Tektronix 4054A with bicolor DVST - bright orange spot when screen dims issue

I made progress on my 4054A display problem today.

The 4052/4054A Technical Data Manual -03 has NO description of the 4054 High Voltage power supply. But it does describe the 4052 High Voltage power supply starting on page 11-18 to 11-22 and it describes the Storage Backplate (or faceplate), Collimation and Flood Guns on pages 11-15 to 11-17. Even though the 4052 operates with a -4KV cathode, the circuit descriptions were helpful to me to understand what was going on with the 4054 Option 31 High Voltage Power Supply.

Another very helpful document on storage CRTs AND Circuits was on this page https://frank.pocnet.net/other/sos/Tektronix_StorageCathode-rayTubesAndCircuits_1971.pdf

Great descriptions and circuits - and described what the collimation electrodes did - basically electronic lenses that straighten the electron beam from the writing gun to hit the faceplate perpendicular instead of at an angle!

After reading this ebook and several from vintagetek.org, I felt confident my issue was in the Control Grid circuit which output voltage is modulated by the Z-Axis Amplifier to turn the beam on and off. In the case of the 4054/4054A Option 31 Color Dynamic Graphics CRT it also lowers the beam intensity for Write-Thru to prevent Write-Thru from storing on the screen (one of my observed issues).

I started checking the high voltage power supply Control Grid power components and R263 was open instead of reading 1M (see HV Pwr Supply schematic for Option 31 Color CRT in first post in this thread).

This resistor limits the current to the Z-Axis amplifier - and according to my DER EE LCR Meter (just received from Amazon today) is an open circuit but has a couple of pF capacitance. Since the High Voltage power supply to the CRT cathode is a 50KHz switching power supply - I think this caused the Control Grid to malfunction.

Since the Control Grid power is tied to the -6KV cathode - the Control Grid voltage is around -5.8KV based on the 230V spark gap E278 between the Cathode and Control Grid.

I didn't have a 1M 1/2 Watt carbon composition resistor listed in the Tektronix Common Resistor and Capacitor Catalog so I substituted two 1/4 Watt carbon film resistors (I know they have some inductance) a 2.2M and a 1.8M in parallel (which I measured with my DER EE LCR Meter and got 999.1K) and the orange bright spot is gone - and now my CRT BIAS pot can adjust out the dot in the corner of the blinking orange cursor - YIPPEE!!

I do have four 1M 1/2 Watt carbon composition resistors coming from an EBAY seller that I will use to replace my 'bodge'.

I'm now running through the 4054 display calibration steps, hopefully my 4054A will be back in full operation after calibration.

Here is a screenshot of my Mandalorian and Grogu bitmap image - NO Orange spot after the screen dimmed and no orange or green retrace lines between the dots!

I do have some issues with the left and right edges - not completely cleared. That's why I'm going to go through the display calibration again.

Don't worry about the red/orange spot at the top of the photo between Mandalorian and Grogu - I have the 4054A in my theater room and the door to the hall was not completely closed.

One of my learnings in reading the Tektronix ebooks on Storage CRTs is - the flood guns are what keeps the stored image visible and therefore are very important!
The writing gun only works when actually drawing lines or dots on the phosphor - which causes the phosphor to charge slightly positive.
The flood gun current is lower than the writing gun and causes secondary electron emission which makes the phosphor glow green in the stored areas.

Writing beam current for Write-Through is too low to store in the phosphor - which is why the BASIC ROM drawing the blinking cursor or the Option 30 Dynamic Refresh coprocessor drawing the visible Write-Through objects must keep refreshing those objects to remain visible.

During one of my calibration runs yesterday I experimented with increasing the CE-2 and CE-1 collimation adjustments by 20V from 55 and 60 to 75 and 80 respectively.

I did this to improve the flood gun pattern when I first powered on my 4054A after replacing the bad 1M ohm resistor.

Here is the flood gun pattern with the default CE-2 = 55V and CE-1 = 60V adjustment. Notice this pattern does NOT cover the edges of the screen - even after 15 minutes.



The calibration card on my Option 31 CRT just had 55 and 60 written - not whether it was CE-1 or CE-2.

