irregular flourescent flicker or blink without load change on circuit

[archibald tuttle]

Think I've been over every connection and wire nut in this circuit. It is largely exposed in a basement so I can trace it fairly easily hits the finish basement from a wall shared with the unfinished side.
It had only 2 duplex outlets on a 20 amp minibreaker / about 50 feet of #12 wire with two junction boxes.

I jumped one of the outlets out of the wall in a wiremolding starter box and added a standard single pull switch and wire mold to an 8 bulb bath bar (it's not a bathroom just looking for a bunch of light). filled the bar with 13 watt (60 watt equivalent) flourescents. everything looked hunky-dory but noticed maybe ten minutes in a single quick blink off or flicker. And once I was paying attention saw that it was repeated but at no regular interval. could go 20 minutes without and then could blink a couple times in 5 minutes.

There is a computer battery back up plugged into one of the plugs and when the lights blinks it is tripped in.

I haven't challenged the circuit with a high amperage appliance. can do that tomorrow to see if it exacerbates matters or causes similar or more sustained symptoms. but do recall that the nature of flourescent loads (for reasons shrouded in mystery to this grasshopper) can challenge switches (and presumably connections) in ways that result in a different rating or recommendation for use of devices with flourescent loads so thought I'd post a question tonite for anything else to look for/at. and any explanation for a guy who lacks an oscilliscope of what phenomenon cause flourescents to test circuits more aggressively than their amperage might otherwise dictate.

thanks,

brian

 

[archibald tuttle]

solved my own problem but . . .

solved my own problem but . . .

it was bad contact of the breaker to the panel buss. fooled me because the other minibreaker in the same single breaker replacement did not exhibit symptoms so didn't check there first. only after I checked all the junctions. but i failed to recall, until I pulled it off to look, that often these breakers are made with two smaller grabs for the buss that individually feed the minis rather than both soldered or fastened to a single grab. this seems like silly engineering more or less designed to fail. not so often that I remembered, I admit, but as soon as I saw it I recalled seeing a similar failure 4 or 5 years ago and thinking at the time what a lousy design that was -- and unnecessary insofar as I can see.

brian

 

[gar]

130405-1133 EDT

archibald tuttle:

Is the "Minibreaker" a screw in fuse replacement? Invented and originally manufactured by Mechanical Products of Jackson MI.

I doubt your CFLs or other fluorescent lamps are the cause. If you have a DVM, such as a Fluke 27, with a min-max hold, then connect this to the circuit at the fuse or breaker panel and monitor the voltage. What happens to the min voltage when flicker occurs?

If you see a substantial dip, then connect a 15 or 25 W incandescent bulb at the same location as you had the DVM. Now put the DVM directly on the service wires, not just the lug, of the phase that has the flicker. When the incandescent flickers does the DVM also show a drop? If so then the problem is external to the main panel on the input side.

An assumption I am making is that there is a farily large voltage dip or I would not expect a CFL to flicker.

A test at my bench. The source impedance is such that a 10.5 A resistive load, a heater, when connected causes a voltage change from 122.7 to 117.8 V.

The flicker test is with a 1/3 HP unloaded mechanically, thus quick start to full speed, used to produce a momentary voltage drop. Three bulbs were used. A 100 W incandescent, a 27 W NuVision CFL, and a GE dimmable 26 W. An old Fluke 27 used for the min-max voltage test. The max reading is essentially the line voltage before the motor load is added.

100 W 122.8 drops to 119.1 see noticable flicker.
NuVision 122.9 drops to 119.3 see a little flicker, less than the incandescent.
Dimmable 122.9 drops to 119.4 hardly detectable flicker.

This is a signal known exactly type of experiment. Meaning I know when to expect the flicker. If I did not know when to expect a change it is unlikely I would notice either of the CFLs flicker.

.

 

[hurk27]

If I'm thinking right you are talking about a GE panel with half space breakers, I have had allot of problems with these panels but they do have an easy fix, since there is a full stab between each set of half space ones you can get the GE full space tandem breaker to replace any Minnie's that have a burned buss stab, the panel will have the same amount of breaker spaces and is listed for both styles.

I think it was one of the worst bus connection designs I have ever seen, there is so small of a contact connection on the buss when using these half space GE breakers that even 20 amp ones fail at the bus connection, they also make 2-pole ones for 240 volt loads, but what I never figured out is why they also have a full stab between each set of two half spaces? luckily they do and can be a fast fix for these panels, as they will take both 1" breakers and ?" but the ?" half space ones only have a little set of fingers to make contact and fail very easily, I think they only make up to a 30 amp half space breaker for these panels (at least I have never seen one larger) so for 40 amp and above you have to use a full space breaker.

