So much for the super beast.

[K8MHZ]

My sister has had problems with the control board in her furnace. The furnace guy told her last time it happened she didn't have a ground. So, I went over there and physically traced the old Romex all the way back to the panel. This involved taking paneling off, moving stuff in the attic, etc. Everything was as I expected from the readings I took before we did the physical tracing. Nothing was wrong.

A couple days ago, she loses another board. Furnace guy says same thing. I reminder her of the above. So I asked her to ask the furnace guy if a bad neutral causing over voltage is known to take out the boards. He said it was possible so I went over there to load things up and check for signs of a bad neutral.

Here are the readings. Voltages are (L1-N) - (L2-N) L readings at the main breaker, N readings at the bare neutral coming into the main panel.

Lightly loaded, 118.3 - 118.4

Moderately loaded 115 - 121

Heavy load 109 - 115

The 109 side is the furnace side. The boards probably don't like low voltage. That I understand.

A call to the POCO gets a linesman out right away.

He pulls the meter and puts this cheesy thing called a 'Super Beast' on the tabs, turns it on and by looking at two crude analog meters he tells me the neutral is almost perfect. Now, this thing is using 12 awg extension cord wire with clamps that look like they belong on a 6 amp charger.

I told him (as nice as I could) what I thought of his Super Beast and told him if I loaded the house up with the range, microwave, etc,. I could get the voltage down to 109. I know he is wrong about the bad neutral, but at least I can show him the sag. So (while the Beast is still connected) he tells me to load the house up.

I told him I would probably need the meter put back in for that to happen. So he puts the meter back in and says he can't connect the Beast at the can because there is no room.

He decides to check some connections at the pole feeding the house. 119 -119 lightly loaded, 114-116 heavily loaded. He did this with his meter, the Beast lay unused in the snow.

While he was doing his thing, I took voltage readings at the meter (there was room for a regular meter probe) and got my loaded voltage down to 109 - 116. (L readings at line side meter tabs, N at bare conductor in can).

The other end of the cable right at the tranny read (we talked via cell phone) 119-119 no load (main off) and almost that loaded.

So he shuts the power down for a couple hours (sister not happy) and replaces the triplex that (to me) read OK. He also replaced a bunch of connectors. He turns the power back on.

I repeat the test and the drop is within a volt of my earlier tests. I showed him, he verified it with his meter and was dumbfounded. So, he changes the connectors at the drop (without testing them). That recovered about a volt, but we were still tilted and low with 116-112. Also our loading gear (the stove) was cycling less as it got hotter. The triplex from the house to the pole was not changed. He said he did change the connections on both ends.

By now my sister has been without steady power for three hours and she is sick of it. I told the linesman that and he said he thought it was the transformer, anyway, and was closing out the ticket. He told me the transformer was a 10kva and it should be a 20. This is for at least four homes. There may be a fifth one, I will check later.

I gave the linesman my card and asked if he would have the area field engineer call me. He said he would.

So, what do you think is going to happen when they heavy up the transformer and don't look for the bad neutral (that is somewhere between the meter can and the pole) ?

If they bring the loaded voltages up, one side may get real high. The most I could get the high side to swing to was 121. But that was with 115 on the low side. If they bring the low side up to 120, that will put the high side at 126. Or worse, as bad neutrals don't ever get better, they just get worse.

But anyway, so much for the 'Super Beast'.

Why would anyone use such a crude instrument for troubleshooting professionally? It looked like it was made from drain pipe and Radio Shack parts.

 

[macmikeman]

It gets used daily around here to locate where the neutral is broken. One of the more useful tools invented.

 

[ptonsparky]

When I look for bad neutrals I expect the L to L voltage to stay the same. Lets say 240. Line to neutral would be equal. 120 & 120. Load one side and the voltage swings on line to neutral but remains steady L to L. 108 & 132 with 240 L to L.

Your Light to moderate load seem to follow that scenario while the heavy load indicates a total VD problem as well.

I would like to know the currents involved as well as voltage.



Doesnt every tech carry three volt meters and three amp clamps to each service call?

 

[K8MHZ]

It gets used daily around here to locate where the neutral is broken. One of the more useful tools invented.

I could see it's use for locating a broken neutral. But not for a high impedance one. I checked the specs on it. It's basically a 1600 watt space heater with volt meters on it.

One one he was using had analog voltmeters on it. 0-150 volts, 5 volts per increment. Hardly a precise tool for checking voltage fluctuation

They do have digital meters on some and they even have one that's 80 amps.

