'Project:' Power Supply 15vdc @ 3 Amps

Audioguru

Joined Dec 20, 2007
11,251
I just noticed that this power supply is for the 12W amplifiers thread.
Audio amplifiers do not use and do not need a voltage regulator.

The amplifiers produce hum maybe because their input bias has absolutely no filtering.
Instead of using a high current voltage regulator to reduce the hum I recommended and showed a simple RC filter for the input bias to reduce the hum.
 

Thread Starter

kbyrne

Joined Dec 10, 2011
92
My fault it should not read 32.5 VAC at all. The correct voltage should read 32.0 VDC at
that spot. The origional cap. called for 47,000uf @ 35vdc. That is what I was worried about
I am unsure weather to scrap this circuit as not feasable at all or not. I can get a
10,000uf cap thou. considering this how does the rest look after a change from the caps.
Thank you I missed a Boo Boo.
 

Thread Starter

kbyrne

Joined Dec 10, 2011
92
This power supply was for a different project but it was a amplifier project. I have a box with two`TDA2003V amplifiers and tone controls, It calls for 15VDC unknown amperage.
No voltage regulators only raw DC for all audio Amplifiers and their tone controls??
All right can I ask Why. They have precise voltage.
 

Audioguru

Joined Dec 20, 2007
11,251
With a 15V supply, a TDA2003 amplifier IC produces 5.2W into a 4 ohm speaker just before the output clips. Its output into an 8 ohm speaker is about only 3W. Your amplifier design will have an output a little less.
The max allowed supply for a TDA2003 is 18V but the graphs on the datasheet stop at 17V. In a car it will be from 12.6V to 14V.

Amplifiers that are designed properly work fine with raw unregulated DC. You can use a low power voltage regulator with a lower supply voltage for the tone controls circuit.

A voltage regulator has a precise voltage because it has a low power precise zener diode to set the reference for a current amplifier.
 

Thread Starter

kbyrne

Joined Dec 10, 2011
92
If I put two TDA2003 amplifier IC'S. Two tone controls in the same box can all of the boards
(4) be Linear regulated. Or do two have to be raw unregulated the amplifier boards which would mean make the tone controls raw unregulated also. A question to clarify something because you make me think. Thankyou
 

R!f@@

Joined Apr 2, 2009
9,682
You cannot use a 35VDC cap. It should be rated 10% more. For a voltage of even around 30VDC the cap should be at 50VDC.
 

Thread Starter

kbyrne

Joined Dec 10, 2011
92
That is one of a few reasons for my being real leary of the origional schematic, and project found by a newbi to electronics on instructables.com. I would have used 50VDC but question also the extremely large ammount of capacitance. One 47,000uf @35vdc, & two
4700uf @ 16VDC capacitors with two positive voltage regulators as a way of increasing
Amperage to 3A from 1.5A. Is that type of circuit even possible?? Thankyou
 

Audioguru

Joined Dec 20, 2007
11,251
The power amplifiers work best and work fine from unregulated voltage. The tone controls circuit is very low power and can be powered with a little voltage regulator IC.
The input voltage to the voltage regulator must be at least 2V higher than its output voltage. Then the tone controls circuit works from a lower voltage than the power amplifiers which doesn't matter.

Please post a link to the Instructables project because it might have many more problems.
 

Thread Starter

kbyrne

Joined Dec 10, 2011
92
I will post a link to the instructables project like you requested and stop trying with the power supply @ 3A project. The TDA2003V amplifiers are rated at 15VDC, The tone
controls are rated at 9VDC to 20VDC so allis fine. The last board in that project is a LED VU
meter rated at 12VDC. A 1.2 Ohm & a 5.0 Ohm voltage devider should give me 12.0 VDC
even with Raw unregulated DC. Thanks for the advise. I also have a post on a 12 Watt Amplifier project here and am looking at all responses as that is a work in progress.
 

Audioguru

Joined Dec 20, 2007
11,251
Two extremely low value resistors in series short the power supply and get very hot. They are not a voltage regulator.
With a 15V supply, a 1.2 ohm resistor and a 5 ohm resistor in series draw 2.42A and heat with 36.3W. Any load reduces the voltage.
 

Thread Starter

kbyrne

Joined Dec 10, 2011
92
For the sake of clarity and to ask a question I have uploaded three gif files.
This box is what the power supply was for. I have scraped the power supply of this thread as useless per your knoledge. I am going with unregulated raw DC.
The problem is One board takes 15.0 VDC & the other board takes 12.0VDC.
A voltage devider as shown should take the voltage from 15.0 VDC to 12.0VDC
But my question is that the best way or is there another way like a 12.0 VDC
voltage regulator additional circuit. I ask this stupid question only as a learning experience for myself as a newbi to electronics. Thankyou
If the voltage devider is set up right that is.
 

