I measure 1200 lux at a distance of 1 meter. The Petromax 829 runs with an operating pressure of 2,5 bar and is equipped with a Helox mantle.Otherwise there are no other lights in the room. How many CP are that ?
You may take a look here: petromax.nl\Lichtmeten ;-) Sorry no German translation yet... Short: Lux at a distance of 1m should equal CP (or candela). But 1200CP from a 500 CP mantle would be darn good.
I have already looked at the page and translated it and I also looked at the measurements from Newark 2008 and was horrified to read that a Petromax 829 only has 255 cp. I forgot to write that 1 meter is equal to 3.28 feet or 39.37 Inches.
That's what I have also figured out by the equations. But there must be an error somewhere, because I cannot imagine that a 500 CP mantle without a reflector would perform as good.
Those Newark meter readings were taken with my photometer and are reasonably accurate. It does incorporate a lux meter but the baffled tube prevents any reflected light from reaching the sensor and in the full bright daylight that day without a lamp it read 1cp. So the figures in that table are the cp from the mantles it was aimed at with maybe a very small percentage error. A lux meter is not designed to read a point light source it is a tool to measure ambient light in a room and will be affected by reflected light which will give an inflated figure. When we were experimenting with these photometers we found that my 150mm dia 500mm long tube gave readings about 10% high but with 50mm baffles at 250mm and the tube end we created a 50mm reflection free virtual tube and the readings became much more sensible and agreed with the rated power of the lamps we were measuring. If you look at these Newark figures you will see a very few lamps gave anywhere near their rated power but the Hipolito and Phoebus 615 did and some got pretty close such as a couple of Optimus another Phoebus and a Geniol. Here is a PDF of a table of lamps I measured some years ago and this shows just how uncommon it is for a lamp to make its rate but it is reasonable to assume any lamp will deteriorate with use so the results were more or less as expected. Neil.
That exactly explains the too high readings. Any lantern straylight reflected from the ambient must be carefully shielded. On the other hand, the 1 m distance makes it easy to calculate.
There can be a number of factors to affect performance in any pressure lamp even from new. Jet size is vital of course but also the configuration of the hole. How round it is in other words, with use the jet will wear and gradually increase in size but also distort and a ragged hole gives a poor irregular gas flow which affects how the flow grabs air and is why lapping a jet surface can improve performance. Also carbon build up in the generator affects the pressure/flow at the tip. So an older generator will probably function better if the inside is polished and with any Preston type generator cleaning the loop is difficult and polishing near impossible. Generators with internal packing will also reduce flow as the carbon builds. The same applies to air and mantle tubes where the pressure is low in operation so any friction in those can have a disproportionate effect. So in practice lamp generator/burners die of old age and a 500 rated Petromax at around 250cp is not so surprising. If I remember right we also measured a Coleman type Milspec lantern at Newark at about 5cp so the effect of age can be drastic. ::Neil::
I remembered reading that somewhere with that particular milspec, the issue was the generator I remembered correctly. Anyways I make it a common habit to clean my petromax generator often as possible and use the cleanest fuel like Kleanheat. It seems Kleanheat is ever so slightly more volatile than kerosene so most of the vaporization happens in the vertical section of the vaporizer making it easy to clean out if I catch it early enough. (If I ran the petromax almost every night, I usually tear it down and clean it at the end of the week.)
