Regulated power supply

[JebemMajke]

I have a 20 volt, 3,5 A transformer which I would love to use for a power supply.

So I was thinking 18 volts, 12 volts and 9 volts outputs. But I would like to keep the 3,5A, or as much as possible.

In your honest opinion, lm317 or 78xx, which is better?

[drea]

lm317 or 78xx, which is better?

I don't think either's going to fly with your current requirement. Getting from 20v down to 9v at 3.5 amps means dropping 38+ watts. That's going to need a monster heatsink.

You might be better going with some of those cheapy ebay switch mode dc-dc converters and some maybe extra capacitance to smooth everything out.

This is the sort of thing http://www.ebay.co.uk/itm/XL6009-DC-to-DC-Adjustable-Step-up-boost-Power-Converter-Module-Replace-LM2577-/261342145695?pt=UK_BOI_Electrical_Test_Measurement_Equipment_ET&hash=item3cd934ac9f

[merlinb]

The LM350 is rated for 3A, and the LT1084 can manage 5A, although you still have the same power dissipation problem.

This is the sort of thing http://www.ebay.co.uk/itm/XL6009-DC-to-DC-Adjustable-Step-up-boost-Power-Converter-Module-Replace-LM2577-/261342145695?pt=UK_BOI_Electrical_Test_Measurement_Equipment_ET&hash=item3cd934ac9f

I bought a couple of those exact modules recently for powering valve heaters. Work a treat!

[Jdansti]

Or use multiple 1A regulators with an independent output jack for each.

[Seljer]

With both types of regulators, you can wire in an external transistors (eg. 2n3055) for increased current capacity. There are example circuits given in the data sheets. But yeah, you'll need hefty heatsinks!

[GibsonGM]

OTOH, obviously you CAN use a regulator to drop down to, say, 9V and keep it below the rating for the regulator (1A...) ..why the need for a monster 3.5A at those voltages??? 

1A will run an AWFUL lot of stompboxes!  Just curious...

[PRR]

What specific thing do you have which needs all 3.5 Amps?

As said, that's big power. Generic solutions may not be best.

At the moment, I have a need for 12V at 2.5 Amps. But it is a very steady load. And it can eat DC or AC. (Tractor spot-lamp.) For *that* case I'd run a dumb *resistor*. 3.2 Ohms (20 Watts!) between the 20V AC and the lamp.

Or if you have 100 pedals each eating 35mA, then five/ten "1A" regulators makes much sense. (Especially since one crushed/shorted power lead would still leave 80/90 pedals running.)

> Getting from 20v down to 9v at 3.5 amps means dropping 38+ watts. That's going to need a monster heatsink.

Actually: 20V AC wants to rectify-out to 28V DC. So he has 19 Volts to waste-off, more power than you figured.

OTOH, 3.5 Amps AC rating will NOT comfortably support 3.5 Amps of _DC_ load. Just as voltage goes up (peak-catching a sine wave), current must go down. Also a rectifier-cap power supply is very hard work for the transformer winding. Expect less than 2 Amps DC without sag (which might regulate-off) or overHEATing.

Any of the "1A" regulator chips can be used with a booster transistor(s) to pass as much current as you can carry. 1,000 Amps is not out of reach (Darlington connection or equiv.). The heat, and heat-sink, problem remains, and can be formidable.

[JebemMajke]

What specific thing do you have which needs all 3.5 Amps?

As said, that's big power. Generic solutions may not be best.

At the moment, I have a need for 12V at 2.5 Amps. But it is a very steady load. And it can eat DC or AC. (Tractor spot-lamp.) For *that* case I'd run a dumb *resistor*. 3.2 Ohms (20 Watts!) between the 20V AC and the lamp.

Or if you have 100 pedals each eating 35mA, then five/ten "1A" regulators makes much sense. (Especially since one crushed/shorted power lead would still leave 80/90 pedals running.)

> Getting from 20v down to 9v at 3.5 amps means dropping 38+ watts. That's going to need a monster heatsink.

Actually: 20V AC wants to rectify-out to 28V DC. So he has 19 Volts to waste-off, more power than you figured.

OTOH, 3.5 Amps AC rating will NOT comfortably support 3.5 Amps of _DC_ load. Just as voltage goes up (peak-catching a sine wave), current must go down. Also a rectifier-cap power supply is very hard work for the transformer winding. Expect less than 2 Amps DC without sag (which might regulate-off) or overHEATing.

