#### dkmac

Joined Jul 11, 2013
32
Hello all,
I am in the process of learning about circuits and how they correlate to each other for a final output. For my next step, I am wanting to build a transformer, more specifically I want to know if it is possible to use one transformer circuit to have two different output? I have been learning a bit about the LM317, LM338, and LM350; however, from what I have found just for a transformer, I am not sure if there is a way to factor in what type of capacitors and their respective values should be? Here is what I am wanting to do, take a 110Vac input, rectify it and have one 12Vdc/2amp output and one 11.5Vdc/5.4amp output. If I had to build two separate transformers, I can do so. The important part of both outputs, they have to be a constant amperage and can be adjustable voltage.

The second part of my question is lets say the 110Vac source has 12amps and I wanted my final DC output to have 5.4amps, does this current drop need to be made before the transformer circuit, or do the values of the circuit take care of this? Any help or direction with this will be greatly appreciated. Take care.

Donny

Joined Jul 18, 2013
27,658
When you talk about capacitors, I assume you mean smoothing capacitor on the output of a bridge or full wave rectification?
The value of the capacitor depends on the maximum current and the required % of ripple.
Oversizing capacitors can have an effect on the required VA of the transformer.
You can have as many secondaries as physically possible, within reason, all contribute the final VA requirement.
When you say 'regulated' is this at the DC level or AC?
the required va at those values is a very minimum op 86va so ~150va would be a practical value to start with.
Any precise regulation is usually done at the DC level and would require a higher initial DC input level.
Max.

#### dkmac

Joined Jul 11, 2013
32
When you talk about capacitors, I assume you mean smoothing capacitor on the output of a bridge or full wave rectification?
The value of the capacitor depends on the maximum current and the required % of ripple.
Oversizing capacitors can have an effect on the required VA of the transformer.
You can have as many secondaries as physically possible, within reason, all contribute the final VA requirement.
When you say 'regulated' is this at the DC level or AC?
the required va at those values is a very minimum op 86va so ~150va would be a practical value to start with.
Any precise regulation is usually done at the DC level and would require a higher initial DC input level.
Max.
Essentially, what I am looking at is taking the the schematic below and modifying it to have an output of 24vdc/8amps to feed two separate sources. What I am really curious about in doing so is how would I know what size capacitors to use for C1-C5? I have a nice little calculator to help me with a fixed values for R2 and the adjustment leg of the LM314 or LM350. Thank you and take care.

Donny

#### Attachments

• 39.2 KB Views: 33
Last edited:

#### shortbus

Joined Sep 30, 2009
9,947
What voltage do you want? First post was 12V and 11.5V, now it's 14.7V?

You do know that the rectified voltage increases in a full bridge by 1.41 -2 diode drops?

You do know that the regulators you are talking about have a needed 'headroom' voltage?

Unless your doing this for a learning experience, you can by a ready made power supply that will work right out of the box, for a lot less money and fooling around. Probably even cheaper than just the cost of a transformer. What it sounds like you need is called a "lab power supply".

Not trying to be mean, just practical.

#### dkmac

Joined Jul 11, 2013
32
What voltage do you want? First post was 12V and 11.5V, now it's 14.7V?

You do know that the rectified voltage increases in a full bridge by 1.41 -2 diode drops?

You do know that the regulators you are talking about have a needed 'headroom' voltage?

Unless your doing this for a learning experience, you can by a ready made power supply that will work right out of the box, for a lot less money and fooling around. Probably even cheaper than just the cost of a transformer. What it sounds like you need is called a "lab power supply".