Yes, 1200 lux measured from a Petromax 829 at 1m from the lantern would be 'super' high if it was the only light source. As mentioned, reflected light needs to be omitted for accuracy. If lux was used the unit of measure, then it was the illuminance being considered. One lux (1 lux) is defined as being equivalent to one lumen spread over an area of one square metre. ** 1lux = 1lm/m2. (lumen or lm is the unit of measure for luminuous flux). That would be equivalent of getting 1200lm of luminuous flux incident over an area of 1metre squared. ** = 1200 lm/m2. Candlepower or cp can be regarded as the unit of measure for luminuous intensity. The current SI System uses candela(cd) as the unit. 1cp is around 0.980 cd. 1cd ≈ 1.02 cp and 1 Hefnerkerze or 1HK is about 0.920 candelas. 1cd ≈ 1.09 HK also, 1 lm = 1 cd.sr (sr or steredian is the unit of measure for solid angles). For measurents in a well-defined cone, the total luminuos flux, Φv in lumens is given by: Φv = Iv x 2π[1 - cos(A/2)] where Iv is the luminuous intensity in candelas(cd) and A is the radiation angle of the lamp—the full vertex angle of the emission cone. Even with this approach, and assuming the Petromax 829 mantle as a point source of light, you'd probably be getting an unbelievably high level of luminuous intensity. I must be missing something here.. Wouldn't correlate or be anywhere near the rated 500cp or 500HK.
On the other hand, if we assume that the mantle is a point light source and using the following:- 1 lm = 1 cd.sr A full sphere has a solid angle of 4π steredians, so a light source that uniformly radiates one candela in all directions has a total luminous flux of 1 cd × 4π sr = 4π cd⋅sr ≈ 12.57 lm ** a 1cd light source produces ≈ 12,57lm., We'd be ending up with some doubtfully low figures of luminuous intensities from a Petromax 829.
I also don't think the Petromax 829 can have 1200cp.I was expecting an out put of 500 to 600cp. I would like to thank you for your interest in light measurement. A very special thank you goes to Neil for the detailed explanations. I have now contacted a specialst company that produces measuring devices. I explained to them that I collect pressure lamps and want to measure the Real light output and whether they can offer me a device ,possibly with equipment, that is suitable for this purpose.
I would agree that professional devices for direct measurement of the luminuous intensities (in cp or cd units of measure) of light sources are pretty expensive, perhaps not so readily available to the general public. The necessary setups for accurate measurements are also rather involved or complex. In many of the commercially available electric lamps for illumination purposes, we would most likely see that they are specified in these three measures: 1. Rated or consumption power in watts(W). 2. Luminuous flux or output in lumens(lm). 3. Luminuous efficacy in lumens per watt(lm/W). For classic, liquid or gas-fueled lamps, we would usually be more interested to know the luminuous flux outputs. However, most if not all, classic products had only been specified with 'candlepower', cp or HK ratings. In a modern sense, or out of curiosity, we tend to compare classic lamps with contemporary illumination products such as incandescent light bulbs, LED fixtures, high pressure mercury/metal-halide lamps, etc. Modern devices are not specified in CP ratings but mostly in the SI units. Therefore, it won't be immediately or conveniently possible. If we assume that the ratings of classic pressure lamps are correct, perhaps we might just simply calculate the lumen outputs? Example, if we have a 500HK lamp and if 1HK is equal to 0.920cd, then it should have a luminuous intensity of about 460 candelas. We would also have to assume that the mantle is light source that uniformly radiates 460 candelas in all directions. (**no reflectors whatsoever involved here). If this manner, it is supposed to produce a luminuous flux of: = 460cd x 4π sr = 460cd.sr x 4π = 5,780.53 cd.sr or 5,780.53 lumens. (1 lm = 1 cd.sr). With the lumen figures, it makes possible, some degree of comparison between classic pressure lamps and modern-day illumination devices.
This photo is taken from Petromax catalogue No.92 see over there: https://classicpressurelamps.com/threads/mid-1930s-petromax-catalogue-no-92.19950/ This is a photomontage like the other images in the context. At least the table in the upper left corner was added. I am not sure about the other image components. So they certainly do something in a photo-metric laboratory but I am not too sure about what they really did. Erik
There are many variables to get a good, reproducible outcome. It starts with the fuel, pressure, type mantle, how does it expand, etc. and ends with a expensixe yearly calibration of your luxmeter. Also like said on my webpage you should measure at different angles after a certain time so all is on temperature. We do measurement on all kind of things all day long at the office so this part is well known. I to had contact with a certified company to check my text and test set up. Like said on my webpage. The setup from Light International and also used by Neil was a good set up: measure at 100cm and block all incoming other light. The only minus point would maybe be that the device used had no digital output back then. In that case and maybe nowadays you can put the lamp on a turntable and rotate it while measuring. Then you get a 360degrees measurement. But remember only in the plane of view.