Attachments

t06afre

Joined May 11, 2009
5,934
For the sake of clarity and to ask a question I have uploaded three gif files.
This box is what the power supply was for. I have scraped the power supply of this thread as useless per your knoledge. I am going with unregulated raw DC.
The problem is One board takes 15.0 VDC & the other board takes 12.0VDC.
I took a look at your schematics. They are quite blurry. I would recommend using pdf file format then uploading schematics. You can download free PDF drivers. They are quite good. If you can reupload the pictures in a better format. Tt would be helpful
I took a look at your VU board. You are using TL071 and the Lm3914. They both work with voltages up to 18 volt. So I think you can use the 15 volt supply on the VU board also. Perhaps add a (approx)1000uF in parallel with a 100nF cap at the voltage input. From the schematic I think the TL071 is used in a simplified precession rectifier setup. Are you sure this circuit will work with a single supply voltage. In this document http://sound.westhost.com/appnotes/an001.htm in figure 7(or perhaps figure 8 is better) you will find a single supply version. You can still use the TL071 opamp.
 

Audioguru

Joined Dec 20, 2007
11,251
DO NOT use PDF to make a schematic. Instead crop it, enlarge it and post it as a PNG file type.
Your schematic has the parts spread apart too much so crop them together then the schematic can be much larger but use the same total space.

A voltage divider cannot power a circuit because its voltage changes when the circuit's current changes. Also your voltage divider uses 2.4A(!) and heats with 36W(!).
Use a 12V regulator IC instead to power the 12V stuff. Where is the preamp and tone controls circuit?

With a 15V supply, a TDA2003 amplifier has an output just below clipping of 5.2W into a 4 ohm speaker. It heats with about 3.3W so the total power is 8.5W. Your stereo amplifier will use 17W. But it operates from 15V so its max current from the 12V transformer is 17W/12V= 1.4A.

Your transformer max power is 12V x 2A= 24W. I can't read your fuzzy schematics to see how much power they all need from the transformer.

An LM3914 will melt if its LEDs are bright, it is set to BAR mode with most LEDs turned on and the supply voltage is higher than about 6V. It needs to be planned properly.
 

t06afre

Joined May 11, 2009
5,934
DO NOT use PDF to make a schematic.
Why not PDF I may ask. A PDF file made from a most modern ECAD systems will be vector based and not bitmap based. A vector based picture is much more easy to zoom. And will probably be smaller or the same size as a PNG file. Anyway no big deal.
So to the next question. I have never used the LM3914 my self. So I did not know it used a constant current arrangement for the LEDs. Hence it will for sure overheat if used with 15 volt in bar mode. But adding a series resistor. And let them dissipate most of power not the chip. Would that be possible?
 

Audioguru

Joined Dec 20, 2007
11,251
A PNG drawing opens instantly. A PDF takes time because the program for it must open.

I have a VU meter that I made years ago with an LM3915 (logarithmic steps). Its LEDs are pretty bright at 25mA. It uses a 9V supply and each LED is actually two 1.8V LEDs in series. The 10 LEDs use 250mA when they are all turned on. Then the IC would dissipate (9V - 3.6V) x 250mA= 1.35W. But its max allowed heating is 1.35W when it has open air cooling which it doesn't because it is enclosed. I use a 10 ohm 1W resistor in series with the LEDs to take away half of the heat.
 

Adjuster

Joined Dec 26, 2010
2,148
:D Question: I have a schematic and need assistance to turn into a reality.
Problem: 1. Origional schematic listed 47,000Uf @ 35 vdc for capacitors
C1, C2, & C3 of attached schematic. What is proper parts count. Am I right
in suspecting something amis due to too large of capacitance. Another book states 2000 Uf per amp. Transformer going into this power supply to be a
24vac @ 3 Amp.
Just in case you have not realised this by now, capacitance adds in parallel, not in series. Thus two 10000μF capacitors in series would be equivalent to only 5000μF. To obtain 20000μF they would need to be put in parallel.

I suspect that you originally put three capacitors in series in an attempt to get a larger total. This would not work.
The effective capacitance Ctotal of a number of capacitors in series is given by Ctotal = 1/((1/C1) + (1/C2).... + (1/CN))

http://www.allaboutcircuits.com/vol_1/chpt_13/4.htm
 
Last edited:

Thread Starter

kbyrne

Joined Dec 10, 2011
92
Just to clarify something. The origional schematic By author G. Randy Sloan of Audio pile Cookbook fame lists the op amp as a LM1458 single supply set up. I have breadboarded it to success. Question: Will a TL071 single op amp work the same as a LM1458 dual op amp.
His book says any garden varity op will work. The LM3914 is a beautiful IC I have used on another VU meter style and it did not overheat but preforms satisfactorly. The preamp &
Tone control will be in another box under this small instrument case box. From your computations I still have 700MA left over for the VU meter. Thankyou for all advise and feedback.
 

Audioguru

Joined Dec 20, 2007
11,251
An LM3914 is a voltmeter with linear steps. The lowest steps (100mV, 200mV, 300mV) are very far apart and the highest steps (800mV, 900mV, 1000mV) are very close together.

An LM3015 is used as a VU meter because it has logarithmic steps like our hearing. Each step is 3dB more or less than the adjacent one. The total range is 30dB for 10 steps.
The datasheet fore the LM3915 has schematics for peak detector circuits.

An LM3016 was called a VU meter IC but maybe it is not made anymore. Its steps in the center are close together and the lowest steps are far apart.

Maybe your circuit is the Velleman VU Meter Kit. It will work with any opamp. I think it sets its LEDs very dim so it does not overheat the IC.
Here is the Velleman schematic:
 

Attachments

Top