Photo Gallery

1972 GE 2500L (6.6 A 147.2 W)

The lamp shown here has become quite a rarity nowadays as municipalities across North America have retired their obsolete incandescent streetlight luminaires with LED counterparts (or are in the process of doing so). The present 2500L is a special incandescent lamp which was designed specifically for an operation on 6.6 A series circuits that once were popular in the USA in street-lighting installations. The low lamp voltage resulting from the elevated drive current results in a higher lumen efficacy thanks to a shorter and fatter filament, which limits heat losses compared to the situation in mains-voltage GLS lamps of comparable life rating. Moreover, the smaller low-voltage filament increases the source’s luminance, which is beneficial to the optical control of the emitted light. Finally, because those lamps are driven at constant current, they have a better lumen maintenance that standard GLS lamps due to an effective wattage rise over time as the filament evaporates and its ohmic resistance rises.

An interesting feature of series-circuit incandescent lamps made after the mid-1950s is their integrated voltage fuse which ensures continuity in the circuit after the lamp failure. This feature was most often implemented in the form of lead wires brought close together so as to facilitate the ignition of an arc in the event of filament rupture (the OCV of street-series circuits is several kV). This arc melts the lead wires which are quickly fused together, thereby quenching the arc and forming a robust short circuit. Such design greatly simplified the maintenance procedure since it eliminated the cut-out device that was usually inserted across the terminals of the lamp socket.

The GE 2500L shown here is the second smallest model of street-series lamps (after the 1000L) that were sold in North America. The lamp has an output of 2500 lumens, which was adequate for the illumination of secondary roads and of (small) town centers. This particular model is rated for a 3,000 h average service life, best suited for the group replacement of lamps during maintenance. Although the number of incandescent streetlight installations began to decline strongly during the 1980s, these lamps remained in production for several decades more. GE eventually discontinued them around 2010, which is quite remarkable given how inefficient the technology is compared to its HID counterparts.

- Not only do incandescent lamps operate more efficiently at low voltage and high current, interestingly their flux output is also more constant through life. Over time, the filament impedance increases as a result of evaporation, causing a slowly increasing wattage and output at source level (at constant current that is) which compensates for the optical absorption resulting from bulb blackening.

What is even more peculiar is the fact that those current-driven lamps are less prone to failure via filament hot spot formation than standard voltage-driven incandescent lamps. That's because when their thicker and stiffer filament breaks following hot spot melting, the resulting arc can weld the filament extremities back together. As a result it will take several of such transient phenomena before the filament fail so catastrophically that an arc will initiate between the lead wires and short the lamp permanently. On that subject, it's interesting to note that most of those lamps were operated in the vertical cap up position, and when a long-enough arc is struck between two filament sections after local melting, then a gliding arc effect (Jacob's ladder) occurs, i.e., convection and self induction force the arc to move upward and to transfer itself to the section where lead wires are close together, thus making the process of end-of-life shorting more reliable than at any other positions.

However, each re-welding of the filament lowers its impedance, causing a decrease in output flux. Lamps have been found with up to ten welds, resulting in a very low light output. Those lamps were referred to as “slumpers” and prompted manufacturers to push for group replacement, typically three times a year, in order to maintain a nominal output flux in installations. Nevertheless, despite the drawback of filament re-welding, this process also curbs early failures, which was a bonus from a safety point of view of a reduced light output is a much better outcome than no light at all in the first occurrence of filament rupture.


Keywords: Lamps

1972 GE 2500L (6.6 A 147.2 W)


The lamp shown here has become quite a rarity nowadays as municipalities across North America have retired their obsolete incandescent streetlight luminaires with LED counterparts (or are in the process of doing so). The present 2500L is a special incandescent lamp which was designed specifically for an operation on 6.6 A series circuits that once were popular in the USA in street-lighting installations. The low lamp voltage resulting from the elevated drive current results in a higher lumen efficacy thanks to a shorter and fatter filament, which limits heat losses compared to the situation in mains-voltage GLS lamps of comparable life rating. Moreover, the smaller low-voltage filament increases the source’s luminance, which is beneficial to the optical control of the emitted light. Finally, because those lamps are driven at constant current, they have a better lumen maintenance that standard GLS lamps due to an effective wattage rise over time as the filament evaporates and its ohmic resistance rises.

