Hello BillyBB
beso's answer quite valid and sound advice, but there might be something a little more complex going on.
If you connect a DC coupled oscilloscope to the test points on the device, you might find that whilst you have a general level of 9 volts DC there is a large AC component (a rapidly varying voltage). This would confuse your digital meter into displaying a varying reading. If it's high frequency, then switching the meter to AC won't help. Your connecting an analogue meter is probably the best way to get a general idea as to what's happening, as it will integrate the AC waveform and give you an average although very approximate reading but the oscilloscope is the only way to be certain as to what's going on.
You can easily check for an AC component on a DC voltage by connecting a non-polarised capacitor of a few hundred nano farad in series with one of your analogue meter test leads and set it to the AC range. If you get a reading, then it must be AC that's causing it as the capacitor will block all DC from getting to the meter. This test will only work with your analogue meter as the digital one will have a very high input impedance which will be a disadvantage in this case.
Regards
Peter