Replacing a Wall Wart Power Supply

[accpf]

Assuming I have the same voltage, correct polarity, and sufficient current, can I generally replace any class 2 power supply with any other? This will be used for common household electronics. In particular, when researching power supplies, I see specifications for ripple, noise, setup, rise, and hold up time, and I'm not sure if these things are important to take into consideration when swapping power supplies. I have access to these specs for the new supply, but not the old one so I can't make any direct comparisons. I'd also like to know if it matters that the power supply types may be different (switching vs non-switching).
Thanks for any responses.

 

[berkeman]

Assuming I have the same voltage, correct polarity, and sufficient current, can I generally replace any class 2 power supply with any other? This will be used for common household electronics. In particular, when researching power supplies, I see specifications for ripple, noise, setup, rise, and hold up time, and I'm not sure if these things are important to take into consideration when swapping power supplies. I have access to these specs for the new supply, but not the old one so I can't make any direct comparisons. I'd also like to know if it matters that the power supply types may be different (switching vs non-switching).
Thanks for any responses.

I have no idea why a power supply would have specifications for setup and hold times. Those are only relevant to flip-flop circuits, AFAIK.

The answer to your first question is generally yes, but you do need to keep ripple and noise in mind. A switching wall wart is nice because it is smaller, but its output noise can cause problems for some devices (like radios).

 

[accpf]

Could any of those variables (ripple, noise, setup, rise, hold up time, switching vs nonswitching) cause the electronics to have a shorter life? I can see how noise would affect audio quality, but that aside can I be hurting my electronics by using a different power supply?

 

[berkeman]

Could any of those variables (ripple, noise, setup, rise, hold up time, switching vs nonswitching) cause the electronics to have a shorter life? I can see how noise would affect audio quality, but that aside can I be hurting my electronics by using a different power supply?

No, not in any way that comes to mind. What's the application?

 

[accpf]

This question applies to things like DSL modems, answering machines, etc.

 

[berkeman]

The answering machine should be pretty vanilla. The modem may be more of an issue. I'd use a non-switching wall wart with a modem, just to be safe.

 

[Bob S]

Two styles of wall-warts are grounded and ungrounded (to the ground in the wall socket). I recall seeing one wall-wart that had a 3-pin wall plug, but neither side of the dc output was grounded. For your application, I would choose a 3-pin grounded wall-wart.

Bob S

 

[accpf]

Berkeman, why do you think the modem would have an issue with a switching power supply? I have access to a spec sheet... is there anything in there that I should look for to determine the "cleanliness" of power coming out of the supply, or to determine if the PS will work in such an application? What do you think about other applications such as a cordless phone, network switch, etc?

Bob S, the old power supply that is to be replaced has a 2-wire connection so I'm not sure there would be a use for the ground in this case.

 

[berkeman]

Berkeman, why do you think the modem would have an issue with a switching power supply? I have access to a spec sheet... is there anything in there that I should look for to determine the "cleanliness" of power coming out of the supply, or to determine if the PS will work in such an application? What do you think about other applications such as a cordless phone, network switch, etc?

The receive section of modems generally has fairly high-gain analog circuitry, and the signal-to-noise ratio (SNR) can be degraded if noise from the switching power supply gets into that circuitry. If the modem was designed originally to work with a non-switching wall wart, then it may not have much filtering between the power input and the analog RX circuitry. It certainly may have filtering (and a good PCB layout), but it's hard to know that without seeing the schematic and layout.

Another potential issue for any application of a switching wall wart is what kind of power circuit it is feeding. If the wall wart has a 5Vdc output and feeds a circuit, then the circuit may be using the 5V straight without any post-regulation. If the wall wart puts out more like 9Vdc, then the thing it feeds likey has either a switching or linear voltage regulator. If it has a switching regulator that expects a clean input voltage, then having the switching wall wart followed by a switching DC-DC regulator could pose stability problems. The switching frequencies of the two switchers, the output ripple of the wall wart, and the types of switchers they are, all affect the stability issue.

 

[accpf]

Thanks berkeman, your in-depth answers are helpful and certainly give me a lot to think about.

 

[SystemTheory]

This reference may be a bit overkill. The output of a switch-mode power supply includes inductors and capacitors which must absorb energy at turn-on, if I recall, it takes cycles of AC power to charge up the components and this would influence setup time:

http://www.smps.us/topologies.html

The input side for an industrial switch-mode supply is described here.

http://www.smps.us/power-supply.html

I am not aware if the 370-400 volts would apply in a class 2 wall wart, my guess is, it would not. One critical factor is the turn-on current drawn by the load when it first connects to the supply. Some loads tend to draw a large inrush current. The setup time and inrush current are associated with turn-on transients because components that are cold and off absorb energy at turn on.

