LED Street Lights Unvieled In Silicon Valley

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
Bigglez
Posts: 1282
Joined: Mon Oct 15, 2007 7:39 pm
Contact:

LED Street Lights Unvieled In Silicon Valley

Post by Bigglez »

A recent article in the Merky News caught my attention this week.
San Jose, California plans to replace many of its aging 62,000 yellow
sodium vapour street lights with 20,000 white LED versions at a cost
of about $1,000 each. Hopefully the new ones will reduce maintenance,
lower operating costs from the current $4million annual street lighting
bill, and improve visibility compared to existing yellow street lighting.

A few street lamps are ready to go in a pilot program. More info here.

According to an LED manufacturer there are about 55 million
street lights in the USA.
jimandy
Posts: 572
Joined: Sat Dec 04, 2004 1:01 am
Location: Birmingham AL USA
Contact:

Re: LED Street Lights Unvieled In Silicon Valley

Post by jimandy »

Several years ago Birmingham replaced incandescent bulbs in all traffic lights with LEDs. Now I notice a number of the LEDS are dark. I suppose the fix, if/when necessary, is to replace the entire array rather than individual LEDs
"if it's not another it's one thing."
User avatar
haklesup
Posts: 3136
Joined: Thu Aug 01, 2002 1:01 am
Location: San Jose CA
Contact:

Re: LED Street Lights Unvieled In Silicon Valley

Post by haklesup »

That's interesting. I had heard that they origianlly chose the sodium lighting over other choices available at the time to reduce light pollution for the Lick astronomical observatory atop Mt Hamelton. I wonder if that was considered in this choice, if that was forgotten or if it's just not important anymore.

http://www.ucolick.org/public/visitors.html

http://en.wikipedia.org/wiki/Lick_Observatory
See the paragraph starting with "Current State" for a citation
Bigglez
Posts: 1282
Joined: Mon Oct 15, 2007 7:39 pm
Contact:

Re: LED Street Lights Unvieled In Silicon Valley

Post by Bigglez »

jimandy wrote: Several years ago Birmingham replaced incandescent bulbs in all traffic lights with LEDs.
Locally this was also done starting in about 1997. The amber signals
were replaced later, but red and green were done first. New installations,
starting in the late 1990s were all LEDs, and the program was sponsored
by the local utility company who didn't need to invest in as much
"outside plant" with the much lower LED consumption.
jimandy wrote: Now I notice a number of the LEDS are dark. I suppose the fix, if/when necessary, is to replace the entire array rather than individual LEDs
Correct. The original housings are retro fitted with LED modules
that include their own power supply. Various techniques are used
to derive the DC LED power supply current, which includes temp
compensation for LED traffic signals in very cold or very hot (or both)
climates.

Over time the BOM for the controller has been refined (cost reduced)
and some LED clusters are now defective. The main cause appears to
be the growth of tin dendrites on the PCBs and results in groups of LED
failures. The original clusters contain about 100 LED diodes on a PCB.

Defective LED clusters are for sale in our local hobby parts emporium
(halted.com).

The filament bulbs were specially produced to last an average of 8000
hours, and were routinely replaced in rotation (red and green was anual
and amber bi-annual). 8000 hours is about one year and one month,
give or take, and don't forget the all lights are never on at the same
time, so annual replacement scheduless are conservative.

The replacement LEDs were expected to last ten years, so failures are
likely to escalate any year now...
Bigglez
Posts: 1282
Joined: Mon Oct 15, 2007 7:39 pm
Contact:

Re: LED Street Lights Unvieled In Silicon Valley

Post by Bigglez »

haklesup wrote:I had heard that they origianlly chose the sodium lighting over other choices available at the time to reduce light pollution for the Lick astronomical observatory atop Mt Hamelton.
Correct and as noted in the article I linked. Also
noted is that new LED street lighting can be dimmed by wireless
command, so light from the city is lowered when the observatory is
in use.
Jim Barrett
Posts: 31
Joined: Fri Sep 12, 2008 3:08 am
Location: Milwaukee WI
Contact:

Re: LED Street Lights Unvieled In Silicon Valley

Post by Jim Barrett »

One of the problems we have here in Wisconsin is lack of heat from the incandescent bulbs.
Under certain (not uncommon) conditions the snow will drift onto the lights and not enough heat is generated by the bulbs to melt them clear. I sat at an intersection & wondered if they would end up putting heaters on the lenses to solve the problem & what that would do to the energy savings realized by switching to LED. :???:
Dean Huster
Posts: 1263
Joined: Wed Dec 05, 2001 1:01 am
Location: Harviell, MO (Poplar Bluff area)
Contact:

Re: LED Street Lights Unvieled In Silicon Valley

Post by Dean Huster »

I had heard that the sodium versions were chosen over the mercury versions because the yellowish color provided better illumination in fog.

