How can I test dubious ground or neutral wires

[ebeowulf17]

I install and repair coffee and espresso equipment and we sometimes run into electrical problems beyond my knowledge (well versed in household wiring, good with fundamental applications of ohms law, etc. but I'm no physics major!) The issue that comes up most often is dubious grounds, or occasionally neutrals.

If there's a loose wire-nut connection, a partially broken wire with one thin strand left, or a very poor connection in a breaker box, I could get good multimeter readings from the building wiring going into a coffee brewer (120vac from each of the two hot legs to ground and to neutral, 240 between them (or 208 if feeding from three phase power), and 0 voltage from neutral to ground... but that thin/loose/weak connection wouldn't be able to carry the current necessary for the machine to operate safely and properly.

So my question is, what can I do, beyond the multimeter testing I just described, to try to root out wiring that's not completely mis-wired or disconnected, but simply undersized or poorly connected?

If it helps at all, the most recent issue to make me wonder about this is a large (240vac, 25amp, brews up to 3 gallons at a time) coffee brewer at a hotel which had run fine for years until it's main circuit board caught fire. We didn't have a replacement board, but had a spare machine, so we installed a different machine (eliminating the likelihood that any problems with internal wiring of the machine could be the culprit), which subsequently blew up its main circuit board within 24 hours. At that point, I checked and re-checked the wiring to the best of my ability and hoped that it was a terribly unlikely coincidence... and installed a new circuit board, which caught fire within about 2 hours of being hooked up! If the neutral wasn't acting like a proper neutral, the relay and a few triacs on the board could see way higher voltages than they expect, which might explain the meltdowns, but I don't know how to find the wiring problem.

Sorry for the long first post. I'd greatly appreciate any insights into testing neutral and ground wiring.

 

[schip666!]

Wow... 6000W making 3 gallons of espresso in one swoop...I have a coffee-nut friend who will be interested in that...

The problem is that you are looking for fairly small resistances. The nominal resistance of your entire 6000w circuit is 10 ohms (Volts/Amps) and you need to have a really good meter-probe connection to get below 1 ohm measurement. One way to deal with it is to measure the voltages on both sides of -- or across -- various connections under full load to see if there is any significant difference -- I once fixed a friends truck-starting-problem, a bad battery connector, by measuring the voltage along the wire while (trying to) run the starter.

For your exploding coffee machine problem, it could be a flaky power connection -- a funky ground connection would only affect electrical safety. But I would mostly expect bad power to make it not work very well (too cool) rather than over-heat and flame-out. Running on two-phase, the neutral shouldn't be carrying any current. Do you get any indication of where the over-heating is on the board? Are you sure you are running on the right power? Try measuring voltage across the power connections, wire to board, to see if there is a significant drop. If there is there is resistance and that makes heat. Can also use a temperature probe to see if anything is hotter than expected.

Oh, another thing...is there anything else (especially new equipment) running on the same power feed? If you heavily load one leg of a two-phase (or three for that matter) feed, the _other_ leg will increase in voltage to compensate.

 

[ebeowulf17]

Sorry, should've been a bit more specific, but was afraid to as my initial post was already pretty long...

The brewer uses both the two-phase (208, fed by two legs of a three-phase system) and single phase (120) power, and so requires the neutral to carry current back from the 120 components. The heating circuit gets the full 208 and the circuit board and everything else it controls (valves, etc) get 120. Because of the way the 208 heating circuit ties in, if the neutral fails, the board sees 208 instead of 120, which fries various components on the board. In all three of the blowouts, it's been one particular triac that's blown first, right next to the board-mounted relay which switches one leg of the heating circuit (the other leg is always available, only one side gets switched.)

We're definitely running on the right power for this machine.

From what they tell me, there haven't been any changes in major equipment nearby. I am curious though - I've never heard of one leg increasing in voltage to compensate for loads on another leg... can you explain in more detail how that works? I've heard vague references to the benefits of maintaining "balanced loads" on distribution transformers, but never heard any meaningful explanation of why it matters or how things can go wrong if you don't. I'm guessing that's basically the same issue. Would love to learn more!

Not too worried about ohm meter sensitivity - no problem at all consistently reading as low as 0.1 to 0.2 ohm resistances between various grounding points, so if sub-1 ohm readings are useful, I can take them. What would you have me measure?

In the end, I still mostly need to know if there's a way to check the quality of the ground and/or neutral wires, as that's the most likely culprit causing these meltdowns.

