Using a switchmode power supply in parallel

[David]

I have the Rocktron take on the 1-Spot switchmode power supply that I have been using for at least four years with no difficulty -- except for the daisy chain power harnesses I have to use with it.  I suspect the daisy chain is set up in series because I tried powering some LEDs that marked amp footswitch settings with it.  The LEDs were in parallel and the power supply most emphatically did not like that.

Anyway, my ramp-style pedalboard is not getting it done.  The pedals are spread too far apart and I want to get rid of the daisy chain.  Yeah, this might be an excuse, but...  What I'm interested in doing is putting a box on my pedalboard that takes the switchmode power supply's 9VDC 1.7A output and distributes it in parallel.  Can this be done? Are there any things to watch out for?

Finally, would this arrangement need noise filtering?  This is something I just could not get straight.  I was considering putting in the 100 ohm resistor and the large capacitor that bridged the power input rails.  I was trying to figure out what wattage of resistor I would need.  I looked up the equations and put in my numbers.  I got three different answers -- and none of them looked viable.  I somehow doubt there are 100 ohm resistors that can handle 81 watts!

[JKowalski]

I have the Rocktron take on the 1-Spot switchmode power supply that I have been using for at least four years with no difficulty -- except for the daisy chain power harnesses I have to use with it.  I suspect the daisy chain is set up in series because I tried powering some LEDs that marked amp footswitch settings with it.  The LEDs were in parallel and the power supply most emphatically did not like that.

Anyway, my ramp-style pedalboard is not getting it done.  The pedals are spread too far apart and I want to get rid of the daisy chain.  Yeah, this might be an excuse, but...  What I'm interested in doing is putting a box on my pedalboard that takes the switchmode power supply's 9VDC 1.7A output and distributes it in parallel.  Can this be done? Are there any things to watch out for?

Finally, would this arrangement need noise filtering?  This is something I just could not get straight.  I was considering putting in the 100 ohm resistor and the large capacitor that bridged the power input rails.  I was trying to figure out what wattage of resistor I would need.  I looked up the equations and put in my numbers.  I got three different answers -- and none of them looked viable.  I somehow doubt there are 100 ohm resistors that can handle 81 watts!

Daisy chain wires are set up in "parallel", they can't really be in "series" because that doesn't make sense here electrically. The wires for the power come out of the supply, are hooked up to the first jack's + and -, then are continued from that jack to the next jack in the same way. The end result is that all of the jacks are powered from the same + and -, therefore in parallel.

I'm not sure what the situation is with the LED powering, I didn't really understand what you described.

Switch mode supplies are very high frequency so the power filtering is a little different. You aren't trying to eliminate 60Hz "hum" as much as 100-1000kHz "whine". If the supply is well built the noise level should not be a problem. The best thing to do is test it and see if the noise is audible - and what type of noise it is. If it is 60Hz then you are likely a victim of ground loops, and if it is a whine then consider getting a new power supply, adding small decoupling caps at each pedal's DC jack, or adding external filtering for the PSU.

The 100 ohm resistor + filter cap you mentioned would only be useful for very small current draw effects and is not a good idea to place on the output of a PSU that will be expected to draw considerable current. Remember ohms law, V=IR. You are creating a voltage divider with the 100 ohm resistor and the circuit being powered that results in any small to medium current causing a massive voltage sag.



By the way:

http://www.avtron.com/grid_resistors.htm

Highest one is 10,700 Watts power handling, trust me they get much larger.

I have a couple fin cooled resistors (8 ohm and 470 ohm) that handle 100 watts no problem, useful for dummy resistance loads for speakers (the 8 ohm ones that is)

[PRR]

For Chris and lurkers (not for David):

> Highest one is 10,700 Watts power handling, trust me they get much larger.

If you have an electric stove, you have two 5,000W resistors right there.

People with electric heat have bigger resistors, possibly 20,000-50,000W in a central furnace or distributed around the baseboards.

Electric hot-water is another many-KW resistor.

They may not be the "best" resistors. Their resistance will rise a lot from cold to red-hot. Depending how precise you need to be, you may have to de-rate them (or oil-cool them). They are inductive, and maybe enough to matter within the audio band. They won't stand 24/7 full-power for years and years (on another forum, the dude's stove-switch stuck, the element overheated, burned a hole in itself).

In the right place and time, you can get high-watt resistors cheap. Acquire a mile of barb-wire or other iron fencing. (Just be careful what gets loose.) Bunch it up so the ends are convenient and the loops,snarls  don't touch. Somewhat awkward but lots of dissipation. And steel is the cheapest metal commonly sold as wire.

