I bought three Tektronix TAS 250 oscilloscope with my friends from local seller for only 80 euros, what a bargain! One of them is working perfectly, but two others have some problems with one channel and cursors. So it was time for some troubleshooting. Here is some pictures I’ve taken. Sorry about blurriness in some pictures, they are taken with my phone camera.
Repairing - Cold joint in CH1 BNC connector
In one of three units we got, the CH1 didn’t show the test signal when probed. It should be 2Vpp 2kHz square wave.
The problem was in the resistor connecting BNC connector to main board (yellow wire in CH2 & EXT Trig picture). I measured resistance from center of BNC connector to the circuit board with my DMM and it showed infinite resistance. I desoldered the resistor, measured it and it showed 22ohm, just like CH2. I soldered it back and everything is now working. It must have been cold joint or something.
Repairing - CH1 voltage range selection
The second unit had problems with voltage range selection knob. It really took some effort to find this problem, because display showed selected range as it should. However it wasn’t actually switching voltage range selection relays on main board. We did some basic tests for relays and transistors driving them and they seemd to be ok. Testing also showed that there was signal on the base of transistors too. Actually that same voltage was in all transistors when it should be in only one of them at the time.
So now we had to locate the chip driving those transistors. It was found from the PCB mountent directly behind the front panel which holds all the buttons and knobs. In picture you can see three 74HC574 D-flipflops, two 74HC245 buffers and PIC microcontroller plus some passives. After some probing we found that 74HC574 (UA23) was driving those CH1 relays through connector labeled as JA23. We measured its output pins voltages and they were all logic high no matter in which position voltage range selection knob was.
We had some problems at removing that 20 pin dip from the PCB because we didn’t have side cutters small enough to cut chip legs. What we came up with was to cut VDD pin of old chip and then bend the legs of new chip and solder it on the other side of the PCB. Not so pretty but hey, the unit works now!