Plugging Power Strips Together in Series?

[LightningInAJar]

TL;DR

Plugging 2 together?

I have read one shouldn't plug a power strip into another power strip. Why might that be? I don't mean to exceed the number of outlets of the first strip. I just don't have enough space between outlets to plug everything in. Like a 1/2" between the outlets would be fine.

 

[DaveE]

Lots of fires are started because of ad hoc electrical setups like this. So the powers that be try to protect you by saying "don't do that stuff". A powerstrip sits behind a desk and collects dust in an unused connector for 6 years and then someone plugs in a big load. This isn't a good scenario regardless of the specifications. Especially after someone moves the desk and makes the plug pull 20% out of the receptacle. Also, more connections means more risks. Fires never start in the middle of a properly specified and built wire run; it's always overheating at a receptacle/plug.

However, if you pay attention to ALL of the current ratings, and provide for cleanliness and mecahnical stability, there should, theoretically, be no problem. However, keep in mind that the receptacle the strip is plugged into may only be rated for 15 Amps, yet each of the 6 receptacles on the strip are also rated for 15 Amps. No, you can't plug in six 15 Amp loads into the strip. Your average user has no clue how many amps are flowing through that thing. Lots of tiny loads? I don't see a problem. A hair dryer, an air conditioner, and a hot plate? Big problem.

 

[Rive]

I don't mean to exceed the number of outlets of the first strip.

As long as you keep that, it's temporarily passable. But every connection is a possible failure point, so one definitely should keep the number of connections as low as possible.

Like a 1/2" between the outlets would be fine.

There are many types of power cords, some of them with decent outlet-to-outlet distance. I would rather try a different (maybe a modular) one than make some kludge with a jungle of unnecessary cables.

 

[anorlunda]

I sympathize. The spacing between outlets can be maddening, when combined with the size and shape of some devices. The reason for the ban, as others said, is the risk of drawing too much power in total.

A better solution is to find extensions that have better arrangements of outlets or increased spacing. Here are a few I found with a quick search. Note that some include USB charging ports as well as regular outlets. Those can eliminate the need to plug USB chargers into the outlets.

 

[berkeman]

Summary:: Plugging 2 together?

I have read one shouldn't plug a power strip into another power strip. Why might that be?

For one thing, it violates National Fire Codes in the US:

https://www.nfpa.org/news-and-resea...table-power-taps-power-strips-firecodefridays

 

[anorlunda]

For one thing, it violates National Fire Codes in the US:

And that is important, because violating the code gives your insurance company a legal excuse to not pay after a fire.

 

[LightningInAJar]

How do computers, printers, and lamps compare to hair dryers, hot plates, and air conditioners as far as risk of overload?

Is there a protocol for cleaning the dust from a power strip?

And is it true there is serious effort to create an air conditioner using the concept used in stretching rubber bands and aligning and dealigning molecules?

 

[berkeman]

And is it true there is serious effort to create an air conditioner using the concept used in stretching rubber bands and aligning and dealigning molecules?

What mainstream journal did you see that in? Do you have a link? (No YouTube videos please)

 

[anorlunda]

How do computers, printers, and lamps compare to hair dryers, hot plates, and air conditioners as far as risk of overload?

Each device should have a label telling how many watts it uses. A hair drier up to 1500 watts, a laptop up to 750 watts. Just look at the labels.

A 15 amp circuit at 120 V, can deliver 15*120 or 1800 watts.

 

[Windadct]

On the elastomeric AC issue - I know Dupont looked at this 20+ years ago, but was not very efficient. So technically it works - but is not really practical.

 

[Guineafowl]

By stringing together power strips, you increase the loop impedance. This is the AC resistance of the circuit from distribution transformer to the appliances, and back again.

The breaker protecting the circuit might be rated 16A, but this is a nominal figure. In the event of a short circuit fault, a typical ‘B’ curve breaker requires 5x that figure (to trip reliably within a reasonable time), ie 80A. For that current to flow, the impedance of the circuit must be low enough, which it might not be if several power strips are daisy-chained.

 

[DaveE]

By stringing together power strips, you increase the loop impedance. This is the AC resistance of the circuit from distribution transformer to the appliances, and back again.

