Power supply quick question

PmCimini · Post by **PmCimini** » 10 Jan 2012, 20:04

Whats the deal with the resistors on the power supply input? A bunch of circuits have really small (10 - 100 ohm) resistors between the jack and the circuit's main line.

rocklander · Post by **rocklander** » 10 Jan 2012, 20:05

I believe that's called decoupling..

https://www.freestompboxes.org/viewtopic ... ple#p17704

PmCimini · Post by **PmCimini** » 10 Jan 2012, 20:25

What about the resistance? Some circuits use 10ohm, some others up to 220ohm... Is there a best value for this?

rocklander · Post by **rocklander** » 10 Jan 2012, 20:31

I'm not sure it matters, but I exhausted my knowledge on the subject with my earlier post

Zokk · Post by **Zokk** » 10 Jan 2012, 23:27

I'm not sure to talk about the same resistors, but in many circuits you can find 10ohm resistors on the DC lines or ferrite beads, the resistors will blow in case of reverse polarity while protecting the circuit itself (mainly when used with +/-15V psu), and act like a filtering add on to clean up DC signals.
Check out the implementation of the psu input on this one:
http://yusynth.net/Modular/Commun/MIXER ... sch-1U.gif
look for R1 & R2.

deltafred · Post by **deltafred** » 11 Jan 2012, 00:04

Zokk wrote:... the resistors will blow in case of reverse polarity while protecting the circuit itself ...

No they won't, the semiconductors will probably be damaged long before the resistors blow. The circuit will see the full reverse voltage which is usually fatal.

If you want to protect from reverse polarity then series diodes are best as they block any reverse voltage.

Parallel (reverse) diodes allow a small reverse voltage to reach the circuit, whatever the forward volts drop of the diodes are. They effectively put an almost short circuit on the power supply to drop the voltage. The diode current rating must be greater than the power supply can deliver.

mictester · Post by **mictester** » 11 Jan 2012, 00:16

deltafred wrote:
Zokk wrote:... the resistors will blow in case of reverse polarity while protecting the circuit itself ...
No they won't, the semiconductors will probably be damaged long before the resistors blow. The circuit will see the full reverse voltage which is usually fatal.

If you want to protect from reverse polarity then series diodes are best as they block any reverse voltage.

Parallel (reverse) diodes allow a small reverse voltage to reach the circuit, whatever the forward volts drop of the diodes are. They effectively put an almost short circuit on the power supply to drop the voltage. The diode current rating must be greater than the power supply can deliver.

You can get "fusible" resistors, and they usually fail fast enough to protect the rest of the circuit.

However, this isn't the usual reason for this resistor - it forms (with the electrolytic capacitor to ground) a filter with a very low corner frequency. It's a good idea to keep audio off the supply rail - the electrolytic will look like a short to ground for audio - and it will frequently aid the stability of high gain circuits.

phatt · Post by **phatt** » 11 Jan 2012, 07:36

+1 what *Mictester* said above

Go look at some Valve PSU's for clues. ( I think the term is R C filters).
Phil.

deltafred · Post by **deltafred** » 11 Jan 2012, 10:19

mictester wrote:You can get "fusible" resistors, and they usually fail fast enough to protect the rest of the circuit.

I very much doubt they will protect semiconductors from reverse voltage. I would certainly not rely on them.

They will protect other components and the power supply from excess current in the event of incorrect power supply connection.

Zokk · Post by **Zokk** » 11 Jan 2012, 13:24

I have to admit that those resistors may not blow fast enough to protect some ICs; but I've experienced the case with one module someday and there was only one TL074 blown, the other discrete trannies were intacts... maybe a question of luck here, I don't know.
But a polarity inversion or short cut with the power rails is always a bad thing for some components.

Also adding a diode in series with the power supply rail may affect the good work of some ICs due to the inherent voltage drop.

There is another possibility: adding a voltage regulator on the circuit for each power rail (I'm mainly talking about +/15V or +/-12V, but for a simple +9V only one reg would be needed of course), so there will be additional filtering (some more caps needed), reverse voltage protection (using the LM3XX for example), and efficient isolation of the circuit and psu. This solution implies another voltage drop anyway, however it's possible to feed the circuit with little more voltage to compensate.

deltafred · Post by **deltafred** » 11 Jan 2012, 14:45

phatt wrote: Go look at some Valve PSU's for clues. ( I think the term is R C filters).
Phil.

I know about the properties of first order RC filters thanks so don't need to look them up.

What I was referring to was the mistaken belief that a resistor in the supply line will protect the circuit itself :-

Zokk wrote:the resistors will blow in case of reverse polarity while protecting the circuit itself

True they will protect some of the circuit especially if as Mictester says they are fusible but applying reverse voltage to semiconductor devices is a bit of a lottery, some survive, some don't. Those that do survive can often have their specs degraded.

I would prefer mine to not be subjected to it, YMMV.

PmCimini · Post by **PmCimini** » 13 Jan 2012, 19:52

mictester wrote:
deltafred wrote:
Zokk wrote:... the resistors will blow in case of reverse polarity while protecting the circuit itself ...
No they won't, the semiconductors will probably be damaged long before the resistors blow. The circuit will see the full reverse voltage which is usually fatal.

