# over voltage operation

#### ship

##### Senior Team Emeritus
So I’m getting really bad at math in some ways I suppose, (took ethics instead of real math in college and its been like probably 25 years since I took such stuff in high school.)

A normal FHM lamp is 1,000 Watts at 120 Volts, has a color temperature of 3,200̊K, 27,300 Lumens and 400 hours of lamp life.

A Philips (Discontinued but advertised in 2003) #7786R lamp was 1,000 Watts at 100 Volts, has a color temperature of 3,200̊K, 27,000 Lumens and 300 hours of lamp life.

Given for every 1% change in voltage, if 120v is applied (well over the maximum 110% recommended safe operation specification for the lamp,) a lamp will have 3.1/2% more luminous output, 0.4% more color temperature, and 13% less lamp life for each percentage of voltage over its rating. Also given 0 for lamp life is zero - dead lamp, and the maximum color temperature of the lamp (short of extreme pressures of xenon in the lamp) is somewhere between 3,500̊K and 3,655̊K, this would also result in lamp failure.

What would be the specifications of this lamp at 120v, 110v and the specifications of the maximum voltage that could be applied to the lamp theoretically for at least a nominal 20 hours of use?

#### ship

##### Senior Team Emeritus
No takers? Where are the board Christmas Vacation computer/math majors? Should be a simple table if not better yet graph to work up.

I'll sweaten the deal in offering a prize for the best info and posted graph showing the effects of voltage at each % between say 80 and 125v on this lamp and the other 120v FHM lamp. Need to show color temperature, luminous output and lamp life matched up with voltage.

Winner gets a (pre-market and possibly not otherwise avalable) Color Kinetics PAR 20 LED with variable burst and pattern.

#### Attachments

• PAR 20 LED.JPG
98.1 KB · Views: 199

#### derekleffew

##### Resident Curmudgeon
Senior Team
Okay, I'll bite. Though I usually recuse myself from ship's questions, but that LED PAR20 looks sweet. Here's the simple table. If I have to make all the graphs to qualify I will, but wanted to get my answer in.

Last edited:

#### soundman

##### Well-Known Member
Okay, I'll bite. Though I usually recuse myself from ship's questions, but that LED PAR20 looks sweet. Here's the simple table. If I have to make all the graphs to qualify I will, but wanted to get my answer in.
I am also working on the problem and I think you might want to look at your work again, I just glanced at your lamp life figure and I believe it is incorrect. At 1% over volt or 101 volts the lamp will loose 13% of its rated life or 39 hours (300*13%) not 3 hours. My pretty chart will be done some time after noon on Friday unless I truly just go out for one drink right now and come back to work on it.

heres a sample

#### Attachments

• Book1.xlsx.pdf
27.2 KB · Views: 210
Last edited:

#### derekleffew

##### Resident Curmudgeon
Senior Team
Thank you, soundman. You were correct, I inadvertently got my relative and absolute references awry. Here's my graphs, but I have a feeling there's info ship is not telling us and that it is a trick question.

#### Attachments

• shipslamp2b.pdf
38.8 KB · Views: 290

#### ship

##### Senior Team Emeritus
nice so far, I'll let it play out a bit in looking for a table and graph that perhaps best is easy to read and perhaps expand the question a bit. Hmm, perhaps have Wolf or Dave or someone else choose the winner after four weeks.

Origional theory for me looking into this lamp was that a 100v lamp operating on a dimmer would provide more UV-A light than that of a 120v lamp. This much less at 120v would provide a lot more UV-A light for a black lighting effect.

Perhaps add this factor of how much black light UV-A light a lamp at voltage, under it and over voltage now provides into the concept if any effect. In other words, will this lamp now be a better black light source at 120v, and how much a drop off in UV-A output does the 120v lamp have with voltage changes? (Should be much harder as there is not a huge amount of info out there on actual or estimated UV output on an incandescent lamp - but it is out there.)

This whole concept as more an educational study that is useful to all now in many ways such as in comparison of 115v lamps to 120v lamp in why they seem brighter or in perhaps getting more UV-A much less light in general at a specific color temperature out of a lamp while working on a dimmer setting or lower specified voltage if that lamp is already close to it.

Perhaps in working with Dave a forum calculator can be made which if you punch in the specs of a lamp, and its voltage or dimmer percent you now have its expected output and lamp life. Don't believe such a thing otherwise exists.

#### ship

##### Senior Team Emeritus
No takers on the graph given the new concept? Let me help with that part in making it fairly easy again in theory at least: (Come on, while not easy, such a chart also showing UV output of a halogen filament lamp could be useful to know about when operated in over and under voltage conditions. After all, put your dimmer on 90% and how much UV output is there? Put the fixture 90' away from the source and is there any UV output? Perhaps this distance concept - law of squares would be the basis for it, should be its own graph.

Ok, I'll give extra points and more swag and stuff to someone that also does this and also comes up with a formula that goes along with the above for every 1% voltage change formula. This based upon a formula for UV-A output of a filament or halogen lamp to start with based off the luminous output / color temperature combination if anyone can calculate one. Hmm, could even find a lighting fixture that might make such a study worth it. Or perhaps my NSI 8x16 light board if the total deal - all questions and concepts answered.

"By far the greater part of the electrical power consumed by the incandescent wire is output as electromagnetic radiation. The total radiation increases as the fourth power of the material temperature. The spectral distribution of the radiation, i.e. its distribution over the ultraviolet, visible and infrared (heat) ranges, shifts to shorter wavelengths in a kind of bell-shaped curve as the temperature increases. The peak moves out of the infrared towards the visible region (see Fig.3). Unfortunately the melting point of tungsten does not permit the peak to be shifted into the visible region of the spectrum.

