Mike Haughton's article from MEW 215
This article by Mike Haughton in Model Engineers' Workshop 215 was well received. It's a very practical and informative survey of the options for lighting your workshop.
|Neil Wyatt||24/07/2014 20:28:19|
15707 forum posts
Mike Haughton's article on workshop lighting in Model Engineers' Workshop issue 215 generated a good deal of interest.
As a piece of particular interest to those setting up their first workshop, as well as those upgrading their existing facilities, here is an online version.
For those who don't subscribe to MEW, other content in issue 215 includes:
so with all that to enjoy, why aren't you a regular reader
|Kris Lehane||20/08/2014 19:29:55|
|11 forum posts|
Very interesting, chaps! Since I spend my life wrangling lamps when not creating swarf.. A few interesting titbits for you. From what I remember:
So what light levels do you need? In Lux, HSE says 1 lux at floor level for a fire exit. (or rule of thumb, to be able to tel the difference between a ten and Five pound note when laid at floor level, for the hieght of an average man. It used to be 10 Lux, and some people insist on 15 Lux immediately the power goes. A shop floor or office is an average of 20 Lux, minimum 5. To work in detail in a machine shop it's 200 lux (average) for detail, minimum 100, And for fine detail, it's average 500 Lux, minimum 200. HSE and USDAW and Gov. regs.
A 2KW Xennon super trooper light is gonna find you, at 133 lux flat field with a field of 4M on a throw of 300 feet, and that's the baby one!
As for kelvin, (Not degrees kelvin!! do that an a GCSE Chemistry paper... fail) Colour temp/mired shift, a standard 40W 240V lamp (not lightbulb... bulbs grow-lamps blow) is just under 1800 Kelvin. To give an example, a theatre T class lamp is 2800K, CP Class lamps are just over 3000K, and that's hot for the eye. Studios take average daylight at 5200K, which is a white film light. Film standard. Average bright sunny day.
Some useful equations:
Lumens to lux: Lux =10.76391 x lumens/(4 pi R squared Feet)
Watts to lux: Lux=10.76391 x Power (Watts) x N(Lumens per Watt) / area (foot squared)
Lux to Candela: Candela=0.92900304 x Lux x distance Squared (feet) or: Candela= Lux x distance squared in Metres.
So providing you can extract the information out of the box or manufacturer, you can work out how close your lamp and what power you need to target! Depending on your eye-sight.
BIG WORD OF WARNING: Avoid 240V lamps anywhere near the work area flying debris and LX shock...
For all the photrographers amongst us.. have a nosey at Lee lighting, Arri, and Rosco's Kelvin charts. I'm never without a swatchbook and a truckload of Technical filters to make all the lamps match the HMI in the sky... and I'm a big fan of LEDs! saves on megawatts. They also have the reflectors built in to the cathode, and the above maths also applies.
Enjoy playing light, be safe.
|Kris Lehane||20/08/2014 19:44:26|
|11 forum posts|
Just a quicke: LED lamps do not suffer the same fate as other type of source, as the quality of Kelvin does not deteriorate with age, unlike a dying filament or arc source, or flourie. I'll grab some details of the latest LED's being fitted to our film lamps...
|1994 forum posts|
I don't buy too many copies of MEW but 215 was one issue I missed that's sadly not available as a back issue.
So, I plan to fit four 6 foot tubes in my 5 x 3 metre shop, or should I fit four doubles instead?
|Oompa Lumpa||21/08/2014 09:55:05|
|888 forum posts|
"old" style tubes or LED tubes?
|1994 forum posts|
Old style tubes Graham.
|Oompa Lumpa||21/08/2014 10:54:46|
|888 forum posts|
Just don't. I changed all my tubes to LED last year and two things, it reduced by quite a bit the power consumption and the lack of flickering is just heaven.
Every light in my Office and workshop is now LED and I wouldn't fit a traditional bulb for any reason. I am just making a fitting for my "new" Drill Press so I can install a nice LED bulb in there. My Den in the house is all LED too. My missus however won't have LED - she was an early adopter when they were, well, crap. What she doesn't realise is that as the bulbs are failing I am replacing them with LED.
One major advantage for us machinist types is that when you reach up for the reflector to reposition the light better, you don't get burned. They are not hot to the touch. A massive plus as far as I am concerned.
