Sizing up power delivery to home lab

[don.r]

I am interested in what people think would be a good amount of power for a home lab. I was thinking of two circuits on 15A breakers each (120V). This should be enough for a few PSUs, CRO, computer, soldering, hot air, etc + any DUTs. Obviously if you have a reflow oven that needs to be taken into consideration. Anyone running close to the limits (tripping their breakers)?

[NiHaoMike]

That should be enough unless you're welding or using some really big power tools. Use opposite phases so you can also have 240V for testing equipment from overseas.

BTW, I suggest 20A as that's the standard nowadays.

[don.r]

Thanks. My home is older (1970's) so its fitted with 15A breakers mostly. I probably could run 20A as the romex looks sufficient. I have a 1000VA autotransformer for the 240V and the arc and propane welders are in the barn, where they shall remain. 

I am interested in what others have as I doubt my lab won't be growing in the future. I do like the older equipment as well (70's, 80's) which tends to be a bit more power hungry.

[nanofrog]

Couple of 15A circuits should do you just fine IMHO. Worst case, sit down and figure out how to balance the load so you don't trip the breakers.

What is the wire size, and length of the run (swap out the breaker & receptacle for 20A)?
If it's 14AWG, it likely wouldn't be enough (15A limit) according to electrical code, so 12AWG would be needed. Basing this on the NEC regs, but I'd be shocked if the rules on this differ in CAN, given the same 120V circuits in place.

In my case, I run a single 20A, and do have to be careful as to what's turned on at any given time. Specifically, I've run into issues for warm-up/inrush current conditions (i.e. running near the limit, and the laser printer starts a warm up cycle and SNAP!).  Lots of crap on the circuit though, as it's an old house (multiple rooms tied to the circuit; house was built in '56).

It's even one of the reasons I've been swapping the lighting to LED (not just power consumption, but less heat generated, as I have to supplement the HVAC system with a window AC unit due to the computer equipment).

[don.r]

I'm fairly sure its 12awg as I have seen that elsewhere but then some of the wiring is also 14awg (bedrooms). Its a farmhouse and likely the farmer did some of the wiring himself so I have to be cautious. Its about 40 ft from panel to room and then another 20 ft in the room. Your experience is exactly why I am asking. Sizing up the equipment I have now means I am getting close to 15A with most of the usual suspects turned on as well as lighting.

The only circuit I can share with is where the home theatre is located. That is set up on two separate 15A circuits with the main amp/tv on one and the subwoofer/lighting on the other. When the sub turns on, the lights dim.

[nanofrog]

I'm fairly sure its 12awg as I have seen that elsewhere but then some of the wiring is also 14awg (bedrooms). Its a farmhouse and likely the farmer did some of the wiring himself so I have to be cautious. Its about 40 ft from panel to room and then another 20 ft in the room. Your experience is exactly why I am asking. Sizing up the equipment I have now means I am getting close to 15A with most of the usual suspects turned on as well as lighting.

The only circuit I can share with is where the home theatre is located. That is set up on two separate 15A circuits with the main amp/tv on one and the subwoofer/lighting on the other. When the sub turns on, the lights dim.

If it's 12AWG on all of it (i.e. no 14AWG legs off of a 12AWG feeder), I wouldn't see a problem with upgrading the breaker and outlet at all (should be within 3% limit for voltage drop). Given the construction is from the '70's, I'll presume it's grounded until you state otherwise.

In my case, the wiring in the house is a disaster, as codes weren't followed (didn't even have code enforcement until 2006). So I've had to deal with crap like a bathroom connected via an 18AWG zip cord (lamp cord) on a 20A circuit.      I'd love to beat the crap out of the a-hole that did that. 

[strangelovemd12]

This should be a fun topic, the power users are consuming must swing by magnitudes.  I wonder what the largest energy demand anyone has for reasonably heating/cooling their space.  There are probably some arctic circle EEs who would literally die if you took away their old scopes and meters.  All those neon indicators and CRTs really warm the place up!

Having just gotten into electronics I've been working on things nomad style throughout the shop.  I don't really have any equipment yet, but even looking at some of the more populated benches around here I can't imagine having enough desktop stuff to need more than a single breaker.  Here my electrical concerns are mostly with outlet count.  Those long strips with outlets every 200mm or something are perfect for running behind almost every bench or desk.   For stacks and clusters of instruments gangs of 2-4 standard outlets are ideal (Ahh, phantom!). Usually the practical solution is to just plug in devices as you need then, since most people don't jump between dozens of devices on a daily/weekly basis, or "the leads don't reach".  Even simultaneously running a fairly high wattage desktop PSU, CRO, 80W iron, computer, etc. should be fine, though maybe close, on a single breaker.  Plenty of power for most home projects I can think of...

