    110 FOR I=1 TO 10
    120 RDRAW 10*I,0
    130 RDRAW 0,10*I
    140 RDRAW -10*I,0
    150 RDRAW 0,-10*I
    160 NEXT I
    170 END

    Here is the result of running that program - lines are also affected by the issue:


    The 4054 calibration instructions start with voltage checks.
    Here is the list of voltage tolerance and allowed ripple on the power supply outputs:

    • Voltage Tolerance Ripple

      +15V (reg.), +/- 0.2%, 5mV p-p max, adjust R156 on Low Voltage board
      -15V (reg.), +/- 0.2%, 5mV p-p max
      +5V (reg.) , +/- 3.0%, 10mV p-p max
      +12V (unreg.), +/- 1.5%, 1.5V p-p max, adjust on I/O board
      -12V (unreg.), +/- 1.5%, 1.5V p-p max
      +20V (unreg.), none, 1.2V p-p max
      -20V (unreg.), none, 1.2V p-p max
      +175V (unreg.), none, 7.0V p-p max
      +290V (reg.), +/- 1%, 100mV p-p max
      +490V (unreg.), none, 5mV p-p max

    Last edited by nikola-wan; May 21st, 2018 at 05:10 AM. Reason: added 12V tolerances

  9. #9

    Default My Tektronix 4054 is really working now!!

    It was a 290V filter capacitor - should be 300uF, but showed 30uF.

    I didn't have the right size but put in a 250uF 250V capacitor - and my 4054 is now working!
    The capacitor only needs to filter 125V - it is between the 175V output and 290V output. There are three caps in series filtering the 490V, 290V and 175V outputs - all rated 300uF 300V.

    Simple test of nested squares and text - no hum bars

    Diagnostic ROM CRC Test

    Even better news - tape drive works and now I can see the menu on the 34 year old "Steve Rice" tape I got from Stan Smith (that I replaced the belt from an Imation DC6250 tape):

    Steve Rice tap autoload menu

    I did not have that menu program - I made custom menu programs.

    This Automatic Tape Directory program must have been an exchange program. It looks for a REM statement at the beginning of each file on the tape and takes the comment as the title of that file - from looking at the listing:

    Automatic Tape Directory menu instructions

    Now I tried an experiment to load a DC600 tape that I had created in 2000 - seemed like the belt was ok, but it wasn't and the tape got tangled

    I thought I was using the backup tape - but it was my 'original'.

    I then took the belt from another DC6250 cartridge and transplanted it into the backup tape and got it working:

    Monty's program tape backup - autoload menu

    Here is the output from running the 4052R11 cartridge verification program with the cartridge plugged in:

    4052R11 Cartridge Verification output

    One more - my 3D Hidden Line plot program from the 70's:

    my 1977 port of Byte Magazine Hidden Line program to the 4051

    Now I should upgrade my 4054 to the 4054A boards

    Last edited by nikola-wan; June 2nd, 2018 at 12:46 PM.

  10. #10

    Default My upgraded 4054A is working too!!

    I followed the instructions in the upgrade kit:
    - removing the board assembly to replace the MAS and I/O boards
    - carefully removing the old ALU ROMs and replacing with new EPROMs
    - removing the extra 32KB of 4116 DRAMs and moving to the new MAS board
    - putting it all back together (tricky, different size screws in some holes)

    Powered it up and the BUSY light stayed on

    Tried the Diagnostic ROM Cartridge - which seemed to run and then stop after clearing the screen

    Finally - it began running and has run fine since - passing all the diagnostic ROM tests

    Here is the CRC list of v1.5 4054A ROMs:

    4052A ROM CRC list with 4052R14 GPIB Enhancement cartridge in slot 2

    After this CRC test was over, the screen cleared and I got this error message:

    4052 R14 ROM Pack is not compatible with this machine.

    I expected that as the new I/O board has a TI 9914 GPIB controller and is supposed to be very fast compared to the original GPIB software I/O design.

    Here is a photo of my "Vipers on Patrol" program - and I added the result of RND(0) which is now 0.89.... which means a real 4054A!!

    Monty's Battlestar Galactica - Vipers on Patrol

    (see my scan of the 405x Ref Guide that came with the upgrade kit on bitsavers)

    Two more photos - I was looking for my Tektronix serial to PC serial cable so I could try to back up some of these tapes to my PC.

    I didn't find that cable - unless it was the short 25pin cable with bare wires

    I did find my 4924 Hard Copy Unit and 4662 eight pen Tektronix plotter that I used to digitize the Vipers for my program

    Tektronix 4924 Hard Copy Unit

    Tektronix 4662 eight pen plotter and digitizer

    Last edited by nikola-wan; June 4th, 2018 at 06:54 AM.

Tags for this Thread


Posting Permissions

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