Tektronix 7904 Oscilloscope – Power Supply Repair

Been looking for a 7844 for a while, but this one came in by surprise. I went there just to see it, as it was posted as a working unit and for sale. Was not sure if I wanted to pick it up. Anyways went there to check it out, and found its dead.That made me pick it up as I never explored a 7000 series before. (Not that I have done rest all, 😉 but the 7000 series are so popular I wanted to see whats inside). Being faulty made me more interested in it. As usual, came back home with a scope in the back of my truck, (again !!), wondering what to expect and what is the damage or did I buy an expensive paper weight.

Reference @ http://w140.com/tekwiki/wiki/7904

Safety first –

 !!!! DANGER !!!!

THIS IS INSTRUMENT USES DIRECT RECTIFIED LINE VOLTAGE INSIDE ITS SMPS. BE EXTRA CAREFUL WHILE WORKING AS THE MAIN FUSE IS THE ONLY PROTECTION AND IT IS NOT MEANT TO PROTECT AGAINST SHOCK, SO BY THE TIME IT BURNS OUT, YOU WILL BE A “WAY TOO WELL  WELL DONE” STEAK. HIGHLY RECOMMENDED TO USE AN ISOLATION TRANSFORMER AND DO TAKE ALL SAFETY PRECAUTIONS. ANODE OF CRT IS AT 21KV. DO FOLLOW INSTRUCTIONS AND DISCHARGE CRT ANODE CONNECTION BEFORE WORKING ON THE SMPS UNIT.

The whole exercise is available as a video, here –

 

 

For those like to read, here is the process.

Here how it was when I acquired it. Someone kept an Ink cartridge on top of the scope, which leaked out, in and around the scope. Otherwise the physical condition was okay.

 

 

 

As always, first task is to clean it, here is the unit after cleaning and before proceeding, quick  anatomy of the Tektronix 7904.

From Top

IN-5

LHSIn-4

RHS

IN-3

Now in to my lab for further investigation and fixing.

 

 

 

No power is a simple problem to solve. This is a SMPS and is showing classic symptom of trying to power on and failing. Tektronix documentation calls it ” burst mode”. The video shows  the sound signature of the problem.

The power supply is a switch mode on the first side, and that also generates EHT for the CRT, which is multiplied by a voltage tripler, also housed inside the SMPS unit. The SMPS feeds second linear regulator, called as LV Regulator in the documentation, which supplies another set of working voltages.

Removing SMPS from the unit is easy, a whole bunch of connectors to disconnect, and screws behind and pull it out from back. Remember the orientation of each one of the connectors (pine 1, there is a mark on the PCB and connector)  and also there are two connectors  deep inside LV board. Do discharge CRT anode.

IMG_1589

SMPS on bench. The top PCB is the LV regulator or the second stage of power supply, linear regulator.

IMG_1591

Since the power supply is not powering on, its not the linear regulator, rather the SMPS section which is faulty. You need to remove the casing around the SMPS to reach the internals of the switch mode unit and HV tripler.

WARNING: The main filter capacitor takes time to bleed and discharge, even if you disconnect power, the caps may still hold the charge. There is a neon on the inverter PCB which will flash if the caps are charged.

IMG_1597

Here is the schematic locationScreen Shot 2018-11-24 at 12.00.35 AM

The internal pictures of the SMPS unit is posted in Tek wiki, I am not repeating them here. Lets head straight to the problem. This is one rare instance where I could visually identify the problem from the rectifier/capacitor board.

See if you can spot it,

IMG_1593

If you can not, here is a clue, the  blue capacitor on the left, standard tantalum problem, shorted and hot, hence color change.IMG_1599

The reason for this color change is, typically once its shorted, it gets hot, and does show up in different color. This may vary from type and make, but this one was easy one for me. Confirmed it as dead short with meter.

There is a row of such tantalum in the SMPS secondary as filters. I see a mix of electrolytic and tantalum, not sure why, may be to save space and low ESR ?

To test, I replaced the capacitor with”color” {{ NO racist attitude here, I promise}} and here is the result {{ sorry for the bad quality, picture taken at midnight as always}}

IMG_1619

Now here is the section in the schematics where these capacitors are placed.

Screen Shot 2018-11-24 at 12.33.33 AM

Removed the power supply again and replaced all tantalum capacitors on the output filter section. I did see more tantalum on other stages/feedback/oscillator control etc. replaced all of them as I didn’t want to come back in again  soon.

And here is the treasure collected after the tantalum hunt

IMG_1647

{{ The fuse was blown by mistake, by a stupid short circuit, totally ashamed to describe here}}

There are two burned resistors – These are the one sitting between the sense lines of the LV power supply and power out.  They got burned as I missed to plug in one connector So here are few important points to remember

  • Remember – Connect all cables to power supply and ensure proper polarity of the connector
  • Power supply will not work unless connected to the load/aka scope. Do connect all cables and ensure they are properly seated. If they are not plugged properly the resistors between sense and power out can burn out as the sense line will draw power instead of main power out – if the power out connector is disconnected or improperly seated.
  • HV Connector – Be careful, discharge CRT every time you disconnect.
  • On the left side of the SMPS, its too close to A4 logic board. Do not throw the transistors on the A4 board while you try to connect the cables.
  • Take care.