Unless you want to put it in a 125*C incubator to test x/T or something silly.  Things like that are what push up the breaker count.  I have a lot of 1.5-2kw equipment, but most of it doesn't get used simultaneously, and shares a single breaker that is separate from my bench and lighting circuit.  I also have a single 20A breaker that runs to an outlet through about 6ft of 12gu that I plug the air compressor into.  Compressors have a sadistic love for kicking in just as you're pushing through a knot of oak.  This is also used as my general purpose dedicated outlet whenever I need one, so mostly for running yard tools

The soldering oven (first project, woo!) I'm working on is >2.4kW, so I'll be driving it off of the 240v split phase.  I'm trying to think of other projects to take advantage of that wealth of power.  Building one of those local on demand water heaters could be fun.  My garage/ climate mix allows for no heating in the winter (though it'd be nice) and in the summer I just plug in a fan that has a couple meters of copper tube strapped to the front that I pump coolant through.  <100W.

To what extent is isolating test equipment necessary to ensure it gets fairly clean AC in?  Do internal filtering and FCC regulations pretty much take care of that?  It's not a problem for PCs, and I haven't heard it come up.

Coldhardstatz (TL;DR):

Breaker 1: 120V 20A
-Lights ~120W
-Computer ~300W
-Audio/Video ~200W
-Battery Charge Station ~50W (in practice)
-Soldering Iron 35W
-PSU 300W
-Remainder over 4 desktop outlets.

Breaker 2: 120V 20A
-Various Power Tools
-Garage Door Opener

Breaker 3: 120V 20A
-Dedicated outlet, used as needed.  (Air compressor by default)

Breaker 4: 120V + 120V 30A (What's the notation for this?)
-Reserved for Future Use 

[pickle9000]

While draw is a problem I have always found that having a separate breaker/line for testing purposes my main issue. Nothing worse than having a breaker blow and shut down your test equipment at the same time. My home lab does not have this nearby so when I need it I run an extension cord across the room. Not the best solution but it's OK. At this time I have lighting on it's own breaker, and two for the bench.

[richard.cs]

This is such an American question  - in the 240V parts of the world no-one would even consider it until they're running environmental chambers or reflow ovens.

I'd be inclined to go with Pickle9000's suggestion of having a seperate circuit for the EUT, mainly so it's less annoying when you short it rather than any concern about power drawn in normal use.

[nanofrog]

To what extent is isolating test equipment necessary to ensure it gets fairly clean AC in?  Do internal filtering and FCC regulations pretty much take care of that?  It's not a problem for PCs, and I haven't heard it come up.

Its more about making sure there's sufficient voltage to the end point (without causing any safety issues such as a fire). Electrical codes require no more than a 3% voltage drop limit on 120VAC (Vdrop = 3.6VAC in this case). So if the wire is too thin for the distance, the voltage drop will exceed this value, and may not get sufficient power to run properly, or not at all (may even have a protection circuit that shuts it down). If such a circuit were missing in a piece of gear/equipment, ...., it could be damaged. So for longer runs, the wire gauge may need to be increased to bring the voltage drop within electrical code so devices plugged in will operate properly.

Now if a device is particularly sensitive to noise (AC mains can be very noisy, depending on your location from the local substation, condition of the infrastructure, ...), a proper design might need to include filtering. There are even emissions standards in place that relate to this (hint: things like FCC A or FCC B statements on the stickers pasted to the back of equipment, such as computers and monitors  ).

This is such an American question  - in the 240V parts of the world no-one would even consider it until they're running environmental chambers or reflow ovens.

I suspect you're right.

To me, it comes down to the amount of power a standard single phase (wall socket/light switch) circuit can provide. Here in the US/CAN, a standard outlet can deliver 1875W (125VAC*15A). I'm under the impression, that in the UK for example, a typical single phase circuit is designed for 13A. Assuming the calculations are done using 250VAC (250VAC*13A = 3250W), then such a circuit would be able to deliver an additional 1375W above a standard wall outlet/switch circuit in the US/CAN.

[richard.cs]

To me, it comes down to the amount of power a standard single phase (wall socket/light switch) circuit can provide. Here in the US/CAN, a standard outlet can deliver 1875W (125VAC*15A). I'm under the impression, that in the UK for example, a typical single phase circuit is designed for 13A. Assuming the calculations are done using 250VAC (250VAC*13A = 3250W), then such a circuit would be able to deliver an additional 1375W above a standard wall outlet/switch circuit in the US/CAN.

