Low Voltage Regulator
I am reorganizing and rebuilding my bench top equipment. I wanted more flexibility and smaller enclosures in order to puzzle them for the momentarily needs.
I recently dismantled my first tube tester project. The main reason I didn’t like the setup was the fact that due to the length and the intricate socket wiring to the numbered bananas ( associated to each of the tested valve pins), the tester wasn’t created the ideal conditions for accurate measurements. In some instances DUTs were oscillating and/or distorted the measuring signal. I reached the conclusion that because there are only few types of tubes that I like and regularly test, it would be more appropriate to have custom setups for each individual case. This way the testing can simulate the real operating condition of the tested tube (valve).
The schematics of the low voltage regulator is based on the IC LM338K. excellent IC offering adjustable out 2-37V at max 5A, thermal stability and short circuit protection. The potentiometer used is a 5K 2W 10 turns Vishay offering very smooth and precise voltage adjustment. There are 2 ranges possible:
- 1.5-17V (switch connects center tap to ground, full wave rectifier configuration)
- 2-32V (switch connects bottom of rectifier bridge to ground, bridge rectifier configuration)
One of the uses for this low voltage regulator is to supply power to the DUT’s filament (heater). Proper range to be selected in order to make sure that the power dissipated by the LM338K does not send the IC to disconnect ( Pmax is thermally limited, specifications are stated for Pmax=50W). One does not want to have the filament disconnected while high voltage is still connected. For example for 1.5 to 12.6 V filaments range 1 is recommended. A CRC(10mF-0.235R-10mF) circuit after the rectifier supplies a low ripple to the IC. For example for a 26 type tube at 1.05A ( in this case the LM338K dissipates about 15W) the ripple before the IC is about 70mVrms. I tried to measure the ripple at the output of the regulator. It was too low to be measured with a RIGOL DS1102E. According to the data sheet the ripple rejection at Vo=10V is 60dB minimum which is 1000 times smaller than 70mVrms (or 70uVrms).
I used a 300W PC power supply enclosure that comes equipped with main AC IEC plug with EMI filter and 12V fan. It looks decent and works great.
Next project is to reconstruct the HV adjustable regulated power supply…