 

[mbrooke]

Doesn't sound like a bad neutral to me more like hair thin conductors and an undersized transformer.

A small neutral can produce similar readings to one that has a slightly poor connection. One side will go down a lot the other phase will climb a few volts. Nothing wild but the voltage will not be stable as much.


Your linemen sounds friendly by the way. Some aren't as willing to change connections or openly admit the transformer is undersized.


To be honest around my place any transformer feeding more than a single house is 25kva, usually doing 5 houses. 50kva if they are electric heat houses. 10 kva is to small for 5 houses, especially if you have an electric range in operation.

 

[K8MHZ]

When I look for bad neutrals I expect the L to L voltage to stay the same. Lets say 240. Line to neutral would be equal. 120 & 120. Load one side and the voltage swings on line to neutral but remains steady L to L. 108 & 132 with 240 L to L.

Your Light to moderate load seem to follow that scenario while the heavy load indicates a total VD problem as well.

I would like to know the currents involved as well as voltage.



Doesnt every tech carry three volt meters and three amp clamps to each service call?

I think there are two problems. One is too much resistance in the neutral, the other is the voltage sag.

The POCO guy didn't take any current measurements. He only had one Fluke voltmeter and the Beast out of the truck. His load, the Beast, was rated at 1600 watts at 120 volts or 13.3 amps. Taking into consideration the 114 volts at the can, the resistive element would have only drawn about 12 1/2 amps.

 

[growler]

My sister has had problems with the control board in her furnace. The furnace guy told her last time it happened she didn't have a ground. .

A couple days ago, she loses another board. Furnace guy says same thing.

The 109 side is the furnace side. The boards probably don't like low voltage. That I understand.

109 volts doesn't sound that low to me.

2 boards go bad. One cause is that they replaced with a defective board (no quality control) or there really is a problem with the furnace and the guy isn't smart enough to trouble-shoot.

If two guys say there isn't a ground and you know there is one that should tell you something. Sounds like standard BS when they don't know.

 

[mbrooke]

109 volts doesn't sound that low to me.

2 boards go bad. One cause is that they replaced with a defective board (no quality control) or there really is a problem with the furnace and the guy isn't smart enough to trouble-shoot.

If two guys say there isn't a ground and you know there is one that should tell you something. Sounds like standard BS when they don't know.

Its mind blowing how poorly trained some technicians are. The only thing they know how to do is blame it on some none existent issue.

BTW, if the control board was looking for a ground reference at most it would flash error codes, not self destruct.

 

[K8MHZ]

Doesn't sound like a bad neutral to me more like hair thin conductors and an undersized transformer.

A small neutral can produce similar readings to one that has a slightly poor connection. One side will go down a lot the other phase will climb a few volts. Nothing wild but the voltage will not be stable as much.


Your linemen sounds friendly by the way. Some aren't as willing to change connections or openly admit the transformer is undersized.


To be honest around my place any transformer feeding more than a single house is 25kva, usually doing 5 houses. 50kva if they are electric heat houses. 10 kva is to small for 5 houses, especially if you have an electric range in operation.

Yeah, he was a nice guy.

At first he was just willing to walk away based on the reading of the 'Beast'. When I told him I was a licensed electrician and showed him the voltage drop on my meter, he got real friendly and started checking and chopping. I really don't think he knew what he was looking for. At least he was nice enough to tell me what he was reading at the connectors. There is a balanced 118 volts on one side of a run consisting of triplex and the meter drop. On the other side is an imbalanced voltage dropping as low as 110 volts on one side and as high as 121 volts on the other.

 

[mbrooke]

Yeah, he was a nice guy.

At first he was just willing to walk away based on the reading of the 'Beast'. When I told him I was a licensed electrician and showed him the voltage drop on my meter, he got real friendly and started checking and chopping. I really don't think he knew what he was looking for. At least he was nice enough to tell me what he was reading at the connectors. There is a balanced 118 volts on one side of a run consisting of triplex and the meter drop. On the other side is an imbalanced voltage dropping as low as 110 volts on one side and as high as 121 volts on the other.

Sounds like hair thin conductors. What gauge are they, or at least roughly?

 

[ptonsparky]

A bad neutral does not mean it has to be open entirely. Poor connections are as guilty as open, maybe more so.

 

[gar]

140310-1555 EDT

K8MHZ:

How I would run your test.

Run a test lead wire from the pole transformer ground back to the house where the measurements are to be made. Use extension cords or any convenient insulated wire #22 or larger for the test lead.