So its all done, and here is a happy 7904, my first 7000 series.

IMG_1697

Disclaimer
I have no commercial affiliation with any of the products/organizations/individuals mentioned in this blog.
The information provided here is for educational purpose only.
You are free to distribute this as long as it stays original with all information as is here and it is free and you don’t scavenge any tubes from any scopes or this blog.

NO electrons were harmed during the repair/filming  of this instrument restoration. All free electrons found inside the unit were  rehabilitated to the nearest vacuum tube.

NO EXTRA SCREWS OR PARTS WERE OFFICIALLY FOUND AFTER REASSEMBLY. UNOFFICIAL EXTRA ITEMS WERE DISPOSED OFF SECRETLY AND DECLARED AS “EXCESSIVE ASSEMBLY” DURING MANUFACTURE.

Please do report any errors or stupidity.

=== THE END ===

Back to Home Page for more restoration projects

13 thoughts on “Tektronix 7904 Oscilloscope – Power Supply Repair

    1. I used electrolytics. They are not an ideal replacement as Tantalums have low ESR, but it was okay. Ideally there should be a low value ceramic in parallel for HF Decoupling to match tantalum.

      Like

  1. I’m seeing C1353 and C1354 on the schematic at 100uf. Is this the main culprit? Dark blue capacitor on mine also. It’s showing a short across one of the capacitor and had the “dead” power supply sound when I fired it up after buying it at a ham radio auction . The board is not labeled not very helpful and the capacitor says K47 +20. I see a parts list for mechanical parts in the manual but nothing corresponding to circuit board parts. Could you list the values you replaced with ? Not a technician but my Dad is a Certified Electronic Technician and I am trying to make the job easier by correctly identifying the bad parts so I can either replace with a substitute electrolytic or order tantalums if I can find them.

    Like

    1. Remove the cap(s) from the board and test for shorts ( or lift one lead up and test).
      Ensure you are referring to the manual for the correct model/SN Range. You can check the Tekwiki manual here – https://w140.com/Tek_7904_OCRed_by_Tabalabs.pdf

      Both C1353 and C1354 are 47uF/20V. Replace with something like 47uF(33v). Electrolytic will do, however if you go the tantalum path increase the voltage rating to 33 or 50V.
      Also be ***careful***, this is a switching power supply + HV unit. It contains lethal voltages.

      Like

  2. Replaced all those caps. Ran the functional check as laid out in service manual. Get a pulsating flashing with the 1khz calibrator signal. I can press the beam finder and see the little blip/line on screen but nothing else. No trace. Check the the output signals on smps power supply pins labeled +5,-5,+15-15,+50,-50, etc. all reading correct. Checked signals on z-axis board and those read correct also. Any suggestion on what to check next?

    Like

    1. Good. check your plug-ins, trigger settings, ensure all plug-ins are seated properly. Measure the voltage on the horizontal & vertical deflection plates of the CRT. That will point you to the problem.

      Like

  3. Got it working. Thanks for all the help. I now have vertical and horizontal working right. Picked up an original instruction manual w/ schematics which has helped a lot. This scope has been a fun little project and I’m surprised by the quality given the age of the scope. I picked up this scope at a ham radio auction for $50 a couple years back and didn’t know the history of it, if it worked, etc.

    Like

  4. It is a DIRE mistake to “repair” such an old unit without replacing ALL the caps.. They cost little to nothing and have far superior specs to old units. Make sure to replace electrolytics in SMPS supplies with units rated at high frequency, not 60 cycles.

    Like

    1. What I have replaced here is not electrolytic, it is rather tantalum. Secondly, no need to replace electrolytic unless they show a leak or high ESR. I have half a dozen tek scopes from 1950-60s and almost all of them are running with original caps, reformed with less than 1mA current to bring them back to spec. You replace what is weak/failed. Ideally on tek board, especially around the 80s era, never replaced those caps, unless they are bad, as you will end up ruing the board in the process.

      Like

  5. I have the exact same model for 7 series and I got the same issue. The scope didn’t turn on and it was in the “burst mode”. Thanks to this website and LAZYELECTRONS I was able to troubleshoot the scope. Interestingly, the exact tantalum capacitor had failed and it was causing the issue. Therefore, I changed all of the K47 +20 capacitors and replaced them with 47 µF Conformal Coated Tantalum Capacitors 25 V Radial. I put the link here just in case someone is interested in:

    https://www.digikey.com/en/products/detail/kemet/T350K476M025AT/4476342

    I haven’t changed the other capacitors but I think at some point I need to change them as well. I pray every time I want to turn the scope on!

    Time to check the plug-ins, trigger and other settings since I am a newbie and trying to get this scope functioning again.

    Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s