Pwr Supply

[timrand]

I have come across a Switch Mode 9v regulated pwr supply does anybody recommend switch mode ? or are they too noisey ?

[kroushl]

I personally don't have any experience with switching power supplies, but I've heard they are hella noisy.

Brad

[hawkeye2u]

You are better off with a small transformer And reg circuitry
What sorta W are u using
I have a Digitech X series dist pedal that wants 4.8W from a 9v source
Now that is just over 500mA! The 9v batt died within 1 hr!!!

If I had a few effects in daisy chain I would get a transf with a sec voltage of say 12v AC/ 1A , rectify that with a full bridge rectifier, BIG filter capacitor, ie 4700-10 000uF 25v ( I know its overkill but when U need the juice U NEED the juice!!
U then have a VERY well filtered dc (unregulated) at 17 v/ 1A, with a peak cap of close to 10A ( only momentarily!!)
Now feed this into a (for a 9v supply) LM7809 ( or eq) ( if U need diff voltages just subst the LAST 2 numbers, ie LM78XX, where the XX stands for the voltage)(also worht mentioning is the 79XX series, they are NEG regulators, DIFF pinouts to the 78XX series!!!!)) into pin1, pin 2 to ground, pin 3 is output, a small filtercap of 100uF/25v volt is fine here , also concider ,in parallell with the electro 100uF, put a 0.1uF for high freq
filtering as electros at high freq tends to get inductive and not filter well for high freq
Presto !!!!VERY highly reg 9v with only mV ripple and steady as a rock
The LM series only prod 1A output (don't forget the heatsink) but the principle is the same forhigher rated regulators
Have fun :twisted:

[downweverything]

i use some switch mode, but if they are regulated and you add a large filter cap across it, i cant seem to tell the difference.  anyone else try it?

[hawkeye2u]

With switchmodes u have Electrolytics in the primary circuit, now  when they age ,usually the secondary voltage goes HIGHER and destroys some components
That will not happen with a supply as I descrived above VERY STABLE
The switchmodes ARE smaller but at a higher repair cost AND lower reliability IMHO

[Peter Snowberg]

If the supply is a good one, a switch mode supply is better than a linear supply in almost every way. If it is a cheap supply, you take your chances.

To take a step back, a switch mode power supply is one that turns the AC input into DC right away at the input voltage and then uses that power to run a smaller transformer with a very high frequency AC go go to the output voltage. The main advantage of this is that you can shrink the transformer to a tiny fraction of it's original size. Military and avionic AC are 400Hz because you can shrink the size of the iron in the transformers by 80% in comparison to 60Hz.

If your supply runs at a high enough frequency, any "hum" will be ultrasonic. This also makes it easier to filter because smaller caps are used at the higher frequency.

Any good switcher will also contain a "crowbar" circuit that will short the supply out if it tries to put out too much voltage. It's like the supply will actually commit suicide to save the circuit.

Because of the way they "switch", a switcher is always watching the output voltage and adjusting their switching accordingly. When it comes to putting out the right voltage, I have never seen a decent switcher fry downstream components.... ever.

Noise is all up to the switching frequency. I have never had the slightest problem there.

Take care,
-Peter

[hawkeye2u]

Peter Snowberg
YOU are Right in everything you say , re eff aspects ,smaller etc, exept the fact that aging caps in the primary WILL cause the secondry voltage to go high
I've repaired to many TV sets ( maybe I'm too sceptical here re the switchmodes) that have Line output stages destroyed, at a high cost for the customer because of a few caps in a switch mode supply
I'm from the old school mate, I like the old RELIABLE and very Stable pwr supplies
Apart from that everything U say is CORRECT
Cheers  

[hawkeye2u]

To add to my prev comment
I've also repaired lots of VCR's. DVD's, CD' players as well
ALL have switch mode supplies!! with supply problems!!!!!!

Ie wont respond to commands, low display brightness, weird problems that occur intermittantly, goes into child lock, etc etc
ALL Failures related to Capacitors in Sw mode supplies

MANY faults, Maybe not breakdown as severe as in TV sets, ARE related to capacitor failing in the prim OR secondary part of switch mode supplies.

IMHO it is to do with the fact that electrolytics are supposed to deal with low freq, not high freq switching. They will do so ,if you bypass in parallell some small value polyester caps( see prev note) AND also that manufacurers only use 85 deg & the closest match in voltage, ie a Voltage rail of 12v Gets a cap of say 100uF/16V,( Chinese prod lines!!!!!!!!!!!!!!!)
I would immediately repl that with a 105 deg cap of 100uF/ 63v (wich will NEVER fail) the diff in cost would be 0.10c at the most
Why not build circuits that will last
Hope this clarifyes my humble view
Please excuse my spelling mistaces, english is not my native language

[Peter Snowberg]

I have to disagree.