You're correct although most of the UK is at 240V (nominally 230V +10%/-6% but in reality 240V from transformers often on the 250V tap). Most UK sockets are doubles, and although they're not really designed for 13A from both at once* there is obviously more power available, and it's common in a work-area or a kitchen to have several double-sockets near each other on a single 32A circuit. Certainly where I work no-one thinks twice about plugging a 2kW heat gun into a bench already loaded with test equipment, even if that bench is hanging off a single 13A plug.

That perhaps indicates that 3 kW is plenty, although bear in mind the UK system is designed with fairly slow fusing curves, so 4 kW for a couple of minutes wouldn't pop any fuses (by design since the cables also wouldn't get dangerously hot in that time). I'm not sure what the current-time curves of a USA circuit breaker are like.

*The specification calls for testing with 14A through the first and 6A through the second, Only the most cheap nasty rubbish should be expected to fail with 13A on each.

[Monkeh]

*The specification calls for testing with 14A through the first and 6A through the second, Only the most cheap nasty rubbish should be expected to fail with 13A on each.

I would not expect any socket, single or double, to actually handle 13A after it's been in use. BS1363 may have a fairly decent plug, but the sockets suck.

[richard.cs]

I would not expect any socket, single or double, to actually handle 13A after it's been in use. BS1363 may have a fairly decent plug, but the sockets suck.

I'm aware that they contact springiness can degrade with time and heat, and that the contact area in some designs is quite small but I wouldn't go that far. I've had 20 amps through one for several hours without any damage or excessive heating. That was with a "double-neutral plug"* though so there was no heating from the fuse. Not a practise I would recommend, not least because having removed the fusing you need to have knowledge of the upstream wiring to determine if it's safe or not.** 3kW heaters in good quality sockets seem ok, but with the plug running warm due to heat from the fuse.

* Double-neutral plug, don't do this unless you seriously know what you're doing: Take two rewirable 13A plugs of the type with loose pins. Throw the live pin and fuse assembly away and replace with the neutral pin from the other plug. Simple unfused plug.

** Usually not

[david77]

My shed's connected to 3 phase 230V with three 25A main fuses, 3 phase in domestic premises is the norm in Germany.
The electronics lab runs off a single 10A breaker, only the lights + PC are on a dedicated circuit. I hate it when I'm left in the dark should the breaker trip.
That's 2300W maximum (or a bit more for some time, same as in the UK scenario) and I've never had a problem (*) with T&M gear drawing too much power, what the hell do you expect to run? 20 old tube era Tek scopes at once?

One thing I find quite handy is to install one master switch that allows you to shut the whole lab/bench down when leaving the room. I've got a switch right next to the light switch - when I leave everything's powered down except the PC.

(*) One thing that regularly tripped the breaker used to be my 500W isolation transformer, it's inrush current must be huge. I solved that with a 555, relay and a resistor in series to limit inrush current.
I could also just swap the 10A for a 16A breaker, the wiring would take that no problem, but then I'd have to replace the master switch as it's only rated fo 10 amps...

[don.r]

That's 2300W maximum (or a bit more for some time, same as in the UK scenario) and I've never had a problem (*) with T&M gear drawing too much power, what the hell do you expect to run? 20 old tube era Tek scopes at once?

On 120V we are on a more limited power budget of 1875W (15A) per circuit. Lights + computer can take 500W alone. Bench PSU maybe 100W, CRO another 100W (I use 2), soldering station+hot air maybe another 100-300W, etc. Like I said, I'm partial to the older stuff (but not vacuum tube). I could probably approach 1875W without much problem before I even plug in any DUT. If I am testing a PSU or amp, I'll need maybe another 500W or so. I could do it on a single circuit if I thought about shutting down everything I didn't immediately need all the time but if I forget just once...POW! There is also the nice aspect of having a DUT on a separate breaker but I could put in a faster blowing, lower amperage breaker in its own panel just for that socket.

I might look more carefully at the wiring to see if I can switch to a 20A breaker and then share the TV's circuit (whose access I can control!) with a purpose built DUT test socket with fast blow 10A breaker.

