EB5AGV's Workbench: Tektronix TDS-524A repair

A color two channel digitizing oscilloscope

Tektronix TDS-524A

This page is devoted to my Tektronix TDS-524A digitizing oscilloscope.

As usual, I got it nonworking. Here you have the steps I took to get it back to life.

Here you have the characteristics of this unit, taken from the Tektronix WEB

Features & Benefits

500 MHz Bandwidth
Sample Rates to 500 MS/sec with up to 100 GS/sec Equivalent Time (ET) Sampling
2+2 Input Channels
Pulse Width, 1 ns Glitch; Runt, Pattern, State, Triggering
1 mV/div -10 V/div Sensitivity
Infinite and Variable Persistence
Option 1M: Record Lengths to 50K points per channel
8-Bit Vertical Resolution and up to 13 Bits with HiRes
3.5 Inch DOS Floppy Disk Drive (Standard)
Vertical Accuracy to 1%
Tek TriStar™ (DSP) Processor
Slew Rate, Setup & Hold Violation, and Time-Out Triggers
Waveform Pass/Fail Testing
Advanced Signal Processing Functions
25 Automatic Measurements
Full GPIB Programmability
Desktop Publishing Outputs
Segmentable Acquisition Memory
VGA Output
Direct Readout in "Amperes" and "Watts" using P5200 and TCP202 Probes
Channel Deskew
HP Deskjet Printer Support

I hope you enjoy the pictures and comments on this page!

Unit as received

The oscilloscope was described by seller as non-powering up... and so it was not a surprise when I got it and did nothing when power button was pressed.

Here you have some pictures of the unit and its innards:

This is the control board:

This is the acquisition board:

And, below the control board, you find the power supply and the display system:

Power supply repair (I)

As fuse was OK and power supply didn't start, I got it out of the unit.

This is a top view:

And this is a bottom view:

I found that there had been some components replaced but work was nicely done and those components seemed to be OK.

This is the manufacturer label on the power supply:

And this other label shows its characteristics:

Thanks to some other Tektronix fans, I got the needed power supply schematics. As the power supply didn't even start, I checked the STANDBY SUPPLY section and, yes, there was a shorted transistor!

And the good thing is that it was a common (in Europe at least) BU-508A... it could not be so easy, right?

I got a replacement next day, thanks to my friend Vicente. This is the unit he gave me:

I replaced it, reinstalled the power supply on the unit, temporarily put everything else on the oscilloscope, powered it... and it started!!!. But just for three seconds. Then it died again :-(

And, as bad things come in groups, when I left the unit aside until I get some more spares, I found this on the display:

This was not visible without an strong front light, so I had not noticed it... asking on the Tek forum I learnt that this was a problem with a kind of gel used between the CRT and the LCD shutter (the so called Nucolor system) which had deteriorated with age and so display was deemed as trash, except that the color shutter could be removed and then use the unit as monochrome, or run it on an external display by means of the rear panel VGA connector. Some more adventurous people had also developed an interface to a color LCD, but of course that option was not an immediate one for me to do. Well, live to learn!

Power supply repair (II)

Checking the schematic I suspected that the problem could be in some bidirectional overvoltage protectors which were fitted on the collector of the BU-508A. If they failed, the transistor would die by overvoltage. So I ordered them and also some BU-508A... and waited for some days until they arrived and then I installed them on the power supply:

Upon power up, the power supply kept working!!!

And then I got a blue flash on display (note the horrendous defective gel effect), but unit refused to boot.

So something should be wrong... I tried several diagnostics charts on the service manual, but they soon ended replacing half scope!

Searching on Tek forums I read something about a D1 BUS BOARD being connected upside down on a TDS-540... and, yes, it was also my case :-)!. I had rushed to check the power supply and had not noticed it was fitted the wrong way. So I put it right... and the unit booted without any error!

As internal display was unusable, I connected an external TFT display, and got this:

This is what could be seen on the internal display:

Not a pretty sight :-(!

The good thing is that the unit seemed to work fine!

So, all in all, it was still a good unit... but, what a pity, it would be perfect if the display was OK!

Working on the display

I am pretty difficult to convince that something is impossible if I don't try it... So, well, I had nothing to lose (except some time), so I dismatled the unit to get the CRT + LCD shutter assembly out.

Fisrt, I got the front bezel out:

Then the front panel controls assembly:

And then I took out the CRT. It had some type of rubber-like material which attached the LCD shutter assembly to the CRT. It was very easily cut so I removed some of it:

Digging a bit more I found the gel material!

So I did the same for all the frame, cutting just the rear side of the rubber. There are plenty of wires running inside it so you need to be very careful!

This is the top part with the rubber cut and removed:

Using always a plastic tool, in order to prevent scratching the delicate LCD assembly, I got some gel out:

Working with that tool all around the space were the gel was located, with lots of care, I was able to sepparate both parts... what a mess!!!

I knew that it would be possible to remove that from the CRT, as it was glass... but, what about the LCD???

I began with the CRT:

Eventually, I got it cleaned!

I used a plastic piece with sharp corners to peel off crud and then, for final cleaning, paper towels with alcohol. The gel is not fully disolved by alcohol but it helps to clean the glass. It is not easy but doable.

So now it was the turn of the LCD... first, with the plastic tool and very carefully, I got out the maximum amount of sticky gel:

And then I used cotton swabs with alcohol to, patiently (it took a couple hours!), remove little by little the gel:

And this is the result!

OK, so now I had two sepparate parts:

How could I rejoin them?. As I left some of the rubber part, the LCD shutter could be kept as same distance that it had originally. But it was not attached to the CRT... so I used some tape to keep it in place and then added some rubber foam bands to the display frame so it would press it against the CRT. Result is pretty decent:

OK, mechanically sound but... would it work?...

YES, IT DID!!!

As you can imagine, I am very happy :-)!

I have also replaced SMD capacitors, which are very prone to leak. Fortunately, this unit seems to have very low hours on it, so they had not yet started to leak.

I will document the procedure here in short time.