Bench Power Supply Design

[MiKro]

Hi All,

I have finally finished the schematic for my regulated bench power supply. Now to build it. LOL.

mk

 

 

 

 

 

[curtisa]

Some Q's and thoughts for you. Bearing in mind the diagram is a bit blurry, even when blowing it up on my screen:

• Input-side filter caps in the LM3x7 regulators are normally connected across the input and ground. Connecting them to the midpoint of the voltage setting divider will reduce their effectiveness for input-side ripple filtering (I've got a feeling you'll probably also get worse output ripple as loading increases too). Although there are those whopping big caps on the outgoing sides of the bridge rectifier (10000uF and 2200uF? Too blurry...) which will swallow up most of the ripple.
• 2K variable pot on the voltage setter dividers for the two +/-24VDC sections give an effective adjustability range of 1.25V - 11.67V (asuming the R1's of each are 240 ohms - hard to make out), which won't be enough to get you to the +/-24VDC you're after.
• 5K variable resistors on the 12V, 9V and 5V sections gives an effective adjustability range of 1.25 - 27.29V. Is it possible the values of the variable resistors on the +/-24V and 12/9/5V sections have been swapped on your diagram?
• The 5, 9 and 12V regulators will be working up a bit of a sweat with 35V at their inputs. The 5V in particular will be having a hard time as it's trying to burn off significant amounts of power when supplying moderate load (eg, 35V input - 5V output = 30V. Multiply by say 300mA of load and the poor old thing will be cooking up 9W). Bank on a fairly substantial heatsink or maybe consider adding a chunky votage-dropping power resistor(s) on the input side to ease the burden on the regulators. Also note that the LM317 is specified for a max Vi -V0 differential of 40V. 30V as designed does appear to give you plenty of room to move, but if your input voltage rises too much and your 5V output voltage sags too much (or you have it set to a lower value using the 5K variable R) there's a chance you're going to exceed that rating.

[MiKro]

14 minutes ago, curtisa said:

Some Q's and thoughts for you. Bearing in mind the diagram is a bit blurry, even when blowing it up on my screen:

• Input-side filter caps in the LM3x7 regulators are normally connected across the input and ground. Connecting them to the midpoint of the voltage setting divider will reduce their effectiveness for input-side ripple filtering (I've got a feeling you'll probably also get worse output ripple as loading increases too). Although there are those whopping big caps on the outgoing sides of the bridge rectifier (10000uF and 2200uF? Too blurry...) which will swallow up most of the ripple.
• 2K variable pot on the voltage setter dividers for the two +/-24VDC sections give an effective adjustability range of 1.25V - 11.67V (asuming the R1's of each are 240 ohms - hard to make out), which won't be enough to get you to the +/-24VDC you're after.
• 5K variable resistors on the 12V, 9V and 5V sections gives an effective adjustability range of 1.25 - 27.29V. Is it possible the values of the variable resistors on the +/-24V and 12/9/5V sections have been swapped on your diagram?
• The 5, 9 and 12V regulators will be working up a bit of a sweat with 35V at their inputs. The 5V in particular will be having a hard time as it's trying to burn off significant amounts of power when supplying moderate load (eg, 35V input - 5V output = 30V. Multiply by say 300mA of load and the poor old thing will be cooking up 9W). Bank on a fairly substantial heatsink or maybe consider adding a chunky votage-dropping power resistor(s) on the input side to ease the burden on the regulators. Also note that the LM317 is specified for a max Vi -V0 differential of 40V. 30V as designed does appear to give you plenty of room to move, but if your input voltage rises too much and your 5V output voltage sags too much (or you have it set to a lower value using the 5K variable R) there's a chance you're going to exceed that rating.

Yep I am aware of the ground. I made it a floating signal ground and not chassis ground.  The caps are 1000uf and 220uf. 

Yes I got the Pots bass ackwards thanks for that. Yes I am aware of the input voltages and will be adding current limiting resisters where needed. I am still bread boarding each now.

Thanks for the input.

MK

[curtisa]

4 minutes ago, MiKro said:

Yep I am aware of the ground. I made it a floating signal ground and not chassis ground. 

Don't think we're talking about the same thing. I mean this:

Should probably be like this:

[MiKro]

1 hour ago, curtisa said:

Don't think we're talking about the same thing. I mean this:

Should probably be like this:

Ah, I will look into that thanks for the extra eyes. I do miss things at times. LOL

MK

[MiKro]

1 hour ago, MiKro said:

Ah, I will look into that thanks for the extra eyes. I do miss things at times. LOL

MK

@curtisa, Yes you are correct I some how moved that over. it should look like you drew it. . The cap to ground and no connection between I and A except at ground. Thanks for catching that. easy to change.

Update: It has been corrected. Damn how did I miss that one. LOL. Thanks again for your extra look over.

I know how I did that I had removed it once and put it at the Rectifier section, then moved it back using a larger one at the rectifier then stepping down at the module.  I must have just joined it in wrong when I went back. DUH!!

As far as heat and voltage I will most likely use lm317 for all of the LM series. I may also up the wattage of the resistor to the trim pots for those

mk

[curtisa]

Looks good

[MiKro]

So after revisting my design  I plan to start the build tomorrow finally.

MK

 

 

 

[Helen09]

Thank you for sharing