EEVblog Electronics Community Forum
Electronics => Repair => Topic started by: tonycox01 on August 15, 2019, 08:00:04 pm
-
The long running saga of my cursed HP 3325A continues. But I am not going to give up!
After seeing problems in the output amplifier I went back to the PSU that I fixed. I disconnected all boards from the PSU and put a DC electronic load across the +/-15V supplies and started to load the PSU to see how it behaved.
I gradually upped the load and all seemed well until at about 440mA/30V (~12W) the voltage suddenly drops to zero. If I reduce the load the voltage comes back up. This behaviour is quite reproducible.
I'm a bit stumped by this - can anybody advise how to investigate further?
thanks
Tony
-
tonycox01,
Unregulated DC is always present at current limiters Q7 and Q4, when the 3325A is plugged in and the power switch is OFF. Same with the +5 Volt regulator input.
Q8 and Q11 are responsible for turning on the +/- 15 Volt regulators, when the power switch is in the ON position. Note the power switch pulls the turn-on line to ground when depressed. Proper power supply operation depends on that one grounding screw marked “NOTE”, being electrically bonded to the chassis.
Best to run a tap into that hole to clean out the threads and add a lock washer under the screw head.
The power supply turn-on line is held up by way of relay coil K1, sourced from the unregulated DC input to the +5 Volt regulator. This maintains the power supply in the OFF state. Monitor the voltage on this turn-on line during your load testing, to see if it suddenly rises from zero Volts.
You didn't mention anything about a test load on the +5 Volt regulator. Do note that crowbar SCR CR10 gets activated when the unregulated input exceeds approximately 17 Vdc, effectively shorting this unregulated supply. This unregulated DC would be another item to monitor during load testing.
If there are signs of corrosion present, indicating this 3325A has been stored in a damp location, then the integrity of electrical bonding between all castings may be suspect.
RF+ Tech
-
Does this PSU have the power transistors soldered into the board, or are they in sockets? Needless to say, the latter was not one of HP's better ideas.
-
Since I solved this I thought I would follow up - the 3325A PSU has overcurrent protection built in according to a detailed reading of the service manual and my DC load tests were clearly causing it to trip. So I had a working PSU.
After all this the unit was still showing no disply or in fact any signs of life - but signature analysis of the main logic board yielded a small pile of dead TTL ICs (including the main 555 clock for the front keyboard). Its the first time I've tried HP's signature analysis system and I must say I am impressed.
After replacing the IC's the unit came alive but was showing amplitude calibration errors during self test and the output signal was absent. Working through the troubleshooting manual I finally tracked it down to a dead differential amplifier IC on the mixer board. After managing to find a spare and replacing it the unit is now working and after going through the calibration process seems to be working very nicely.
I added a 10MHz internal over controlled oscillator to improve the accuracy and I am very happy with the unit and how it is performing.
I was a *tremendous* struggle to get this thing working and many times I felt like throwing in the skip and moving on. But after chipping away at the manuals and learning an awful lot about mixing circuits (which I didn't expect to have to do) I can say I get a great deal of satisfaction from using this bit of resurrected kit. It wasn't economical to repair, and there are certainly better things out there, but I just love using it!
Tony