Edited By Oompa Lumpa on 21/08/2014 10:54:57
30 forum posts
That sounds like a major investment Graham, which make of tube did you go for?
|1994 forum posts|
Yes, do you have a link for the type you used, are they expensive?
207 forum posts
I've been a follower of LED lighting for many years now - only wished I'd patented the LED traffic light mod idea when I had first thought of it when I first started using multi chip LED indicator lamps! For LED lighting have a look at http://www.bltdirect.com/products.php?adcid=adwords&gclid=CNjOs9HkpMACFZTLtAodLT4A9A (Usual disclaimer) I've bought some LED GU10 halogen replacements which are still going some 5 years ago.
Edited By Johnboy25 on 21/08/2014 17:50:15
Edited By Johnboy25 on 21/08/2014 17:51:29
|Ed Duffner||25/08/2014 06:49:25|
|715 forum posts|
Whilst in college doing my Electrician's 236 courses we were told, when installing fluorescent lighting in a machine shop the luminaires should always be twin fittings. Single tube fittings can sometimes give rotating machinery the appearance of being stationary. It may be a different story with todays' high frequency luminaires.
There are a number of different colour temperature fluorescent tubes available for different applications e.g. White, Cool White, Warm White, Daylight etc. and even some special ones if you're a tropical fish or a Butcher!
Replacing the standard bimetallic type of starter with an electronic one will remove flickering when a fluorescent light is switched on, but they're expensive.
It was interesting to read in this article that normal tungsten filament lamps have been removed because of legislation. I wondered why B&Q didn't sell them any more! Instead they have the tungsten-halogen/quartz envelope inside the glass bulb/lamp. That design must generate quite a bit of heat and reduce lamp life considerably I would think.
Edited By Ed Duffner on 25/08/2014 06:58:10
|3062 forum posts|
I think you will find that the halogen quartz envelope types are in fact rated longer, at 2000 hours, whereas the "old"lamps were merely 1000 hours.
I have never had any problems with strobing causing machines to look stationary with Fluorescent tubes, even with variable speed drives in use.
|Ian S C||25/08/2014 13:15:18|
7262 forum posts
The old fluro tubes operated at 50 Hz, the compact Fluro lamps operate somewhere around 35 kHz, no strobe troubles.
Ian S C
207 forum posts
"Ah... An interesting topic stroboscopic effects with different types of lighting" - as my college lecture said... Perhaps this is why I can remember some of the details after 40 years! Well yes it is desirable for industrial premises if not an HSE requirement to have twin style fluorescent light fittings installed where rotating machinery is being used. Single tube fittings will produce a 100 Hz ripple effect with the lamps peak light intensity output on the positive & negative part of each cycle. i.e. double the 50 Hz mains frequency. And to quote my old lecturer "... you can't have flicker" which quickly became that years most used phrase! I digress, twin tube fittings are or were produced with one tube leading the supply and one tube lagging thereby reducing the stroboscopic tenancies. If the light fitting has a modern electronic high frequency ballast fitted the stroboscopic is virtually eliminated.
The electronic switch starter replacements are more expensive but they do extend the tube life by reducing the starting filaments from burning out prematurely if the old type of starter fails.
Another point of interest before you all fall asleep with boredom... Have you ever wonder what other benefit there is with low voltage lamps on machine tools other than 12 or 24 Volts being considerably safer from electric shock than 230 Volt mains? Well the answer is that a filament lamp being a reasonable Wattage will have a larger thermal inertia thereby reducing a stroboscopic effect of the lamp.
One last point if your building your own LED lighting some LED's are designed with their own circuity to enable them to run off AC or DC supplies - Please be aware that if your supply is an AC type powered by a transformer, the stroboscopic effect could be considerable. Ideally your LED's ought to be powered by a DC regulated and smoothed supply. I've also experienced early GU10 LED replacements having major strobing effects.
I hope this throws some light on the subject! From an electronics & electrical engineer of many years standing!
587 forum posts
Edited By Nobby on 25/08/2014 14:57:21
2904 forum posts
I got sick of replacing the halogen lamps in our study. We have a dozen or so on one of those track things and they only seem to last a thousand hours or so - and cost an arm and ten legs. IKEA now sell LED replacements that are just as bright and use a fraction of the power (= heat and cost). They should also last considerably longer: 50,000h is often specified.
