Simple DC voltage selector circuit to drive motor

Hi All
I am new to this forum but hopoe some of you may be able to help me, thanks in advance if you can.
I need to design DC a voltage selector circuit for driving a loaded DC motor.
I need to select a range of set voltages between +10V and +32V @ ~500mA driven from a DC power supply.
My initial design was a series of voltage dividers tapped on a wafer switch, however I have run in to two problems here, loading (should have thought of that) and either current limiting in the divider cct or bleed resistor running hot.
Next I thought I could use a basic zenner clip / clamp type cct but seem to be having similar problems as I wish the non loaded side of the cct to be low current.
I didn't think this would be too difficult but its puzzling me, or am I having mental block :-?
Thanks all

DO you already have a power supply that has multiple outputs (voltage levels) and you want to select any of those outputs?

Hi Prototyp
Thanks for your reply, I do have a variable power supply (single output), but I want this unit on a test centre where the set voltage levels have to be precise and repeated regularly. So the idea is to have a switch box to save time, effort and error.

If you use a potential divider running at say, 10mA, then using a power transistor as a source or emitter follower, will result in a satisfactory set up. At lower voltage outputs the actual output voltage would be lower then the theorectical, but it will be repeatable, so you just select the potential divider resistors to compensate for the added volt drop.
Frank

Aditional comment - use opamp

Hi.

In adition to a common colector, you can use a opamp to make a "reinforced" voltage folower.

If you're not familiar with opamps, a single transistor should do the job.

Anyway, make sure that the transistor can whitstand 16 watt heat generation.

The suggested emitter follower can work as a simple voltage regulator with moderate precision. A variable voltage regulator like LM317 would however set the voltage independent of load changes and also protect itself against shorts and thermal overloads. Finally, you can of course use a power supply with digital programming or analog remote control capability. The latter would be the usual test solution in industry.

Thanks everyone for the advice, I know that a programable PSU would be the most straight forward way forward, but I fancied a little project to liven up the week ;-) I had been looking at an adjustable voltage regulator as auggested (LM317) and think I will take this idea forward, it also gives me room to add a few more handy features to the design.
Many Thanks