JSR04, +12V, +5V (from the negative side)

[peranders]

I'm not sure a CA3140 will work but you can try.

About the R8, value?, how many volts across it, both AC and DC voltage?

[builder brad]

I measured 1.4v Ac across R8 previously, I am unable to do so again as T10 has just blown again, and I am not sure about the DC reading.

when I put T10 to ground via that 2.2 ohm resistor I measured 0.7 volts across all ot the bc337 and bc327 transistors except the bd140 (T4) transistor which measured 1.6v - is this correct?

[peranders]

Have you really checked the transistor types?

NPN for the current limting and series transistor and _PNP_ for the drivers. The same kind of types as for the +12 V section.

[builder brad]

as far as I can see all the transistors are correct.

Originally in order to change the negative section over to positive you sugested that I reverse all of the transistors in your updated schematic.

recently I replaced all of the transistors in order to try to fix this problem and then changed all of the transistors to the same type as the + section, obviously being mindfull of their orientation.

[peranders]

... but you mention BD140? I'll hope you are using BD139 for a positive regulator.

BC327 = PNP should be used for the drivers (two of them)

BC337 = NPN should be used as current limiting transistor

BD139 = NPN should be used as series transistor

[builder brad]

the section is now working - I had previously swapped the bd139 for a bd140 (reversed) as I misunderstood your instructions to "reverse" this component in the schematic.

I now have +5.01 volts out which does not drop, even with a 12 ohm load connected

BUT the 12v section is still buggy, I am seeing a 4volt drop with the 12 ohm load connected to the 12v outputs. I have almost 27 volts unregulated, so there is plenty of headroom. I have tested all of the transistors for the 0.7v difference and they all seem to be working.

[peranders]

I'm glad you finally got it working. You can experience weird phenomia when you swap npn-pnp as you may have noticed.

Good work!

For the 12 volt, when the voltage starts to drop, measure the voltage across T3/T5. When it is starting to change you have reached the limit. One way to increase the "headroom" is to decrease the series resistors at the rectifying diodes. Test with the load you plan to use and some more, but not _much_ more.

[builder brad]

I have measured the 12v section with various loads and have observed the following

unloaded regulated output 13.27 v

12 ohm load causes 4 v drop
15.5 ohm load causes 2 v drop
23.5 ohm load does not make a difference to the output

so, from the readings I can see that I am ok up to 560ma, maybe a little more, but certainly not 700ma.

the unregulated supply voltage dropped from 27v to 22v when I added the 15.5 ohm load, so it apears that the regulator is strugling to supply the current and not the transformer.

what can i do to get closer to 1000ma, or more?

would larger capacitors help in this case?

even under load there is no temperature increase

[peranders]

It seems that the section before the regulator not is what you ought to have.

Since you have halfwave recification you must have twice the amount of smoothing caps but first, how must does your AC voltage drop when you load?

You must tune three parts:

1 AC voltage in

2 The resisitance of the series resistors

3 Smoothing capacitance

Those three parameters determines the lowest voltage into the LM317 including ripple, don't forget that.

[builder brad]

I measured 0.6v drop in the AC supply voltage with that 12 ohm load connected, so I guess that I need to look at the caps.

I have just replaced the 4700uf x 2 main caps with 2 x 22000uf and tried the 12 ohm load. Unfortunately I am still seeing exactly the same voltage drop with that 12 ohm load. I have .5 ohms in series at the moment - could this be lowered any further?

Brad

[peranders]

Where do you have the voltage drop? Have you adjusted the series resistors at the diodes? The original values were 2.2 ohms and 10 ohms and as you can see you must how lower values if you want to take out a larger current. See also post 69.

[builder brad]

ok!

I was originally using a 1 ohm 4watt resistor at R5 which has now been reduced to 0.5 ohms by placing another 1 ohm resistor in series - this has not made any difference to the output under load.

I see 19 volts ac input which does not drop when I load up the output with a 12 ohm resistor. I am getting 27 volts dc at the + side of diode D5 which does drop to 23 volts with the 12 ohm load.

PS. The power supply looks great I have 30mm heatsinks on the regulator and the transistor with 4 x 30mm diameter x 30mm high 22000uf supply capacitors - it really looks the part, if only I could get higher current output!

[peranders]

Which value does R27/R29 have?

Current limitation at 0.7/R27 => 1 ohms => 700 mA

1 A = 0.68 ohms

[builder brad]

R29 is currently 1 ohm

[peranders]

Ok, then. You will get 650-700 mA as max current.