Fortunately for David and all of us, we never need 10,700 W resistors on our pedalboards.

[David]

Daisy chain wires are set up in "parallel", they can't really be in "series" because that doesn't make sense here electrically. The wires for the power come out of the supply, are hooked up to the first jack's + and -, then are continued from that jack to the next jack in the same way. The end result is that all of the jacks are powered from the same + and -, therefore in parallel.

I'm not sure what the situation is with the LED powering, I didn't really understand what you described.

I have three 3PDT switches set up in a separate box.  Two of these replace the SPST non-momentary footswitches that must have come with the amp.  The other opens a connection between the reverb tank's output and ground.  I have them wired so that a LED indicator comes on when the effect comes on.  Power for the LEDs is supplied to them in parallel from a Boss-type female connector.  The power supply does not like powering my pedals and powering these LEDs at the same time.  It was necessary to use a secondary power supply to the LEDs.  While not convenient, this is not impossible.  However, if I do this again, I want to correct this irritation.

And, just to set the record straight, the daisy chain has two wires on it, like old-fashioned Christmas lights, which were series -- not three wires, indicating that each jack is wired in parallel.

Remember ohms law, V=IR. You are creating a voltage divider with the 100 ohm resistor and the circuit being powered that results in any small to medium current causing a massive voltage sag.

Yes, Mr. Ohm and I were introduced many years ago.  However, and I looked this up, this is not the equation for power consumption.  I recall that one of the terms is squared, but not which one.

[valdiorn]

lets say each pedal is drawing 20mA on average.

V = I*R
9 = 0.02 * R  => R = 450 ohm

this is the equivalent resistance that the pedal introduces to the power supply
lets say you have 10 pedals, that's like having 10 450 ohm resistors in parallel, which would be 1/10 the resistance, or 45 ohm

now if you were talking about adding a 100 ohm resistor at the front of the power supply in series with the load (all the pedals), you have a problem:

Ignore the capacitor at DC (as if its fully charged) and you have a voltage divider:
45 ohm / (45+100 ohm) * 9 volt = 2.79 volts.... that's not enough for your pedals, now is it?

if you were thinking about putting it in parallel (across the power supply connectors) then I don't understand why you'd want to do that, it would serve no purpose but to draw more current, and would actually make regulation harder as a result.

also, how the F... did you figure you'd need 81 watts??


what I'd do (apart from building my own power supply, which I did):
use no resistor and a big big cap. the output impedance of the power supply will be enough to provide a low pass filter (along with the cap. I suggest 4700uF). The pedals are dissipating about 2 watts total (yeah, pedals don't draw much if they're designed good, usually the LED eats half the current ), 200mA current. it should be good. Add 5 ohm and another cap on top of that for more regulation (but more voltage drop).

[R.G.]

V = I * R and I = V/R

P = I * V, so

P = I * (I*R) = I²R  and
P =  V/R *V = V*V*R = V²/R

[David]

V = I * R and I = V/R

P = I * V, so

P = I * (I*R) = I²R  and
P =  V/R *V = V*V*R = V²/R

P = 1.7A * 9V = 15.3 watts if my math was right.  OK, not 81 watts.  I was wrong.  However, I doubt I could walk into the local Rat Shack and pick up a 100R 16 watt resistor -- even when they did have components.
Guess maybe I need to dig out that old dusty breadboard.

[R.G.]

Let me add a bit to this.

Details matter. Your problem *may* be with the power supply itself, not the peripheral stuff. I have personally run over 200 pedals from *one* 9Vdc switching power supply, all in parallel through daisy chains. All switchmode 9V power supplies are not alike. Figuring out how to find a 16W 100 ohm resistor may be like putting a bandaid on a deep stab wound. It'll only cover the hole, not fix the deeper issues.

[David]

Let me add a bit to this.

Details matter. Your problem *may* be with the power supply itself, not the peripheral stuff. I have personally run over 200 pedals from *one* 9Vdc switching power supply, all in parallel through daisy chains. All switchmode 9V power supplies are not alike. Figuring out how to find a 16W 100 ohm resistor may be like putting a bandaid on a deep stab wound. It'll only cover the hole, not fix the deeper issues.

