voltage~current regulator 48V 2A

Thread Starter

podgymax

Joined Oct 19, 2008
25
I am no using 1N4007 and elec. cap with 50V
Sorry for my wrong drawing in the picture.
Actually, I have brought a 220 TO 24-0-24 transformer
the ele. cap on this circuit is 2200uF/100V, not 50V)
IN5401 diode and 220-48 transformer)
 

Wendy

Joined Mar 24, 2008
23,408
OK, the diode is cool, 3A (be better if it were more amps, but it will work).

48VAC X 1.4 = 67V DC, so we're good to go there. Let me get back with you on the regulators.

Question: Do you want 2A max, or do you want this variable?

I assume you want the 48V DC fixed, not variable.

Question: How quiet do you want this power supply to be? Digital switchers (aka buck boost) are much more efficient, but they have noise. Analog designs are quiet, but they get hot. I'm assuming you want analog (which is what I'll look at).

**************************

OK, you probably won't be able to use the LM317, but the TL783 looks very promising. I linked the datasheet to this post, the application notes that are part of the datasheet cover most of the ground you are interested in. I'll post some circuits, but I suspect the other guys will do better since mine won't be tested.
 
Last edited:

Thread Starter

podgymax

Joined Oct 19, 2008
25
Thx for your reply~ Actually, I want to make a switch-mode charger at last. So, I don't need to make it become very quiet.

I want 2A max and 48V DC fixed. Does any chips available for my circuit? or I need to design a regulator?
 

Wendy

Joined Mar 24, 2008
23,408
I think Wookie might have some better ideas on this subject. We've pinned your unregulated source down enough to know it is practical.
 

Skeebopstop

Joined Jan 9, 2009
358
Just to jump in here. Been a while since I worked with transformers but isn't it something like:

V2/V1 = N2/N1 and I2/I1 = N1/N2. So 220/48 = 4.58
I2/I1 = 1/4.58 and assume I1 has a maximum of 10A, so I2 has a maximum of 2A?

If so I don't know if the cap will charge enough if the load out of the regulator pulls 2A?

I'm probably just thinking wrong, not very power supply savy I'm afraid.
 

Wendy

Joined Mar 24, 2008
23,408
About the cap value, that is a good question. I think it is good enough, especially with a regulator (switching or analog, doesn't matter) after it. The regulator will finish the process of ripple removal.

The part about the transformer, that is a store bought component, so most of the math on that has been done.
 

eblc1388

Joined Nov 28, 2008
1,542
There is a very simple rule for power supply. One cannot get a DC regulated voltage same as the AC secondary output voltage at rated current.

A 48V AC secondary CANNOT provide a 48V DC regulated output, based on the following consideration:

a) there are diode drops(1.4~1.8V in this case with bridge) during rectification process
b) there must be some "headroom" for the regulator to regulate, a LDO regulators helps but still some volts is needed
c) the mains voltage can varies up to 5%, more in some cases

One common mistake is to assume the rectified voltage increased after rectification. This is only true if a small current is taken from the capacitor.

I often laugh when someone post a 30V 5A "regulated" power supply schematic or project, using only a 22v secondary transformer and two 2N3055s.

To be qualify as a true 30V 5A power supply, the PS should be able to provide 30V at 5A or near 0V 5A for an unlimited time.

Anyone who have built power supplies can actually post and say his power supply can actually do that?

A real 30V 5A power supply uses transformer with 36~38V AC secondary, 7+1 2N3055s for the output stages and comfortably provide either 30V at 5A or shorted output at 5A for unlimited time.
 

Wendy

Joined Mar 24, 2008
23,408
I think we're going to have to disagree strongly at this point. A 48VAC secondary translates to 67V DC (48VAC X 1.4, this formula works) when full wave rectified and filtered, providing plenty of headroom. I have built many power supplies from the ground up, many without regulators. A 6.3VAC transformer works great for a 9VDC output (real is 8.9V DC). The circuit under load still measured 9VDC. I have done this and can say that, many times, many ways. The voltage drop under load is dependent on how heavy duty the transformer is, but the translated DC voltage will be very comfortably over the secondary AC voltage. If the capacitor is insufficient for the load their may be transient dropouts, but 2200µF is plenty for 2A.

I was bringing up the headroom issue when we were told it was going to be 52VDC. It isn't, it is 67VDC, a different story. The only debate in my mind is what the regulator will look like.

48VDC regulated, 2A max (current limited). OP prefers switching. Thoughts?
 
Last edited:

Wendy

Joined Mar 24, 2008
23,408
I could draw something up, but it would be untested. I've seen Wookie point to some buck boost regulators.
 