We had a 30 unit strip mall that has these in each unit they were 200 amp 20/40 space panels, and we were asked to replace all the panels but they were so embedded it would have been a nightmare as they were located flush mount in block walls with conduits coming in both the top and bottom, but I was able to find the full space tandems that was listed in the label and told the owner I could just replace them with the full space tandems as in one of the stores that had allot of coolers (Liquor store) had a constant problem of these breakers failing, we found a listed GE full space tandem breaker listed on the panel label and was able to get enough to replace every one in all the units, the owner gave me a $2,000 cash bonus because I saved him quit a lot of money over changing out all these panels, also it was at a time we were swamped with work so I didn't really have the time or man power to do all these panels.

 

[hurk27]

130405-1133 EDT

archibald tuttle:

Is the "Minibreaker" a screw in fuse replacement? Invented and originally manufactured by Mechanical Products of Jackson MI.

I doubt your CFLs or other fluorescent lamps are the cause. If you have a DVM, such as a Fluke 27, with a min-max hold, then connect this to the circuit at the fuse or breaker panel and monitor the voltage. What happens to the min voltage when flicker occurs?

If you see a substantial dip, then connect a 15 or 25 W incandescent bulb at the same location as you had the DVM. Now put the DVM directly on the service wires, not just the lug, of the phase that has the flicker. When the incandescent flickers does the DVM also show a drop? If so then the problem is external to the main panel on the input side.

An assumption I am making is that there is a fairly large voltage dip or I would not expect a CFL to flicker.

A test at my bench. The source impedance is such that a 10.5 A resistive load, a heater, when connected causes a voltage change from 122.7 to 117.8 V.

The flicker test is with a 1/3 HP unloaded mechanically, thus quick start to full speed, used to produce a momentary voltage drop. Three bulbs were used. A 100 W incandescent, a 27 W NuVision CFL, and a GE dimmable 26 W. An old Fluke 27 used for the min-max voltage test. The max reading is essentially the line voltage before the motor load is added.

100 W 122.8 drops to 119.1 see noticeable flicker.
NuVision 122.9 drops to 119.3 see a little flicker, less than the incandescent.
Dimmable 122.9 drops to 119.4 hardly detectable flicker.

This is a signal known exactly type of experiment. Meaning I know when to expect the flicker. If I did not know when to expect a change it is unlikely I would notice either of the CFLs flicker.

.

Gar I have noticed that in using the Sylvania's CFL's the dimable ones will flicker but the non-dimable ones didn't over a large voltage range, mainly I used the Micro-Mini 13 and 23 watt (60 and 100 watt equivalent) it was a restaurant with many heavy cycling loads like A/C dishwashers with electric boost heaters along with other constant switching loads and a very maxed out 1200 amp service.

I was hired to try to lower their electric bills and allot of the usage was all the wall sconces and hanging lights that used 100 watt halogen T-3 lamps, they didn't want to change out the fixtures as they fit the decor they had to a T, I was able to install a medium base socket in each of the wall sconces and hanging lights and use the Sylvania Micro-Mini CFLs and they gave almost the same light out that the 100 watt halogens did, and using the 2700k ones you could not tell they were CFL's behind the frosted glass, for the can lights I used the 23 watt dimmable floods, but these we still notice the flicker but not as bad as the incandescent lamps, also changed many of the electric kitchen appliances to gas and had 3 250kbtu tankless type water heaters installed replacing a 250 gallon 350kbtu WH along with a few other changes we were able to get their electric bill down from over $12 to 15k a month to around 3 to 4k the biggest savings came by having makeup air installed in each set of hoods over the cooking area as before they relied on a 20ton HVAC unit that brought in outside air to make up the hood exhaust, this was causing a very negative pressure that was drawing in outside air into the condition space making the HVAC units for the table area to run constantly.

I guess that the Sylvania's uses a AC to DC conversion then a well regulated 25khz inverter to supply the voltage to the lamp which is why I think they have good voltage control, I also use these here at home, I put them in back in 2008 when I purchased the house and have yet to replace a single lamp, also my laser printer doesn't flicker my office lights like it does my desk lamp that has one of those high intensity 12 volt lamps in it.

Another thing is the Micro-Mini's have a smaller diameter glass tube in the coil that makes up the lamp so it takes allot less time for it to reach full brightness unlike some of the ones with a larger diameter tube.

Just wondering if you have ever tried the Sylvania Micro-Mini's? there about $9.00 for a package of 3 lamps at Menard's so they do cost a few buck's more but to me they are worth it.