The Product - LED Icicle Stars

Christmas 2005

Christmas 2006

Inventory

Who Made The Product?

Boxes

Inliten probably designed the product and had it manufactured overseas by Excellence Optoelectronics Inc.

Inliten licensed the NOMA brand and uses it on some of its products. Other products may be carry the house brand of prominent retailers.

Along with the NOMA brand name, Inliten got the rights to use the NOMA company identifier, which saved Inliten from having to register new barcodes (which costs money).

Mechanical Design

Lamp Assembly

Note the plastic tab in the lower left. This helps keep the lamp in place, and insures correct polarity.

(Actually, the clip doesn't insure correct polarity; you can push a lamp into the socket rotated 180 degrees, and it would make electrical contact. But the clip wouldn't fasten. So this feature makes incorrect polarity evident, not impossible.)

The exact angle at which the star is mounted on the LED lamp is not important.

This photo shows two lamps with the retaining/polarizing clip facing the same direction (toward the camera), but the stars are mounted at different angles.

But for the most part, the mounting of the stars on the LED lamp is quite uniform.

This is what the LED lamp assembly looks like out of its holder.

There are actually two versions of the lamp assembly. This one has a short plastic base. Another version has a long plastic stem sticking out of the base of the lamp holder. (I'll call these "short-base" and "long-stem" lamp assemblies.)

Here are samples of short-base and long-stem lamp assemblies.

The long stem is presumably to prevent the use of a LED lamp in a socket intended for common incandescent lights.

Note: a short-base lamp can fit in a long-stem socket, but a long-stem lamp can't fit in a short-base socket.

It seems that the first and last LED lamp in each series circuit of 35 lamps is short-base, and all the remaing lamps are long-stem.

This is a simplified drawing of how the LED lamp assembly breaks down.

Both short-base and long-stem lamp assemblies are polarized the same way: LED cathode towards the tab.

The overall light set contains exactly four sockets that have three leads sticking out. These sockets have two of the three leads connected together.

In this case, the two leads on the right are in parallel.

These special sockets are used to begin and end each series circuit of LEDs.

Here are the first two sockets in an average light set.

Note the first 3-lead socket and the 2-lead socket, instances of which make up most of the series circuit.

The whole light set is wired like this:

We actually took the time to count the number of LEDs in several sets, noting how many lamps were in each icicle "drop".

	cable layout
tracking number	connector	number of LEDs in each icicle	wire	number of LEDs in each icicle	connector	comments
1	plug	5,2,4,3,5,2,4,3,5,2	2-wire	4,3,5,2,4,3,5,2,4,3	socket
2	plug	5,2,4,3,5,2,4,3,5,2	2-wire	4,3,5,2,4,3,5,2,4,3	socket
3	plug	5,2,4,3,5,2,4,3,5,2	2-wire	4,3,5,2,4,3,5,2,4,3	socket
4	plug	5,2,4,3,5,2,4,3,5,2	2-wire	4,3,5,2,4,3,5,2,4,3	socket
5	plug	5,2,4,3,5,2,4,3,5,2	2-wire	4,3,5,2,4,3,5,2,4,3	socket
6	plug	5,2,4,3,5,2,4,3,5,2	2-wire	4,3,5,2,4,3,5,2,4,3	socket	one LED completely missing plastic star
7	plug		2-wire		socket
8	plug		2-wire		socket
9	plug	5,2,4,3,5,2,4,3,5,2	2-wire	4,3,5,2,4,3,5,2,4,3	socket
10	plug	5,2,4,3,5,2,4,3,4,2	2-wire	4,3,5,2,4,3,5,2,4,3	socket	one string contains 34 LEDs instead of 35

Electrical Design

Overall

• 120V
• 40 mA
• 60 Hz

The household line in the United states is sometimes called 110 VAC and sometimes 120. Here are some details:

location	nominal RMS voltage	peak voltage	peak-to-peak voltage
United States	120 V	170 V	340 V

With 35 LED lamps in a string, each lamp is subject to 4.86 volts, which is the peak voltage (170 V), divided by 35. This is both the forward and reverse voltage. Since I didn't find any rectifier diode, it would appear that the reverse current is limited to whatever leaks through the LEDs.