Any of the "1A" regulator chips can be used with a booster transistor(s) to pass as much current as you can carry. 1,000 Amps is not out of reach (Darlington connection or equiv.). The heat, and heat-sink, problem remains, and can be formidable.

I've measured 3,5 A and 20 v from it. Because I couldn't find any markings on it, I';ve placed a rectifier bridge and a big cap ( around 2200u marked for 100v, because I wasn't sure of the voltage, so I wanted to make things safe ). And DMM said 20,01 v and 3,5A.

And since it's only 20V it's not going to be enough for Tiny Giant to have enough watts ( 50+ ). Or maybe I am wrong?

Well, I wanted to make a big power supply for all my pedal/preamp needs. From fuzz face to valvecaster, delays, jfet preamps, doubling voltage of some boosters/distortions  and etc. Something like 10-15 outputs.

[Jdansti]

How did you measure 3.5A?  Did you have your meter in series with a load?  What was the resistance of the load?

Also, are you measuring 20VAC or 20VDC?

[JebemMajke]

My Chinese made DMM has an option for measuring Amps, It read 3,5A.

As for volts, well it has written V-- next to it. But when I measure a fresh 9v battery ( duracell ) it says 9,5v. So I guess it measures DC.

[GibsonGM]

Ohhh.....I see, Jebem.     Sometimes there is confusion about the difference between voltage and amperage.
Amperes are units of electrons that flow when you place a load on a power supply...they are the current that powers things.
Voltage is the PRESSURE the electromotive force is under.   A resistance (load) is like a valve that slows the flow of current.

We can only measure the voltage on a power supply that has no load connected - you need a load to measure current.   If you were to place a 1K resistor across your 20 V supply, you should then be able to measure  .02A, or 20 mA, by placing the meter in series with that load.   That would be current flowing THRU the resistor.    The resistor (called the 'load') DRAWS the current out of the supply; it is not pushed into it, and won't flow out of the supply until such a load is connected.   All our pedals are 'loads' for our power supplies.    Google "Ohm's Law" and read up for more info!  And "how to use DMM".  Remember - never try to measure resistance with the power supply turned on!!  Might fry your meter or at least its fuse.

V---  means DC voltage;   V~  or so, that means AC voltage.

Your power supply can output UP TO 3.5A  AT the 20 volts, is what those figures mean.   YOU can use it for anything you like within reason.    Open, it may measure a higher voltage because...there is no load on it.  That is the "Unloaded voltage"

Setting up a 1 or 2A supply at 9V and/or 18V would be very, very useful for what we do here, and can be done fairly easily with just a few parts (caps, resistors, regulator ICs).     

[duck_arse]

^ what gibson says, also Mr Ohm says slightly differently.

IF you do connect a 1k across your supply, and then measure the voltage across that resistor, you can then calculate the current flowing and the power dissipated. if you get less than 20V, your load is too great.

[JebemMajke]

I did 1k resistor thing, and I;ve measured voltage across the resistor and it measured 19,5-19,6 volts dc. ( resistor connected one side to positive and other to negative and than lugs of the dmm conneceted to the eather side )

But when i measure it in series with the resistor its' 20,6 v dc. ( one side of the resistor connected to the positive other to one lug of dmm and other lug of dmm goes to negative of the transformer )

So, what way is the correct one, and what does those figures mean?

[Seljer]

You approached something in the wrong way because the numbers aren't adding up. If hooking up a 1kiloohm load that draws roughly 20mA causes a 1V voltage drop from the open circuit voltage, that means the internal resistance of the transformer+rectifier is in the range of 1V/20mA = 50ohms which is way too much for a transformer that should supply 3.5A

A good way to ballpark transformer power ratings is by weight. 20V*3.5A = 70Watts. That'd probably be about 1kg of metal! Is your one that big?

[JebemMajke]

Nope,

But is big and heavy.

Just measured the thing on scales. 0,86 kgs

[GibsonGM]

1)  The first way you measured the voltage across the resistor was correct - that is how you measure voltage. In PARALLEL with the load.  Your reading seems a little low for such a small current output, but we will get to that in time. 

You measure current (Amperes) in SERIES, which you should not do until you read up on using your DMM     You have to learn about what things you might do wrong to burn up the meter or its fuse.   Trying to measure currents that are too high for the range on the meter will pop the internal fuse.  Trying to measure resistance in a circuit under power will probably do the same or WORSE, fry the meter! 

