Sallala

0-30 Vdc Stabilized Power Supply

Recommended Posts

Thanks, the post ended up out of place and it was not clear to whom you were answering.

I agree that it is not good practice to parallel the power supplies, but you should know that it is general practice to parallel regulators such as the 78XX or 79XX series to get more current rating from them. In this case, it does work. When doing this, you should use a small resistance like 0.1 or 0.22 ohm between the output pins because all regulators are going to be a little different and this will keep the current flowing equally between them. This is the same practice as paralleling power transistors.

MP

Share this post


Link to post
Share on other sites

MP, to boost the current output of a regulator, a power PNP transistor and input sense resistor is recommended to be wired across the regulator, as shown at fig. 10 in the datasheet of the MC78XX series by ON Semi here:

http://www.onsemi.com/pub/Collateral/MC7800-D.PDF

This way, their superb load regulation (the output drops only 1.3mV typ. with output changes of 5mA to 1.5A) is kept.
If you parallel regulators as you mentioned with 0.22 ohm resistors on their outputs, the output will drop up to 330mV.

You are correct that transistors can be paralleled this way and are equalized with minimal consequence.

Share this post


Link to post
Share on other sites

The boost with transistor(s) is how I do it, also. Although I use a NPN instead. Either method can be used. The reason I jumped in was your statement "CANNOT". And as you see, it CAN be done.

You are correct that you will have some loss due to the resistors, but overall, you will have a boost in current handling. If a 12 volt regulator is rated at 1 amp, then this would mean the internal resistance can be resembled as 12 ohms. 12/12 = 1 amp. If you add the 0.22 resistance, then 12/12.22 = 0.98 amps. If you have two regulators in parallel in this manner, you will now have the capability of 1.96 amps, which is not double, but is close. If someone is needing to get exactly 2 amps out of 1 amp regulators, then this will not quite get them there. But if they need to get more than 1 amp or need to push this regulator a little over it's limit, then it is certainly a solution. Especially for a beginner who is not up on transistor formulas and cannot build a device with transistors. All they need is ohms law, a couple of resistors and another regulator.

I am not telling you your method is wrong. Just that the other might be an option for some. ;)

MP

Share this post


Link to post
Share on other sites

I had an earlier problem with some componemts getting very hot which I eventually traced to a broken trace (pun not intended).
Everything is "cool" now but I have another problem that I cannot figure out.
I(max) varies with the load. For example, the supply will crank out more than 3.5A with a 5R resistor across the DC output. However, with 15R I(max) drops to abot 1.6A; 20R=1.2A; 32R=0.8; 75R=0.36A. AS it can be seen the relationship is not linear but follows a parabolic curve.
The V(max) is "approached" with increasing R but never reached.
I am using a 24V 8A tansformer, so the problem is not here. Any idea? I am lost and ready to give up.
Thank you,
Mastrila. ???

Share this post


Link to post
Share on other sites

Hi Mastrila,
Maybe when your parts got hot they were damaged.
The guys who are talking about modifying this power supply to get 5A of current output are having problems similar to yours with the original circuit. Their post is here:
http://www.electronics-lab.com/forum/index.php?board=13;action=display;threadid=196
Those guys and I believe that the circuit depends too much on "selected" parts as follows:
1) A transformer that has really good regulation (low resistance windings) so that its voltage doesn't sag too much at the voltage peak when it is sourcing many amps to charge the filter cap.
2) Rectifiers that have a low voltage drop during #1 above.
3) A filter cap. that is at its extreme max. tolerance (+100% is double its rated value) and has a very low ESR.
4) Q1 and Q2 with more hFE than minimum or typical spec.
5) A power utility that provides 120VAC at all times (don't your lights dim when your computer is turned-on?).
6) Good luck.
Such a circuit can be described as having a "minimal" design.
A 2N3055 might be older than me. A newer PNP output transistor or power FET might be better.
I haven't built this circuit and won't unless it is upgraded so that it can meet its own specs.