Good to know. All of my life I've been doing it wrong. Plugging in all kinds of things everywhere throughout the house. Now I know that they all need to be located really close to the breaker panel since the AC impedance must be minimized. Kind of scary when you consider that adding 4 feet or so of distance is bad. My house is much, much longer than 4 feet.

 

[berkeman]

Plugging in all kinds of things everywhere throughout the house.

The NFPA issue and home insurance argument are probably the more important considerations...

 

[sophiecentaur]

However, if you pay attention to ALL of the current ratings, and provide for cleanliness and mecahnical stability, there should, theoretically, be no problem. However, keep in mind that the receptacle the strip is plugged into may only be rated for 15 Amps, yet each of the 6 receptacles on the strip are also rated for 15 Amps. No, you can't plug in six 15 Amp loads into the strip. Your average user has no clue how many amps are flowing through that thing.

As UK resident. I have a great temptation to say Ya Boo Told You So. The 32A fused Ring Main that's the standard way of supplying appliances gets a lot of stick from non-users. "Installation is too cheap and cheerful.", "The plugs are too lumpy", and many more such comments. Fact is, while it's not impossible to give yourself problems, anything plugged into a UK ring main socket will have a fused plug (Max 13A) and that includes multi-way splitters. Every appliance, plugged into the splitter will also have a fused plug. To much of a load will blow the upstream fuse(s). Low current devices will (should) have 2A, 3A, 5A etc fuses in their individual plugs. That protects you from power loss, higher up the chain and only a partial inconvenience. Also, you only find 240V circuits (no sign of 120V, anywhere - not approved) so the current demand is half that of many (most?) US transportable appliances.

Any properly wired UK home will have several double outlets around each room, so there's seldom any need for 'Christmas Trees' with crazy numbers of outlets (see above). 32A at 240V is over seven and a half kW which would leave the room occupants gasping with the heat so the 32A choice was pretty sensible.

I agree that any plug and socket connection can overheat and cause a fire but those 13A mains plugs are mostly pretty robust with plenty of brass in good contact. Everyone should be aware of the 'dark brown or fishy' smell of overheated plastic and investigate further.

 

[Averagesupernova]

The problem I have with the ring system is a break in a connection anywhere within the ring goes unnoticed while potentially overloading it.

 

[Guineafowl]

Good to know. All of my life I've been doing it wrong. Plugging in all kinds of things everywhere throughout the house. Now I know that they all need to be located really close to the breaker panel since the AC impedance must be minimized. Kind of scary when you consider that adding 4 feet or so of distance is bad. My house is much, much longer than 4 feet.

I sense a hint of sarcasm...

‘Don’t daisy-chain power strips’ is a rule that everyone seems to know, but no one knows quite why, hence the lively discussions. Surely if you have a 16A breaker protecting the circuit, and the power strips are appropriately rated, there should be no problem - if you overload the setup, the breaker will trip.

It’s not so much the length, but the smaller conductor size (0.75mm2 on the one next to me, versus 2.5mm2 for the fixed wiring), and the infrequently used connections, which often make poor contact.

A typical UK ring main, protected by a 32A B breaker, has a max loop impedance of 1.16 ohms (BS 7671). If the external loop is at the 0.8 ohm maximum for TN-S supplies, that doesn’t leave much to play with before exceeding the specs for disconnection times, adding in the fixed wiring and two or three power strips.

If it’s not that, then what is the problem with doing this?

 

[anorlunda]

‘Don’t daisy-chain power strips’ is a rule that everyone seems to know, but no one knows quite why, hence the lively discussions. Surely if you have a 16A breaker protecting the circuit, and the power strips are appropriately rated, there should be no problem - if you overload the setup, the breaker will trip.

That's a rather flippant reply, and very irresponsible as advice. The dangers are real. From my local news this morning:

HIGHGATE, Vt. (WCAX) - Police say an elderly woman is dead after a fire in Highgate.
It happened Sunday at about 9 p.m. on St. Armand Road. State police say when crews arrived, Keith Ploof, 64, told them his mother, Sylvia Ploof, 89, was still inside. They got her out but she died at the hospital from her injuries.