At the highest practicable temperatures, this peak is at about 850nm (nanometers). The visible region ends at 780nm. In this case about 20% of the total radiation is given off as “light”, about 0.3% in the UV region, and the rest (the majority) as heat." Page 4: http://ecom.mysylvania.com/sylvaniab2c/b2c/z_login.do;jsessionid=ID4001DB0.9892852556966844End go to the PDF:
Engineering Bulletin - Technology and Application - Tungsten Halogen Low Voltage Lamps Photo Optics

"Substantial heat is generated in all halogen lamps (90% of their light is infrared and a small amount is UV which can be protected against by almost any screen or lens)" G.E. Lighting Spectrum , 9200 Catalog, General Electric 1993 “Principles of Lighting” p.17

"The X-rays and most UV are absorbed by a couple of feet of air." Biomedical Effects of Light: The Beauty of Light, by Ben Bova; John Wiley & Sons, Inc., N.Y. 1988 pp81-87

Certainly a study to study into further as if its own question of the day. Not just a "roadie..." a scientist.

Last edited:

#### derekleffew

##### Resident Curmudgeon
Senior Team
I answered your original question to the best of my limited ability and then you went and changed the rules. I will not participate further as it's a purely academic exercise and I gave up academia years ago. No incandescent source with standard theatrical color media, even L181, R59, or AP#3850, is going to produce satisfactory results compared to an arc-source (carbon-arc, Xenon, MSR/MSD or HMI, or MH) with a Wood's glass or glass UV filter.

Over-voltaging the incandescent lamp is going to produce more UV, but it will never have enough UV to be seen through a UV filter. Since L181 transmits only 0.78% of light across the visible spectrum, one would need at least a 5K HMI Fresnel to even approximate the effect of a 250W Metal Halide Wildfire™ fixture. No incandescent source is going to transmit enough energy below 300nm to be usable. The physics just aren't there.

Send me the CK PAR20 or not.

#### ship

##### Senior Team Emeritus
what makes you assume it's woods glass Used instead of quartz glass on a halogen lamp with no woods glass lens mentioned?

Sorry to offend at expanding the question a bit, my intent was scientific.

#### derekleffew

##### Resident Curmudgeon
Senior Team
I mentioned Wood's glass as Devon Glass #139 or #146 I don't think would be appropriate (colors on their website are not at all close, BTW), as neither comes close to L181. I know the client will not/should not spend the money on dichroics for this, as it's not going to work and just wasting money. Plus dichroics for cyclights would be cost prohibitive, possibly so would Wood's glass. I think even HT181 would burn out in seconds, so you're looking at heat shield, which cuts down the intensity slightly, and then a custom stand-off color frame for the 181. Just too many obstacles to overcome, when an affordable solution (Wildfire™) is at hand.

#### ship

##### Senior Team Emeritus
Nope, I was more looking to compare a
#7786R (3/03) Philips #923877130900 CL, Quartz, Single Coil 1Kw/100v T-11mm c-65mm MOL 117.6mm R-7s Horz. Burn ±15̊ 3,200̊K 27,000 Lum 300hr

with a

FCM G.E. #23797 (P2/28) 4CL, Quartz 1 Kw T-3 c-8 MOL 4.11/16" RSC (Burn Horz ±4̊) 3,200̊K 28,000 Lum 400hr

lamp in an open faced cyc light fixture. Quartz glass and before any concept of gelling, lensing or dichroicing' them which might or might not in the "best for UV output" range have an effect on UV output of the lamp itself.
More less a simple question of the effects of over voltaging a halogen lamp, its output at voltage than some theoretical over-voltage output in the UV-A range. This in addition to the output question already plotted out.

Sorry to offend. See that commercial on TV where some drips of paint from an abstract artist on a floor above travels thru the floor, onto the light fixture than drips onto the not so able differentuate graph. Something like that perhaps with four graphs in one I was thinking. Below graph X range measures voltage, above is flexible on a few scales, first lamp life, second color temperature, next luminous output and added to these three already worming their bell chart like way thru, UV-A output of a halogen lamp with quartz glass under over-voltage operation.

Should in theory given the Y-graph side would have four scales in one to it, and the X side of the graph only track voltage, still it should be possible to plot out.

Don't worry, unless someone else produces the above concept you will still get your reward, and otherwise you will get something out of me. E-Mail me with shirt size at least.

I mentioned Wood's glass as Devon Glass #139 or #146 I don't think would be appropriate (colors on their website are not at all close, BTW), as neither comes close to L181. I know the client will not/should not spend the money on dichroics for this, as it's not going to work and just wasting money. Plus dichroics for cyclights would be cost prohibitive, possibly so would Wood's glass. I think even HT181 would burn out in seconds, so you're looking at heat shield, which cuts down the intensity slightly, and then a custom stand-off color frame for the 181. Just too many obstacles to overcome, when an affordable solution (Wildfire™) is at hand.

#### ship

##### Senior Team Emeritus
derekleffew no doubt in origional work has the lamp but I'm still not where I would want to be either in origional concept for presentation or end result once UV-A was added to the mix. Want that commercial with this doesn't make sense, than the drips of paint commercial in presentational type chart including all. Of note is that Wildfire just published and sent out a newsletter on the subject of blacklight and it could be helpful in starting to determine the UV output of a incandescent filament lamp.

So, I'll take the LED lamp out of the equasion for reward.

Really it is more about that of a HPL say 575w/115v lamp in UV output over that of a 500w/120v lamp for UV output as a general concept, and both under a dimmer percentage. Very useful concept I feel to study into, and one that has not been to date sufficiently.
.

Last edited:

CB Mods