Share this post


Link to post
Share on other sites

Hello all, I hope someone can help. Im new to electronics, but want to build this power supply as a first project.
I have studied the circuit and Im beggining to understand bits of it, but I want to build the PSU with a fitted moving coil volt and ammeter. Please can someone point me in the right direction as to where and how to connect these items to the power supply PCB?
I know this subject hes been looked at by a few people in this thread, but I need a few clear instructions....Im a newbee.
Thanks
Ian

Share this post


Link to post
Share on other sites

Hi Ian, welcome to Electronics-Lab and to the hobby.
1) Connect the positive terminal of a 3A DC ammeter to the positive output of the circuit.
2) Connect the positive terminal of a 30V DC voltmeter to the negative terminal of the ammeter. This junction becomes the positive connection of the power supply to your load.
3) Connect the negative terminal of the voltmeter to the negative output of the circuit. This junction is the negative connection of the power supply to your load.
This way, the voltmeter shows the voltage across your load. The voltage will drop a little when the load current is increased due to the voltage-drop across the ammeter.
In order to avoid the voltage-drop across the ammeter, a low-current ammeter in series with a resistor can be connected across R7 to indicate the output current. But the resistor must be correctly calculated to match the ratings of the low-current ammeter. Also its scale markings will require changing to "amps".

Share this post


Link to post
Share on other sites

Ian, do you really need this old technology? Mixos has posted a diagram of the current meter and voltmeter connection on page 2 of this post. This is with the 7107 IC, which will give you a much better digital display. Just a suggestion.


MP

Share this post


Link to post
Share on other sites

Hi, thanks for the info. I re read the posts, and was thinking along those lines myself this afternoon. I think it was just a nostalgic thought on my part wanting to see the old 'moving needle'...lol.
Im going to start hunting for parts to make this up, and I will have a go at the digital meter as well. Ive made simple things in the past, but this is a bit of a challange for me. Hope no one thinks Im being a pain with the Newbee questions.
Ian

Share this post


Link to post
Share on other sites

it sounds like you've got some oscillation in the circiut somewhere. Check your rectifier, but apart from that, i'm not sure what else could cause it (maybe a short somewhere feeding back?)

Share this post


Link to post
Share on other sites

juanpmoron, C7 is just a bypass cap on the output. Most likely, you have either connected it wrong or you are using a capacitor with too low of a voltage rating. Was it a used capacitor? What do you have connected to the output that audioguru thinks might cause your problem? In an earlier post you mentioned a problem with another part failing. You should check your circuit with the schematic.

MP

Share this post


Link to post
Share on other sites

Good news, Guys,
I found a high-voltage, readily available and inexpensive replacement opamp for U2 and U3, which are operating in this project with a supply voltage beyond the maximum specified ratings of the original TL081 opamp. It is ONSemi (formerly Motorola) part number MC34071AP. Its maximum supply voltage rating is 44V (this project gives 38V plus).

MP,
Do you recall that a couple of years ago, Chinese electrolytic capacitors were very unreliable? Maybe Juan got some. I know some techs who made a fortune relacing defective computer capacitors.
ONSemi is not recommending the TL081 for new designs (will soon be obsolete).
Please see ONSemi's spec sheet for Juan's MC34063, where they say that maximum ratings are not to be used continuously nor simultaneously. Their wording for the TL081 is the same as TI's (just a copy) which is very confusing.

Juan,
A-HA, I saw your other post about the 24V to 12V switching power supply that is heating too much when powered by this project. My suspicions are correct that C7 is too small physically (low ripple-current rating) and value (not enough storage to provide high-current pulses that are drawn by the switcher). Hopefully, a big C7 will solve both problems.