A 3.4 Volt forward voltage drop is a reasonable figure, since many blue LEDs drop 3.4 Volts and 430 nM bright blue types range between 3.5-3.8 Volts.

The rest of this section is reserved for actual measurements of LED electrical characteristics.

year		lamp base	forward voltage	reverse current
2005	good LED	short-base
2005	good LED	long-stem
2005	dim LED	long-stem
2005	dead LED	long-stem
2006	good LED	short-base
2006	good LED	long-stem
2006	dim LED	long-stem
2006	dead LED	long-stem

Notes:

Technical Information From Inliten

• There is current limiting in the string. As part of the string, there are there are "safety bulbs" which have a chip resistor inside the package. These are not visible and are part of a patented technology.
• LED lights have a patented socket design so that bulbs can not be inserted in the wrong direction (polarity).
• All of these lighting sets are hand assembled, it is possible that one might have slipped though missing a socket.
• LEDs are rated at 3.4V for 70-light LED light strings. 3.4V is an average voltage.
• Our LED lights with "Safety Bulbs" have a "Current Limited Design", so there is no concern about the AC peak power.

• Data on all components, especially LEDs.
  • Technical information on the LEDs: forward voltage, forward current, reverse voltage and current.
  • Source data on the LEDs: manufacturer and part numbers.
• Circuit information
  • schematic
  • information on power conditioning (rectification if any, and current limiting)
  • Information on "safety bulbs" with a chip resistor: which lamp assemblies contain the "safety bulbs"; how does this bulb work.
  • What features of this design allow it to accommodate manufacturing variations (e.g. a series circuit missing a lamp or two) and operation variations (e.g. a lamp burning out)?
• What is different between the lamp assemblies with the long stem on the base and those without the stem?
• Optical data on the LEDs.
• Technology information: a list of all patents used in this product.

It might seem that this list is rather long. That's becase we started out with a short list, and lack of answers or unsatisfactory answers led us to new questions. As an example, when we discovered a light set with one less lamp than there should be, we came to be concerned about what design features might protect against this situation.

How We Used The Product

Christmas 2005 Purchase

The lights were left on 24 x 7, through the Christmas season and beyond. We noticed that a few lamps went out, but didn't worry much about them (other than promptly replacing them with the spares that were included). Eventually, we ran out of spares and just let the lights run. Finally, turned off the lights and unplugged them. But we left them in place for the next year. We estimate that the lights were in continuous operation for four months.

No failures were noted.

Electrical Failures

Failure Modes

1. Reduced light output and color shift.
2. Flickering.
3. Lamp goes dark.

This process may take days or even weeks to complete. Which lamps in the string fail seems to be random - some never fail.

Frequency Of Failures

• indoor use only
• continuous operation, 24 hours a day, 7 days a week
• estimated run time 4 months

Of the five strings that failed:

	dead lamps
best string	13
worst string	27
average string	20.8

Could It Just Be A Bad Batch?

Is That Really A Big Problem?

• The color shift detracts from the beauty of the LED lights.
• The flickering worries me because solid state devices shouldn't really do that. It makes me think of heat problems and suggests safety issues. But I have no proof of what is going on.
• The obvious problem is that the LED lamps go dark.

LED lighting products sell for a premium price over the traditional incandescent products. This premium price is justified many ways on the box:

• energy savings - the 2006 box claims to save 76 percent on power
• long life
• stays cool

There are many ways to interpret such lifetime figures. To me, I take the consumer view that the entire string should last approximately 25,000 hours. That's 2.85 years of continuous operation, 24 hours a day, 7 days a week.

As the end of life approaches, it would be reasonable for a few lamps to burn out, but not so many that the consumer would think "It's ugly/broken/exhausted."

And that's the big problem - too many LED lamps have burned out too early. To me, the light string looks ugly/broken/exhausted after just a few months, not after the promised time span of years.

There's just no way that the claim of 25,000 hours is true.

And since the long lifetime is a substantial part of the reason for buying this product, the whole thing seems like a fraud.

Failure Analysis

We suspect failures on several levels.

Electrical Issues

We think that it's just a case of bad electrical design - or several design errors.

Excess Voltage

Interestingly, 3.4 Volts times the 35 LEDs in each series circuit is equal to 119 Volts, which is close to the nominal RMS voltage in the United States. But the peak voltage in the United States is 170 Volts (not counting the occasional surge).

