lights dimming

Status
Not open for further replies.
160906-2120 EDT

mgookin:

What I was trying to do with the above data using a Fluke 27 or 87 was to give an illustration of what such an instrument could do.

In the past, in particular 2011, I collected a lot of data at my home on voltage and total power consumption vs time at a resolution of 1 second. This produced a couple megabytes of data per day. On plots I made averaged over one hour periods, for a 4 month period, and plotted over a one day period the voltage remained mostly within 123 to 125 V with no time of day variation. See photo P-28 at http://beta-a2.com/EE-photos.html . The voltage plots over the next 3 months were a tighter band, and more straight. This plot, P 28, did show a small time of day variation.

Since my last reading the results are:

Date & Time ----- MAX ----- MIN ----- Instantaneous value at end of time period

160906-1145 --- 126.9 --- 123.2 --- 123.7 ---- present temperature 80 F.

160906-2143 --- 124.5 --- 121.0 --- 122.7

Ambient temperature and how it varies thru a day may show up in a particular day as correlating with the time of day, but over many days this seems to level out.

.
 
gar said:
160906-2120 EDT

mgookin:

What I was trying to do with the above data using a Fluke 27 or 87 was to give an illustration of what such an instrument could do.

In the past, in particular 2011, I collected a lot of data at my home on voltage and total power consumption vs time at a resolution of 1 second. This produced a couple megabytes of data per day. On plots I made averaged over one hour periods, for a 4 month period, and plotted over a one day period the voltage remained mostly within 123 to 125 V with no time of day variation. See photo P-28 at http://beta-a2.com/EE-photos.html . The voltage plots over the next 3 months were a tighter band, and more straight. This plot, P 28, did show a small time of day variation.

Since my last reading the results are:

Date & Time ----- MAX ----- MIN ----- Instantaneous value at end of time period

160906-1145 --- 126.9 --- 123.2 --- 123.7 ---- present temperature 80 F.

160906-2143 --- 124.5 --- 121.0 --- 122.7

Ambient temperature and how it varies thru a day may show up in a particular day as correlating with the time of day, but over many days this seems to level out.

.
Click to expand...

Interesting. We don't get spikes like that. Around here it's more gradual and stepped.
 
160906-2154 EDT

Last night I went to the shop and ran tests on a Quincy two stage 5 HP air compressor.
The source is a 3 phase delta with a wild leg. The 240/120 transformer is 100 or 167 kVA, and the wild leg 50 or 75 kVA. Various size wires are used from the transformers to the sub-panel where I made measurements. Probably nothing smaller than 00 copper. The cable length is probably 300 to 400 ft. My measurements were on the 240/120 phase which is the most stiff.

The measurements are startup current and voltage drop. The first measurements were at zero back pressure, and the second set with 75 PSI in the tank.

Anytjoing after the first two plots is missing after I post.

0 PSI tank pressure:

Fluke 87 MIN/MAX response time at 100 mS showed a voltage change from 244.8 to 237.6 during startup. A voltage drop of 7.2 V.

Current was measured via a Fluke Y8100 with DC coupling on the 200 A = 2 V range.

The current plot of one leg is:


View attachment 15620
.
The start time is 1 major division from the left. QO breaker contacts first close, then bounce for about 10 mS and then stay closed. Time to steady state current about 400 mS. Peak current about 175 A and converted to RMS about 124 A.

The corresponding voltage plot is:
View attachment 15621

.
A 10X probe was used on 1/2 of the single phase transformer.

The following plots are missing on posting.


Next the same two plots with 75 PSI in the tank.
DS2_QuickPrint10-I.JPG

.
Note the duration is slightly longer and steady state current slightly greater.


DS2_QuickPrint11-I.JPG

.
 
Last edited:
Here are the missing plots:

Anytjoing after the first two plots is missing after I post.

0 PSI tank pressure:

Fluke 87 MIN/MAX response time at 100 mS showed a voltage change from 244.8 to 237.6 during startup. A voltage drop of 7.2 V.

Current was measured via a Fluke Y8100 with DC coupling on the 200 A = 2 V range.

