Need some clues

[colin55]

It would be better to use both halves (sides) of the secondary. Use 2 diodes and the centre-tap. Or you can use the bridge, but it introduces two diode-drops.
Using the two windings improves the output enormoulsy.

 

[xtrmi]

I actually thought the same thing but when I tested it I got a bigger drop. If you want I can redo the connections and take that data also. I will need to find the diodes at the store or in some broken stuff I got laying around here.

 

[Bob Scott]

I recently built a linear power supply and I expected 1.25 - 28v from it @1.5amps. I'm using a Triad F7-28 transformer I get about 34 vac from it.

I'm surprised that no one has mentioned......You must to multiply the RMS AC voltage by 1.414 to get the peak value, the resultant DC voltage is derived from the peak value of the AC waveform:

(1.414 X 28) = 39.6 VDC
less a couple of volts for rectifier drop to 37.6 VDC when the circuit is loaded.

A transformer's voltage is rated at its rated current, so if your circuit does not draw 2 amps, the voltage will rise slightly. Small transformers have more of a voltage rise than larger ones.

If your regulators are overheating, add heatsinking and consider switching to a lower voltage 24 VAC transformer. It should have adequate output voltage when rectified to produce 28V, even after losing a couple of volts in rectification and 3V for regulator dropout.

It gets more complicated. You have to derate the current capability. Here is a guide from Hammond regarding rectifier design with all the answers:


Bob

 

[xtrmi]

Thanks for the pdf Bob. I've actually solved that problem after doing the calculations. I changed r1 from 100ohm to 300ohm. So Now I'm at 20V max, but I'm having an issue with a large voltage drop. I haven't done the calculation yet, but it was suggested to lower the resistance across the whole divider to ease the drop. I'm not sure if that will help, but I'm willing to try it. I have a very large heat sink on the regulator now. It has about the same amount of aluminum as a stock cpu heatsink. I got it out of an old psu and was too lazy to cut it down. I cant remember what the conditions were but it did feel warm when I was testing it at one time; but usually it is room temp.

 

[marcbarker]

Are you sure it's a 317K ?

If the input voltage is really 19.2 and the output is 11 V, the regulator looks like it's overloaded.

 

[xtrmi]

Its not a 317K its a 317T, but that isn't in my component list. I still need to learn spice so I can make my own models. When you say overloaded do you mean I drawing more than 1.5amps from the regulator? I believe the K is a can type package. I think that is what it is called.

 

[marcbarker]

When there's a possibility the regulator will be required to run unloaded, I use 120Ω because 240Ω might not provide enough of a minimum load current.

For 15V use 91R and 1k, 180R and 2k or 200R and 2.2k for R1 and R2 respectively.

The design needs a variable control though, currently it's by a 5 K lin. pot.

That's interesting point H999 about minimum load and choice of current-setting resistor. Inexperienced (or unimaginative!) designers will use only the values given in the Datasheet, rather than to pick values better suited to their application. Not being very imaginative when I suggested use 240 ohm!

 

[marcbarker]

Its not a 317K its a 317T, .<..>When you say overloaded do you mean I drawing more than 1.5amps from the regulator? .

One thing is for sure,... whatever regulator you are using, if you are drawing more current (or make it dissipate more power) than it is designed for, it's not likely to behave as you might expect!

 

[audioguru]

I don't know if SPICE knows that the output voltage of an LM317 will bounce around all over the place if the resistor from its output to the ADJ pin is higher than 120 ohms. The output voltage rises without a load when this resistor value is higher than 120 ohms. Then of course the output voltage sinks down to what it is supposed to be with a load.

So get rid of the 298 ohm resistor and use the proper 120 ohms. The resistor can be 240 ohms for the more expensive LM117.
Also re-calculate the value of the pot to match the 120 ohm resistor.

The output current limiting reduces the output current when there is more than 15V from the input to the output. The minimum output current is only 150mA when the input-output is 40V.

 

[tariq7868]

I recently built a linear power supply and I expected 1.25 - 28v from it @1.5amps. I'm using a Triad F7-28 transformer I get about 34 vac from it. I'm using a LM317T regulator and at the output I've seen up to 43-44 volts DC. I've been able to draw down the voltage to 37volts when drawing .5 amps. I assume once I draw 1.5 amps I'd see around 28v or so. I'm not really happy with this so I'm needing some clues on how to bring this down to 1.25 - 28v range and not see such a big voltage drop based on the load.

i was having the same problem with my ps which i also regulated through 317.. and i thought it was due to larger internal resistances or lower current rating of the trasnsformer .. and i also do think it is normal ... am i right???

 

[marcbarker]

Last time I did a bench variable supply, I used a 723 + '3055. I've not used a 117 for a long time.

Did someone just say a 117 needs a different resistor to a 317 just now? I thought a 117 and a 317 were the same chip but hermetic ceramic instead of moUlded plastic?

 

[audioguru]

A voltage regulator regulates the output voltage so it barely drops when loaded.
If the voltage from the transformer collapses with a load then of course the output of the regulator powered by it will also drop.

