Anyone heard of a Mac Pro tripping a breaker when powered on?

[velocityg4]

OP

I didn't notice it asked or stated. But do you still have Applecare on the Mac Pro? If so take it to the Apple Store for diagnostic and repair before tearing it apart and running all these tests.

 

[nanofrog]

My house was built in 2005.

I've indicated it before, my suspicion is the PSU, the caps in particular.

But to be sure the wall power isn't an issue (circuit load at a minimum), we've asked for some simple information that's not been provided/clarified. Apple would likely ask the same of you anyway.

They'll still need some diagnostics, whether it's covered under Apple care or not (you'd just have to pay for a new unit if the coverage has expired, and unfortunately, they're not inexpensive- nearly $300USD last I checked).

Dudes, this is a simple black and white problem; he has two Mac Pros, one which trips circuits and one that doesn't. Yet you draw the conclusion that his house wiring's at fault.

No, we're after proper diagnostics. Just because you suspect something, you still need to rule other possibilities out first, as leaping to a premature conclusion that it's the PSU could prove wrong, and end up costing the OP money that's not necessary (assuming it's out of warranty). And the PSU's aren't exactly inexpensive if that's the case.

 

[-aggie-]

I've indicated it before, my suspicion is the PSU, the caps in particular.

But to be sure the wall power isn't an issue (circuit load at a minimum), we've asked for some simple information that's not been provided/clarified. Apple would likely ask the same of you anyway.

They'll still need some diagnostics, whether it's covered under Apple care or not (you'd just have to pay for a new unit if the coverage has expired, and unfortunately, they're not inexpensive- nearly $300USD last I checked).


No, we're after proper diagnostics. Just because you suspect something, you still need to rule other possibilities out first, as leaping to a premature conclusion that it's the PSU could prove wrong, and end up costing the OP money that's not necessary (assuming it's out of warranty). And the PSU's aren't exactly inexpensive if that's the case.

There won't be any leaping to conclusions on the PSU. You may have not read about swapping the PSU's to see if the other Mac does this. That will establish the PSU is the problem and will cost the OP nothing.

 

[nanofrog]

There won't be any leaping to conclusions on the PSU. You may have not read about swapping the PSU's to see if the other Mac does this. That will establish the PSU is the problem and will cost the OP nothing.

Check one of my previous posts. Swapping the PSU's is the easiest way to determine if that's the case, but it also has to be attempted on a circuit that has little to no load (other than the MP PSU) to be sure circuit load isn't figuring into the test (prevents a false positive).

My suspicion is the swap will prove the rest of the system is fine, which means the original PSU is the culprit. Specifically bad caps, as the Voltage Regs are in tact (otherwise it would just be DOA).

But the test is necessary IMO, but the circuit load condition must be adhered to in order to make sure there's no other influence that could render the results inaccurate. False positives can be a PITA, even if the system is still under warranty (time lost as the system spends more time in repair).

 

[Lordedmond]

is it tripping the ELCB or the normal MCB

as your house was built 2005 then you could have a ELCB


it would still point to a PSU fault , but this would be a small earth fault leakage less that 20ma will trip a ELCB

this is usually caused by the delta suppression caps

 

[nanofrog]

is it tripping the ELCB or the normal MCB

as your house was built 2005 then you could have a ELCB


it would still point to a PSU fault , but this would be a small earth fault leakage less that 20ma will trip a ELCB

this is usually caused by the delta suppression caps

I'm under the impression they're UK/Europe centric, and are older than RCD (GFCI here, I presume you call them something else).

Another newer type here, is the Arc Fault types (AFCI), spec'ed out in the 1999 NEC (bedroom use). Not all jurisdictions have adopted them though (just adopted in 2009 where I am), and I'm not sure about the OP's location.

 

[Phantom Gremlin]

I'm under the impression they're UK/Europe centric, and are older than RCD (GFCI here, I presume you call them something else).

Another newer type here, is the Arc Fault types (AFCI), spec'ed out in the 1999 NEC (bedroom use). Not all jurisdictions have adopted them though (just adopted in 2009 where I am), and I'm not sure about the OP's location.

