Hi,
If you scope across a DC power supply (watch out when doing this- one side of your scope is connected to mains ground!) there are two modes of connection. Check out the switch next to the BNC probe connector- it should be marked AC/GND/DC or something similar.
When set to GND it gives the scope a 0V reference so that you can allign your trace to a good zero point. When in DC coupled mode the display will show the amount of DC voltage present across the input. In this example, if you were measuring the 24V power supply, you could have the sensitivity set to 5V/div and if all was well the trace would jump up almost 5 divisions on the graticule.
Now, in a case like this you posibly have an AC signal riding on the DC voltage. Say this signal is 100mV. In the above reading 100mV would be 1/50th of a single division on the scope- the trace would just look slightly blurry- you'd know something was up, but you wouldn't be able to measure its frequency or amplitude accurately.
If you switch the input-coupling switch to AC, this inserts a capacitor in series with the input BNC, and allows you to measure an AC voltage in the presence of DC. With the DC potential being blocked, you can "zoom in" on the AC signal- so you could wack up the y-axis sensitivity and really take a look at the signal.
Apologies if you knew this already, I wasn't sure from your post if you'd tried this!
If it's a zener ref/741/2N3055-or-similar-pass-transistor setup, it will be easy to fix. Just replace the faulty components. What is the case history? Did it suddenly start oscillating, or had a previous fault occured? What is your actual DC output with this fault condition?
Good luck with it!
Mark