Fire investigators say the blaze started in the living room where an overloaded power strip failed and ignited nearby combustible materials.

From an earlier post:

By stringing together power strips, you increase the loop impedance.

Surely the impedance of two power strips in series is not nearly as large as the impedance of my 200 foot (60 meter) extension cord. That cord is allowed by the code. I don't believe that impedance has any practical meaning in the context of this thread.

 

[sophiecentaur]

The problem I have with the ring system is a break in a connection anywhere within the ring goes unnoticed while potentially overloading it.

Sounds like a possible scenario but do you have any evidence of records of this problem? It wouldn't be convenient to test for a break without physically breaking the ring somewhere. If broken rings were at all common, wouldn't you expect a 'ring break tester' to be installed as a matter of course in each ring circuit? UK regs are quite stiff and I have a fair level of confidence in this well tested wiring technique.

It’s not so much the length, but the smaller conductor size (0.75mm2 on the one next to me, versus 2.5mm2 for the fixed wiring), and the infrequently used connections, which often make poor contact.

Not only the shorter length of thinner cable but the presence of extra sliding and dirty contacts can localise the region of high resistance. local I²R heat can often be detected particularly in cheap mains connectors. The equivalent loss along 10m of fat cable is not relevant - except to your electricity bill.

I always feel bad about these frequent set-to's that we have on PF. But I still take part.

 

[DaveE]

I sense a hint of sarcasm...

‘Don’t daisy-chain power strips’ is a rule that everyone seems to know, but no one knows quite why, hence the lively discussions. Surely if you have a 16A breaker protecting the circuit, and the power strips are appropriately rated, there should be no problem - if you overload the setup, the breaker will trip.

It’s not so much the length, but the smaller conductor size (0.75mm2 on the one next to me, versus 2.5mm2 for the fixed wiring), and the infrequently used connections, which often make poor contact.

A typical UK ring main, protected by a 32A B breaker, has a max loop impedance of 1.16 ohms (BS 7671). If the external loop is at the 0.8 ohm maximum for TN-S supplies, that doesn’t leave much to play with before exceeding the specs for disconnection times, adding in the fixed wiring and two or three power strips.

If it’s not that, then what is the problem with doing this?

Just a hint? Sorry, kind of rude in retrospect.

Yes I agree that the primary issue is with high resistance at the connectors. But, except in unusual cases, I don't think it's about the fault currents. I do agree that's technically correct, as in, it's a possible failure mode. But, as @sophiecentaur said, the real world issue is slowly overheating with the larger loads. It's about temperature at the connectors, more than tripping a circuit breaker when there's a failure. This is partly because overcurrent failures of just the right sort don't happen often, but slowly overheating is a constant risk that can develop over a long time period. It also has a more significant thermal runaway problem with corrosion increasing with temperature creating positive feedback. It's a silent problem that will continue until/if someone notices, somehow.

 

[Averagesupernova]

Sounds like a possible scenario but do you have any evidence of records of this problem? It wouldn't be convenient to test for a break without physically breaking the ring somewhere. If broken rings were at all common, wouldn't you expect a 'ring break tester' to be installed as a matter of course in each ring circuit?

That's a bit of a straw man. I do not have any evidence of this. That's my point. We can have many many failures without ever knowing it because the load might not ever exceed safe limits. The circuit is protected at more than safe limits in the event of a broken ring. Just because I have no evidence of something does not mean it should not be a concern.

 

[LightningInAJar]

Maybe in the future power strips will have digital meters on them to show how close to disaster? Lol. Can a power strip be built to withstand the worst case scenario at any cost?

 

[Tom.G]

Can a power strip be built to withstand the worst case scenario at any cost?

Maybe.

Write me a $2 000 000 contract with a 50% non-refundable cash deposit and I will research the possibility.

 

[Guineafowl]

That's a rather flippant reply, and very irresponsible as advice.

It wasn’t advice, but rather a means of introducing the argument, which was my interpretation of why NOT to daisy chain power strips. Please note the word ‘surely’.

Also, your 60m extension doesn’t have multiple connections to get dirty.