Get two or three similar 1500 W space heaters. Test each one to determine its hot resistance. The ones I have used tend to be about 1200 to 1300 W when hot.

Measure at the main panel busbars L1-N, L2-N, and Test Lead to N under the following conditions:
One heater L1-N.
Same heater L2-N.
One heater L1-N, and a second heater of nearly the same resistance L2-N.

From these measurements we can learn much about the transformer, service hot wires, and neutral.

Initially one assumes negligible current on the transformer grounding wire, that all connections are secure, and that thru the meter and main fuses there are no problems.

Make a judgement if the service wiring is of equal wire size and material.

With only one 120 V load the measured voltage drop across the neutral should be about 1/2 the change in voltage across L1-N from no load to loaded.

With two identical series loads connected from L1-L2 and the mid point to neutral the change in voltage from no-load to loaded across L1-L2 should be the same as the change in voltage across L1-N when when only one test load is used.

If a furnace circuit board can not properly operate from at least 95 to 135 V. then it is badly designed. Also there should be protection from high voltage transients.

The impedance measurement of the neutral from pole to busbar is mostly the wire resistance and a few connection joints.

The effective impedance measurement on the hot wires includes the transformer internal impedance, wire resistance, power company meter resistance, and main breaker or fuse resistance.

.

 

[K8MHZ]

Sounds like hair thin conductors. What gauge are they, or at least roughly?

6 awg ALU.

 

[K8MHZ]

140310-1555 EDT

K8MHZ:

How I would run your test.

Run a test lead wire from the pole transformer ground back to the house where the measurements are to be made. Use extension cords or any convenient insulated wire #22 or larger for the test lead.

Get two or three similar 1500 W space heaters. Test each one to determine its hot resistance. The ones I have used tend to be about 1200 to 1300 W when hot.

Measure at the main panel busbars L1-N, L2-N, and Test Lead to N under the following conditions:
One heater L1-N.
Same heater L2-N.
One heater L1-N, and a second heater of nearly the same resistance L2-N.

From these measurements we can learn much about the transformer, service hot wires, and neutral.

Initially one assumes negligible current on the transformer grounding wire, that all connections are secure, and that thru the meter and main fuses there are no problems.

Make a judgement if the service wiring is of equal wire size and material.

With only one 120 V load the measured voltage drop across the neutral should be about 1/2 the change in voltage across L1-N from no load to loaded.

With two identical series loads connected from L1-L2 and the mid point to neutral the change in voltage from no-load to loaded across L1-L2 should be the same as the change in voltage across L1-N when when only one test load is used.

If a furnace circuit board can not properly operate from at least 95 to 135 V. then it is badly designed. Also there should be protection from high voltage transients.

The impedance measurement of the neutral from pole to busbar is mostly the wire resistance and a few connection joints.

The effective impedance measurement on the hot wires includes the transformer internal impedance, wire resistance, power company meter resistance, and main breaker or fuse resistance.

.

It would probably take 150 feet or more of cord or cable to do that.

I have the bad board. It's a Honeywell S8610U

 

[dfmischler]

I have the bad board. It's a Honeywell S8610U

That board family is apparently often used as a "universal replacement". Can you tell what board was originally spec'ed from the wiring diagram and verify that it is an acceptable replacement? The S8610U3009 PDF lists a bunch of boards Honeywell says it can replace. Pexsupply.com is listing that board for around $90 before tax/shipping.

 

[gar]

140310-2348 EDT

K8MHZ:

I would not think 150 ft would be a problem for you. Almost any wire would work that has some insulation and adequate mechanical strength. CAT-5, twinlead, enameled magnet wire, Belden 8723, and previously mentioned extension cord are possibilities. I believe you could use a 500 ft roll of something and roll off only what was needed. I don't think the inductance of the remaining coiled wire would present a problem.

A 1000 ft roll of Belden 8525 is about 100 mH at a Q of 3 at 1000 H. 100 mH at 60 Hz is about 40 ohms from a Shure Reactance Rule. Just on the bench with no large local current the overall, no filtering, induced voltage is about 1 mV. If you had induced voltage problems you could change the orientation of the coil. But you are looking for volts of difference on the neutral, and I don't think you will see any substantial error from an induced voltage relative to the 1 V level.

.

 

[PetrosA]

I don't think this is a simple neutral issue either. You start out with 236V and under load that drops to 224V. In this area 236V is only seen out in the boondocks and, if I recall correctly, is pretty close to PECOs minimum tariff voltage (I think their tariff voltage is 228V). Your POCO's tariff voltage may be different.

