DIY Home Improvement Forum banner

dimming lights from large geothermal heat pump

16K views 65 replies 14 participants last post by  atuel 
#1 ·
Having some light dimming for a second or two (60% of full brightness) when a large geothermal heat pump kicks in (70A max in rush over 4 Cu cable).

I had the utilities monitor my voltage over the past several days and they see drops in voltage of up to 3-5V for a second or two and tell me its with in their legal requirements (even if it is just barely). No phase imbalance detected. Doesnt appear to be loose connections, burns, or over heated breakers either. Though was standing there as it kicked in and may have heard a pop.

Its a 400A service on 4/0 AL cable coming into a single meter and then split from there into two 200A panels with direct connection to the meter with 4/0 AL hots but a smaller gauge neutral cable... maybe 2/0 AL but cant get a good look at it.


So sort of running out of things to check here... I'm wondering there could be an issue with a split 2 panel setup causing issues over a single 400A panel.

Could the smaller neutral be causing some resistance to large in rush currents?

There are a couple visible nails that came in through the exterior siding that dented or puched out one of the knockouts. Nothing visible close to any of the cables but havent pulled the breakers to look underneath.
 
See less See more
#2 · (Edited)
Usually any motor will briefly dim lights due to the inrush current to bring motor up to speed. My 1 HP sump pump dims mine for a split second. It sounds like your heat pump has a pretty good sized motor. Sounds right if only dimming lights for a second. Did u say u have 4/0 AL feeding meter? If so that is only good for 200A unless they paralled two 4/0's on each leg.
 
#4 ·
All I know for sure is its 4/0 AL coming into my panels with a 2/0 neutral. I wasnt actually there when they opened up the meter box to install the monitor, so I'm just going off what they were telling me that its a 400A service with just my neighbor and I served off the one platform transformer. I thought they said its 4/0 feeding the meter, but I'll have to double check with them.

Should the neutral be undersized like this?
 
#3 ·
The voltage drop could have occurred all the way back inside the pole transformer.

If you measure the voltage at various locations such as at the receptacle or junction box where the pump is wired in, at the breaker terminals for the pump circuit, also at the big lugs for the service coming in, you can get an idea of whether the drop is happening in the branch circuit wiring or in the service wires.

You probably won't be able to measure voltage at the meter box or further upstream.
 
#6 ·
If this wire belongs to the poco, then it may be sized right.
If you had to supply the wire to the meter, than that wire is undersized for 400 amps.
The neutral is allowed to be undersized based on load calculations.
 
#8 ·
Just called the utilities (medium sized town co-op) back and they confirmed its 4/0 triplex direct burial cable not paralleled (with a 2/0 neutral) from transformer to a platform that services both my neighbor and I. Neighbor has 1/0 service lines and I have 4/0 with 2/0 neutral. Its all underground including the feed to the transformer. No above ground cables.

The utility engineer is indicating the 4/0 cable is rated at 386A. Does that sound right? Found some manufacturers that rate at over 300, but not that high and thats at 90C.


So correct me if I'm wrong here, but seems my neighbor and I are sharing and even at 386A it would seem easy to get close to maxing that out. They measured I was drawing 20-30A just with my normal tv, lights, etc non-major appliances running. I could easily run washer/dryer, water heater, stove, and dishwasher as I often do and easily be eating up 100-150A before my geothermal kicks on. The inrush would put that up over 200A fairly easily for an instant. And thats before we factor in my neighbor is sharing off the same 4/0 feed to the platform. Seems undersized to me if I keep any sort of concurrency to my activities when its cold outside.
 
#27 ·
Just called the utilities (medium sized town co-op) back and they confirmed its 4/0 triplex direct burial cable not paralleled (with a 2/0 neutral) from transformer to a platform that services both my neighbor and I. Neighbor has 1/0 service lines and I have 4/0 with 2/0 neutral. Its all underground including the feed to the transformer. No above ground cables.
Has your neighbor said anything about their lights dimming?
 
#11 ·
Yes, but it just seems unlikely that NEC would place a nearly 2X overhead on de-rating the 4/0 AL cable. A 386A rating for utility runs seems high when NEC (even if it doesnt apply) would only rate it at around 200A.

So really, trying to find out if 4/0 is enough to supply (2) 200A panels where they are actually in use vs just needing extra breaker capacity for seldom used circuits.

And if so, is it enough to feed the dual 200A panels and my neighbor?
 
#14 ·
Going dim to 60% brightness for a couple seconds isnt normal I dont believe. I'm also having a higher failure of equipment (DLP bulb and light engine, wireless router, cable modem all going at different times) now than before. I dont believe any of this is normal. Never had anything like this at other houses.
 
