I would also get hints from the deflection plate drivers of an oscilloscope. Preferably an inexpensive, low bandwidth design.

 

Hi All,

I wanted to write and provide some feedback on the circuit diagram for the bipolar power supply suggested by specie. We have built the supply, and it works well. There were two modifications that were needed following testing;

1) Two capacitors were added near the output of each channel. These were needed as there was an oscillation found in the circuit following construction (15kHz). One was right at the output, 0.86 µF (to ground). The other was connected before R14, and between R12 and VR2, 0.47µF. See the schematic.

2) We also replaced R14 with a 330k resistor as the voltage regulation was not optimal when a 1 Mohm load was put on all of the outputs. Varying one output affected the others by a few volts.

Initially we tried to use four LM324's, the thinking being that if one blew up then we would still have three viable channels to work with. However, there was a problem then with R1, that it was overheating from too much current being drawn. We returned to the original schematic, were one LM324 was used for all four channels.

Finally, we added some nice lcd displays (Dick Smith Australia, PN Q-2220) , 10 turn pots, and BNC outputs on the back (From RS). It is a thing of beauty! I have attached some pictures of the unit, and also some doodling on the provided schematic.

So all in all, it is a major success! Thanks to all that helped along the way.

Best wishes,
Dr Dunk

Attached Images

	BipoloarSupply 003.jpg (40.9 KB, 15 views)
	BipoloarSupply 005.jpg (49.5 KB, 16 views)
	BipoloarSupply 006.jpg (100.3 KB, 19 views)

Attached Files

Schematic.PDF (128.0 KB, 13 views)

 

Looks a nice job, well done!.

__________________
PIC programmer software, and PIC Tutorials at:
http://www.winpicprog.co.uk

 

Glad to be of help.

 

Looks, good.

What about voltage stability?

__________________
I also post at the following sites:
http://www.stop-microsoft.org http://www.heated-debates.com
Screen name: Aloone_Jonez

 

Hi there,
I was thinking of hooking the supply up to a chart recorder and monitoring the output over a couple of hours. How else would you suggest to test the stability?

Cheers!
DrDunk

 

You could manually measure it at regular intervals.

You could leave it connected to a multimeter with a minimum amd maximum setting for a few hours.

__________________
I also post at the following sites:
http://www.stop-microsoft.org http://www.heated-debates.com
Screen name: Aloone_Jonez

 

A chart recorder will be better if the resolution is high enough to show minor variations clearly.