Yes, I think this is the answer. It's very unlikely that high current is tripping the breaker. It's either an arc fault or a ground fault. E.g. leakage to ground of only 30 mA can cause a trip. As discussed previously, a leaky cap (or maybe a MOV if Apple uses those) can cause a trip. Unfortunately a kill-a-watt wouldn't be useful in diagnosing this problem; the much more painful suggestion of swapping power supplies would probably give clearer results.

A quick google found an article that discusses these breakers. Note that combination AFCI/GFCI breakers are available.

 

[nanofrog]

Yes, I think this is the answer. It's very unlikely that high current is tripping the breaker. It's either an arc fault or a ground fault. E.g. leakage to ground of only 30 mA can cause a trip. As discussed previously, a leaky cap (or maybe a MOV if Apple uses those) can cause a trip. Unfortunately a kill-a-watt wouldn't be useful in diagnosing this problem; the much more painful suggestion of swapping power supplies would probably give clearer results.

A quick google found an article that discusses these breakers. Note that combination AFCI/GFCI breakers are available.

High startup currents can trip a breaker, no matter the type, which is why Soft Start circuits (different ways of doing it) are used (charges the caps,... slowly compared to just "ON", which can appear as a dead short).

But in this case, I suspect bad caps, as they'll cause such issues, no matter the breaker type. It appears as a dead short (as you mention, AFCI can include other types, as I'm familiar with those that go in the panel which do). AFCI + MCB at a minimum (though you could call it AFCI + GFCI as well, but I separate GFCI to mean outlet, not panel). This may be just my way of thinking though.

 

[nardeezy]

I had a similar issue with my Mac Pro. However, I figured out the root cause was the AFCI circuit breaker. Is your office a new add on? Newer construction after 2002/2003 requires AFCI circuit breakers in all bedroom type dwellings (including an office/den). I replaced my AFCI circuit breaker with a direct replacement (newer) and it seemed to have solved my problem since Dec. Interestingly, my new AFCI circuit breaker tripped last night and I'm not sure why but I'm thinking it did so because the Mac Pro woke from sleep? Anyways, good luck!

 

[mason.kramer]

There is no way the extra 4-5 amps from "normal" starting of a computer is going to trip ANY breakers.

Of course there is. I live in apartment with two circuits. Each is rated at 15 amps. Startup current accounts for fully 1/3 of the allowed current on one of my breakers.

 

[product26]

the lights dim when i wake or turn on my mac pro. This happens in both a house built early last century that still has old school fuses (not a breaker box) and where i am now which was built in the last 10 years

 

[s.m.t.]

Yes, I think this is the answer. It's very unlikely that high current is tripping the breaker. It's either an arc fault or a ground fault. E.g. leakage to ground of only 30 mA can cause a trip. As discussed previously, a leaky cap (or maybe a MOV if Apple uses those) can cause a trip. Unfortunately a kill-a-watt wouldn't be useful in diagnosing this problem; the much more painful suggestion of swapping power supplies would probably give clearer results.

A quick google found an article that discusses these breakers. Note that combination AFCI/GFCI breakers are available.

This is what I ran into in my office. Code says they have to be AFCI breakers have to be installed in rooms that can be bedrooms. I had to replace one for my office with a standard breaker or my UPS would trip the arch fault...

 

[MitchLewis]

I would advise connecting your MacPro up to a UPS. A UPS has a battery inside that isolates it from the power problems in your home. It tries to protect you from brown outs, etc..

We recently bought this model for our MacPro:
http://www.bhphotovideo.com/c/product/613559-REG/APC_BX1500LCD_Back_UPS_XS_1500VA_LCD.html

I like it because it has voltage/amperage readouts displayed on the front so you can monitor what's going on with your power.

 

[Phantom Gremlin]

you should not see nuisance tripping

First, AFCI type breakers can be "touchy", so what I say following doesn't apply.

I did some more Googling.

Apparently manufacturers follow UL standard 489 as to how circuit breakers should behave. There is something called "bulletin 1489" which I think is from ANSI and is similar but not identical.