The current plot of one leg is:


View attachment 15620
.
The start time is 1 major division from the left. QO breaker contacts first close, then bounce for about 10 mS and then stay closed. Time to steady state current about 400 mS. Peak current about 175 A and converted to RMS about 124 A.

The corresponding voltage plot is:
View attachment 15621

.
A 10X probe was used on 1/2 of the single phase transformer.

The following plots are missing on posting.


Next the same two plots with 75 PSI in the tank.
View attachment 15623

.
Note the duration is slightly longer and steady state current slightly greater.


View attachment 15624

.Now they may show up.

.
 
160906-2355 EDT

I give up on pictures.

The point of comments I have made in my previous posts is that if you understand how various devices work, how to make measurements, how electrical circuits work and are analyzed, then you can use some intelligent procedures to troubleshoot and solve problems. Just guessing and doing trial and error should only be used as a last resort.

JDB3 has done a good job so far with the tools and information he has been given. I am quite suspect of the homeowner being a good source of information other than he thinks he has a problem. The homeowner has to define what he means by dimming, what is its duration, and how is it correlated with other events. Many that have responded to this thread are making suggestions of what might cause problems, but not how to determine if their suggestion is the cause. By using appropriate loading and measurements along with general a priori information one can probably converge on the cause quicker than just using trial and error methods.

.
 
160907-1204 EDT

gadfly56:

There were a total of 4 plots. Attachments 15620 and 21 got lost. They were present while working on the post and visible in preview. The two lost ones were at 0 PSI tank pressure and only slightly different than the visible ones at 75 PSI.

.
 
gar said:
160906-2302 EDT

I don'y know the problem, but this wesite has problems with pictures.

.
Click to expand...
This site has always been horrible for uploading pictures to post, for one thing they have limitations not only on individual image file sizes but also on how much storage there is for each user.

What works the best is to use a host site such as photobucket to store your image and then link to that image with
 
160907-2230 EDT

kwired:

I may have to go that route.


I had a problem sometime yesterday, overnight, or this morning. I believe it had to be overnight, and if so the Fluke 27 did not detect a severe overvoltage on the supply to my home. There is nothing in my home that could cause the damage.

Two different transient limiters fried. One was an APC unit at one end of the house, and the other was a Belkin at the other end. These were on different sub-panels, and may or may not have been on the same phase. Phase was probably unimportant.

Both units had burned MOVs and blown internal fuses. The APC used screws to hold it together and was not too hard to get apart. The Belkin is a snap together unit and I had to use a bandsaw and big chain cutters to get it apart. Not good for failure analysis, but low cost assembly and it keeps the customer from opening the box.

To cause the failure there had to be a substantial voltage, probably at least 400 V, with a good source of energy.

I have not detected any unprotected equipment having been damaged.

I believe the spike occurred while the Fluke 27 was monitoring. Note the Fluke 27 is not designed to detect short spikes.

Both transient limiters did their job so I have no complaint. It is not worth the trouble to send them to the manufacturer.

I won't buy another Belkin. It is too hard to get apart.

.
 
Something changed...your circuit resistance

Something changed...your circuit resistance

It doesn't have anything to do with the load in the entire house, just those 2 circuits (if not the same)simple test...check your line (hot ) to neutral and to ground. If you have 120 on the ground, but not the neutral, ( with the load on)you have a voltage drop because of a loose connection or a break. If you're losing voltage to both neutral and ground, you have a loose or broken hot wire. Breakers tight? Neutrals tight on your panel?
 
Guess I can't give you troubleshooting methods on here, but to me it sounds like if the two rooms are the only ones experiencing the dimming lights, and they are close together in the home, they may be sharing a neutral if they're on different phases. Sound like a loose neutral to me. Either in a junction box or even on the neutral bar in the panel. Common effects from the sound of it. ALSO remember that ita the neutral, with no breaker protecting it. It can melt without tripping your breakers...find it. Keep the switches off and hunt that fugger down.
 
If all that has happened since the dulling livhts, I would make certain that the sevice is bonded to ground. If you're experienced blown fuses in devices, that could be from low voltage drawing higher amps also. Sounds to me like a neutral or ground at your service at this point. You should measure voltage with the load off, then on. If voltage drops going to neutral but not to ground, there you have it
 
Status
Not open for further replies.
Top