An interesting feature of series-circuit incandescent lamps made after the mid-1950s is their integrated voltage fuse which ensures continuity in the circuit after the lamp failure. This feature was most often implemented in the form of lead wires brought close together so as to facilitate the ignition of an arc in the event of filament rupture (the OCV of street-series circuits is several kV). This arc melts the lead wires which are quickly fused together, thereby quenching the arc and forming a robust short circuit. Such design greatly simplified the maintenance procedure since it eliminated the cut-out device that was usually inserted across the terminals of the lamp socket.

The GE 2500L shown here is the second smallest model of street-series lamps (after the 1000L) that were sold in North America. The lamp has an output of 2500 lumens, which was adequate for the illumination of secondary roads and of (small) town centers. This particular model is rated for a 3,000 h average service life, best suited for the group replacement of lamps during maintenance. Although the number of incandescent streetlight installations began to decline strongly during the 1980s, these lamps remained in production for several decades more. GE eventually discontinued them around 2010, which is quite remarkable given how inefficient the technology is compared to its HID counterparts.

- Not only do incandescent lamps operate more efficiently at low voltage and high current, interestingly their flux output is also more constant through life. Over time, the filament impedance increases as a result of evaporation, causing a slowly increasing wattage and output at source level (at constant current that is) which compensates for the optical absorption resulting from bulb blackening.

What is even more peculiar is the fact that those current-driven lamps are less prone to failure via filament hot spot formation than standard voltage-driven incandescent lamps. That's because when their thicker and stiffer filament breaks following hot spot melting, the resulting arc can weld the filament extremities back together. As a result it will take several of such transient phenomena before the filament fail so catastrophically that an arc will initiate between the lead wires and short the lamp permanently. On that subject, it's interesting to note that most of those lamps were operated in the vertical cap up position, and when a long-enough arc is struck between two filament sections after local melting, then a gliding arc effect (Jacob's ladder) occurs, i.e., convection and self induction force the arc to move upward and to transfer itself to the section where lead wires are close together, thus making the process of end-of-life shorting more reliable than at any other positions.

However, each re-welding of the filament lowers its impedance, causing a decrease in output flux. Lamps have been found with up to ten welds, resulting in a very low light output. Those lamps were referred to as “slumpers” and prompted manufacturers to push for group replacement, typically three times a year, in order to maintain a nominal output flux in installations. Nevertheless, despite the drawback of filament re-welding, this process also curbs early failures, which was a bonus from a safety point of view of a reduced light output is a much better outcome than no light at all in the first occurrence of filament rupture.

Philips_200PS21DB_-_NL_1968.jpg Philips_6131_300W_-_NL_1941.jpg GE_2500L_145W_-_USA_1972.jpg Xycarb__255860_2000W_120V_-_FR_1995.jpg Thorn_M5821D_-_EN_1990.jpg
Lamp/Fixture Information
Manufacturer:General Electric
Model Reference:2500 L
Lamp
Lamp Type:Incandescent
Filament/Radiator Type:Incandescent tungsten filament
Base:E39
Shape/Finish:Pear shaped clear
Service Life:3.0 kh
Electrical
Wattage:147.2 W
Voltage:22.3 V
Current:6.6 A
Optical
Lumen Output:2500 lm (initial)
Lumen Efficacy:17.0 lm/W (initial)
Physical/Production
Factory Location:Cleveland, OH, USA
Fabrication Date:April 1972
Application/Use:Streetlighting (series circuit)
File information
Filename:GE_2500L_145W_-_USA_1972.jpg
Album name:Max / Incandescent lamps
Keywords:Lamps
Filesize:570 KiB
Date added:17 Aug 2024
Dimensions:1500 x 785 pixels
Displayed:510 times
DateTime Original:2010:11:21 10:06:34
Exposure Time:1/125 sec
FNumber:f/5
Flash:No Flash
Focal length:35 mm
ISO:800
Model:Canon EOS 450D
Software:Adobe Photoshop CS5 Windows
White Balance:1
URL:https://trad-lighting.net/gallery/displayimage.php?pid=258
Favourites:Add to Favourites