Here is an Allen-Bradley page on breakers. See the chart at the bottom.

The standards provide some leeway, it's not possible to cost effectively build a breaker that trips exactly at a certain current.

Looking at the chart, see circle #4. That means the breaker must not trip below 500% of max current for at least 0.1 seconds (6 cycles of AC). That's a long long time. A properly behaving power supply will not draw 15x5=75 A of 120 V for 0.1 seconds. It wont. More typically a breaker won't trip for at least 0.7 seconds at 500% of max current. E.g. A-B breakers won't.

See circle #3. That means the breaker must not trip below 200% of max current for at least 10 seconds. It has some leeway. It can take as much as 120 seconds to trip at 200% of max.

So, as -aggie- said, "There is no way the extra 4-5 amps from "normal" starting of a computer is going to trip ANY breakers." You either have a touchy AFCI breaker or an older breaker that isn't following the spec.

 

[nardeezy]

I would advise connecting your MacPro up to a UPS. A UPS has a battery inside that isolates it from the power problems in your home. It tries to protect you from brown outs, etc..

We recently bought this model for our MacPro:
http://www.bhphotovideo.com/c/product/613559-REG/APC_BX1500LCD_Back_UPS_XS_1500VA_LCD.html

I like it because it has voltage/amperage readouts displayed on the front so you can monitor what's going on with your power.

That UPS is okay. I had one then returned it. The best UPS is a true online UPS. This particular APC model is not a "true" online UPS.

 

[Photomatt70]

Same issue

I have and still do experience the same issue. I have had an electrician come to my home (new renovation in 2005 with new wiring) and replace my 15 amp breaker with a 20 amp breaker. Along with this he did a diagnosis on the wiring and checked for and shorts. nothing found.

With my research, I found two things: One, being the power supplies that are installed in Mac Pros (as well as higher end HP machines) and the other was the AFCI circuit breakers not being able to jive with the power supply on the Mac Pro.

As far as that, other locations are okay, but recently had my machine out on a job location, indoors, and saw a drop in the lights as I powered up my system.

I am curious to know the outcome of this.

 

[mason.kramer]

First, AFCI type breakers can be "touchy", so what I say following doesn't apply.

I did some more Googling.

Apparently manufacturers follow UL standard 489 as to how circuit breakers should behave. There is something called "bulletin 1489" which I think is from ANSI and is similar but not identical.

Here is an Allen-Bradley page on breakers. See the chart at the bottom.

The standards provide some leeway, it's not possible to cost effectively build a breaker that trips exactly at a certain current.

Looking at the chart, see circle #4. That means the breaker must not trip below 500% of max current for at least 0.1 seconds (6 cycles of AC). That's a long long time. A properly behaving power supply will not draw 15x5=75 A of 120 V for 0.1 seconds. It wont. More typically a breaker won't trip for at least 0.7 seconds at 500% of max current. E.g. A-B breakers won't.

See circle #3. That means the breaker must not trip below 200% of max current for at least 10 seconds. It has some leeway. It can take as much as 120 seconds to trip at 200% of max.

So, as -aggie- said, "There is no way the extra 4-5 amps from "normal" starting of a computer is going to trip ANY breakers." You either have a touchy AFCI breaker or an older breaker that isn't following the spec.

Hey man, thanks for this awesome post. I learned something really useful from it. I guess my breakers are out of spec, because they did occasionally trip when, for instance, experiencing the startup current of my Mac Pro while my AC was running, or vice versa.

 

[Phantom Gremlin]

Hey man, thanks for this awesome post. I learned something really useful from it. I guess my breakers are out of spec, because they did occasionally trip when, for instance, experiencing the startup current of my Mac Pro while my AC was running, or vice versa.

If your apartment has only two 15 Amp breakers, it's probably "vintage". I'm not surprised that you have problems. I'm sure that older breakers were much less technologically advanced, plus they may simply be worn out.

Reminds me of my youth in NYC. One screw-in fuse for the entire apartment. The lights flickered when the refrigerator kicked in. Of course no AC, so that wasn't a problem in terms of electrical load.