#16 ·
I got the poco putting the monitor back on later this week and leaving it on so we can demonstrate the problem. They only monitored for a few days where it was 60 degrees out and we didnt even have a chance to run the heat pump but maybe a little bit in low during the night when nothing else was on.

Now if I could only find a decent electrician around here... really wish I could just drop a chunk of change and have it fixed, but lately, the contractors I get out here seem to just be a waste of money. Half of me just wants to spend a little more to get a meter and debug it myself in my non existant free time rather than waste time/money on another crappy contractor.
 
#18 ·
What would you like fixed, specifically?
What if there is nothing to be fixed? What meter would you propose to use to "debug" this yourself?

An electrician cannot do anything for you, unless the problem is in the house or with the service. This house seems very new so I doubt there is very much wrong there.

If the POCO said there is only a 3-5V drop then that's that. Did the heat pump come on that night? If it did then that is a reading point for them. Doesn't matter if it was warm or not. If it came on it came on.

Where are you getting this ambiguous 60% number from? I HIGHLY doubt anything is dimming 40%. :whistling2:

I don't want to sound rude, but you are not going to stop asking until you get the answer you want to hear. This is a common phenomenon on forums like this. There is a very real chance that answer is not coming.
I understand that you are upset about something, but sometimes that something is not anything.
 
#19 ·
That might be a bit drastic at this point. So far you have told me that NEC doesnt apply and its expected/live with it. And now an extreme to the other end with dedicated everything. This is not really helping me.

I'm getting monitoring put in to help diagnose... poco shouldnt have pulled it so soon last time and should have configured it to get the data we find ourselves asking for now. So right now I'm leaning to the supply being an issue, but I could be wrong as its really just barely an indication at this point.

I'd really like to hear from people that have "stretched" the limits of 4/0 to 400A and either had or not had issues. Or others that always do 4/0 paralleled or use larger cable and never had an issue. Or maybe others that have used split 200A panels like I have instead of a single 400A and gotten burned on it. Really, any related stories/situations with heavy loads like this or similar issues.
 
#21 ·
A single 400A panel in a residence is an absolute rarity. Your dual 200A panels is the norm.

Bottom line is what is the voltage drop reading? Regardless of what size wire and what panels you have.

If they are coming out again this will confirm or deny the last reading. I'd wait and see what that shows.
 
#24 · (Edited)
Thank you... that is helpful to know 400A panels are not normal.

Well this is different than what you said before.



Why not crank the t-stat to 80 deg for 10 or 15 minutes to be SURE it gets into high?

Well, two reasons...

I was trying not to skew the original results. Poco originally said they'd keep it on there a week and would call before coming out to fetch it, so I was waiting for the colder weather that we're now back in. Unfortunately, they came out and got it only after a couple days and did not call so I couldnt tell them it had been off for the several days of high temperature weather. Poco also forgot to configure their monitor to record current along with the voltage, so still left speculating on what I'm actually drawing. They assure me they will configure it this time. They also assure me they'll leave it on there for a week this time.

Other reason was its not quite as easy as turning up the tstat as I have this too damn smart for its own good zone controller with multiple tstats non here that trys to prevent it from ever getting into high. It likes to play games with enabling one zone and seeing how it affects the other zones and running the fan only to circulate air around from other sources (stove, dryer, etc). Under the right continued failures to maintain temperature, it'll give up with the games for a while and kick it in high right away for a while before it goes back to playing games again.

So I was being a little ethical and a little lazy. I can try to force it and probably will at some point this time around. Maybe just leave the windows open so it cant maintain temperature.
 
#25 ·
Ateul, the power company uses a different set of rules for sizing cables up to your meter. From that point into the house, sizing is based on the NEC residential code, and it sounds like at least your wire sizes are fine. I doubt the problem is on the neutral, since most heat pumps run on 240V as has already been indicated.

If you're really getting a 60% drop in your lights, and are getting equipment failure, but your POCO has only measured a voltage drop of a few volts back at the transformer, then you could have a problem anywhere from that point to the loadcenter that feeds the heatpump, but the problem might also be with the POCO.

You might have a licenced electrician who knows how to use a scope or a true RMS multimeter with a max/min capture feature try the following:
During heat pump startup, measure max/min voltage from line to line, and each line to neutral at each of the following points:
1) At the heat pump (the heatpump end of the #4 Cu cable).
2) At the heat pump circuit breaker (the loadcenter end of the the #4 Cu cable).
3) At another 240V circuit breaker in the same loadcenter (such as your dryer or stove. Make sure it's not running when you're measuring).
4) At a 240V circuit breaker in your neighbor's loadcenter.
The voltage drop should be the most for #1, and least for #3. And if #3 is more than a few volts sag for a second or two, then your problem is probably upstream of the loadcenter. It could be as simple as the loadcenter wires haven't been properly torqued into their lugs-- which is potentially a very dangerous situation. But my guess is that it is a POCO problem.
 