Comment 1 to 15 of 15
Page: 1

Ria   [Sat 17 Aug 2024 at 11:00]
I've always been intrigued by series incandescent lighting, it's something we never had here in the UK. We only have one example of a series lamp here, which I'll have to try and find the photo of. It's an Osram-GEC one as I recall, made in England, but exported to Australia. So it's been all the way around the world and come back home again..!
Sammi   [Sat 17 Aug 2024 at 13:56]
It would be interesting to have a Jacob's ladder as part of our collection. Wonder
Max   [Sun 18 Aug 2024 at 08:03]
You sure had series incandescent lighting if you ever had those old miniature incandescent Christmas light strings Wink
Ria   [Sun 18 Aug 2024 at 09:39]
Oh those things, yes I remember them..! We actually have an American C7 set that Robert sent us, but I don't think it's series fed, they're 120V lamps, I must check at some point.
I do vaguely remember incandescent street lighting being replaced with MV when I was about 5 years old or so, it was one of those defining moments that I think went towards my becoming a collector Incandescent lamp HD
Max   [Sun 18 Aug 2024 at 13:39]
I was born too late to witness the transition from incandescent to mercury in my neighborhood, although I do remember the incandescent streetlights in some of the most historical parts of Paris. Interestingly those skipped the mercury phase and went straight to white sodium and ceramic metal halide, the condition being that a warm-white light similar to that of incandescent lamps be emitted (the city of Paris has a really strict code).
wide-lite 1000   [Sat 14 Feb 2026 at 05:11]
Nice lamp . I have a 600L version .

I need to dig it out at some point .
Max   [Tue 24 Feb 2026 at 07:17]
It's quite mind boggling that 600-lumen lamps were once considered sufficient to light a street...
Sammi   [Fri 27 Feb 2026 at 18:42]
And now we're back to the same level of illumination... Sad
Ria   [Sat 28 Feb 2026 at 00:07]
Ain't that the truth... Confused
Max   [Sun 01 Mar 2026 at 20:03]
Except for bicycle and dirt paths, we certainly have not returned to 600-lm light sources/luminaires. It is a fact that with the advent of HPS and MH lamps, people became used to flooding places and roads with way more light than necessary, so returning to pre-1970s illumination levels is not really a bad thing overall. That also brings me back to the mercury lighting days, before everything became too bright, so I quite like this recent change.
Tuopeek   [Sun 01 Mar 2026 at 20:37]
I feel CCTV was the later driver for the whiter brighter lighting we have today but with camera sensor improvements we probably could reduce the lighting now or even go IR. BTW, interesting use of a Jacob's ladder to insure the circuit remained connected
Ria   [Mon 02 Mar 2026 at 01:05]
The problem with the erm.. "street lighting" we have now is that it is (a) the wrong colour temperature, so people like Sammi who suffer from night blindness, simply can't see under it, and (b) the "lanterns" have just been put on existing poles, so that you get pools of light underneath them and little light everywhere else. We rarely go out after dark now because of this Mad

For what it's worth, the best street lighting in our opinion is LPS LPS Linear Sodium Lamp SOX Lantern GoldFinger
Max   [Mon 02 Mar 2026 at 17:25]
I hear you about the horrible quality of some illuminations, but in that case that's clearly because of an improper use of technology rather than shortcomings from the technology itself.

Tuopeek - Unlike the UK, the massive usage of CCTV came very late over here, and it is localized to dense cities mostly. The trend towards over-illumination began much earlier and became widespread, so I am not sure it has to do with CCTVs overall. I think it's more a case of Jevons paradox coupled to a feeling of safety associated with brightly lit places. The latter was certainly a low hanging fruit for municipalities who wanted to be/look proactive on the subject of safety, and the lighting industry was all too happy to cater to their needs, naturally. There's also an image/reputation aspect to it as the brighter a city the more rich and prosperous it must be. This association began typically during the 1950s with the whiteways in America.
Ria   [Tue 03 Mar 2026 at 00:29]
@Max - I think we're going to have to agree to differ on this one..! I will say no more Neutral
Max   [Tue 03 Mar 2026 at 06:44]
I didn't bring this topic here... Fact remains that we don't have any problems with LEDs over here (no more glare than HPS and MH luminaires, acceptable CCT, decent/good color quality, even road illumination) precisely because municipalities make the right investments which result in the proper usage of the technology, instead of going for the cheapest luminaires they could find on aliexpress. Solid state lighting is just a tool, it's not intrinsically good or bad, like with most technologies its outcome depends precisely on how we use it.

Comment 1 to 15 of 15
Page: 1