[sdiy] Power Supply Completed (Load Tested)

[Michael Ruberto]

Load testing has gone very well. Testing was performed with a 30.1 Ohm 75W 
1% resistor connected from each output to ground. The PSU meets the designed 
current rating of 500MA easily. The output voltage didn't even fluctuate, 
same readings as unloaded. Tracking is the same but, I was testing with 
equal loads on each rail. Overall the unit seems bulletproof. As you see in 
my pics the transistors and regulator are attached to a heatsink - damn good 
idea. These devices get moderately hot but not too hot to touch. Q1 gets the 
hotest and clearly this is due to it's double role as a preregulator for the 
+15 rail. Conversly, if the +18.5 rail is not used or lightly loaded then Q2 
is the hotter device because the positive rails have both Q1 and U2 to 
dissapate the power. If you decide to build this supply and use it at 
constant full current, use adequate heatsinks and airflow. Employ a cooling 
fan if the case housing the supply is small.

FYI: R3 & R11 are both simply a 3K and 620 Ohm resistor in series (it's what 
I had). R7 & R8 should be matched as tightly as you can measure. R6 & R14 
are Ohmite "Brown Devil" 10W units and they only get luke warm so you should 
be able to get away with a lower wattage here. U3 may be substituted as long 
as you choose a device with very low offset and good ripple rejection. I 
recommend the 308 in this app and you can get them at Jameco Electronics. I 
used the 308AH metal can version, it was actually cheaper... :)

M. A. Ruberto


>From: "Michael Ruberto" <frankentron at hotmail.com>
>To: synth-diy at dropmix.xs4all.nl
>Subject: [sdiy] Power Supply Completed
>Date: Sat, 18 Jun 2005 20:24:17 -0400
>
>Today I got the sucker working. Beware, sometimes brand new MOVs EXPLODE in 
>your face! I guess it must have been defective...
>Anyway the PSU is a dual tracking 15V with an 18.5V output for use with 
>modules that have onboard regulators and the SEM VCO I have in the design 
>stage. I have not load tested it yet but my tests of the tracking circuit 
>have been better than expected. I calculated a 0.1% error between the 
>outputs.
>here's the schematic:
>
>http://frankentron.freezope.org/source/diy_blue/images/PSU_REV1.gif  
>(29,281 bytes)
>
>Here are some pics:
>
>http://frankentron.freezope.org/source/diy_blue/images/PSUPCB.jpg   (35,043 
>bytes)
>
>http://frankentron.freezope.org/source/diy_blue/images/PSU1.jpg   (33,269 
>bytes)
>
>http://frankentron.freezope.org/source/diy_blue/images/PSU2.jpg  (57,381 
>bytes)
>
>As you can see from the last image it's not a looker. The chassis and 
>trannies are from an old ionic air purifier I got from the garbage. I had 
>to steel wool the heck out of the box to make it shiny again. I mounted 
>binding posts for the outputs so I can use it with a breadboard setup 
>considering none of the rest of the synth has been built yet. I got juice 
>though, and its a start!
>
>M. A. Ruberto
>
>