Share this post


Link to post
Share on other sites

Mastrila,
Your variac, isolation transformer and rectifier diodes probably are not made to have a low voltage drop during the brief but HUGE current surge when the filter capacitor, C1, is re-charged. Therefore the voltage across C1 (at the lowest point of its ripple) probably is too low. Let's call this voltage, V+. V+ must be more than the project's output voltage by at least the total amount of voltage-drops that follows:
1) The base-emitter voltage of Q4. (0.9V)
2) The base-emitter voltage of Q2. (0.8V)
3) The voltage-drop across R15. (1.0V)
4) The output-high voltage of U2. (1.2V)
With the project providing 3A, voltage drops above add up to 3.9V.
Therefore if you want 30V output, V+ must be at least 33.9V.
A 24VAC transformer, low-current rectifiers and a low-value for C1, supplying 3A, won't give you this. What do you measure across C1?

Share this post


Link to post
Share on other sites

concerning REPLY#26
I believe you will need a power supply which is isolated from the voltage which you are measuring for the panel meter to work (ie:the power source should be a 9v battery per manufacturer). This is causing me problems as I have 25 students encountering this problem (we added a panel meter to their project this year).
Does anyone have any suggestions/ alternatives to the 9v battery other than adding a transformer.

thanks

Share this post


Link to post
Share on other sites

EIST, the solution of using a battery would be a simple solution. As you have stated, the alternative is to use a transformer. The voltmeter needs a different source than the voltage which it is measuring.
If you could describe something about the rest of the project, maybe we can come up with another alternative. It sounds as if the project has some voltage in it to be measured. Perhaps there is a place where you can tap some of it.

MP

Share this post


Link to post
Share on other sites

EIST,

The alternative to a second transformer is to put some turns of insulated wire on top of your transformer winding. I have done this on several occasions when I needed an extra output. It can be tight sometimes, but you don

Share this post


Link to post
Share on other sites

HI
I AM NEW TO EL-LAB
I READ ABOUT 0-30V POWER SUPPLY I AM ALSO INTRESTED TO MAKE IT
1st OF ALL TRANSFORMER SHOULD HAVE1.6 TIMES HIGHER :o CURRENT RATING THEN SPASIFIED (NOT 3A IT SHOULD BE 3.6 A) HEATSINKING OE 2N2219 IS NOT EASY. IT SHOULD BE REPLACED BY TO-220 TYPE OF TR . DIODES OF BRIDGE RECTIFIER SHOULD ALSO BE CHANGED TO 4A TYPES. THE FILTRING CAP SHOULD ALSO BE CHANGED TO 1OOO MF TO HANDEL LARGE CURRENT SURGES.

Share this post


Link to post
Share on other sites

Hi Zia, welcome to our forum. About this power supply:
1) We have already determined that the transformer's 3A rating is too low because the output of the power supply is 90VA (30V X 3A).
If you use a much higher current rated transformer, then its voltage won't sag as much during rectifier conduction, and its voltage won't rise as much with no load. Maybe a 5A rating is good.
2) I don't remember if the heatsink problem with the little 2N2219 was discussed, but I agree with you that a T0-220 would be much better.
3) We also determined that using a higher current bridge rectifier instead of those little diodes will reduce losses and allow for a heatsink.
4) We have had many suggestions for the size of the filter capacitor. In order to reduce loss due to ripple voltage, make it really large, such as 10,000uF.
5) As you have read here, the minimum and maximum voltages at the filter capacitor are crutial:
a) A minimum voltage of at least 33.9V when the project has a 30V, 3A load.
b) A maximum voltage of less than 36.4V when the project has no load, when using the 44V replacement opamp.
If you include a 10% mains voltage variation then these voltages are impossible. Perhaps this project should be rated to only 25V output, and when using changes as above should perform very well with a 24V/4A or 5A transformer.

Share this post


Link to post
Share on other sites

Hello All I have built this power supply but I got a problem. As soon as I switch it on, the volt meter reads 23.5v and then I try to vary the Volt Pot, it goes up to 35v! In other words, my PSU is only variing from 23.5v to 35v !! I have changed the op amps twice and even tried the TL071 (which is better acording to a friend of mine who read the datasheets) Could anyone help me PLEASE?

thanks a lot

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now