SgtWookie

Joined Jul 17, 2007
22,230
Arguably, one of the easiest ways to generate a schematic for a switching-type regulator circuit would be to use National Semiconductor's WebBench Designer.

It can be accessed from their home page:
http://www.national.com/analog

However, I don't know where our original poster lives, so they may or may not be able to obtain all of the components in the schematic that it generates, and creating a good circuit board for a high-frequency switching supply may not be within our OP's capabilities. It's rather advanced for a first project.

Besides, if it's a lead-acid battery, it would not be 48 volts. It's voltage would depend upon charge percentage and internal temperature, but fully charged at 25°C it would measure about 51.2v, it's "float charge" voltage would be around 54v and absorption charge voltage would be around 56v. But, that's only at one internal temperature; there is a negative temperature coefficient of (roughly) 3mV/°C per cell when you stray from 25°C. A 48v lead-acid battery would have 24 cells, so the coefficient would be roughly 72mV/°C. Manufacturers have different specifications for the temperature coefficient; it depends upon the individual battery chemistry - that's why I said "roughly". If you don't account for it, your battery will have a short life.
 

eblc1388

Joined Nov 28, 2008
1,542
I think we're going to have to disagree strongly at this point. A 48VAC secondary translates to 67V DC (48VAC X 1.4, this formula works) when full wave rectified and filtered, providing plenty of headroom.
No problem with that.

It is not so intuitive to understand especially when a meter would really indicate that it is 67V DC on the capacitor.
 

Skeebopstop

Joined Jan 9, 2009
358
Last edited:

Wendy

Joined Mar 24, 2008
23,408
Something that hasn't been talked about here, and SgtWookie mentioned to me. I'm passing it on to ya'll because it is true.

48VDC is dangerous voltage. It is possible to hurt or kill yourself with it. I've seen techs knocked on their butts by it, and if the current goes the wrong way, you could be facing EOL (End of Life).

Be careful with it.
 

Wendy

Joined Mar 24, 2008
23,408
No problem with that.

It is not so intuitive to understand especially when a meter would really indicate that it is 67V DC on the capacitor.

The reason the meter measure that voltage is because it is there. During the part of the cycle where the peak voltage is dumped into the cap it is very low impedance. A cap has no problem providing the 20ms of time that the transformer isn't providing that voltage and current.

Like I've said, I've built a lot of them, because I was into electronics before the age of wall warts (bless them). Back in the day designing and building a power supply was de rigueur. I still use the variable I built using a 24V transformer, it goes to 30V quite nicely (after regulator circuitry).
 

Skeebopstop

Joined Jan 9, 2009
358
Something that hasn't been talked about here, and SgtWookie mentioned to me. I'm passing it on to ya'll because it is true.

48VDC is dangerous voltage. It is possible to hurt or kill yourself with it. I've seen techs knocked on their butts by it, and if the current goes the wrong way, you could be facing EOL (End of Life).

Be careful with it.
Aye, 50VDC is the cutoff for EVL (extra low voltage). Beyond that electrocution comes into play.

I did always find it funny how VL (low voltage), goes up to 1000VDC :)

Might I recommend you tie your secondary center tap off to a referenced earth ground somewhere so that it is +-24VAC so you should end up dealing with +-24VDC after regulator.
 

eblc1388

Joined Nov 28, 2008
1,542
Like I've said, I've built a lot of them, because I was into electronics before the age of wall warts (bless them). Back in the day designing and building a power supply was de rigueur. I still use the variable I built using a 24V transformer, it goes to 30V quite nicely (after regulator circuitry).
I'm speechless. :(

To your comments I can only provide you with a simulation, which sort of reinforce my previous explanation.

The circuit applies a 24Ω loading to the capacitor, which simulate a 2A loading. You can see how the voltage on the capacitor drops from 67V to become some 48V under load, which is sort of expected.


 

Attachments

Wendy

Joined Mar 24, 2008
23,408
In theory there is no difference between theory and practice.

In practice there is. And I think there are holes in how you are applying theory.

Been there, done that, still using the results (my personal power supply). It's not a fluke, but has been very repeatable for the many I've built.

So where is all the noise on your simulation coming from?

I took this from the AAC book, which I happen to agree with. This is what you should have had...



http://www.allaboutcircuits.com/vol_3/chpt_3/5.html

A full wave bridge fills the gaps even better.

For what's it's worth, I've commited to writing a power supply section for the book.

Anyhow, this is getting off topic. If you're wanting to continue the discussion we can start a new thread.
 
Last edited:
Top