LEDs will soon replace those awful "low energy" CFLs too. There was a big song and dance about phasing out incandescent filament bulbs not so long ago but the CFLs did little more than reduce the cost of the electricity. Some of them take ages to warm up and wear out pretty quickly, like their larger cousins.
I'm looking forward to LED prices coming down. And hopefully we will see dimmable LEDs becoming more readily available soon. Already there are direct replacements for the T5, T8 etc tubes that can be driven by existing reactive and electronic ballasts.
The thermal time constant of a filament bulb is surprisingly short. You can pass music or voice audio from a bulb to a photocell. If you drive a bulb with a PWM voltage, the frequency needs to be in the hundreds of Hz before the lifetime is as good as what you'd get driving it with clean DC. That's because the lifetime of the filament is inversely proportional to something like the 5th or 6th power of the temperature (I forget which). The lifetime mechanism is fundamentally driven by evaporation of the filament, if that's the right word.
|Billy Mills||27/08/2014 01:28:27|
|377 forum posts|
The thermal time constant of the bulb is dependent on filament thickness and on the operating voltage. When talking pictures with optical soundtracks were developed it was quickly found that low voltage heavy filaments produced far lower hum levels. The Stroboscopic effect is wildly misunderstood. A real stroboscope produces a very brief flash- around a microsecond. A slight ripple at 100 Hz from incandescent or florescent lights will not "freeze" rotating parts but will produce a slightly blurred image at sync rates.There is a very clear difference between a moving part and a stationary one. If your visual accuity is so low then you should not operate a machine tool.
Incandescent bulbs appear to be very simple basic devices but they are somewhat deceptive. The amount of light for the electrical power input is very low - around 2%, most of the rest being heat. Changing the voltage reduces the lifetime extremely quickly, reducing the voltage and they can last much longer than the nominal 1000 hour life. Depending on the bulb type a 1% increase can reduce bulb life by 10 to 16%. A Photoflood bulb is essentially an over voltaged bulb, it has a life of around 50 hours. Decreasing the voltage is a bit of a miser's folly, light output falls off and you spend a lot more on electricity that you save on bulbs.
If you take a 100W incandescent, over 1000 Hrs that's £140 of electricity. Although a LED replacement will cost more than a 100W incandescent it uses 20-25% of the power so say £30-35 of electricity for the same amount of light but very few replacements is a very good deal.
contrary to popular myth, LEDS do get dimmer with time, their lifetime is rated to 75% of their initial output in some standards. This depends -as you might expect- upon how hard the LED is driven. Most LEDS can be expected to reach their quoted lifetimes however it is somewhat of a scandal that many CFL's fail far short of their rated lifetimes although reliability has improved.
My workshop has it's old 5' flouro's with localized LEDS, when the LEDs can match a 58W tube then they will be swapped but at the moment the replacements seem to need a bit more work.
Edited By Billy Mills on 27/08/2014 01:31:39
|Billy Mills||27/08/2014 02:18:18|
|377 forum posts|
Should have said 10 bulbs at 100W which is the kind of consumption that a lot of houses - mine included- runs as a lighting load. That's at 14p per KW/Hr. So if you pay £10 per replacement bulb after changing 10 bulbs you break even at 1,000 Hours then save around £100 per 1000 Hours and don't have to replace bulbs. You do loose all that heat however..
|Oompa Lumpa||27/08/2014 03:48:00|
|888 forum posts|
They are Philips branded and were £21 each - they fit right into the original fittings.
2904 forum posts
That price will be coming down soon and the LED replacements are improving rapidly If you look at the trade rags and see what is coming through from the suppliers, you can see that what we get will soon be a lot better - or they could be once they've recovered their pound of flesh from the retail market.
Please login to post a reply.
Want the latest issue of Model Engineer or Model Engineers' Workshop? Use our magazine locator links to find your nearest stockist!
You can contact us by phone, mail or email about the magazines including becoming a contributor, submitting reader's letters or making queries about articles. You can also get in touch about this website, advertising or other general issues.
Click THIS LINK for full contact details.
For subscription issues please see THIS LINK.