If it were a neutral issue, I would expect the loaded L voltage to drop in relation to N, and the unloaded L voltage to rise above by a pretty large percentage - at least 15-20%. If it's a combination of neutral and undersized capacity somewhere, then loaded L should drop, and unloaded L should rise, but L to L voltage would drop somewhat as well. In your case L1 and L2 are staying within 5-7% of each other which leads me to believe that your neutral is ok, but capacity is low. The fact that your phase-to-phase voltage is dropping indicates that you either have too small of a transformer, too small a triplex, or both. Let us know what they find.

 

[mbrooke]

6 awg ALU.

Ok. On the small side, but often the norm, as for the pig, roughly how far from your service drop to it. how far from the drop to the house?

How is it spliced to the pole run?


POCO connections can become intermittent anywhere too.


And all honesty if you really want to test for an open neutral or a neutral thats to small; disconnect you water bond, get some heaters, about 4 or 5 of them and load one leg down to about 48 or 60 amps, place a 60 watt incan on the other leg with a meter attached.

Keep it there for some time, looking to see if the voltage changes or the bulb begins to pulse in brightness or remains steady. Amp clamp on the heater circuit is a good idea.

If you have a separate meter you can compare the 240 volt voltage as well, which will help to a degree to determine voltage drop.

Anyway, do you have a water bond? Is the municipal water pipe metal? Are your neighbors well or city water?

 

[mbrooke]

I don't think this is a simple neutral issue either. You start out with 236V and under load that drops to 224V. In this area 236V is only seen out in the boondocks and, if I recall correctly, is pretty close to PECOs minimum tariff voltage (I think their tariff voltage is 228V). Your POCO's tariff voltage may be different.

If it were a neutral issue, I would expect the loaded L voltage to drop in relation to N, and the unloaded L voltage to rise above by a pretty large percentage - at least 15-20%. If it's a combination of neutral and undersized capacity somewhere, then loaded L should drop, and unloaded L should rise, but L to L voltage would drop somewhat as well. In your case L1 and L2 are staying within 5-7% of each other which leads me to believe that your neutral is ok, but capacity is low. The fact that your phase-to-phase voltage is dropping indicates that you either have too small of a transformer, too small a triplex, or both. Let us know what they find.

My thoughts exactly. If the line is older from some time ago this is not an uncommon problem at all. 10kva for a row of houses was the norm more than 40 years ago, as well as an open air #6 or even #8 cu LV run. Load demand for houses grow, and new ones might have been added also. As I said earlier, even for my poco when houses are all gas the pigs are sized at 25kva feeding only 4 or 5 houses. The national average for most pocos, 4 to 6 houses on a 25kva.

 

[hurk27]

When I look for a bad neutral problem the first thing I do is to check if there is a GEC bond to a city water pipe that can mask a weak or faulty neutral, also cable and shielded phone drops can also add to masking a bad neutral as they are bonded at both the pole and the house, after turning of the main, I lift these connections and make sure that no one is in a proximity of an electrical appliance and a Earth reference point as the grounding can become energized above earth if there is a bad neutral and you load down one leg of the service as there will be no return path back to the source, after lifting the above points, I turn off all the breakers in the panel and any sub-panels, I turn on only the breakers on one leg of the service or I use a separate load such as was mentioned a space heater of at least 1500 watts, then and only then I take my voltage readings being careful to not make a path from anything bonded to the service neutral and earth or other grounding, if I get more then a 10% voltage drop, or even a difference of 100% between the legs of the service, I check to see if it is also at the meter, then I call the POCO and set up a time to be there so I can also lift the masking return paths I mentioned above so that their beast can do its job without the cable, phone shield or water ground or any other return paths that could be masking the bad neutral.

Yes it's a few more steps but that's what we are paid for and we should understand how to safely load only one side of the service after lifting any possible masking return paths that could otherwise let a lost neutral go unfound.

Keep in mind that a RG6 or RG11 cable drop can handle quite a bit of current before you will see a voltage drop, I have seen RG6 catch fire because of a bad neutral, but that was on a very overloaded 120 volt only service that the POCO had already tested with the beast and said it was a good neutral, I wasn't involved and the previous electrician didn't know to lift the cable drop bond at the service before loading the service down with a 100 amp load bank, luckily he put out the burning cable but the homeowner freaked out on him and I was called in.

Just remember that if the neutral connection is lost that doing the above will also bring all grounding of this house or building up to 120 volts to earth, so I can't say enough how much care you must take to make sure no one is in harms way when doing these kinds of test.