Dean
Dean Huster, Electronics Curmudgeon
Contributing Editor emeritus, "Q & A", of the former "Poptronics" magazine (formerly "Popular Electronics" and "Electronics Now" magazines).

R.I.P.
User avatar
MrAl
Posts: 3862
Joined: Fri Jan 11, 2002 1:01 am
Location: NewJersey
Contact:

Re: LED Street Lights Unvieled In Silicon Valley

Post by MrAl »

Hi,


I've heard a lot of different things about LEDs for street lighting.
There are numerous opinions and things that could go wrong that
were not even considered for a fraction of a second when a change
like this is made.
There are some locations with lights now, but their climate conditions
might be different from that where the new install is to be located.
Other conditions could change from location to location also which would
null any comparison between the two.

The smart money goes into a prototype system first, which is a
much smaller scaled down version of the intended final project.
For example, if 500 lights will be replaced then replace only 50 of them
and TRY this out for a few years BEFORE committing to a full install.
The experience learned could make a huge difference with the remaining
450 lights. Yeah, it takes a little longer to get the full system up and
going but there is more certainty that the final system will work as
expected once it's up and running.

Changing the entire technology is certainly not the same as changing a light bulb
and thus it has to be handled differently.
LEDs vs Bulbs, LEDs are winning.
User avatar
dacflyer
Posts: 4748
Joined: Fri Feb 08, 2002 1:01 am
Location: USA / North Carolina / Fayetteville
Contact:

Re: LED Street Lights Unvieled In Silicon Valley

Post by dacflyer »

here in Fayetteville NC all our traffic signal lights are 100% leds, as described earlier some of
the leds appear to be out or flickering in the string. i hate them, most of them are made by Leotec , the problem is, that the leds are being slightly overdriven, also the strings have no series resistors at all, there is a main resistor, but there should be one on each string, and theres over 100 strings in modual. i do however like the newly designed Dial lites they are great, and they mimic the incadescent lamp type signals, they have a light engine in them, typically 7 of the lumiled type leds, so if one does go out, the whole thing will go, and not have that crappy look of partial leds missing.
Bigglez
Posts: 1282
Joined: Mon Oct 15, 2007 7:39 pm
Contact:

Re: LED Street Lights Unvieled In Silicon Valley

Post by Bigglez »

Jim Barrett wrote:One of the problems we have here in Wisconsin is lack of heat from the incandescent bulbs.
Under certain (not uncommon) conditions the snow will drift onto the lights and not enough heat is generated by the bulbs to melt them clear. I sat at an intersection & wondered if they would end up putting heaters on the lenses to solve the problem & what that would do to the energy savings realized by switching to LED. :???:
Good point! Some signal heads include heaters to
reduce the turn on thermal surge that can easily
break the lamp glass. These obviously add to wasted
power, but a non-functioning signal is a hazard, too!
Bigglez
Posts: 1282
Joined: Mon Oct 15, 2007 7:39 pm
Contact:

Re: LED Street Lights Unvieled In Silicon Valley

Post by Bigglez »

Dean Huster wrote:I had heard that the sodium versions were chosen over the mercury versions because the yellowish color provided better illumination in fog.
Yes, some cars have/had yellow 'fog lights' for the same reason.
Sodium lighting was introduced after WW-II when discharge
lighting replaced incandescents. The sodium lighting gives
much greater efficiency (watts to lumens) as the energy is
in narrow spectra. Mercury vapour is similar, but much of the
energy is in the non-visible UV spectrum, and requires a
"photon-shifter" to give visible light. Same deal with fluorescent
tubes and white LEDs.

The yellow sodium lamps require high temps to liberate the
sodium from liquid state, and reduce the striking voltage.
This is done with a partial fill of neon gas which accounts for
the red glow at turn on. Once the heat from the neon warms
the sodium it's vapour has a lower work function than neon
and so the current flows in the sodium and no longer in the
neon, the colour changes to orange.

By cooling a section of the tube it will be red (neon) in that
section only. In practice the sodium tube is held in a vacuum
jacket to keep the sodium hot enough to work.

