tektronix 2710 spectrum analyzer

Finally, the project is finnished. The 2710 is now working like it should, no errors.
The complete story: with photos: https://www.pa4tim.nl/?p=4522

The first measurement after repair:

Hi PA4TIM, I have watched with great interest all your steps done in order to repair this equipment.
I am also among those users owning such old equipment and trying to fix it. :grin:
For my equipment many problems have been found.
(One lesson leaned: do not to buy untested/old equipment from some electronic repairing companies, as you may get instead a full collection of defective boards…)

The most annoying problems was related to intermittent electrical contact problems at the keyboard PCB. There was a lot of bad solder joints on a surface mount resistors, so I had to do all the re-soldering on this board.
Another problem is related to the elastic membrane used for the keyboard and I must find a way to change it. I think it makes sense to replace this membrane with a series of individual electrical contacts, but there may take a lot of work for positioning and soldering of each unit.

- - - Updated - - -

The big problem I have encountered was classical Display Storage board defective. I understand this is a common problem for older PLS173 type circuits.
Based on a.cz
schematics posted here on edaboard forum: https://www.edaboard.com/threads/17974/
I was able to implement some additional logic circuits in order to simulate faulty signals.

I have noticed that some of PLS circuits heat up very much. To avoid this situation, I have mounted some additional miniature fans on board. For more efficient cooling and I have mounted also a small fan between digital boards and I have drilled supplementary vent holes into the case for extra ventilation.
At this point I think that most of the problems were solved by the improvisation I have made. I have obtained the necessary on screen display and I plan starting calibration procedures this week.

This thing (and especial the PLS and some 74 logig indeed get very hot. I like your solution with the minifans. But isn't that to much load for the , in my opinion, allready stressed 5V. I was thinking about placing a bigger fan on the top of the instrument above the boards so they pull cool air from extra holes in the bottom over the cards. Normaly I do not like that buit the SA get to hot to place something above it anyway. My 2712 also gets very hot.

I have no problems with the switches. Is it just dirt or are the contact pads worn ? maybe you can glue copperfoil under the membrane to get better contact. I have not opend the keyboard but I know people who had problems with keyboards in things like calculators that did not react anymore and they solved this by painting the contacts of the membrame with conducting paint like they use to repair heated glass in cars.

I'm glad my digital problem were not as big as yours. I'm good in analog electronics but not in digital.

Nice sharp pictures!

The additional holes have improved a lot the air flow ventilation and the mini fans draw about 0.08A each, and I think that’s not so much load for the 5V power supply. The keypad it’s completely damaged, when you press some of the key it remain mechanically blocked. I have tried to repair some of the push buttons, but there’s no way to change all of them. I have ordered and expect to receive 50 pcs momentary tact switch (SMD Ultrathin Tactile Switches 5x5mm) to be used as replacement.

I look forward to see the pictures and hope you succeed.

Yes, it's a work in progress but there are some things to solve in first place. :wink:
I have received the tactile switches. I have ordered 3 different switch types in order to use the most suitable as replacement.

- - - Updated - - -

I have prepared the PCB by cutting in half each contact area. I have soldered the switches.
My plan was to keep only the first layer of the keyboard membrane by removing the inner conductive area and the elastic metal membrane included.

No problems up to this moment, but considering Murphy's Law is that anything that can go wrong will go wrong, I have discovered that actually there’s no way you can use the tactile switches under the keyboard membrane. :grin:
To compensate I need to place small spacers in-between. See the attached pictures for more details.

- - - Updated - - -

The easy way to add some “spacers” it’s by using a drop of glue on top of each tactile switch. :-D
Done that and I’m now waiting for the glue to harden in order to do some tests.

Nice solution. Good to know if my keyboard gets a problem. I will cross my fingers for you.
Did all switches stopped working or just a few because I would not expect that. Many keys you almost never use.

Actually, a single broken keyboard switch may remain stuck down, rendering the keyboard useless. :roll:

You are right, I did not think about that. But your solution looks like a good improvement. I think the switches will be more responsive. Now you need to push rather firm.

The keyboard story never end. :lol:
I have found after the switches replacement, there’s a strange keyboard behavior. In a short period of time the keyboard stop responding, and it was strange some of the matrix transistors begin to receive parasitic currents via the pcb.
I think that was the main reason for why they were start misuse this keyboard in the first place. To solve this issue I have replaced all the resistors with lower values, in order to compensate pcb parasitic resistance.
The 22k resistors were replaced with 2.2k and the 100k resistors by 9.1k resistors. After that I have tested for a long period of time and it seems the problem was solved. Starting now I may focus on most important activities, as starting the calibration of the unit.

I have the R&S AFGU function generator. That looks like it is from the same series as your signal generator. Nice SA on the left. Is it HP/Agilent ? A friend has a HP scope that looks like that.



I see your 2710 also has the TG. I have a 2712 too, but without the TG. I was thinking about swapping the tracking generator from the 2710 to the 2712 but I have not examined if this is possible.