0-30V Stabilized Power Supply

[nemeszoli]

Hi audioguru,
              I used TLE2141 opamps ordered from farnel.com. Without a load, on R7 I have 2,2mV,
wich means about 10mA passing trough it, I think this is normal because the current supplying U1 is
passing trough R7 and 0V wire. I tried taking out U1 from the circuit, in this way there vas no voltage on
R7 so I think, with this design it is normal to have some current on the 0V wire

 

[audioguru2]

The TLE2141 opamp for U1 has a typical current of 3.5mA when warm and a max current of 4.5mA.
The zener diode D8 adds 5.6mA to the current in U1 so its total current is 9.1mA to 10.1mA. Then the voltage across your R7 is 2.5mV to 2.7mV which is nothing. The input offset voltage of U1 is a max of only 1.4mV. 

 

[nemeszoli]

Hy, I made some measurements on my power supply but unfortuneatly I saw that I have a ripple of about 0,8V  ??? at 6A tested with a load of 1.5ohms. I have a transformer with 26V output, I would like to have on the power supplys output at least 24V without ripple I tried also using a 22000uF capacitor but the result was the same as with a 10000uF one

View attachment 41224

 

[audioguru2]

Ripple on a power supply is in triangular waves with a quick charge then a slower discharge.
Yours has square-waves maybe from an overloaded transformer or from transformer current in the amplifier circuit or from pickup if the transformer or rectifier is near a sensitive amplifier part.

With an output of 6A then the average transformer current is 8.5A because the rectifiers conduct for only a small portion of the waveform. 

 

[nemeszoli]

Hy, I don`t realy understand why but, by changing the place where the 22000uF C1 capacitor is connected my power supply seems to be working now.  I designed a connector for C1 wich was connected on, and in my earlier post I attached a picture of the output voltage form (square wave), now I connected C1 directly on the wires connecting the rectifier (black and red on the picture) and this way the output voltage looks nice, the triangle wave is the voltage on C1, but it doesn`t  seem to influente the output voltage. I gues the problem should be in my PCB design

View attachment 41231

View attachment 41232

 

[audioguru2]

Hi Nem,
I am glad to see that you found the problem in the pcb wiring.
But the traces on the pcb seem to be much too narrow for the very high charging current of C1 that might have 15A peaks.

 

[tdx]

Hello everyone. I am looking at building this power supply for a specific use which requires 0-20 volts and typically up to 2 amps, so my question is can I save some money and take some weight off by stepping down the transformer just a bit? Would a 24v 3.3 amp such as the Hammond 185E24 work if I never need to exceed 20 volts?

 

[audioguru2]

Hello everyone. I am looking at building this power supply for a specific use which requires 0-20 volts and typically up to 2 amps, so my question is can I save some money and take some weight off by stepping down the transformer just a bit? Would a 24v 3.3 amp such as the Hammond 185E24 work if I never need to exceed 20 volts?

The original circuit uses many old and parts too small that overheat then the circuit does not make 30VDC at 3A.
The improved circuit will probably work well with a little 24V/3.3A transformer then its max output might be 24VDC at 2.3A.
You should adjust the calibration so that when the voltage pot is at max then the voltage is almost at max.

 

[tdx]

Thank you for the feedback audioguru. I am using what appears to be the latest revision you have made with the MC34071/TLE2141 opamps with the parts list using red text to indicate the part changes, I've attached them below to verify they are the correct revision. So if I use a 24v 3.3A transformer and set the voltage pot to my 20v max needs I would not need to make any further changes as far as parts go correct? But could I feasibly reduce the C1 capacitor to less than 12,000uf? Also I'm pretty sure its safe, but using 1/2 watt resistors in place of 1/4 watt would be fine in this circuit as in most, correct?

Thanks for your help

View attachment 41237

View attachment 41238

 

[audioguru2]

For a max output current of 2.3A then C1 can be reduced to 2.3/3 x 12000uF= 9200uF which is not a standard value. Use 10000uF.

I haven't seen a 1/2W resistor for about 45 years.  If yours are that old then they probably have their values changed.

 

[tdx]

I haven't seen a 1/2W resistor for about 45 years.  If yours are that old then they probably have their values changed.