 

[Euphues]

Not Just Old Wiring

I live in an apartment complex in Manhattan that was opened in 2003. I had never had electrical troubles until my Mac Pro arrived. It's a beast of a machine, but unfortunately trips the breaker in my office every now and then. I think I might try to ditch the energy-hogging floor lamp for a less energy-intensive LED model. I probably ought to get a beefier surge protector as well.

 

[nardeezy]

I live in an apartment complex in Manhattan that was opened in 2003. I had never had electrical troubles until my Mac Pro arrived. It's a beast of a machine, but unfortunately trips the breaker in my office every now and then. I think I might try to ditch the energy-hogging floor lamp for a less energy-intensive LED model. I probably ought to get a beefier surge protector as well.

I had the exact same problem as you and the most probable root cause is an older AFCI circuit breaker. My home was constructed between late 2003 and early 2004. Back then, AFCI was still a little new to the electrical code. Since then, the AFCI circuit breakers have gotten better and are less sucseptible to false trips. I replaced my office AFCI circuit breaker with an exact replacement (same mfg. model#, etc.). The only difference is the one that I previously used was mfg'd in 2004 and the current one was mfg'd in 2008. I would try replacing the circuit breaker.

 

[Tiberinus]

I want to just mention this - don't swap a 15AMP breaker to a 20AMP without ensuring the wire gauge is adequate. Better yet, get someone licensed to do this for you if you are not comfortable with it. Putting too big of a breaker on a circuit that has inadequate wiring for it can lead to bad things, very bad things.

You could have something as simple as a bad breaker, bad power supply, weak connection on the outlet you're using inside the box.

My 10 year old house, 3 days into living there once finished the toaster and microwave would trip the GFCI. Pulled every outlet in the kitchen - one had so much insulation removed from the wires punched into the back they were shorting out - I can only assume heat from current draw of the toaster caused them to move just a little and touch. Had nothing to do with equipment load being too high.

 

[MysticColby]

I've been having a similar problem:
got a new Mac Pro
gave my old G4 to my grandmother (lives in a condo, medium old, maybe 1970s)
she has a PC (I easily converted her to mac after showing her the basics for 30 mins), and it has worked without problems for years
I plug the G4 into the same power strip as her PC and it won't turn on (while PC is on or off)
I plug the G4 into another outlet in her condo and it trips the breaker (did not try to turn it on!)
I plug the G4 into yet another outlet and it won't turn on
ideas?

 

[zorinlynx]

This could also be caused by a defective breaker.

Breakers often have two trip mechanisms, a magnetic trip meant to shut off power instantly on a short circuit, and a thermal trip meant to shut off power on a marginal (not excessive) overload. Thermal trip mechanisms can be destroyed by short circuit current, which is why the magnetic trip exists.

Some older circuit breakers might be marginal enough that inrush current from a Mac Pro is enough trip them magnetically.

Try changing the breaker; they're not that much money and a lot less than redoing your entire electrical system. It may resolve your problem.

 

[cohen777]

Call an electrician. My new house had the rooms wired with 15 amp fuse/wiring for the whole room(s). My Mac Pro and, or laser printer would pull so much current that I would have problems running the equipment. I had a dedicated 20 amp fuse/wiring put in for the computer and laser printer. The Mac Pro does pull a lot of current.

 

[nanofrog]

I had the exact same problem as you and the most probable root cause is an older AFCI circuit breaker. My home was constructed between late 2003 and early 2004. Back then, AFCI was still a little new to the electrical code. Since then, the AFCI circuit breakers have gotten better and are less sucseptible to false trips. I replaced my office AFCI circuit breaker with an exact replacement (same mfg. model#, etc.). The only difference is the one that I previously used was mfg'd in 2004 and the current one was mfg'd in 2008. I would try replacing the circuit breaker.

Good call.

I was going to mention this, as AFCI's tend not to play well with some energy saving devices, particularly CCFL bulbs (devices that have a high startup current). The MP's PSU is likely "seen" as such a load, and trips.