Have built scientific meter with 18vdc dual rail supply which uses 4-9v batteries.
The voltage is split and fed to 7815 and 7915 regulators. The negative regulator is balking at the 18 volts it is getting. Isn't 3v difference sufficient to get the 7915 to work?
If so, any suggestions/tricks as to what I can do to either hold the voltage up. Batteries show 18v input (open circuit) and 16.5 when in use ( so that should not be enough). Does this mean the 7915 is really substandard? Thanks.

 

Both the regulators and the meter have to use power to operate. That load may be more than your 9 volt batteries can supply.

Can you post up a schematic of your power regulators and give an accurate figure for the current draw of your "scientific meter"?

 

Your negative 15V regulator has an input of only 16.5V which is too low and will get worse as the batteries run down to make only 12V since a little 9V battery quickly drops to only 6V.
Re-plan the circuit to use a lower supply voltage and/or AA cells.

 

The 78xx/79xx series of regulators have an approximate 5.5mA "tax" current that they send from the IN terminal to the GND terminal in order to (somewhat) regulate their output voltage. Their output voltage is not guaranteed until there is an additional 5mA load on the output.

If you are using alkaline 9v batteries, they'll likely measure about 8.6v with a 25mA load on the output (a 360 Ohm resistor).

Alkaline 9v are usually rated for about 500mAh; or 25mA for 20 hours.
NiCD or NiMH are usually rated about 150 to 200mAh, or 7.5mA to 10mA over 20 hours.

However, they will fall below your required input voltage very shortly after you install them.

78xx/79xx and 9v batteries do not play well together.