#26 ·
We didnt capture a significant drop in the poco monitoring the first time since it was a bit of a warm stream where we didnt have the heat pump on. I've currently got the poco monitoring it for a week (ending tues or wed next week). They are actually monitoring it at the meter vs out at the transformer. They've got a box made by PMI duct taped to the top of my meter box with cables going inside the meter box. Should they be monitoring out at the transformer instead?
 
#35 ·
So got a WTF question here...

I just went out to double check the tightness on the breaker terminations at the breaker at least. Saw my multimeter while I was grabbing a screw driver and through what the heck... i'll just check the voltage while I'm looking.

So checked the voltage and it read about what it should... I cant really vouch for the accuracy of my multi meter (its a craftsman model 82026 digital multi meter circca the late 90s). It was reading about 123-124 on phase to ground at both panels (makes me wonder if that is accurate a drop to 116-117 that the poco measured is actually an issue). I forgot what the exact numbers for the phase to phase, but I recall thinking nothing hugely out of the ordinary there.

Then I checked the terminations of the 4 awg cable for the heat pump at the load center. They were loose and a few strands of the THHN at the back of the neutral were not actually in the neutral bar. It was very hard to see... almost missed it except I dropped my screwdriver and happened to look up at a very sharp angle at it from below. My load center neutral holes are pretty much exactly 4 awg with no room to spare so guessing it was forced in and bent those back. They were bent perpendicular to the cable up against the side of the neutral bar. I re-terminated this including cutting the fowled end off and staring a new.

Also torqued the connections at both the neutral bar (which was actually pretty tight to begin with) and at the breaker. The breaker connections seemed loose though. I sort of noticed that ever time I torqued down the cable, the stranded THHN distorted a little bit and each time the cable was "wiggled", it got a little loose. The 70A breaker is supposed to handle 2 awg directly so the "clamp" so it was crushing the stranded THHN flat and any movement of the cable lets the strands move and then it needs to be tightened again.

So now the WTF moment. After doing all that and with the 70A breaker still off, I checked voltage again. It was doubled from my previous readings. Phase to Phase to phase was over 400 and phase to neutral was over 200 in both panels. My house isnt burning down here, so not sure what just happened. Even pulled the entire breaker for the heat pump out of the load center to break any circuit there, but still reads double what it should. So did my multimeter just die or is there something that would cause this?
 
#38 ·
I went and checked it about 30 minutes ago and it had dropped back down, though seemed like it was trending to increase a few volts every minute. It started at 123 or so and went up slowly to 129 in the space of a few minutes before I quit looking at it.

However, the meter has a low battery indicator on it. I just turned it on to check again and the battery indicator came on came on a minute later.

Changed out the battery and just checked again... With the same whatever load is running (at a steady state... no heat pump coming on), its just under 124V for each P2N and 246V P2P. Both are holding steady.

How accurate are these multimeters on voltage? Its a digital model that reads in tenths of a volt for up to 200V and in full volts up to 500V. At a glance, its not exactly a cheap walmart one, but not sure how good it is either. Are even the fairly cheap ones pretty accurate on reading voltage?

Basically, I'm wondering how much we can compare these readings to the first set the poco took with the PMI recording meter. They said lowest drop they saw was 116V in the first measurement they did where the heat pump wasnt running but in our conversations we assumed normal current was 120V. I didnt ask him what the voltage was before the drop occurred.
 
#39 ·
In all likelyhood the poco is using a true RMS DVM and yours is not. That could explain the 10V difference. But not the 200V difrerence.

Low batteries causing a 2x swing in the AC measurement seems really unlikely. You might try putting the old batteries back in and see if you get 200V line to neutral. If not, then I'd be inclined to believe that 200V reading. If the poco is really giving you that much jump in your voltage, no wonder you've been burning thru appliances like cheap candles!

Do you really have a #4 coming back from the heat pump to neutral??? Most 220V heat pumps (and 220V appliances in general) don't have a neutral, just L1, L2 and ground. And if they do, it's frequently one wire size down.

If you've got three #4's coming back from the heatpump, could it be that one of the lines & neutral have been reversed? As long as neutral and the ground aren't connected at the heatpump (and they shouldn't be) you wouldn't get a short circuit, but it could seriously mess up the heatpump, and maybe explain why it's bringing down the rest of the house on startup!

Given all your trials, maybe you are just unlucky, but I doubt it. I feel for you!

Good luck, and be safe!!
 