 

[schip666!]

I would look into the situations leading up to "If the neutral fails..." Probably should have completely separate wiring for the 120v control board as well.

Also, back to power legs, see if something else is running on your three-phase and maybe mis-wired too. Are you SURE that nothing was changed in the power distribution? Someone added or subtracted a big bread mixer or something like that? Inductive kick might be bad enough to cause trouble if a connection back to the service entrance is failing.

To the varying voltage on opposite phases thing. You can try it out with a center tapped transformer, put a load on one leg and see what happens to the voltage on the unloaded side. It's self-inductance or something scientifically sounding like that...

 

[Dmytry]

Sorry, should've been a bit more specific, but was afraid to as my initial post was already pretty long...

The brewer uses both the two-phase (208, fed by two legs of a three-phase system) and single phase (120) power, and so requires the neutral to carry current back from the 120 components. The heating circuit gets the full 208 and the circuit board and everything else it controls (valves, etc) get 120. Because of the way the 208 heating circuit ties in, if the neutral fails, the board sees 208 instead of 120, which fries various components on the board. In all three of the blowouts, it's been one particular triac that's blown first, right next to the board-mounted relay which switches one leg of the heating circuit (the other leg is always available, only one side gets switched.)

hmm, how exactly is that heating wired in so that it'd put 208 on the board?

For the testing - with everything disconnected, you can measure small resistances by hooking up a known value resistor in series with your wire, and powering that with a battery.
You measure the voltage across the resistor, then you measure the voltage across the wire, the voltage in such circuit is directly proportional to resistance so e.g. your resistor is 10 ohm, you have 12v on the resistor, and you have 12 millivolt on the wire, your resistance is 0.01 ohm. Or you can simultaneously measure current in wire and voltage over it, with big enough current (e.g. using a car battery). But it needs to be done with the circuit completely disconnected.

For practical solution - I would look into measuring the voltages during operation to see if something is wrong. As stated elsewhere it is in principle possible that something else has changed about the load. I'm unfamiliar with US systems. My apartment is wired with 400v three phase.

If the neutral is disconnected from supply, your voltage on neutral will depend to dis-balance between the loads connected between neutral and the phases. It is also possible that something like a fridge motor startup briefly 'shorts' the neutral to one phase, putting 208 volts to the board.

 

[Roar]

It IS somewhat difficult to measure the voltage drop in a ground leg caused by there being a high resistance connection but since you're near plumbing which is supposed to be connected to ground in addition to usually being grounded through several other grounded items any water pipes should provide a reference to ground, Take your AC voltmeter and measure the voltage between all points TO the plumbing including the pot ground. I'd guess that you're connected to the WILD leg of a three phase system which tends to be around 277 VAC.
In that you're in an industrial site, you might be exposed to some BIG transient spikes from nearby starting motors etc. If all the voltages are seemingly good, hang some TRANSORBS or such on each leg to ground. There IS a possibility that you're so far from an Earth ground that the impedance of the ground leg is close to infinite at some frequencies where a transient is. I might cheat and connect the green wire to the plumbing also, I don't recall anything saying that you can't ground the nearby sink or metal counter also-- just to make it safe--and also lower the impedance to ground through the piping.
Michael

I install and repair coffee and espresso equipment and we sometimes run into electrical problems beyond my knowledge (well versed in household wiring, good with fundamental applications of ohms law, etc. but I'm no physics major!) The issue that comes up most often is dubious grounds, or occasionally neutrals.

If there's a loose wire-nut connection, a partially broken wire with one thin strand left, or a very poor connection in a breaker box, I could get good multimeter readings from the building wiring going into a coffee brewer (120vac from each of the two hot legs to ground and to neutral, 240 between them (or 208 if feeding from three phase power), and 0 voltage from neutral to ground... but that thin/loose/weak connection wouldn't be able to carry the current necessary for the machine to operate safely and properly.

So my question is, what can I do, beyond the multimeter testing I just described, to try to root out wiring that's not completely mis-wired or disconnected, but simply undersized or poorly connected?

If it helps at all, the most recent issue to make me wonder about this is a large (240vac, 25amp, brews up to 3 gallons at a time) coffee brewer at a hotel which had run fine for years until it's main circuit board caught fire. We didn't have a replacement board, but had a spare machine, so we installed a different machine (eliminating the likelihood that any problems with internal wiring of the machine could be the culprit), which subsequently blew up its main circuit board within 24 hours. At that point, I checked and re-checked the wiring to the best of my ability and hoped that it was a terribly unlikely coincidence... and installed a new circuit board, which caught fire within about 2 hours of being hooked up! If the neutral wasn't acting like a proper neutral, the relay and a few triacs on the board could see way higher voltages than they expect, which might explain the meltdowns, but I don't know how to find the wiring problem.