From the slightly harsh comments on here, I think I can pick out a summary:

- Increased number of sockets increases opportunity to overload the circuit (not so much in the UK/equivalent countries as all power strips are fused).

- Multiple plug/socket connections increase risk of poor contact, high resistance and overheating.

- This last point will also contribute to increasing fault loop impedance, along with smaller conductor size, reducing the chance of a reliable breaker trip and increasing touch voltage in the event of a line-earthed casing fault.

I chipped in the last point. For those who marked my comment ‘sceptical’, do you really think it’s wrong, or are you taking it as THE reason rather than A reason? There are often several valid answers, after all.

 

[sophiecentaur]

That's a bit of a straw man. I do not have any evidence of this. That's my point. We can have many many failures without ever knowing it because the load might not ever exceed safe limits. The circuit is protected at more than safe limits in the event of a broken ring. Just because I have no evidence of something does not mean it should not be a concern.

You are right in many ways because failure of continuity won’t stop supply to sockets. I think that a standard ring test after installation includes continuity testing between the two ends of the ring.

“Straw Man” perhaps - I can’t really argue but it’s reasonable to say that tests must have thrown up few enough fails both to raise concern. (Including testing after adding ring sections.)

we’re into a touch of culture wars here so this won’t be resolved now, I fear. I guess the insurance companies are the final arbiters here. If they had been stopping paying out because incomplete rings were found then that would find its way into the regs, I’m sure.

 

[hutchphd]

Just to beat this dead horse a little longer:

This last point will also contribute to increasing fault loop impedance, along with smaller conductor size, reducing the chance of a reliable breaker trip and increasing touch voltage in the event of a line-earthed casing fault.

My reading of this is that you are implying the Reactance (Inductance presumably) would be sufficient to be a problem here . Hence my skepticism.
Just replace "impedance" with the more correct "contact resistance" and wholehearted agreement ensues.

 

[sophiecentaur]

But, as @sophiecentaur said, the real world issue is slowly overheating with the larger loads. It's about temperature at the connectors, more than tripping a circuit breaker when there's a failure

It's worth putting some numbers into a possible scenario. If you had an HR 'bad' connection of 2Ω in series with a 1kW heater @240V (or devices to the total of 1kW) you could expect it to dissipate about 35W. That's an uncomfortable amount of heat inside a cheap, imported power strip with flimsy plastic parts. Even with 0.5Ω, you could still expect around 8W. Makes you wonder whether they should be putting thermal fuses in power strips. It's done in all cable extension reels these days.

 

[hutchphd]

My (uneducated) concern, when there are contacts, is the development of growing arcs in the box. These may be far from any thermal fuse.
Is ohmic heating the biggest safety problem or is it the initiation of a growing contact arc?? I lack sufficient experience here.

 

[sophiecentaur]

My (uneducated) concern, when there are contacts, is the development of growing arcs in the box. These may be far from any thermal fuse.
Is ohmic heating the biggest safety problem or is it the initiation of a growing contact arc?? I lack sufficient experience here.

Vicious circle imo.

 

[anorlunda]

Where a safety standard exists (see #5) and the standard says “Do not do X,” should we continue discussion about the safety of X?

Suppose someone says “I think X is safe/unsafe because of reasons A and B.” The committee that voted to adopt the standards may have considered reasons C, D, E, … Their actual reasons are forever unknowable. Safety standards never explain themselves. That has nothing to do with the rationality of A and B.

So, is it bad to publicly say “X is safe/unsafe because of reasons A and B”? I think it is harmful. Contemporary readers and future readers who might come to this thread 10 years from not, may see that and think, “I disagree with A and B.” or “A and B do not apply to my case.” and thus come to the opposite safe/unsafe conclusion. That's faulty because it ignores possible reasons C, D, E, …

We can never know the standard committees’ entire reasons, so we should just accept the standard and comply. We should also advise others to comply, and refrain from trying to reason out our own conclusions by logic because that encourages the others to also make their own conclusions ignoring the standard.

 

[berkeman]

I agree, so this is a good place to tie off the thread. Thanks everybody for an interesting discussion. (and remember to never series-connect power strips)