#40 ·
The "neutral" I referred to is actually the ground. I misspoke. There are 3 THHN lines in a water tight conduit.. all are black insulated though one is marked with a piece of elec. tape. I confused myself since its not the typical bare ground when I was thinking before. Probable also means the re-termination I did will only help in the case the heat pump shorts out and needs the ground.

The only 220V with neutral I have is my new dryer upstairs that is a 10-3 with ground and 4 prong outlet. Old dryer outlet is the 3 prong setup.

Anyway, I'll keep monitoring here every so often and see what happens. Hopefully poco caught that as well since the monitor they're using it out there now.

I assume the only way to get that much voltage would be from a bad transformer? Or maybe somehow pushing too many phases down the wire somehow?

Found the manual for my multimeter... doesnt say true RMS anywhere, but has a specifications page saying voltage measurement has an accuracy of 1.2% with a 10 digit accuracy of the analog to digital converter.


BTW, I've been considering putting a full branch monitoring setup in place sort of like I have in my data centers so I can go back and review when my bill comes up too high or we have lame problems like detailed in this thread. Maybe even have it send me email to tell me to turn the damn lights off or something. I have not found one with the features I'm looking for at a price that I'm willing to pay for it yet though. The branch monitoring solutions are getting better and cheaper with the push for green everything.

In the mean time, I'd consider purchasing a better multimeter that would measure true values of current, power, and voltage. Maybe one with a recording history or at minimum a min/max/average over time and an ability to monitor both legs at once. I've been lightly looking for a while now, but if anyone has a suggestion on a certain meter, I'd appreciate it.
 
#41 ·
i read someone else suggest this and you blew it off, but i feel this is very important. Find out if your neighbor (that is being fed off the same transformer) is having the same lighting issues. If so, the problem may be too small of transformer.

i think you finally got over this part, but the nec does not apply to the power company. They ALWAYS use much smaller wires. I think on my house i have 4/0 aluminum on my side being fed by 1/0 on their side. Free air conductors just don't get as hot. There is very little air circulation inside a conduit and tons of air flow when its free air.

As others have said, there may be no "issue" to fix. The pump in question just might take ridiculous amounts of power to get moving. Im no pump or motor expert, but maybe their is something that can be done to lessen the current draw. Some sort of soft start.
 
#43 ·
Yeah, I know the smaller than NEC for the service lines is a common practice now for the poco.... however, it bring up a question I had on the service line being in the ground vs in the air. All my subdivisions's lines are burried. Nearest line in the air is a couple miles away along the state highway.

So the question is... is being in the ground better or worse than in the air in regard to heat dissipation. Manufactures of this 4/0 triplex seem to rate the direct burial higher than the air.
 
#42 ·
I swapped in the old battery... its not ready the double the expected that I had caught. Its about the same as the reading I did before I changed the battery. Its a little higher than the new battery at pretty much 129/130 P2N and 260 P2P.

Starting to wonder if that 3-400V P2P stuff I was getting was real or an odd symptom of the multimeter battery getting low (though the battery low light was not on when I got that high reading and it was when I put it back in just now).

Short of the fact I've been staging my own effort to restore the ecomony with all the new **** I'm buying, are there other symptoms of significant over voltage that I might be able to match up?
 
#44 ·
Well let me chime in on this one. Why not fix the problem at the source?? Install capacitors or VFD at the motor! ALSO remember that the incoming feeders are not sized by the size of panel but the load calc, and by table 310.15(B)(6). which you seem to be with in, not with standing that POCO is not covered by NEC. Larger cables might help the problem, but will not fix it- just install starting help for the motor. Cheers
 
#45 ·
Thats certainly an option... supposedly the heat pump already had one internal to it. I was reading up on them and there were somewhat mixed reviews on them with many examples of people installing them only without the issue being resolved. Anyone have any luck with these?

I've also considered using some power regenerators on my more sensitive equipment.
 
#46 ·
So found something that I'm not sure if it could be related or not.

I was in the crawl space to check things out and make sure everything was tight on the heat pump end.

While I was down there, I noticed an old utility light was busted (socket broken out of the white plastic base) and there was an old 24V transformer that used to feed the old doorbell and zone controller. I disconnected these quick and was going to put a cover on the box when I noticed it didnt seem to be fed from the load center that was just a few feet away.

The run actually goes from loadcenter at the back of the house all the way to the front of the house and then back again pick up the jct box I was at and then another bit to outlets in my sunroom.

Its all done in 12awg NM on a 20A circuit with 15A outlets (5 at front of the house and 4 at back)... total length is a good bit over 100'. Correct me if I'm wrong here, but isnt 12awg not rated for that?
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top