0-30V Stabilized Power Supply

[effenberg0x0]

Thanks Audioguru. The problem is I have to use the pieces available at the course inventory.

Redwire's 5A project suggests RS1506 (15A 600V) or GBJ2506 (25A 600V). They're also not available to me, so I am going for KBPC-3510 35A/1000V.

 

[Enforcer83]

In the Schematic posted by audioguru at the beginning of this topic, there is a 10V 10W Zener diode between the output of the Bridge Rectifier and the emitter of the PNP transistor.  I am confused with its purpose.  I have tried to analyze it using my circuit theory but have failed miserably, partially because I have always used regulators instead of diodes for voltage regulation.  So, what is its purpose?  A PSpice waveform would be very useful to help me visualize what is going on, if at all possible.

 

[Hero999]

1200

 

[audioguru2]

In the Schematic posted by audioguru at the beginning of this topic, there is a 10V 10W Zener diode between the output of the Bridge Rectifier and the emitter of the PNP transistor.  I am confused with its purpose.  I have tried to analyze it using my circuit theory but have failed miserably, partially because I have always used regulators instead of diodes for voltage regulation.  So, what is its purpose?  A PSpice waveform would be very useful to help me visualize what is going on, if at all possible.

The 28V transformer might be 30V without a load on the project. Then it makes an unregulated positive voltage of 41V plus a negative -1.3V supply for U1. The total of 42.3V is very close to the max total allowed voltage of 44V and heats the IC. So the 10V zener diode reduces the max voltage and shares in the heating.

 

[effenberg0x0]

1200

 

[Hero999]

It really isn't. I could order from Farnell but it would more expensive than all the other components (about 120USD!!). As I have to use the components they have at the course anyway, I selected 4,700uf+4,700uf+3,300uf = 12,700uf / 63V.

Electrolytic capacitors are normally only made in E6 values, i.e. 10, 15, 22, 33 and 68 but some manufacturers only make E3 values i.e. 10, 22 and 47 which is why you're having problems finding 12,000

 

[Enforcer83]

The 28V transformer might be 30V without a load on the project. Then it makes an unregulated positive voltage of 41V plus a negative -1.3V supply for U1. The total of 42.3V is very close to the max total allowed voltage of 44V and heats the IC. So the 10V zener diode reduces the max voltage and shares in the heating.

Be patient with me on this.  Power and power electronics are not my strong suit.  Low voltage battery systems are.

So you mean to tell me the Zener is adding 11V to the system.  That does not make sense.  Wouldn't the better option be to place say a 36V Xfmr with a no load rating of  38.52V in the system and connect the transistor, the resistor R19 and the V+ of the opamps to this rail instead?  Even figuring a Voltage drop of 1.05V at 5A through a bridge Rectifier (GBJ10005 is my planned rectifier) you are still dealing with 37.47V.  Yes it may be more expensive but at least the math seems logical.

I think I need a tutor to help me relearn this area of Electrical Engineering.  Doesn't help my circuit Prof. couldn't teach.

 

[audioguru2]

So you mean to tell me the Zener is adding 11V to the system.

No.
A 28VAC transformer might be 30VAC without a load. Then its peak voltage is the root of two times 30V= 42.4V and the full-wave rectifier charges the main filter capacitor to +41VDC. IC U3 also has a negative -1.3V supply so its total would be 42.4V which is very close to its max allowed total voltage of 44v and it has a lot of heat.

The 10V zener diode reduces its positive supply to +31V and reduces its heat.

Wouldn't the better option be to place say a 36V Xfmr with a no load rating of  38.52V in the system ....

Then the positive unregulated supply would be +53VDC and everything will blow up.

Even figuring a Voltage drop of 1.05V at 5A through a bridge Rectifier ....

When the project has a 5A load then the average current in the bridge rectifier is also 5A.
But it conducts only on the peaks of the transformer's sine-wave for only maybe 10% of the time, so its peak current is 50A.
It has two diodes conducting in series so its voltage drop is about 2V.

 

[Enforcer83]

I knew that pesky root 2 was going to come up and bite me in the arse one day.  :-[

Now I see whats happening.

Edit:

Like I said, AC power not my strongest area.

 

[effenberg0x0]

>>A 28VAC transformer might be 30VAC without a load.

Noob question: Why?
Are you considering the AC power oscillation in the primary?

Thanks,
Effenberg

 

[audioguru2]

>>A 28VAC transformer might be 30VAC without a load.

Noob question: Why?

Because the windings have resistance. The output voltage drops down to the transformer's rating only when it has its rated load.

Test the output voltage of a 9V/200mA wall-wart. It is 9V only when it has a 200mA load but is 15V or more without a load due to the resistance of its windings.

 

[effenberg0x0]

Audioguru, I am a little confused.

I was thinking about Ohms law. If I = V / R and then V = I * R, decreasing the value of R would decrease the value of V, assuming constant I.

Then why would the transformer voltage be higher without a load?

Thanks,
Effenberg

 

[Hero999]

The transformer is not perfect and has an equivalent series resistance which drops the voltage when a current it drawn.

 

[audioguru2]

If a 28V transformer is rated at 3A then it is made as a 30V transformer because the designer knows that its voltage will drop 2V due to the resistance of its windings.

The cheap little 9V/200mA wall-wart is made as a 15V one because ..... (same thing).

 

[effenberg0x0]

I did the tests (wall-wart). You are right.

Another stupid one: I've been checking projects of other students. Some of them have a fuse. On the other hand, I could not find a single power supply project on the net that has a fuse. Where do you stand? Should there be a fuse?

Also: It doesn't seem like eagle (last version) auto-track feature takes under consideration a decent space between tracks (potential capacitance?). Or does it?

 

[audioguru2]

A fuse should be installed in series with the on-off switch in the mains wiring to prevent your home from being burned down if something in the project goes wrong.

But I think England has a fuse in the electrical plug.

 

[Hero999]

If a project is low powered <1.5A then it's a good idea to add a fuse because the lowest rating plug fuse you can normally buy is 3A and won't be enough to protect the circuit.


Unfortunately most cheap transformers have one shot thermal a one shot, non-replaceable thermal fuse which is why it's important to add an external fuse.

Another thing is that lots of projects use regulators such as the LM317 or LM78xx which have built-in thermal protection so there's a very small risk of them going wrong.

 

[effenberg0x0]

The input voltage here is 120V. My trafo is 30V, 5A. 150VA / 120V = 1,25A Fuse. Is this the correct thinking?

Another thing: I've noticed some of the few pics of the assembled boards in this thread show the ceramic resistor (10W) soldered like other components, I mean they're touching the board. Since this thing is gonna heat significantly, wouldn't it be better to solder it higher, like 1cm from the board?

 

[audioguru2]

For the 3A version of this project, R7 is 0.47 ohms/10W. With 3A in it then it dissipates 4.23W and will be hot enough to melt plastic and maybe char the pcb.

The original project used a 5W resistor. It would nearly smoke.

 

[Hero999]

I normally put the fuse on the primary side and use a slow blow so the surge current won't be a problem.

For 120V operation, I wouldn't use anything less than 1.5A, 3A would be my first choice.