Yes, they do.  And, I may have obscured some because I am trying to kill multiple birds with one stone.  My thought was that if I am going to go to the trouble of building a parallel power supply bus fed by some kind of switchmode power supply, I would like it to be as quiet as possible.  Hence the discussion about the resistor and capacitor.  In truth, however, the power supply as it is currently being used is not causing noise.  It looks like I was inventing a problem where none currently exists.

The only extant issue I can see here is that I was supplying power to all LEDs in the amp switch box in parallel.  I should probably treat each LED as if it were an effect.

It looks like I unintentionally threw mud in the river.  My apologies.

[R.G.]

The only extant issue I can see here is that I was supplying power to all LEDs in the amp switch box in parallel.  I should probably treat each LED as if it were an effect.

Ah. OK, I understand now.

You really have to use a resistor per LED from the power supply buss.

[David]

There are resistors.  One for each LED.  Positive 9V goes through the resistor, into the switch and then to the LED.

[R.G.]

Sorry. I must be extra thick today.

Let me recap, you tell me if I have the picture in mind.
- You have this Roctron 9V power supply, which is switching style.
- You power effect pedals with it.
- You also want to power LEDs on your amp's remote footswitch with the same 9V adapter.
- You have wired this up and

The power supply does not like powering my pedals and powering these LEDs at the same time.

Can you refine what "does not like" means more accurately? Loud hum? hiss? crackling? No sound or power at all?

I would be very hesitant to do what you're doing unless
(a) I was **sure** that the wires to the LED sections of the switch were not connected to the wires going to the amp in any way
-or-
(b) I knew the inside wiring of the amp very thoroughly and could determine that the amp connections would not cause AC power problems with the power adapter.

There is no reason for your power adapter to have trouble running three switched LEDs. There *may* be a reason it does not like being connected to some of the wiring from your amp.

[David]

I'm the thick one, R.G.  OK,  the amp is set up so that the chorus and distortion functions are controlled by SPST non-momentary footswitches.  I implemented the footswitch functionality with 3PDT switches and hung LEDs so they would come on when the switch was in whatever position turned the effect on (I figured this out by experimentation).  Power for the LEDs has nothing to do with the amp.  It's quite possible I did something stupid when I built the switch unit and wired power to the LEDs.  The situation is that everything is peachy if the switch box's power input is not connected to the Rocktron.  If I connect it to the Rocktron along with everything else,  all power to the pedalboard fails.  I have only allowed this to occur twice (to make sure what I saw really happened) when I first installed the switch unit on my pedalboard.  When I confirmed it, I disconnected the LED power input from the Rocktron and supplied power to it from another wall wart.  It was no longer "broke", so I didn't fix it.  Now I would like to go back and do it right.

I apologize for any obfuscation.  And also, the music store didn't have a One Spot.  I would have preferred that, but I had limited shopping time.

[R.G.]

... I implemented the footswitch functionality with 3PDT switches and hung LEDs so they would come on when the switch was in whatever position turned the effect on (I figured this out by experimentation).  Power for the LEDs has nothing to do with the amp. ... The situation is that everything is peachy if the switch box's power input is not connected to the Rocktron.  If I connect it to the Rocktron along with everything else,  all power to the pedalboard fails.  I have only allowed this to occur twice (to make sure what I saw really happened) when I first installed the switch unit on my pedalboard.  When I confirmed it, I disconnected the LED power input from the Rocktron and supplied power to it from another wall wart.  It was no longer "broke", so I didn't fix it.  Now I would like to go back and do it right.

From that description, I think it's "get out your ohmmeter" time. If there was really no connection between the LEDs and the amp wiring, then you could force the power supply to fail by plugging LEDs and their series resistors right onto the daisy chain of the pedalboard, no amp involved. The only way I can see that happening is if your power supply is right on the slippery edge of shutting down anyway and the extra load just pushes it over the edge. That's quite unlikely, if possible.

I think it's more likely that there is an unseen connection between power adapter power and the amplifier, perhaps in the power jack you put into the footswitch setup. If that happens to have one side of the incoming power connected to the amp ground or some other voltage, the power adapter could merely be saving itself by shutting down instead of producing sparks like the bridge of the Enterprise when hit by Klingon weapons. Could also be an errant wire somewhere.

But I think the ohmmeter test for is the incoming power in the amp footswitch unit **really** isolated from the amp wiring is a good place to start.

[David]

Agreed.  It is highly possible -- in fact, likely -- I screwed the grounding up.

[R.G.]

I only mention the possibility of wiring issues and grounding issues because... er... I've... yeah! - done RESEARCH in that area before.