The tube must cool down before restriking, so these lamps
are neither instant-on nor instant restart!

Also, the tube has a 'cold-spot' - usually the U-tube furthest
from the electrodes. When turned off the sodium condenses
to metal at the coldest place, and when restarted the orange
glow starts there and creeps to the electrodes.

With a bit of practice it's possible to know how long the
street lights have been on and not fully warmed up by their
colours.

Dead control gear (bad ballast) can result in just a neon glow
as there is not enough current to raise the temps and get the
sodium active.

Something so simple and abundant in our world is actually quite
involved and clever!
Bigglez
Posts: 1282
Joined: Mon Oct 15, 2007 7:39 pm
Contact:

Re: LED Street Lights Unvieled In Silicon Valley

Post by Bigglez »

MrAl wrote:The smart money goes into a prototype system first, which is a
much smaller scaled down version of the intended final project.
For example, if 500 lights will be replaced then replace only 50 of them
and TRY this out for a few years BEFORE committing to a full install.
The experience learned could make a huge difference with the remaining
450 lights. Yeah, it takes a little longer to get the full system up and
going but there is more certainty that the final system will work as
expected once it's up and running.
As noted in the Merky News piece that I linked, the city of San Jose
has two pilot LED street lighting programs of about 900 (out of 62,000)
street lamps.

These will be fired up in a few weeks, so I'll be heading over
there to check it out for myself.
Bigglez
Posts: 1282
Joined: Mon Oct 15, 2007 7:39 pm
Contact:

Re: LED Street Lights Unvieled In Silicon Valley

Post by Bigglez »

dacflyer wrote: here in Fayetteville NC all our traffic signal lights are 100% leds, as described earlier some of
the leds appear to be out or flickering in the string.
Good point. Another observation is that not all LED signal
heads come on at the same time. I can clearly see one or more
lagging the others, perhaps only for a fraction of a second.

I suspect that this is because the LED cluster is DC driven
and requires a short time to 'strike' once AC power is fed
from the controller cabinet.

At first I though it was perhaps a delay in the cabinet, which
I believe uses solid-state relays. I'm not sure the human eye
can see a delay of once AC cycle, assuming the relays are
zero-crossing types and some may not fire at the same time.

When the UK replaced its mechanical traffic light controllers
in the 1970s the new triac controlled circuits had a deliberate
slow ramp up to the bulbs. This was to save the lamps from
switch-on fatique and reduce EMI-RFI.

It made the new signals appear to be many hundreds of watts
with very long thermal delays, which in reality where probably
under one second, but still very noticeable.

Turning off a car headlight (of the filament bulb variety) after
dark has the similar long tail as the filament cools off.
User avatar
haklesup
Posts: 3136
Joined: Thu Aug 01, 2002 1:01 am
Location: San Jose CA
Contact:

Re: LED Street Lights Unvieled In Silicon Valley

Post by haklesup »

I had heard that the sodium versions were chosen over the mercury versions because the yellowish color provided better illumination in fog.
Probably true of Low Pressutre sodium, not so sure about high pressure sodium.

Consider the light spectrum of the three light sources, this link gives a good picture.
http://www.nezumi.demon.co.uk/nonad/spectra.htm

LPS is essentially monochomatic having one strong emission line. The other types of light have varying amounts of most of the rainbow.

As we know a prism will bend each color a different amount. Since fog is essentially a field of prisms floating in front of you, it goes to reason that a single color would be less distorted than multi color light.

So I don't think its the Yellow but the monochomacity of the light that improves fog visibility. An LED might come close to this but at a different color it will be interesting to see. In any case LEDs offer more opportunities to engineer the spectrum of the source for the application. LEDs (devices and the chips that drive them) are truly the fastest advancing part of the semiconductor world right now (processor speed? who cares about that anymore)
User avatar
dacflyer
Posts: 4748
Joined: Fri Feb 08, 2002 1:01 am
Location: USA / North Carolina / Fayetteville
Contact:

Re: LED Street Lights Unvieled In Silicon Valley

Post by dacflyer »

bigglez >>said.... Another observation is that not all LED signal
heads come on at the same time. I can clearly see one or more
lagging the others, perhaps only for a fraction of a second.

i see that here too, but its mostly with mixed brands of leds moduals,,here the DOT buys what ever is cheapest. when a color comes on, mostly with the greens, there can be a split second between the heads, one will light before the other. sometimes as long as 1-2ms.
Post Reply

Who is online

Users browsing this forum: No registered users and 31 guests