LED bulb construction. Comparing types of bulbs.

Status
Not open for further replies.

11bgrunt

Pragmatist
Location
TEXAS
Occupation
Electric Utility Reliability Coordinator
I was helping with a complaint where the customer sees his lights dip when his 4 ton condensing unit starts and when the neighbor's 5 ton unit starts. POCO transformer serves both houses and is large enough to handle the LRA from either unit. 350AL from the transformer to a pedestal then 4/0 AL serves each house. All are relatively short conductor runs.
I preach that clear/unfrosted incandescent bulbs under a white ceiling are the worst for showing small voltage drops. When compared to CFLs, most CFLs were worse than incandescent bulbs at showing the perceivable flicker. Voltage drop at this main is less than three percent for the entire house during these condenser start ups.
I had been hearing that LED bulbs were being recommended by our folks to help hide the voltage drop so today I bought several samples and tested. The online reviews talk about energy use, color and longevity. No one talks about how they react to change in voltage.
I assumed :ashamed1: that LED bulb construction would be similar so I standardized on 60 watt, daylight. One brand was at the top of most reviewers' list and performed as poorly as CFLs with a three percent drop at 120 volts. There was one stick shaped LED light that performed very, very well. The voltage drop was changed to fall from 123 to 105 volts at the main. Not one person in the group could perceive the flicker with a 14% voltage drop. I would have bet money that could not happen. All other bulb types and brands did not come close to hiding the voltage drop as the stick shaped LED bulb.
Has anyone tested like this before? How can these LED bulbs be designed so differently? This stick bulb is probably too ugly for open air use so trying to suggest as the fix for all problems in all fixtures will be tough.
I plan to buy many other brands and types to continue testing for a Best Bulb.
Thanks.
 

Smart $

Esteemed Member
Location
Ohio
I think you'll find it's not the LED itself that has better Vd performance, but the associated electronics (aka driver). In most cases, the driver acts as a voltage regulator. How well it regulates will show on spikes and dips.
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
I think you'll find it's not the LED itself that has better Vd performance, but the associated electronics (aka driver). In most cases, the driver acts as a voltage regulator. How well it regulates will show on spikes and dips.
If the bulb is dimmable there is a good (50%?) chance that the driver is not constant voltage.

(Some dimmable drivers may only be phase sensitive.)
 

Smart $

Esteemed Member
Location
Ohio
If the bulb is dimmable there is a good (50%?) chance that the driver is not constant voltage.

(Some dimmable drivers may only be phase sensitive.)
Doesn't matter how it performs it's function. It boils down to how well it regulates the power transferred through the LED(s)... right?
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
Doesn't matter how it performs it's function. It boils down to how well it regulates the power transferred through the LED(s)... right?
All I am saying is that a dimmable driver will deliberately do a poor job of regulation. Even a worse than linear job if it is trying to match an incan's dimming curve.
And recognizing that some dimmable drivers (phase sensitive) may still have very good regulation when presented with sine waves of different voltages.
 

gar

Senior Member
Location
Ann Arbor, Michigan
Occupation
EE
150923-0636 EDT

Intensity vs voltage is driver dependent and a function of both input voltage and input waveform.

I ran some tests on different bulbs several years ago. The results are shown at my website --- http://beta-a2.com/EE-photos.html .
See photo P 14 in particular for a GE dimmable CFL. When fed by a sine wave the output intensity is moderately constant with change in input voltage, Compare this with photo P 17 where the input is a phase shift dimmer.

No photos, but I have recently tested a Cree LED and it dimmed quite well with both a variable voltage sine wave source (Variac) and a phase shift dimmer.

.
 

Smart $

Esteemed Member
Location
Ohio
All I am saying is that a dimmable driver will deliberately do a poor job of regulation. Even a worse than linear job if it is trying to match an incan's dimming curve.
And recognizing that some dimmable drivers (phase sensitive) may still have very good regulation when presented with sine waves of different voltages.
I'd expect non-dimmable to track best, with sideline dimmable tracking better than inline, sheerly on the nature of the driver design.
 

GoldDigger

Moderator
Staff member
Location
Placerville, CA, USA
Occupation
Retired PV System Designer
I'm curious to know how you came up with this percentage.
SWAG. Based on anecdotal reports from members.
Probably on re-analysis the number of drivers designed to work only by analyzing the phasing from inline dimmers is probably quite small except in very high end systems.
 
Status
Not open for further replies.
Top