The vast majority of switching supplies use PWM and a feedback loop to actively control the output. The only example I have ever come across where this was not the case was a switching wall-wart that was free running. You will never find a free running switcher powering logic circuits unless it is followed by a linear regulator which defeats most of the inherent advantages of having a switcher.

The switching controller is comparing the output voltage with a local reference voltage that is generated by a regulator (zener or otherwise). That regulated reference voltage does not shift except for a very tiny thermal gradient. This is how they can take such a wide range of input voltages and self-adjust.

The most common switch mode controller is the LM3524 (and it's cousins). Here is the datasheet http://cache.national.com/ds/LM/LM2524D.pdf

Capacitor failure can result in ripple as it will in any style of supply, but I have never seen it affect the peak voltage. I've fixed a lot of gear and all the CD players and VCRs I've ever been inside used linear supplies.

Take care,
-Peter

[Boofhead]

I've seen quite a few switchmodes go up.  The flyback types used in monitors and small often fry the main switch transistor, presumed cause is a line spike - haven't seen too many of these take out the circuit it was connected to though.  I've seen quite a few PC power go up too, a number have cause secondary side caps to vent and fry the motherboard.  Overall I'd say modern switchmodes are fairly reliable but not as reliable as linears.

I don't think any switcher will match the noise of linear.  Modern switchers operating at very high frequencies are pretty good  but the noise can be a problem is some applications.

[hawkeye2u]

Peter
I will NOT, as I said disagree with You!
My statements ARE however rel to dayly occurrances as a service tech
Maybe its because I live in 240V 50Hz land downunder (AUSTRALIA), our suppliers of 240 varies greatly!!
But the above findings are fault finding units for the last , say 6 -7 years
I have no doubt in your experience but I only stated my humble experince as a tech in the field & at a dayly basis of faultfinding
Lets leave it as a diff of oppinion hey! ;)

My pref is the old (I admitt) low eff transformer, rectifier, BIG filtercaps & a reg circuit with filtering, prob cheaper to produce theese days than sw modes
I got that one in good did'nt I LOL
CYA Peter

PS I'd love to have some other electronics chats with U if U dont mind , since U DO know what U're on about

[Peter Snowberg]

Thats OK Hawkeye, please feel free to disagree with me.

I will freely admit my preference for big, heavy iron and mondo sized caps too. (but I do prefer FREDs or any fast schottky rectifier)

>PS I'd love to have some other electronics chats with U if U dont mind , since U DO know what U're on about

Any time you like. I'm just a Bozo (an equipment designing Bozo ), and this place is here for chatting about electronics. We have the benefit of so many people with vastly different experiences. This is what makes this place so great IMNSHO.

Electronics is such a vast field. When it comes to RF, I might as well be a plumber. I always enjoy learning new stuff.

It's good to have you aboard.

Take care,
-Peter

[Paul Perry (Frostwave)]

I had a lightning strike take out a switchmode wart here (in Australa). I notice a lot of switchmode suplies are "80v to 260v input" which sure is convenient... but, I think they are going to be likelier to be spiked out here in Oz (230v nominal) than USA (115 nominal).
I have three monitors with failed switchmode supplies as well.

[hawkeye2u]

Need I say More?, more negs than positives!!!
The GREAT adv of a sw mode is just that that
they will just "fire up " at 80- 90 v & with the latest designs, just select what incoming voltageis there and adj accordingly
AS FOR RELIABILITY??

[Paul Perry (Frostwave)]

The bottom line for me is, that it is much easier to make a 'traditional' supply than a switchmode, and so the cheapest traditional PSU is likelier more reliable than the cheapest switcher.
A WELL DESIGNED switchmode is IMHO superior, but a crappy switchmode is a spawn of Satan.
In any case, as switchers use less materials (total weight) they will drive out the others pretty soon.

[hawkeye2u]

As a FINAL nail in the coffin of the sw mode supplies!
I was involved in a major update of the Language laboratories of The Univerisity of Stockholm in the late 80's, i managed to escape to Australia after this  
We had 200 student desks, devided into 8 class rooms,with Revox B77 Tape decks , reel to reel (spec modified) we also could switch a bank of 30 tape decks into ANY classroom, for copying purposes etc, so all in all 230 Revox B77's, every student desk had a uPC, every B77 had a uPc, teachers desk had a uPc, so all in all there was LOTS of 5V needed

The 5V regulators for driving all this was highly "stable" , proffessional 5 v switch mode supplies, we also, I might add had 9v & 12 v reg (all sw mode) which didn't give us any problem
We had ALL sorts of problems until we monitored the 5v rail
Solution
Old Heavy duty Transformer, hefty recitier, AND LARGE capacitors
( 47 000uF/16v), we monitored the current, doubled the current capacity of the transformer, Bobs Your Uncle
NO MORE RESETS, NO MORE INTERMITTANT  PROBLEMS, NO LOCKUPS ETC
Need I say more :twisted: