1590A/LM7809 regulated power supply

[antonis]

YOU'VE JUST PROCEEDED TO YOUR MYSTERIOUSLY BURNED REGULATOR COLLECTION ENRICHMENT.. !! 

@Plexi: With nothing burned - nor even hot -, no reverse polarity, no short-circuit, no continious overcurrent e.t.c the only explanation is your mains power supply over 70% low fluctuation.. 

[Plexi]

Yep... 
I'll use sockets where the regulators are 

no reverse polarity, no short-circuit, no continious overcurrent e.t.c the only explanation is your mains power supply over 70% low fluctuation.. 

Sorry my ignorance: what does it means?

What would prevent it?

12v zener at the input, instead the 33R resistor?

9.1v where the 100nf cap is (or parallel to it), after regulator, would help if regulator fails and start to over feed?

That's what I meant with 'safety' 

[antonis]

 

I mean that, according to your description, everything seem to be OK..

But your regulator failed so, for sure, something caused it..

Regulators are prone to damage either from overvoltage or reverse polarity..
(assuming they have internal current limitation in the form of thermal shut-down..)

Some days ago, I destroyed a 7815 regulator WITH reverse polarity protection (diode from OUT to IN) by just momentarily feeding it with reverse supply..
(with no visible results but with a very audible "tsaf"..)

[Plexi]

I mean that, according to your description, everything seem to be OK..

But your regulator failed so, for sure, something caused it..

Regulators are prone to damage either from overvoltage or reverse polarity..
(assuming they have internal current limitation in the form of thermal shut-down..)

Some days ago, I destroyed a 7815 regulator WITH reverse polarity protection (diode from OUT to IN) by just momentarily feeding it with reverse supply..
(with no visible results but with a very audible "tsaf"..)

Again, you're right!
Now I remember.
I'm 90% sure the short did it when I was dealing with a non-functional MKII Tone Bender (PNP, + ground... a lot of wires from transistors bias).

Is there any protection type to avoid that shortcuts/fails?

It's my main power supply in my workshop to test all the specimens on progress (you know...guts all around, test caps and mods, etc) 

[antonis]

Is there any protection type to avoid that shortcuts/fails?
It's my main power supply in my workshop to test all the specimens on progress

There is, of course, but you'll have to deal with current limiting circuits..

A simple power diode on PS out should be acceptable in most cases, also.. 

[Plexi]

You're right.
The design didn't have "power diode" (would be 1N4007 with anode to ground and cathode to +?)

[antonis]

No, because in such a case you would short-circuit your power supply..

To avoid reverse voltage applying to some circuit INPUT you have to place a diode just after possitive IN with anode to IN..
(something like uni-directional jumper..)

To obtain the same for circuit OUTPUT (i.e. on a battery charger..) you have to place that diode just before possitive OUT with cathode to IN..

[Plexi]

Do you have an schematic?

[ElectricDruid]

I can't see any reason for R1 & R2, other than for "adding" their 5mA current draw to regulators quiescent current (?) (and maybe for C10 & C5 bleeding resistors when power off) so I don't take a stand.. 
(maybe a more experienced person in PS design could enlighten us..)

I don't claim to be that "more experienced person", but some regulators have a minimum current draw requirement to meet the claimed specification. E.g. under very light loads, you might well see out-of-spec voltage or ripple. You sometimes see resistors in the regulator circuit to guarantee this minimum load. Again, check the datasheet or nuke the site from orbit. It's the only way to be sure.

HTH,
Tom

[Plexi]

Every info are welcome 
I spent a lot of time searching, and found a lot of power supplys.

As I said: my principal worry is not to burn any more pedal 

I left the board with 3 slots free, soon I'll add one stage for 18v, another for 12 just in case.
I'm waiting my 4 years old haircut machine die, and feed this little boy with 24v and 2A power supply.

[antonis]

As I said: my principal worry is not to burn any more pedal 

As we said: Use power diodes in series with correct polarity..
(if you REALLY mind of their forward voltage drop, use LM317/337 instead of 78/79 XX series..)

