Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

LM317 power supply

Status
Not open for further replies.

bananasiong

New Member
Hi,
From the datasheet, the recommended R1 is 240:eek:hm:, can I replace this with other value like 270:eek:hm: or 220:eek:hm:? I plan to make a power supply with a 24-0-24v transformer (8VA). I'm using 5k:eek:hm: variable resistor and a 500:eek:hm: variable resistor in series to adjust the output voltage (R2). And a 220:eek:hm: as R1. So I can adjust the voltage up to 30V? I don't use high current, only below 1A.

Thanks
 
bananasiong said:
Hi,
From the datasheet, the recommended R1 is 240:eek:hm:,
No, 120:eek:hm: is recommended for the LM317, 120:eek:hm: is recommended for the LM217 and LM117.


can I replace this with other value like 270:eek:hm: or 220:eek:hm:?
The LM317 is just a fixed 1.25V regulator with a minimum load current (see datasheet). You can replace it with any value you like providing it fulfils the minimum load requirement.

I plan to make a power supply with a 24-0-24v transformer (8VA). I'm using 5k:eek:hm: variable resistor and a 500:eek:hm: variable resistor in series to adjust the output voltage (R2). And a 220:eek:hm: as R1. So I can adjust the voltage up to 30V? I don't use high current, only below 1A.
I take it you mean using you 24-0-24V transformer as a 48V transformer.

That would work but the maximum current will be well under 1A and the LM317 will be toasted if it's short circuited - the maximum input-output differential is 40 and your transformer will supply 66.5V.

If you're planning to use just one 24V winding then you'll need a huge filter capacitor because the input will be only 32.5V+ ripple.

Also be warned that AC voltages above 25V and DC voltages above 60V are classified as dangerous.
 
An alternative solution would be to use the 24 - 0 - 24 with a 2 diode rectifier rather than a rectifier bridge.

If you connect the anode of one diode to one 24V winding and the anode of the other to the other 24V winding and connect the cathodes together you have a full wave rectifier with only one diode drop rather than 2.

Connect a filter capacitor from the cathodes to gnd and connect the 0 to gnd.
 
No, I don't use bridge rectifier. Like ljcox said, I'm using only 2 diode to rectify. And I'm using a 1000 uF filter capacitor, should be enough right?

About PTC, I ask it from a electronics shop and they say this is only used for air conditional and television but not in a power supply o_O?? The PTC they have looked like a relay (the size and shape), some with 3 pins and some with 2. They ask me whether I want a NTC or not, then I don't buy either from them, confusing. So which one should I use?

Can I make a power supply like a PSU, when +ve and -ve terminal are shorted, the relay inside is switched and a led indicator shows that the current is not enough. Is this possible? So that the LM317 will not be toasted.

If I use a 24-0-24 transformer compared to a 30-0-30 transformer, for the same application, which one uses more electricity? Is is the 30-0-30 transformer? Because the power dissipated by the voltage regulator is more.

Thanks
 
Hi Banana,
Use 120 ohms for the fixed resistor. If it has a higher value and the load is suddenly reduced, the output voltage will rise and blow up the circuit you are powering with this supply.

The LM317 won't be toasted because it is smart:
1) If the output current is too high it limits the current .
2) If it gets too hot it shuts down then stsrts working again after cooling.
3) If it has more than 15V across it then it reduces the output current. Yours will have an input voltage of 33V. If its output is adjusted to 5V then it has 28V across it and it reduces the current to from 400mA to 700mA.

Your little 8VA transformer is a fire hazzard. It should be 75VA (34V x 2.2A).
 
Oh, I just bought a 8VA transformer. If I want to remain using this, the load current mustn't be greater than 235 mA right?

EDIT: 1 V drop after crossing the filter capacitor, is this true for all the load current or all the value of capacitor?
 
Last edited:
bananasiong said:
Oh, I just bought a 8VA transformer. If I want to remain using this, the load current mustn't be greater than 235 mA right?
Correct. Don't overload the output of the supply or the little transformer might catch on fire.

EDIT: 1 V drop after crossing the filter capacitor, is this true for all the load current or all the value of capacitor?
The amount of ripple voltage depends on the value of the filter capacitor and the amount of load current that discharges it between pulses of current from the rectifiers. With a 235ma load, your 1000uF capacitor will have a ripple voltage of about 2Vp-p. The LM317 needs an input voltage of about at least 2V higher than its output so if the resistor and pot have the correct values then the supply will have a max regulated output of 30.0VDC.
 