If you want to protect from reverse polarity then series diodes are best as they block any reverse voltage.

Parallel (reverse) diodes allow a small reverse voltage to reach the circuit, whatever the forward volts drop of the diodes are. They effectively put an almost short circuit on the power supply to drop the voltage. The diode current rating must be greater than the power supply can deliver.

You can get "fusible" resistors, and they usually fail fast enough to protect the rest of the circuit.

However, this isn't the usual reason for this resistor - it forms (with the electrolytic capacitor to ground) a filter with a very low corner frequency. It's a good idea to keep audio off the supply rail - the electrolytic will look like a short to ground for audio - and it will frequently aid the stability of high gain circuits.

Iget that the electrolytic capacitor will aid the stability, but is the resistor really that necessary? I mean, wouldn't the power supply cable itself work as a resistor for the RC filter?

PmCimini · Post by **PmCimini** » 13 Jan 2012, 19:56

Also, could someone point me to a good power supply protection setup that I should put in my circuits? I mean, I don't really like the series diode idea as it lowers the supply from 9V to about 8.4V (depends on the diode's forward voltage)... What's the deal with the Zener diode in parallel? I've seen this sometimes...

PmCimini · Post by **PmCimini** » 13 Jan 2012, 20:47

PmCimini wrote:Also, could someone point me to a good power supply protection setup that I should put in my circuits? I mean, I don't really like the series diode idea as it lowers the supply from 9V to about 8.4V (depends on the diode's forward voltage)... What's the deal with the Zener diode in parallel? I've seen this sometimes...

Couldn't edit the earlier posts....

Anyhow, after a little search, I found this topic (which I was looking for in the first place, but never found: https://www.freestompboxes.org/viewtopic ... 22&t=10056), and it seems that the zener diode thing is completely cleared out.... Bad memory haha

Anyhow, I still want to know a bit more about the current-limiting resistors... Are they there just so they can burn in case of reverse polarity?

kleuck · Post by **kleuck** » 13 Jan 2012, 20:59

PmCimini wrote:Also, could someone point me to a good power supply protection setup that I should put in my circuits? I mean, I don't really like the series diode idea as it lowers the supply from 9V to about 8.4V (depends on the diode's forward voltage)... What's the deal with the Zener diode in parallel? I've seen this sometimes...

With a 9V1 zener, you have a very simple regulation (all voltage above 9V1 is cut) and if a reverse voltage occurs, the zener becomes a regular diode, and limits the reverse voltage for the circuit to 0,6 volts...until the limiting resistor burns

I personally use series forward protection diodes, but Schottky ones, you can easily find some with 0.2 to 0.3 direct drop.

PmCimini wrote:
Anyhow, I still want to know a bit more about the current-limiting resistors... Are they there just so they can burn in case of reverse polarity?

No, they are here to limit the current in the zener (if present) too when it's acting as a regulator, and to make the time constant of the RC filter higher.

phatt · Post by **phatt** » 14 Jan 2012, 08:22

PmCimini wrote:Also, could someone point me to a good power supply protection setup that I should put in my circuits? I mean, I don't really like the series diode idea as it lowers the supply from 9V to about 8.4V (depends on the diode's forward voltage)... What's the deal with the Zener diode in parallel? I've seen this sometimes...

Use 1N4001,, Not 1N4004 = less voltage drop.
Phil.

Zokk · Post by **Zokk** » 16 Jan 2012, 00:23

Are you sure about the different voltage drop when using a 4001 or a 4004? To me the differences are related to the max reverse voltage they can stand before blowing.
I've never seen such info about that in the datasheets... and I use a lot of 4007UF (cause I've a huge stock in my drawer) with the same behaviours as the other refs but with high rating power (they are nice in small PSUs). Maybe I missed something.

amptramp · Post by **amptramp** » 16 Jan 2012, 02:11

The circuitry inside a wall wart may suprize some people. On one that I took apart, there was the transformer, the diodes (two, in push-pull) the electrolytic capacitor and to my surprize, a resistor in series with the capacitor but not in series with the output line. The idea was obvious - the diodes could not tolerate the inrush current to the internal capacitor. If you connect a pedal to this wall wart, a capacitor right on the input would draw a lot of inrush current. But more importantly, the diodes conduct a large amount of current for the brief time when the diode input voltage is above the capacitor voltage. The current times duty cycle is equal to the average current. A large capacitor would require a large spike of current for a short time. And current is divided between the wall wart internal cap and the cap in the pedal if there is no series resistor in the pedal, so the resistor keeps spikes from coming down the power line.

phatt · Post by **phatt** » 16 Jan 2012, 05:48

Zokk wrote:Are you sure about the different voltage drop when using a 4001 or a 4004? To me the differences are related to the max reverse voltage they can stand before blowing.
I've never seen such info about that in the datasheets... and I use a lot of 4007UF (cause I've a huge stock in my drawer) with the same behaviours as the other refs but with high rating power (they are nice in small PSUs). Maybe I missed something.

Can't remember exact readings,, but I clearly recall noting the loss difference between a 1N4004 and 1N4001 when used in series with the battery supply.
you can of course strap the diode across the supply. As to Which way is better? That will need better minds to comment.

Phil.