Read this to become more familiar with how to make measurements:
https://learn.sparkfun.com/tutorials/how-to-use-a-multimeter


-------------------------------------------------------
2)  Now - this is only a transformer?  No rectifier, no diodes?  You plug one end into a wall socket and the other is just bare wires?  If so, you must set your meter on AC unless it is automatic.   Transformers output AC, not DC, voltage.   If you can post a photo of the transformer, with your meter next to it, that would be nice!   We can help you turn that AC into DC, if it is necessary.  Not very difficult, and this is a good project.

From the sound of it, your transformer (?) is more than capable of making you a good hobby power supply, known as a 'bench supply'.   As suggested, maybe for 1A, and not the whole 3.5A, which you probably will never use in this line of electronics.   Working this kind of thing little by little will help you understand exactly what is going on, which is a nice way to learn!   

[Jdansti]

Just to add to Gibson's last post, the output of a transformer has AC and DC components. You'll probably get a reading with your meter set to AC and set to DC. We try to minimize or eliminate any residual AC component (ripple) when we convert AC to DC for pedal power supplies.

[GibsonGM]

Just to add to Gibson's last post, the output of a transformer has AC and DC components. You'll probably get a reading with your meter set to AC and set to DC. We try to minimize or eliminate any residual AC component (ripple) when we convert AC to DC for pedal power supplies.

+1,000.   Ripple (AC "pollution" on the DC output) usually manifests itself as a BUZZZZZ that is incredibly annoying.
Using a rectifier to "cut" the AC properly, and caps with a regulator to filter and control the voltage, will get you what you want.

Read this...this is the basic idea of what you're looking at doing:  http://www.instructables.com/id/Make-a-simple-12-volt-power-supply/    There is more to learn than just the 'parts' shown in this, but the basics are here...

[PRR]

> Amps, It read 3,5A.

You connected the amp-meter right across the output?

That is a Dead Short. Since voltage is zero, power output is zero.

Also all 20V*3.5A= 70 Watts is now dissipated INSIDE the transformer. It will get hot.

A gross guide to *useful* power is maybe 1/10th of MAXimum current. Say 0.3 Amps. Then you may still have 18 Volts, and can get some work done, without smoking the transformer.

You say "large". How large is large? Get the weight then look through Hammond's line of transformers, find something of similar voltage and weight, see what current they say.

20V DC is plenty for Tiny Giant. It internally regulates to 12V, so you only need 15V DC. However it isn't clear you have enough current for LOUD output. You could try it, with your voltmeter on the raw DC, see if it drops close to 15V.

[JebemMajke]

1)  The first way you measured the voltage across the resistor was correct - that is how you measure voltage. In PARALLEL with the load.  Your reading seems a little low for such a small current output, but we will get to that in time. 

You measure current (Amperes) in SERIES, which you should not do until you read up on using your DMM     You have to learn about what things you might do wrong to burn up the meter or its fuse.   Trying to measure currents that are too high for the range on the meter will pop the internal fuse.  Trying to measure resistance in a circuit under power will probably do the same or WORSE, fry the meter! 

Read this to become more familiar with how to make measurements:
https://learn.sparkfun.com/tutorials/how-to-use-a-multimeter


-------------------------------------------------------
2)  Now - this is only a transformer?  No rectifier, no diodes?  You plug one end into a wall socket and the other is just bare wires?  If so, you must set your meter on AC unless it is automatic.   Transformers output AC, not DC, voltage.   If you can post a photo of the transformer, with your meter next to it, that would be nice!   We can help you turn that AC into DC, if it is necessary.  Not very difficult, and this is a good project.

From the sound of it, your transformer (?) is more than capable of making you a good hobby power supply, known as a 'bench supply'.   As suggested, maybe for 1A, and not the whole 3.5A, which you probably will never use in this line of electronics.   Working this kind of thing little by little will help you understand exactly what is going on, which is a nice way to learn!   

I'll respond to no2, because no 1 has to be done

This is with the rectifier bridge ( 4 diodes, ) and a big ass cap ( 2200 uF, 100v ).

Don't get me wrong, but this is not my first power supply. I've made others, based on geofexs  spyder.

But i am open for learning new stuff, and correcting mistakes that I am making

The only thing that is different, between power supplies I've made before and this one, is that this is not a new transformer, but one which I've salvaged from an old cassette/vinyl deck.

It has no markings on it, so I've placed rectifier bridge with a big cap and measured volts, and tried to measure amps. But failed with the amps

Anyhow the size of the sucker is 60mm*40mm*50mm, weight 0.86kgs

Thing is, I don't have to use it's whole amperage power, just to make it usable for 18v and 9v projects. Be it 1A or 2,5A it's all good.

I was just curious of it's max power.