Not trying to be mean, just practical.
Hi, thank you for the reply. I know that you are not trying to be mean. I am wanting to learn about this, this is why I want to build it. I know that I did put 12vdc and 11.5vdc then changed it to 14vdc (I actually meant 24Vdc not 14Vdc sorry). My reasoning for this is initially I was thinking that I would have to have two different dc rectifiers for my outputs, then it hit me that as long as I have enough coming out of one rectifier I could feed both components (14vdc/8amps to be safe). Each component that is connected to my output will only use the wattage that the device requires and not the entire 24vdc/8amp. So this is why I changed from my original posting. Additionally, I think that all that I need is a rectifier and not a transformer because I can dial in the Vac that I need to pick off, for example I have the capability to tap only a 24Vac source, thus eliminating the need to step down voltage from 110Vac to 24Vac. Please let me know if I am looking at this correctly. So now if I am wanting to convert the 24Vac to 24Vdc/8amps, how would I go about calculating the needed capacitance within the circuit? I have a calculator for the resistance for the circuit and I know that the voltage is adjustable because of the potentiometer. Thank you again for your help.

Donny

#### SgtWookie

Joined Jul 17, 2007
22,230
Well, we should really step back a moment, and ask what your application is? I mean, do you plan on powering two different types of circuits, or charging a couple of batteries - what exactly is the purpose for building this supply?

If you wish to build an adjustable linear power supply, you are really better off sticking to just an Ampere or perhaps two of output current. Otherwise, your power dissipation (ie: waste heat generated in the supply) can quickly get out of hand, and you need some radical cooling like liquid nitrogen or something to keep the regulator from shutting down due to overtemp.

It's good to build a linear power supply for educating yourself, and for having something to power other projects with. However, when you're dealing with a fair amount of current, you really should consider the benefit of using a SMPS; a switch-mode power supply. These SMPS's are very efficient compared to the old-fashioned linear power supplies, and generate far less waste heat. They are more complex than linear supplies and more difficult to troubleshoot - but from the sound of your requirements, you may very well get by splendidly with using a PC ATX power supply converted to a bench supply. Google "ATX Bench Supply" for lots of ideas.

However, if you are planning on charging a battery or batteries with it, you need to tell us - and the chemistry of the battery - as battery charge requirements are very different from powering a typical circuit.

#### dkmac

Joined Jul 11, 2013
32
Well, we should really step back a moment, and ask what your application is? I mean, do you plan on powering two different types of circuits, or charging a couple of batteries - what exactly is the purpose for building this supply?

If you wish to build an adjustable linear power supply, you are really better off sticking to just an Ampere or perhaps two of output current. Otherwise, your power dissipation (ie: waste heat generated in the supply) can quickly get out of hand, and you need some radical cooling like liquid nitrogen or something to keep the regulator from shutting down due to overtemp.

It's good to build a linear power supply for educating yourself, and for having something to power other projects with. However, when you're dealing with a fair amount of current, you really should consider the benefit of using a SMPS; a switch-mode power supply. These SMPS's are very efficient compared to the old-fashioned linear power supplies, and generate far less waste heat. They are more complex than linear supplies and more difficult to troubleshoot - but from the sound of your requirements, you may very well get by splendidly with using a PC ATX power supply converted to a bench supply. Google "ATX Bench Supply" for lots of ideas.

However, if you are planning on charging a battery or batteries with it, you need to tell us - and the chemistry of the battery - as battery charge requirements are very different from powering a typical circuit.
Hi, thank you for your reply. I am not planning on charging any batteries with this circuit. If you are thinking that this is too much wattage for the circuit, I can break it down into two different regulator circuits. As stated before, if I do this I will need one that powers 12Vdc/3amps and another that will power 11.5Vdc/5.5amps to be exact. I am really wanting to build this for two reasons, A) I want to learn and B) I think that it will be very fun. I just built my first circuit a few weeks ago from scratch and had a blast and learned a lot from both Wayne and Bill. On a side note, would the circuit that I posted earlier work for the two separate voltage regulators, and if so will all I have to do is compensate for the Capacitor values C1-C5? Thank you for your help as it is all very much appreciated.

Donny

#### GopherT

Joined Nov 23, 2012
8,009
Ok, a multi-amp power supply s a big project for a beginner. I would suggest, for a lab-power supply, out start out more modestly, and safely with a 500 to 1000 milliamperes power supply. Heat is not so much of a problem and it will work with 95% of the projects you will build until you realize that you will essentially build a new power supply for each mains powered project you eventually build and that a commercially built power supply is a better bench power supply than something you make yourself.