[nanofrog]

To me, it comes down to the amount of power a standard single phase (wall socket/light switch) circuit can provide. Here in the US/CAN, a standard outlet can deliver 1875W (125VAC*15A). I'm under the impression, that in the UK for example, a typical single phase circuit is designed for 13A. Assuming the calculations are done using 250VAC (250VAC*13A = 3250W), then such a circuit would be able to deliver an additional 1375W above a standard wall outlet/switch circuit in the US/CAN.

You're correct although most of the UK is at 240V (nominally 230V +10%/-6% but in reality 240V from transformers often on the 250V tap). Most UK sockets are doubles, and although they're not really designed for 13A from both at once* there is obviously more power available, and it's common in a work-area or a kitchen to have several double-sockets near each other on a single 32A circuit. Certainly where I work no-one thinks twice about plugging a 2kW heat gun into a bench already loaded with test equipment, even if that bench is hanging off a single 13A plug.

That perhaps indicates that 3 kW is plenty, although bear in mind the UK system is designed with fairly slow fusing curves, so 4 kW for a couple of minutes wouldn't pop any fuses (by design since the cables also wouldn't get dangerously hot in that time). I'm not sure what the current-time curves of a USA circuit breaker are like.

*The specification calls for testing with 14A through the first and 6A through the second, Only the most cheap nasty rubbish should be expected to fail with 13A on each.

Is modern panel tech using fuses or circuit breakers in the UK (new construction)?

Figured fuses would only be found in older systems (generally the case here), though still relevant.
Do the breakers replicate the fuse timing-current curves exactly or are they different (assuming this is the circuit protection method used)?
Or where the timings changed (shortened) with the expectation that fused systems will eventually be eliminated due to remodels/refits for increased safety margins?

BTW, if you're curious, here's a link (.pdf) to curves for a line of breakers available here (18pgs).

In the case of the current ratings mentioned, I was going on the basis that the ratings across the board may be x Amps, but is also the value circuit breaker/fuse was designed for (13A for UK, 15A here). So no matter if the load was on a single outlet or split over the max number of outlets/switches allowable (switches & outlets can take the sustained values for safety purposes), it opens when the nominal current draw exceeds this value (not brief overloads that fall in the timing limits).

Or does it differ in the UK in this regard from the US?

[don.r]

Are they talking maybe about the plug fuses?

[Monkeh]

Is modern panel tech using fuses or circuit breakers in the UK (new construction)?

Figured fuses would only be found in older systems (generally the case here), though still relevant.

Breakers. We have fuses in plugs.

Do the breakers replicate the fuse timing-current curves exactly or are they different (assuming this is the circuit protection method used)?

They're different and several profiles are available to suit the intended load.

In the case of the current ratings mentioned, I was going on the basis that the ratings across the board may be x Amps, but is also the value circuit breaker/fuse was designed for (13A for UK, 15A here). So no matter if the load was on a single outlet or split over the max number of outlets/switches allowable (switches & outlets can take the sustained values for safety purposes), it opens when the nominal current draw exceeds this value (not brief overloads that fall in the timing limits).

Or does it differ in the UK in this regard from the US?

We differ, as our plugs are fused, we can have much higher rated circuits with multiple outlets. 20A is the norm for radials, 32A for ring finals (look them up).

[nanofrog]

Are they talking maybe about the plug fuses?

Not sure, and I'm a bit confused as a result. 

I'm not accustomed to fused receptacles, just older panels that use fuses (in terms of structural wiring), or end-point devices. In terms of end-point devices, I've seen plenty of fused IEC sockets, or the older round fuse holders. Neither are part of the structural wiring though. Closest equivalent structurally that comes to mind, would be GFCI's where required rather than a standard 3 prong NEMA 5-15R.

[don.r]

Are they talking maybe about the plug fuses?

Not sure, and I'm a bit confused as a result. 

I'm not accustomed to fused receptacles, just older panels that use fuses (in terms of structural wiring), or end-point devices. In terms of end-point devices, I've seen plenty of fused IEC sockets, or the older round fuse holders. Neither are part of the structural wiring though. Closest equivalent structurally that comes to mind, would be GFCI's where required rather than a standard 3 prong NEMA 5-15R.

In the UK, the plugs themselves are fused. Not the sockets (they have a switch sometimes, though). I believe this lets them size the fuse to the appliance. Its the same fuse as my UT61E uses for the amps range.

[IanB]

In the UK, the plugs themselves are fused. Not the sockets (they have a switch sometimes, though). I believe this lets them size the fuse to the appliance. Its the same fuse as my UT61E uses for the amps range.