[Plexi]

As I said: my principal worry is not to burn any more pedal 

As we said: Use power diodes in series with correct polarity..
(if you REALLY mind of their forward voltage drop, use LM317/337 instead of 78/79 XX series..)

Do you have an schematic?

[Plexi]

Bump with a dumb-question!

If I have 4 stages, and I'm only running one: the other unused ones will affect the overall functioning?

[Plexi]

Here's an update, my dear folks:




Questions:
- Is normal that when it's unplugged, the voltage out 'oscillates'?
Between 8.84v to 8.87v.
- If I have 4 stages, and I'm only running one: the other unused ones will affect the overall functioning?

[antonis]

0.3% shouldn't be considered as "oscillation" - especially under floating output (no load)..
(it should be a good idea to firstly refer to technical specifications and then worry about any deviation...) 

No unused stage will affect the overall functioning - as long as their individual outputs are seperated..
(do your home unused mains outlets affect the overall functioning..??) 

[Plexi]

0.3% shouldn't be considered as "oscillation" - especially under floating output (no load)..
(it should be a good idea to firstly refer to technical specifications and then worry about any deviation...) 

No unused stage will affect the overall functioning - as long as their individual outputs are seperated..
(do your home unused mains outlets affect the overall functioning..??) 

You're right.
Thanks, as always.

With 'stage', I mean another caps/resistors/regulator: are we talking about the same?

[antonis]

are we talking about the same?

We are, aren't we..??

[Plexi]

are we talking about the same?

We are, aren't we..??

...don't get me wrong.
>No unused stage will affect the overall functioning - as long as their individual outputs are seperated..
(do your home unused mains outlets affect the overall functioning..??)
I get confused about the meaning of "stage" to you: 'multiple outs from one regulator' or 'multiple regulators'.

I don't know about that, but maybe if one regulator stage is unused: isn't the caps 'sucking' the overal + feed?

[antonis]

<...don't get me wrong.>

If I got you wrong I would gave up many pages ago... 

To make things clear:

Your layout indicates many individual regulators outputs, isn't it..??

You call them "stages" (and I also agree) and your query consists in possible reacting between them, isn't it..??

All those stages are fed from a common unregulated power supply, aren't they..??

That common supply "sees" many stages inputs INDIVIDUALLY, isn't it..??

As long as the total current need (with every stage loaded) doesn't overcome unregulated supply's capability it shouldn't have any problem, should it..??

In case of some "open" outpus, total current should be less than maximum one, shouldn't it..??

Given that, all things should work more than fine, shouldn't they..??

P.S.
Caps are great suckers only when they're uncharged - the major cause of blown fuses..

[anotherjim]

An off load series regulator like 78xx isn't usually a problem. If it worries you, at least an "output good" LED can be fitted on the output so there is always a little current drawn and adds a pretty light!


From stuff I've read, it took a few goes to get the design of those regulators to a stage where they were mostly situation/idiot proof. That includes being able to tolerate the local capacitors.

They are a servo feedback controller. It compares a reference voltage to the output voltage and changes the conduction of a series output transistor to make the output voltage match the reference. This sounds like a flawless approach, but what if there is a time lag between it commanding a change in output voltage and it getting there? What if, changing to meet increased or reduced current demand causes a change in the input voltage which means another adjustment - and with some time delay in that?
The effect of time delayed reaction is to make the regulator "hunt" around without ever giving a completely stable DC output voltage. At best slight ripple is present, at worst it can be full on oscillation. Frequency of ripple can be AC supply or cyclic load related or, if unstable,  according to values of the reactive elements in the circuit such as inductors and capacitors.

It's the same with op-amps. Many are prone to oscillating if there is too much capacitance on the output. Cure is some resistance between output and that capacitance, but it's not an ideal thing to do with a power supply as we don't want the voltage to drop a resistor would surely cause.