Sorry for the long first post. I'd greatly appreciate any insights into testing neutral and ground wiring.

 

[Roar]

It IS somewhat difficult to measure the voltage drop in a ground leg caused by there being a high resistance connection but since you're near plumbing which is supposed to be connected to ground in addition to usually being grounded through several other grounded items any water pipes should provide a reference to ground, Take your AC voltmeter and measure the voltage between all points TO the plumbing including the pot ground. I'd guess that you're connected to the WILD leg of a three phase system which tends to be around 277 VAC.
In that you're in an industrial site, you might be exposed to some BIG transient spikes from nearby starting motors etc. If all the voltages are seemingly good, hang some TRANSORBS or such on each leg to ground. There IS a possibility that you're so far from an Earth ground that the impedance of the ground leg is close to infinite at some frequencies where a transient is. I might cheat and connect the green wire to the plumbing also, I don't recall anything saying that you can't ground the nearby sink or metal counter also-- just to make it safe--and also lower the impedance to ground through the piping.
Michael

I install and repair coffee and espresso equipment and we sometimes run into electrical problems beyond my knowledge (well versed in household wiring, good with fundamental applications of ohms law, etc. but I'm no physics major!) The issue that comes up most often is dubious grounds, or occasionally neutrals.

If there's a loose wire-nut connection, a partially broken wire with one thin strand left, or a very poor connection in a breaker box, I could get good multimeter readings from the building wiring going into a coffee brewer (120vac from each of the two hot legs to ground and to neutral, 240 between them (or 208 if feeding from three phase power), and 0 voltage from neutral to ground... but that thin/loose/weak connection wouldn't be able to carry the current necessary for the machine to operate safely and properly.

So my question is, what can I do, beyond the multimeter testing I just described, to try to root out wiring that's not completely mis-wired or disconnected, but simply undersized or poorly connected?

If it helps at all, the most recent issue to make me wonder about this is a large (240vac, 25amp, brews up to 3 gallons at a time) coffee brewer at a hotel which had run fine for years until it's main circuit board caught fire. We didn't have a replacement board, but had a spare machine, so we installed a different machine (eliminating the likelihood that any problems with internal wiring of the machine could be the culprit), which subsequently blew up its main circuit board within 24 hours. At that point, I checked and re-checked the wiring to the best of my ability and hoped that it was a terribly unlikely coincidence... and installed a new circuit board, which caught fire within about 2 hours of being hooked up! If the neutral wasn't acting like a proper neutral, the relay and a few triacs on the board could see way higher voltages than they expect, which might explain the meltdowns, but I don't know how to find the wiring problem.

Sorry for the long first post. I'd greatly appreciate any insights into testing neutral and ground wiring.

 

[jim hardy]

I recently used daughter's hair dryer as a Poor Man's Power Outlet tester. Her emergency generator had, it turned out, an bad neutral wire connection at one of its receptacles.

Imagine for your trouble calls a simple test box with a cord that mates with whatever receptacles your customers plug their machine into,

your test box having 4 meter test points for both hots, neutral and ground,
and a receptacle EDIT make that a PAIR of receptacles - from each hot to neutral for about a one kw load like a hair dryer or huge lamp.. you could split a standard 15 amp outlet.
Now with your voltmeter measuring voltage between neutral and ground you apply the load to each hot side in turn and measure how much neutral elevates above ground.
It should elevate less than a volt i'd think for a 1kw load.
That checks neutral wire's condition.

Now measure voltage to neutral on each hot wire with and without load, difference tells condition of hot wires.


I don't know conditions in these places. If it's safe you could apply a small load maybe ten watts from hot to ground, simulating a mild ground fault, and see if ground elevates more than a few millivolts relative to neutral.
If it elevates there's trouble in customer's ground wire and i'd worry about customer's installation meeting code requirement that a ground fault trip the overcurrent device. If it's GFCI ten watts should trip it and you've tested the GFCI function...

Either way tell him you're checking his wiring for protection of your equipment as well as his employees.

Sorry for the excess detail - you doubtless got the idea at first line. maybe it'll help somebody else.