Oh!! I should plan this well! I thought I can have 1 A maximum load for this power supply, but I need at least 34 VA transformer.. the size of the transformer becomes bigger if the VA rating is larger right?
 
bananasiong said:
Oh!! I should plan this well! I thought I can have 1 A maximum load for this power supply, but I need at least 34 VA transformer.
The LM317 has a max output current that is typically 2.2A, 1.5A is guaranteed. If its load draws that much then the transformer must be 75VA.

If it has a 34VA transformer then its output current can be 1A only for high output voltages. At voltages below 19V it will have more than 15V across it and it reduces its max current. It also gets pretty hot.

If you use a 12V-0V-12V transformer then all of it can make 32.5VDC so the LM317 can provide 17.5V to 30.0V at 1A. Then half the transformer can make 15.5VDC so the LM317 can provide 1.25V to 15.5V at 1A. The amount of heat will be reduced a lot.
You could make a circuit that automatically switches the transformer to the proper voltage depending on the setting of its voltage adjustment pot.
 
audioguru said:
If you use a 12V-0V-12V transformer then all of it can make 32.5VDC so the LM317 can provide 17.5V to 30.0V at 1A. Then half the transformer can make 15.5VDC so the LM317 can provide 1.25V to 15.5V at 1A. The amount of heat will be reduced a lot.
using bridge rectifier? Then the center tap is not used? How about the VA? is it around 17 VA for the load of 1 A? So my 24-0-24 8VA transformer is not suitable?
You could make a circuit that automatically switches the transformer to the proper voltage depending on the setting of its voltage adjustment pot.
Using relay like a PSU?

Thanks
 
A 12V-0V-12V 34VA transformer can use all its secondary with a bridge rectifier to make 32.5VDC. Then a relay can switch the bridge rectifier to use only half the secondary to make 15.5VDC. Then the LM317 will provide 1A at any voltage and will not get too hot.

A comparator can sense the output voltage to switch the relay.

Your 24V-0V-24V 8VA transformer is not suitable.
 
Ya, I didn't calculate well. I've done a 5 volts power supply with a 9-0-9 3VA transformer to run a PIC, I can even adjust it to maximum to program a PIC at 13V. The PIC doesn't use much current.

So I have to change my transformer, a 12-0-12 with at least 34 VA using bridge rectifier. Use all the secondary? Means including the center tap?

Thanks
 
bananasiong said:
So I have to change my transformer, a 12-0-12 with at least 34 VA using bridge rectifier. Use all the secondary? Means including the center tap?
It is 24VAC from end to end and is 12VAC from its center-tap to one end.
 
bananasiong said:
No, I don't use bridge rectifier. Like ljcox said, I'm using only 2 diode to rectify.
Of course, a two diode, bridge, it should've been obvious.

And I'm using a 1000 uF filter capacitor, should be enough right?
No, use a 45VA transformer minumum and a >12000:mu:F capacitor.
 
If I don't use the center tap, then the frequency is lower (60 Hz) but not twice of the incoming frequency. So I need a larger capacitor right? But I don't get a 12000 :mu:F from the lab, the maximum I think is only 1000 :mu:F so I have to buy it:(
 
Use a 4-diode bridge rectifier module for 120hz full-wave rectification. The center-tap and one end is 12VAC and end to end is 24VAC.
1000uF will create 2Vp-p of ripple when the circuit has a 235mA load and way too much ripple when the circuit has a 1A load. Use two 4700uF, three 3300uF, five 2200uF or one dozen 1000uF capacitors in parallel.
 
bananasiong said:
If I don't use the center tap, then the frequency is lower (60 Hz) but not twice of the incoming frequency. So I need a larger capacitor right? But I don't get a 12000 :mu:F from the lab, the maximum I think is only 1000 :mu:F so I have to buy it:(
It does not matter whether you use 2 diodes and the centre tap or a rectifier bridge. Either way it is full wave rectification.

So if your mains frequency is 60 Hz, then the output of the rectifier will 120 Hz in both cases.

Only half wave rectification will result in 60 Hz.
 
Hi,
For a variable power supply, it is always better to have theree rages with secondary tapped at three ranges (0-12-18-24) and switched, thereby minimizing power loss across the device(LM317). Use of single winding whethwr with or without center tap, and working at low output with high currents --we deifinitely waste power and further we try to cool the device ba a fan, to incerase wastage.

I had seen bench top power supplies that have automatic switching of secondary tap as you try to incerase out put voltage adj POT. They have used relays.
 
Status
Not open for further replies.

Latest threads

Back
Top