An analogy is pushing a swing, it takes a few pushes at high force to get the system oscillating (storing energy) at the resonant frequency. In a switching power supply the high inductor currents during turn-on transients are similar to the force pulses to get the swing set in motion, then a steady state condition with less force can be sustained. This is just to give you some general insight ... not intended to be rigorous but hope it is helpfull.

 

[berkeman]

This reference may be a bit overkill. The output of a switch-mode power supply includes inductors and capacitors which must absorb energy at turn-on, if I recall, it takes cycles of AC power to charge up the components and this would influence setup time:

http://www.smps.us/topologies.html

The input side for an industrial switch-mode supply is described here.

http://www.smps.us/power-supply.html

I am not aware if the 370-400 volts would apply in a class 2 wall wart, my guess is, it would not. One critical factor is the turn-on current drawn by the load when it first connects to the supply. Some loads tend to draw a large inrush current. The setup time and inrush current are associated with turn-on transients because components that are cold and off absorb energy at turn on.

An analogy is pushing a swing, it takes a few pushes at high force to get the system oscillating (storing energy) at the resonant frequency. In a switching power supply the high inductor currents during turn-on transients are similar to the force pulses to get the swing set in motion, then a steady state condition with less force can be sustained. This is just to give you some general insight ... not intended to be rigorous but hope it is helpfull.

Neither link that you posted says anything about setup time (use the ^F find feature). Please do not confuse folks with incorrect use of nomenclature. "Setup time" is a technical term referring to FFs and some other logic situations. Not to any power supply applications.


EDIT -- Most of the rest of your post is helpful, however. Thank you for that.

 

[SystemTheory]

The concept I tried to describe relates to turn-on transient conditions. This pdf is a checklist for switching power supply specification including the term "turn-on delay."

http://www.umecintl.com/pdffiles/lcd/UP1232A-01.pdf

This link has a 138 page pdf by On Semiconductor regarding Switching_Power_Supply, it is a link well down the list:

http://new.freeradio.org/documents/

What I'm saying is that "setup time" may appear on some manufacturer specifications and I think it is meant to convey the turn-on delay time, but one must be sure to check the use of terms for consistency as berkman cautions.

 

[accpf]

Okay, I have a new but very-related question. Can anybody see any potential problems with swapping out AC-to-AC transformers? I am considering using a transformer that is used by a security camera system, on a sprinkler system. In addition to powering the timer electronics, the power is used to open the sprinkler valves. The voltages match up, and the new transformer has a greater current output (830mA) than the old one (650mA).

With DC power supplies, switching can cause noise and interference. Is there any scenario in which the new AC transformer might be of a different type and cause problems? For example the new transformer was made to power electronics instead of mechanical actuators so pulsing might be an issue, or perhaps there are different AC transformer types that should not be swapped.

 

[berkeman]

Okay, I have a new but very-related question. Can anybody see any potential problems with swapping out AC-to-AC transformers? I am considering using a transformer that is used by a security camera system, on a sprinkler system. In addition to powering the timer electronics, the power is used to open the sprinkler valves. The voltages match up, and the new transformer has a greater current output (830mA) than the old one (650mA).

With DC power supplies, switching can cause noise and interference. Is there any scenario in which the new AC transformer might be of a different type and cause problems? For example the new transformer was made to power electronics instead of mechanical actuators so pulsing might be an issue, or perhaps there are different AC transformer types that should not be swapped.

I don't see any electrical problems offhand (maybe others can think of something), but be sure that the safety ratings are okay for the swap. Are both transformers made for double-insulated operation?

 

[accpf]

I have no idea if either transformer is double-insulated. In terms of safety, the new one is UL certified and fused. Both the old and new ones are ungrounded. However, I don't know if the primaries and secondaries are isolated within the transformer.

Here's another question: With an AC-DC power supply, noise and ripple are factors, whereas they are not with AC-AC. Does that mean that AC-AC is generally more universal? Is the E-I type the only ones found in a wall-wart, or are there other types of small AC-AC transformers? (Toroidals aren't used in wall-warts and there isn't an AC-AC equivalent of switching is there?)

 

[Bob S]

Okay, I have a new but very-related question. Can anybody see any potential problems with swapping out AC-to-AC transformers? I am considering using a transformer that is used by a security camera system, on a sprinkler system. In addition to powering the timer electronics, the power is used to open the sprinkler valves. The voltages match up, and the new transformer has a greater current output (830mA) than the old one (650mA)..

It's OK, unless the new transformer is a 60 Hz (not 50/60 Hz), and you plan to use it on 50 Hz. In that case, you will probably exceed the volt-second rating, and drive the transformer core into saturation.

Bob S