So, at the peak of the A.C. cycle, each lamp is subject to 4.86 Volts - well over their 3.4 Volt rating.

In other words, each failure promotes more failures in a chain-reaction or avalanche.

This is probably why one of the tags says that burned-out bulbs should be replaced promptly to avoid danger of overheating.

Of course, we did replace burned-out lamps - until we ran out of spares!

But the concern here is that each failure that is not replaced will cause the next failure to happen sooner.

No Current Limiting

There might be something in the circuit to serve this purpose - perhaps in the special short-base lamps. But if current limiting is implemented, we haven't found it yet.

No Reverse Bias Protection

The lighting string appears to be just a bunch of Light Emitting Diodes in series, hung across the A.C. line. We see no rectifier diode that keeps the LEDs from being reverse-biased; the LEDs themselves are being used as rectifiers, but they leak when reverse-biased, causing a substantial reverse current flow.

There might be something in the circuit to serve this purpose - perhaps in the special short-base lamps. But if protection is implemented, we haven't found it yet.

Manufacturing Quality Issues

• Some LED lamps had no stars attached; or the stars were attatched so loosely that they easily fell off before the product was put into use.
• At least one of the lighting strings was built with the wrong number of LEDs in the string.

Business Procedures Issues

We have come to the conclusion that, in addition to the electrical design issues, Inliten has a problem with its business procedures. Inliten evidently did not test their completed product to insure that it met the claims on the package (e.g. 25,000 hour product life).

This is rather like selling a car with a miles-per-gallon claim that was calculated from the efficiency ratings of individual components. If the components were assembled properly, the calculation might well be in the ballpark. But, in order to be sure, you actually have to take the car for a drive!

Other Possible Issues

What Is This Web Site Doing About It?

• find out what really went wrong with this product
• verify that the product is safe to use
• replace/repair product that failed
• insure that the product will not fail in the future

action	status
Contacting Inliten	begun 12/09/2006
Contacting Wal-Mart	begun 01/14/2007
Putting Up This Page	done

Contacting Inliten

So we tracked down the e-mail address for Inliten, sent them an e-mail, and got no reply.

After waiting a while, we sent another e-mail, which got no reply.

By the time that we wrote the third e-mail, we were getting concerned that the strange flickering of some lights in the string might indicate something more serious than a light just going out.

Since Inliten has not replied to any of our inquiries, we have taken additional steps, as described below.

Summary of communication with Inliten:

date	we sent	response
12/09/2006	e-mail to service@inliten.com	none
12/14/2006	e-mail to service@inliten.com	none
12/21/2006	e-mail to service@inliten.com	none
01/14/2007	paper mail sent to Inliten	too early to tell

Putting Up This Page

Contacting Wal-Mart

What Will This Web Site Do Next?

But they have not provided technical information that we need to feel comfortable that the product is properly designed to be safe and long-lasting (as they advertise).

Our highest hope is that our attempts to contact Inliten will work out, and they will both replace the product and give us confidence that it is safe and won't fail again.

But, as time marches on, we do have numerous alternatives.

action	status
More Research And Web Site Expansion	pending result of e-mail exchange
Federal Trade Commission (FTC), Bureau Of Consumer Protection	pending Inliten response
Underwriters Laboratories Inc.	pending Inliten response
Consumer Product Safety Commission	pending Inliten response

More Research And Web Site Expansion

And we love to publish our results.

What's next?

Federal Trade Commission (FTC), Bureau Of Consumer Protection

Since Inliten advertises that its LED lights last for 25,000 hours, but but the product fails after a few months, their advertising claim is clearly false.

We might have to contact FTC soon.

Underwriters Laboratories Inc.

All of the holiday lighting products in question bear Underwriters Laboratories Inc. certification. They are all marked "DECORATIVE LIGHTING STRING", "FOR INDOOR OR OUTDOOR USE", and have a red UL holographic seal.

We might have to contact UL soon.

Consumer Product Safety Commission

We might have to contact CPSC soon.

How You Can Help

Investigation Time Line

Other Web Sites

Terry Ritter's "LED Christmas Lights and How to Fix Them"

Related Pages