Did anybody ask about your main filter capacitor? Maybe it is too small so it produces a high ripple voltage.

 

[audioguru]

Last time I did a bench variable supply, I used a 723 + '3055. I've not used a 117 for a long time.

Did someone just say a 117 needs a different resistor to a 317 just now? I thought a 117 and a 317 were the same chip but hermetic ceramic instead of moUlded plastic?

The LM117 has a max required no-load current of 5mA which is provided by a 240 ohm resistor from its output to its ADJ pin.
The less expensive LM317 has a max required no-load current of 10mA which is provided by a 120 ohm resistor as above.
"If the value of the resistor is higher then the output voltage will rise without a load" as is said in the datasheet.

 

[marcbarker]

The LM117 has a max required no-load current of 5mA which is provided by a 240 ohm resistor from its output to its ADJ pin.
The less expensive LM317 has a max required no-load current of 10mA which is provided by a 120 ohm resistor as above.
"If the value of the resistor is higher then the output voltage will rise without a load" as is said in the datasheet.

I'm not sure what the value of Current-Programming Resistor had to do with the question I asked?

Well of course '240ohm' & '120ohm' will [try and] program '5 mA' & '10 mA', hardly a surprise!.

Both the 117 and 317 have the same "1.2 V regulation" function and it's Ohm's Law in action, it's not something special about the IC is it? (edit: if > NoLoadCurrent, o/p voltage loses regulation)

How much is the Adjust Pin current of a 117/317 anyway?

 

[audioguru]

The ADJ pin current (100uA max for both the LM117 and the LM317) is almost nothing and doesn't change. The entire chip's idle current must flow through the output which must be drained to ground through the voltage-setting resistors. The idle current is half as much for the more expensive LM117.

 

[audioguru]

Is XTRMI going to make the LM317 circuit with a 120 ohm resistor replacing his 300 ohm resistor so that it regulates properly?

 

[marcbarker]

The ADJ pin current (100uA max for both the LM117 and the LM317) is almost nothing and doesn't change. The entire chip's idle current must flow through the output which must be drained to ground through the voltage-setting resistors. The idle current is half as much for the more expensive LM117.

By "idle current" you mean Adjust Pin Current?

The adjust pin current can be from "50 to 100 uA" between different ICs, which amounts up to a 2% error in voltage regulation if a 5 mA (240ohm) programming resistor is used. In practice (i.e. in a mass-produced circuit with fixed resistors) this'll contribute this much error in initial voltage setting accuracy. Some people think that specifying 0.1 % resistors prevents this occurring, well it doesn't!!

As for adjust pin current variation, both the 117 and the 317's APC can change by 5 uA which amounts to 0.1% variation in output voltage (for a 240 ohm programming resistor). Not a lot, as you say. Though the 117 technically has a greater variation in APC, because of the higher operating temperature range.

But if application is likely to cause the APC to significantly increase over life (such as for spaceflight, where there is ionising radiation degrading hFE), then decreasing the CPR to below 120 ohms helps a lot.

Anyway.. Isn't the 117/317 both come from the same silicon wafer? I'll guess the cherry-picked ones from the middle of the wafer become classed as 117 and the runner ups become 317s. That'll mean that some "317"s are almost "117"s. Also I wouldn't be suprised if all the H/T/M/etc suffix current rating 117/317s all actually used the exact same chip inside, and the max output current being to do with the amount of metal in the package.

So... there still is the OP problem: why's the output voltage collapsing under 'load'? I think the IC's overloaded myself.....

 

[Hero999]

The design needs a variable control though, currently it's by a 5 K lin. pot.

The trouble with pots is they have a fairly wide tolerance and normally only come in E3 values or near E3 values, i.e. 1, 2 (sometimes 2.5) , 5. This doesn't give you much choice so it's better to choose R1 and uses a slightly higher value pot than you need.

That's interesting point H999 about minimum load and choice of current-setting resistor. Inexperienced (or unimaginative!) designers will use only the values given in the Datasheet, rather than to pick values better suited to their application. Not being very imaginative when I suggested use 240 ohm!

Yes, noobies generally only use 240R.

240R doesn't provide enough current for an LM317 (which needs 10mA, not 5mA), notice how the example circuit on National's datasheet is for the LM117?

The circuit on the LM337 datasheet shows a 120R for R1.

 

[audioguru]

The 10mA max idle current of the LM317 is its max operating current.
The more expensive LM117 is selected for a max idle current of 5mA.
If the resistor from the output to the adjust pin for an LM317 is higher than 120 ohms then the 10mA in it causes the output voltage to rise when there is no other load.

A 78xx regulator has all its idle current in its ground pin, not in its output pin so its output voltage does not rise without a load.

 

[xtrmi]

Yeah I will try it, but I just didn't understand the situation here. There is a max of 10ma to flow between out and adj. Can it be lower? Can it be higher? I was under the impression that there needed to be a higher current through the whole divider so I tried with a 10 ohm resistor between out and adj and a 110 ohm between adj and ground. Hopefully I didn't damage it.