Hmm... the parts list calls for a 1/2W resistor on D3 (220 ohm). I'm looking at all new parts from Newark, it's just that some values for resistors have a better selection in 500mW or are actually cheaper than 250mW.

Also I noticed the schematic and parts list don't quite agree on the Zener and LED, the parts list calls D12 the Zener and LED1 as the LED, but on the schematic D12 appears to be the LED and the Zener is unmarked? I assume its just the labeling and the schematic placement is correct?

 

[audioguru2]

R3 gets warm so it should be 1/2W or 1W.

The schematic and parts list should call the LED D12 and the 10V zener diode D13.

 

[tdx]

Am I correct in thinking that if this power supply is turned on even with no load, or a very light load the power transistors will actually be hotter than running at higher output voltages correct? Say if it were at 5 volts of output they would have to burn off more than if they were cranked up to get rid of the difference between their input from the transformer and the output load correct?

 

[audioguru2]

Heating in the output transistors, the driver transistor Q2, bridge rectifier and R7 is caused by the voltage across them times the current through them. If there is no load current then there is no heating in these devices. The most heating is produced when the current is the highest (3A) and the output voltage is low or is shorted. The the output transistors have 35.7V across them x 3A= 107.1W. The driver transistor Q2 has 34.7V across it x as much current as 75mA= 2.6W. The bridge rectifier heats with about 2V x 3A= 6W. R7 heats with 3A squared x 0.47 ohms= 4.23W.

 

[tdx]

Starting to make more sense now. Now if the intended load will be typically run at 4-12 volts 50% duty cycle (electromagnetic machine), never exceeding 18v then would it make sense to further step the transformer down to 18v 4a? The load shouldn't exceed 1 amp, 2 at the most. Would 18v with at least 10,000 uf C1 keep the remainder of the circuit functioning properly?

 

[audioguru2]

An 18V transformer produces a rectified and filtered full current voltage of 23.4V but with a couple of volts of ripple so the troughs of the ripple are at 22.4VDC.
The output transistors, their emitter resistors, the driver transistor Q2, the opamp U2, the bridge rectifier and the current sensing resistor R7 need a total of about 5V so for an output of 18VDC the unregulated positive supply must be at least 23VDC and the transformer must be a little higher than 18VAC.

A 4A transformer with a load of only 2A has an output voltage a little higher than its voltage rating so your transformer might or might not work. I would use a 20VAC transformer.   

 

[tdx]

20v transformer it is. Now just a few quick questions before I order parts because of some parts list and schematic conflicts:

1) Schematic shows a 100nf poly cap (labeled C10 on some schematics) between the negative line and the emitter of transistor Q3 & opamp U3's Vcc, but is not on parts list, should this be removed completely?

2) Resistor R15 (100 ohm) is nowhere to be found on schematics, delete?

3) The 10v 1w Zener is missing from my parts list, but this should be added right? There should be a 5.6v and 10v Zener?

Thanks again for your guidance.

 

[audioguru2]

1) Schematic shows a 100nf poly cap (labeled C10 on some schematics) between the negative line and the emitter of transistor Q3 & opamp U3's Vcc, but is not on parts list, should this be removed completely?

When I added the 10V zener diode to reduce the supply voltage to Q3 and U3 I added the 100nF capacitor because all opamps need a supply bypass capacitor. But I forgot to number it.

2) Resistor R15 (100 ohm) is nowhere to be found on schematics, delete?

The oroginal schematic used R15 but it was useless. Delete it.

3) The 10v 1w Zener is missing from my parts list, but this should be added right? There should be a 5.6v and 10v Zener?

The 10V zener diode is important. The 5.6V zener diode D8 is extremely important and should be one that is rated at only 5mA. Many zener diodes are rated at 45mA and should not be used.

 

[LEECH666]

Hello again,

I got a question regarding the heatsink of the BD 139 transistor. I think "a pretty big heatsink" is a bit vague/relative. Could someone translate "pretty big heatsink" into a

 

[audioguru2]

The BD139 is in a very old case so its thermal resistance is high (it is difficult to cool).
The 2N3055 output transistors have a lot of power to dissipate.

It is easy for you to calculate the max power of a transistor, look up its max allowed temperature, decide how high you want the max temp to be, allow thermal resistance for insulators then select a heatsink.