Interesting photo, Erik. It is astonishing that they went to such great lengths on the output measurements back in those early years. I wonder if the other manufacturers did anything similar. Controlled measurements requires rather elaborate setups and precision instruments. Not to mention the amount of pre-defined limits applied to ensure the results are as accurate as possible and repeatable with a minimum of errors. Like WimVe mentioned, equipment calibration can be expensive, especially if when done by an approved certifying body. This is a general requirement when specifications are laid for commercial products.
Besides that: nowadays everybody can buy a luxmeter at a chinese webshop and is then "certified" to measure.
A grease spot photometer works fine and will always show comparisons very well although it does involve some minor calculation but the main problem is calibration. To get a cp reading you need a known power source on one side and a certified light is not a simple item to find and also needs a specific regulated electric supply so the simple device starts to get very technical and therefore expensive to calibrate. To compare two lights simple and very effective; to give a cp figure much more complicated. ::Neil::
From my understanding, the CP(candlepower) rating would correlate closest to the 'luminuous intensity', which is measured in candelas(cd) instead of lux or lumens. I'm guessing it won't be as simple as what we would like it to be. The equipment that can actually measure and provide accurate figures in candelas would be some form of goniophotometers instead of a simple hand-held photometer. And it is most certainly not going to be cheap. For instance, these kind of equipment:-
Most luxmeters in use are used to measure , well, Lux on a certain spot. This can be a workbence, tunnel or any other place. But then you measure all the light from all the directions. If you want to do a spot measurement, you have to block all other directions of light coming in. If you want CP you need a distance of 1m. Nothing more or less. Its all on my webpage.
Accordingly, this is supposed to be sufficient for us to determine fairly accurately on the figures of the luminuous intensity in cd or cp without getting overly involved with elaborate or expensive equipment.
Very interesting and helpful, confirming my "gut perception". Thanks Neil. We must remember; brightness is logarithmic... Double the cp is far from double as bright!
When I was making the photometer I and the Guild were experimenting with in the photos above I discussed the detail with Mike Sharman who was an engineering design professor in Cambridge as well as a lamp collector. Considering the design and the initial resulting measurements we agreed the conclusion that as the readings were in an expected range for the rated power of the lamps they were probably reasonably accurate. In fact subsequent factors made me realise those initial readings were probably around 10% high. My photometer uses a piece of 150mm PVC water pipe I found in the trash at work but the two Fil Graff made used 4 inch tube and his readings were wildly inaccurate. Fil experimented with baffles and found that inserting a 2 inch hole baffle half way down the tube and another at the end created a “virtual non reflective” tube and much more reliable readings. I remember drawing the lines of light and reflections to scale and concluded that my 150mm pipe was a happy accident that was only about 10% out and that Fil’s 4 inch tubes allowed a huge amount of light bounce down the tube both of which were controlled to be near zero with the baffles. The readings are never going to be 100% accurate though because the light source is not a point but a slightly irregular ball or two of light with a centre measured at 1 meter but the sides maybe + or – 10mm and with light intensity varying with the square of the distance plus the difficulty of measuring the 1 meter accurately around 10%cp either way is perhaps reasonable. It’s a home made empirical design and build photometer which is accurate enough for our purposes and likely a lot cheaper to make than a commercial light meter. The lux meter was the only expense at around £15 I think plus some time and bits of scrap material. So not perhaps definitive industry standard but good enough for our purposes. ::Neil::