Calibration for 4054 Storage Board step 11 indicates:

"Check for oscillation in the background illumination. If background oscillation is present, readjust CE-1.
CE-1 is typically 5 to 10V higher than CE-2. Note that there may be some interaction between R198 and R197."

CE-2 is the first collimation adjustment in the Storage Board and adjusts the size of the flood gun pattern.
CE-1 is the second collimation adjustment for the brightness of the outer corners of the flood gun pattern.

By increasing CE-2 and CE-1 yesterday I got this fairly even flood gun pattern at 15 minutes after power on - without erasing the screen (other than the slight burn-in of one of the PLOT-10 CAD drawing outlines). This is a screenshot from my iphone in video mode - the brightness was actually much more like the photo above - I'll take another photo today.



I did notice last year after regular use almost every day that the CRT phosphor erased more completely and quicker (5 minutes) - so I will be keeping an eye on this and may reduce the CE-2 and CE-1 flood gun voltages back to the factory settings by recalibrating in a day or two - since increasing the collimation voltages increases the flood gun current and I notice it is harder to adjust the Write-Through brightness and not get storage on those vectors.

Since I have NOT been using my 4054A for very long to avoid burn-in of that bright orange dot - my CRT phosphor may need some time to recover from lack of power on time.

I believe I have fixed by 4054A bright orange spot and weird retrace line issue!

After lots of searching on how the Tektronix direct-view-storage-tube (DVST) works and lots of trial and error troubleshooting, I found a 1M ohm resistor in the High Voltage Power Supply control grid circuit that was open-circuit! I didn't have a 1/2 Watt 1M carbon composition resistor for replacement so temporarily created one with a 2.2M and 1.8M carbon film 1/4W resistors in parallel while I have a proper replacement in shipment.

While reading and rereading storage tube docs on the web, I reread the Tekniques article on the introduction of the Option 31 Color Dynamic Graphics DVST and it indicated that this CRT could display three colors: green for stored vectors, yellow-orange for write-thru or refresh vectors and yellow-green for stored vectors that were overwritten.

I created a BASIC program to test how many colors and got this photo - and I see FIVE different colors:





I see FIVE different colors:

• Orange (first test line)
• Orange-Yellow (second line)
• Green-Yellow (Lime) in the third test line
• Yellow-Green in the fourth test line
• Green - all the stored text

I am puzzled by the last word "refresh" in the lime-green and yellow-green lines is clearly green - although both those lines were overwritten in write-thru mode including "refresh".

The last test line was simply a PRINT statement to store it like the title of the test.


Program attached.

I found out why the "refresh" text was green on two of the test lines by trying to shorten both lines a character at a time I could see that the Option 30 text object must be limited in number of characters or vectors because shortening the lines eliminated the issue.

Here is my latest photo:




When the screen dims after 2 minutes of no keyboard activity - or if you type RINIT to reset the Option 30 coprocessor you only see the bottom three stored text lines because the flood gun current is cut in half to dim the stored vector brightness.

I mentioned that my 4054A still has a little bit of "fade positive" on the edges of the CRT - outside the vector drawing area.

Chris Curtin (Director of Tektronix CRT Engineering for 26 years) just gave me a tip on clearing up this issue - should work on Tektronix 4051, 4052 AND 4054 computers:

Regarding the areas on the CRT that “fade positive” with time, here is the “rework” procedure we would use:
1. Increase the OP Level so those areas are fading positive (a little higher than the level you had before lowering it).
2. Set a trigger so that an erase pulse is generated every 10 or 20 minutes. (i.e., not single-digit minutes and not hours).
3, Check it every 10 or 20 hours, you should see some improvement in 20-40 hours.
Of course, this works best when you are not using the instrument, like overnight.

Step 2 can be accomplished with a simple BASIC program like this one:


Let this program run for 20-40 hours.

My 4054A has been off for almost 4 months until my repair on Monday and I have been leaving it on for about 8 hours a day for the past three days and seeing a little improvement each day - but I suggest anyone with background glowing spots (fade-positive) try this tip from the expert to see if it clears up that issue!

Tektronix Storage CRT phosphor is magical!