Right. In theory you fit a 3 A fuse in the plug for a low power device like a lamp or television, and a 13 A fuse in the plug for a high power device like a kettle or clothes iron. However, it is not uncommon to see 13 A fuses used everywhere as it is easiest to keep one value of fuse in the spares drawer and use it for all duties.

[Monkeh]

Right. In theory you fit a 3 A fuse in the plug for a low power device like a lamp or television, and a 13 A fuse in the plug for a high power device like a kettle or clothes iron.

Said theory is crap. The fuse is there for the cable, not the device.

[richard.cs]

Is modern panel tech using fuses or circuit breakers in the UK (new construction)?

Figured fuses would only be found in older systems (generally the case here), though still relevant.

It's already been said but in the UK all plugs have sand filled ceramic fuses in the range 3-13A (although they are available down to 1A they are rarely used). This is intended to protect the appliance flex, it allows a table lamp to have a long thin flex that if shorted at the end wouldn't pass enough current blow a 13A fuse or trip the 32A MCB in a reasonable time. One area of value is extension leads which I believe are a bit of a fire hazard in the US - nothing stops someone using a thin cord for a 20A appliance. In modern practice distribution boards use similar circuit breakers to the US with a thermal and a magnetic trip.

Do the breakers replicate the fuse timing-current curves exactly or are they different (assuming this is the circuit protection method used)?
Or where the timings changed (shortened) with the expectation that fused systems will eventually be eliminated due to remodels/refits for increased safety margins?

BTW, if you're curious, here's a link (.pdf) to curves for a line of breakers available here (18pgs).

Picking your 15A breaker as an example it is pretty close to out D-curve 16A: http://wiki.diyfaq.org.uk/images/thumb/6/69/Curve-MCBTypeD.png/640px-Curve-MCBTypeD.png
These breakers are almost never used in domestic situations here as it's difficult to get the loop impedance low enough that a short circuit will clear in the time required by the regulations. Most breakers in domestic use are B-curve, http://wiki.diyfaq.org.uk/images/thumb/d/d4/Curve-MCBTypeB.png/640px-Curve-MCBTypeB.png note the 16A MCB becoming instantanious (well, constant-time on the magnetic part) at 80 A in contrast to the 320A for a D -curve or the 280A for a US 15A breaker.

[nanofrog]

Is modern panel tech using fuses or circuit breakers in the UK (new construction)?

Figured fuses would only be found in older systems (generally the case here), though still relevant.

Breakers. We have fuses in plugs.

Thanks for the clarification.

I had forgotten that the plugs are fused there (structural). 

It's already been said but in the UK all plugs have sand filled ceramic fuses in the range 3-13A (although they are available down to 1A they are rarely used). This is intended to protect the appliance flex, it allows a table lamp to have a long thin flex that if shorted at the end wouldn't pass enough current blow a 13A fuse or trip the 32A MCB in a reasonable time. One area of value is extension leads which I believe are a bit of a fire hazard in the US - nothing stops someone using a thin cord for a 20A appliance. In modern practice distribution boards use similar circuit breakers to the US with a thermal and a magnetic trip.

Thanks.

I would agree that extension cords can be a hazard here due to users not understanding that you can't just plug anything that fits the receptacle, or how distance affects voltage drop (i.e. 13A saw on a 100' 16AWG cord, and it gets warm enough you can feel the temp in your fingers).

Picking your 15A breaker as an example it is pretty close to out D-curve 16A: http://wiki.diyfaq.org.uk/images/thumb/6/69/Curve-MCBTypeD.png/640px-Curve-MCBTypeD.png
These breakers are almost never used in domestic situations here as it's difficult to get the loop impedance low enough that a short circuit will clear in the time required by the regulations. Most breakers in domestic use are B-curve, http://wiki.diyfaq.org.uk/images/thumb/d/d4/Curve-MCBTypeB.png/640px-Curve-MCBTypeB.png note the 16A MCB becoming instantanious (well, constant-time on the magnetic part) at 80 A in contrast to the 320A for a D -curve or the 280A for a US 15A breaker.

Only know the curves exist, so details of B, D, ... timing-current curves (would assume those used in low voltage applications would use different models yet), exceed my knowledge/experience, as my education and OJT are in computer engineering. 

CPU's, ...., and such, I can manage. So regarding power, I resort to ready-made PSU's, hand it off to someone with the experience if available, or outsource. I just don't have the experience in this area, or time to learn during a project (consuming company time).  At least I'm slowly getting a bench together to pick up in areas I'm interested in and short on knowledge and experience. 

[Monkeh]

(would assume those used in low voltage applications would use different models yet)

These are low voltage applications.