LM317 can tolerate 40Vmax at its input.

Actually 37 V. It's in the details. 40 refers to the max differential voltage. There's 3 Volt drop across the regulator, so then the input to the regulator has to be <=37 V.

I remember that number from the National Semiconductor Databook. When I read the 40, I wondered where it came from.

 

Actually 37 V.

No please read post 119 (yeah ikr over a hundred posts) max input is in the hundreds of volts.

so then the input to the regulator has to be <=37 V.

did you mean greater than or equal to?

 

Hi,

Yeah i rem reading 40v differential i think, so that means we can have 100v at the input as long as we dont let the output go below 60v. But with capacitors we'd have to be careful i think, so i would think it would be better to use a max of say 35v for this project as that will easily allow a 32v output. We have to take into account that the real DC input is not regulated either, which means if the AC line voltage goes up (a condition known as "high line") the DC goes up by as much as 15 percent, which brings 35v up to about 40v.

 

Bingo! Hence the use of the suggested ZNR's.

But you can also use the LM317T-HV http://www.ti.com/lit/ds/symlink/lm317hv.pdf
for 60 V differential.

Then there's the power dissipation which hasn't come up yet.
30 V drop at 1A for say 7 V out is like 30 Watts for the package to dissipate.

 

No please read post 119 (yeah ikr over a hundred posts) max input is in the hundreds of volts.

That lm317 sounds so out of this world.

You probably didn't understand that app note correctly.

 

That lm317 sounds so out of this world.

You probably didn't understand that app note correctly.

LOL yeah you keep telling yourself that. lol figures are quite clear and the fact I've build one of course. Meh whatever I tried, your chronic denial is beyond my help I'm afraid.

 

Bingo! Hence the use of the suggested ZNR's.

But you can also use the LM317T-HV http://www.ti.com/lit/ds/symlink/lm317hv.pdf
for 60 V differential.

Then there's the power dissipation which hasn't come up yet.
30 V drop at 1A for say 7 V out is like 30 Watts for the package to dissipate.

That high voltage circuit would work without any of those additional components, read the whole document before you pick on part to jump on, it was mainly referring to measures intended to avoid shorts. As long as the minimum component count is rated for the voltage the LM317 will be fine as it is FLOATING. Talking about power dissipation at this point is pointless because the TS was been very vague about his intended range of voltage. So, FAIL, but nice try though.

Why don't you go make one to prove your point, I have proven mine. Spamming NO NO NO doesn't make you right, sorry to burst that bubble.

 

That high voltage circuit would work without any of those additional components, read the whole document before you pick on part to jump on, it was mainly referring to measures intended to avoid shorts. As long as the minimum component count is rated for the voltage the LM317 will be fine as it is FLOATING. Talking about power dissipation at this point is pointless because the TS was been very vague about his intended range of voltage. So, FAIL, but nice try though.

Why don't you go make one to prove your point, I have proven mine. Spamming NO NO NO doesn't make you right, sorry to burst that bubble.

I have no idea how this post happened.

 

Hi

I can confirm that the lm317 will tolerate voltages far in excess of 40V applied to the input terminal provided the differential voltage to other two terminals remains below the magic number.

Anyone who doubts that the voltages which matter are the voltages at the device terminals and not the maximum voltage anywhere in the circuit might like to think about the base-emitter of any transistor.

 

I have no idea how this post happened.

Frustration with a side of irritation was how this happened, so I'm tagging out.

 

the fact I've build one of course.

If you did have a lm317 that takes voltage differentials into the hundreds of volts by itself, your lm317 is made by God.

 

If you did have a lm317 that takes voltage differentials into the hundreds of volts by itself, your lm317 is made by God.

I never mentioned a differential over 40V for the LM317 and 60V for the LM317HV. I was talking about INPUT voltage, LEARN TO READ.

 

Here is a discussion about the use of lm317 and the limit on its input voltage (vs. ground as the common). And you come in and started to talk about how your super-duper LM317 can take hundreds of input volts, without mentioning that it is due to its pre-regulator.

Maybe next you can be more in context next time.

 

OK, but the input is a transformer with a bridge rectifier in this case, so "differential" means single-ended in this special case.

 

Hi again

So how does the LM317 know what potential exists across the grounded resistor? Why does the voltage across the resistor affect the LM317 in any way?. It only feels the voltage at its terminals. I am not surprised that Marcus2012 is getting a bit annoyed - he is quite correct in what he says; he also claims to have built a circuit that relies on the operational mode he describes. And you basically accuse him of lying.

I have a dual adjustable power supply not 10 metres from me as I type. In each half of the system is an LM317 providing the reference voltage. The things survive because although the input voltage is up to 50V (approx it is the rectified output of a transformer so it tends to sag a bit, but maybe it exceeds 50V on a good day), the output is 20V, the reference terminal is 18V75 and so no voltage exists across the device terminals in excess of 30V. Initially I was going to use a 7815 (do 7820s exist?) but no way would one of those survive.

Again I say consider the base-emitter of a transistor. Is it the voltage at the base which counts or the 5, 50, or 500V at the other end of the base resistor? I assume that transistors cannot operate on more than 0V7 in your world.

 

he is quite correct in what he says; he also claims to have built a circuit that relies on the operational mode he describes.

One can claim whatever one wishes, but end of the day, reality is reality.

Take your transistor example. If I take a transistor, unconnected to anything, and then connect a 100KV source to one of its pins, do I have a transistor that can take 100KV input?

To put it another way, if I have a 5v max transistor, and I use a pre-regulator to reduce input voltage to that transistor from 100KV to 5v so my transistor can survive, do you think it is accurate to say that my 5v max transistor can take 100KV?

If he indeed has a LM317 that can take 100V+ on its own, well, he has a LM317 made by God.

 

Here is a discussion about the use of lm317 and the limit on its input voltage (vs. ground as the common). And you come in and started to talk about how your super-duper LM317 can take hundreds of input volts, without mentioning that it is due to its pre-regulator.

Maybe next you can be more in context next time.

For example:
the basic circuit with R1=220ohm, R2=15kohm, input voltage=100V, output voltage=86.5V
voltage at adj pin= 85.25V, no limits exeeded, no pre-regulator used.

 

Maybe next you can be more in context next time.

Pot, kettle, black

Believe what you want

 

no limits exeeded, no pre-regulator used.

When you power up the device, what happens to the output voltage?

 

Believe what you want

Claim what you want, the reality will always be reality.