I'll post a schematic, but my main problem is that as I draw a load on the power supply it is dropping volts. When I turn on the load (radio) it drops but quickly returns. When I key however it drops a 1/2 a volt then when I talk it drops another volt. Leaving me somewhere between 12.5 and 13.5 volts. This on it's own would not be a problem however I am soon going to be adding an amp and would like to have this figured out by then. I assume I am not putting out enough amps, but don't understand why. The radio does not draw more than 5-10 amps at 13.8 volts. I am currently using three transistors that should supply 8amps apiece if I am not mistaken.
Ok enough jibber jabber here is a schematic
Oh and here is the NPN datasheet http://pdf1.alldatasheet.com/datasheet-pdf/view/78029/AUK/D13007.html

 

You have the voltage regulator connected incorrectly. What you are doing is putting 15 V on the gates of the transistors, which should saturate them and open the emitter to base. Look at some high current voltage controllers to rework your supply.

 

Cool how do you mean not connected correctly? The voltage on the base leg does not drop. Only on the emitter does it fluctuate.
So a high current voltage controller does what different supply more amps to the base?
Please explain. Thank you!

 

Here is a circuit for a 12 V 15 A power supply. To make it 15 V, change the voltage regulator.

 

Ok cool I'll probably get some pnp transistors and rework the layout, but I would like to understand why the configuration I have is not doing what I thought it should.

 

In your circuit, the regulated output of the regulator is connected to the base of the transistors. Since only 0.7 V is needed to turn the transistor on, the transistors are saturated and so are fully on. This allows the current to flow, but the voltage is unregulated. I'm not sure why you didn't see a higher voltage for the output.

 

My bet is that the problem is your transformer, which you haven't said anything about. You should measure the rectified voltage during the sag event.

 

My bet is that the problem is your transformer, which you haven't said anything about.

If the voltage across the bridge is 24.5V, why is that not seen on the output since the saturated transistors offer very little impedance ?

 

I'm guessing it's not 24.5V under load. Easy enough to check.

 

I'll post a schematic, but my main problem is that as I draw a load on the power supply it is dropping volts. When I turn on the load (radio) it drops but quickly returns. When I key however it drops a 1/2 a volt then when I talk it drops another volt. Leaving me somewhere between 12.5 and 13.5 volts. This on it's own would not be a problem however I am soon going to be adding an amp and would like to have this figured out by then. I assume I am not putting out enough amps, but don't understand why. The radio does not draw more than 5-10 amps at 13.8 volts. I am currently using three transistors that should supply 8amps apiece if I am not mistaken.
Ok enough jibber jabber here is a schematic
Oh and here is the NPN datasheet http://pdf1.alldatasheet.com/datasheet-pdf/view/78029/AUK/D13007.html

You need to use PNP power transistors to make that kind of circuit work.

 

What kind of regulator do you use?

 

Also in this case maybe the LM723 regulator may work better. The LM723 regulator is cheap and common.
www.ti.com/lit/ds/symlink/lm723.pdf

 

the regulator i'm using is an7815
http://pdf1.alldatasheet.com/datasheet-pdf/view/111558/PANASONIC/AN7815.html
I have thought it might be the transformer. The rectified voltage does dip as well, down to 22.5-23vdc range. How would I remedy that, by using a larger gauge wire on the output wind?

As far as the transistor output, I figure it is doing exactly what it is supposed to do. The 0.7 is lost on the output of the emitter, and the amperage is fed from the collector.
It makes sense that the voltage drop on my rectified current means I'm also dropping amps. Because it doesn't drop the voltage with a light load.

On the transformer, I did not do the primary windings but the secondary windings which I did about 20 (12ga). So I am assuming the primary windings are about 120 (since I'm running off 120 mains) So, to increase the current capacity I should do what?
Rewind the primary winds to 240 wraps and secondary to 40 or so. Or could I just use heavier gauge wire?

 

I think even in Honduras the LM723 is easy to get hold on. But if you insist on using the LM7815. Take a look here for proper design ideas http://www.fairchildsemi.com/ds/LM/LM7805.pdf

 

The two stores I went to this was all they carried. Basically if they don't have it nobody does. I'll ask specifically for that piece next time I go. I'll also probably get some pnp and hook it up the other way, if no one can help me get this circuit right. I don't think this circuit is just inherently wrong is it?
I think I am right on the cusp of getting it, however I could be wrong, I'm just learning all of this.
Thanks for the help everyone!

 

The two stores I went to this was all they carried. Basically if they don't have it nobody does. I'll ask specifically for that piece next time I go. I'll also probably get some pnp and hook it up the other way, if no one can help me get this circuit right. I don't think this circuit is just inherently wrong is it?
I think I am right on the cusp of getting it, however I could be wrong, I'm just learning all of this.
Thanks for the help everyone!

Your setup is not a regulator setup. So the output will vary with load. Your circuit is more a simple current booster circuit. So as a regulator setup it is kind of inherently wrong. But do not worry about that we all have to learn

 

OK, so your transformer is adequate.

Your 100Ω resistor feeding the regulator is a problem. Your arrangement requires a base current on those power transistors, as much as 10% of the load current. If the load increases and drops the voltage, the regulator tries to supply more current to them, but this would cause a voltage drop across that resistor. You should measure the input voltage that the regulator sees during a sag.

 

The circuit will work as drawn, not with great regulation but you can use NPN transistors as an emitter follower after a regulator like this and get better performance than you have seen. But, as wayneh says you need more base current - lower your 100 Ohm resistor or eliminate it altogether. You should allow for an amp from your regulator - it will need to be on a good heatsink.

Be careful with the current sharing between the transistors - perhaps add a small resistor in series with the collectors of each - 0.3 Ohm would drop perhaps 1V at full current and swamp out the differences between the transistors ensuring they share properly.

 

I agree with richard.cs if you need to use NPN's. In the voltage follower configuration you could remove the 100 ohm resistor completely, but you really need the low ohm balancing resistors.

If you can use PNP's the circuit posted by LDC3 is very good. I do have one concern. I think there may be a typo regarding R1. For the circuit to work as intended the voltage drop across R1 must turn on the PNP's before the max current of the 78xx regulator is reached. I think it should be 1 ohm.

The LM723 chip is very good and proven if you can get it.

 

Cool thanks for the responses. I'll be going to town soon so I'll see what I can get. I'm sure I can get PNPs not sure the ratings on them. We will see. Last time I tried to get 10w resistors they only had 1 and it was 1 ohm... some things you just have to do without or order from the states.

But that makes sense to balance them out as it did seem one was getting hotter than the others.