Nexus 5 LG with a dodgy power switch.
I had to use a microscope in order to see and solder the new switch in.
Photo shows a pin next to the switch for scale…
Here’s a macro view of the old switch. Soldering to the two pads at the back was entertaining!
After putting the scope back together, I thought, perhaps it would be a good idea to copy the EPROM data off in case they loose their programming/get corrupted.
Date code is week 8 of 1989 which makes them nearly 27 years old. Bit of a struggle getting to them without taking the whole lot to pieces again. But images are now safely copied off…
Dating from around 1989 a piece of test gear that I imported from the US at great expense about 12 years ago. I recently dug it out to try and it was failing it’s self test. The calibration values are stored in non-volatile RAM, which is a memory chip with built in battery.
The repair was successful, although the specifications and size of this machine are pretty dated when compared with my modern Rigol scopes.
Taking the scope apart and removing the main board:
This is the original Dallas NVRAM. I desoldered the chip and put a 28 IC pin socket on the board. Fitted a replacement chip, which was a cheapo Chinese eBay special, so not sure of its age. However the chip socket will make future replacement simple.
On power up. The scope started rebooting at random intervals. So I took the power supply out and checked some of it’s output capacitors. Two were found with very high ESR values and a further one with slightly high value. All were 220uF 50V. The other capacitors tested ok. There was some electrolyte leakage from the worst caps. And slight board damage. So this was cleaned up:
New capacitors fitted:
After a reset and restore to default calibration, all the self tests passed:
Finally the calibration routines were done:
A tin of rosin. Crush some up and mix with isapropyl alcochol to make liquid soldering flux. What a nightmare! Sticks to your hands the bench, anything else it touches. Soap won’t budge it. Used up a bottle of IP alcohol just cleaning the mess up.
Latest repair for another telephone exchange enthusiast. These provided metering pulses for the old analogue exchanges circa 1980’s/90’s Total of 4 blown up chips a 7 segment display and a zener diode needed replacing. Note the PCB track that acted as a fuse!
Nice piece of test equipment with the optional ovened crystal reference.
This was an intermittent issue, where the gate and external standard LEDs would blink wildly and the counter would lock up. Tapping the case would bring it back to life and it tended to settle down after being powered on for a while. However it would still fail at random intervals.
Gerry Sweeney did a video on a similar counter (model 1999) where he concluded that the reference oscillator connector was at fault. Things weren’t so simple with mine!
Taking to the oscilloscope and circuit diagram soon relieved some strange goings on.
The internal reference is used as a source to drive the counters electronics. Even when applying an external reference signal the fault was still occurring. A nice clean sine wave could be seen on the scope from on pin 4 of PL14. IC17 provides signal conditioning and converts the sine wave to a square wave clock. If the fault was present, probing any point on IC17 would reveal a 4 kHz signal superimposed on the negative portion of the wanted 10MHz reference.
Checking the -5.2 V supply rail confirmed that this 4kHz parasitic oscillation was coming from the regulator formed of IC8b Q15 & Q16. A bit of carefully aimed tapping and prodding pinpointed C59, the main reservoir capacitor for this stage, starting or stopping the parasitic almost on demand.
On the bench, as you can see, the ESR on this capacitor is very variable. Tapping or shaking it caused the value to change dramatically.
A new electrolytic has been fitted and the counter is now working perfectly. 🙂
As I mentioned in an earlier post. I have a GPS disciplined frequency source. This controls a dual oven crystal oscillator very precisely against the 1 second pulse from a GPS receiver.
I’ve just finished building a G4HUP kit which will feed this reference signal to my signal generators and frequency counter.
Lot’s of eye straining and tongue at the right angle to solder all the surface mount components!