Lab lights, how much are you happy with lights in your lab

[sairfan1]

Recently i start feeling pain in my eyes specially when i sit on computer and work, I recently moved to a new house where i have energy savers installed,  it looks me like the reason is energy savers, some of my friends says they are not good in such environment i also saw some debates on internet regarding LED lights and energy savers.

Previously i had tube lights in my lab and i never had any issue, I'm thinking to change lights in my lab but before getting started i thought lets discuss with you guys' experience. 

what do you use for light in your lab, any good ideas to share, thanks.

[floobydust]

LED lighting is shit for task lighting. I haven't yet found something that works as good as incandescent.
Why? Newer is better?

It has to do with how the eye micro-focuses and locates, apparently warm-white LED's are better as there is more red in the spectrum that the eye needs for its servo systems.

I just got some 10W LED's from china and their light looks so bad, yellow fringing and I can't see anything well under them. It can be super bright but that means nothing.
It's the phosphor recipe I am noticing makes a huge difference. All the peaks in spectral output seem to clash with what the eye needs.
Next I'm going to try Cree or Bridgelux LED's and see how they do.

[rstofer]

You want to get lamps with a Color Rendering Index up in the 90s.  An incandescent is 100 and is the baseline for all color rendition measurements.

[richnormand]

Also be careful with LED conversion for your lab, thre are other points to consider too.

I changed several fluorescent tubes for LED conversion (not over the bench). Noticed a high noise floor on the spectrum analyser while reading a weak signal. Fired up the SDR and there was interference all the way to more than 150MHz.
The AM band was totally unusable, even for local station.
Used ferrite rings, grounded each light frame, capacitors, shielded chokes, etc... All helped a little but I ended up going back to the old tubes.

May not matter much for the average user, but in a lab setting it might ruin your day. Some DC converters for LEDs are really noisy.

For the bench I have some Phillips floods with CRI about 90. Seems to be OK. For the 3D microscope I went more blueish and bright as it helps resolution quite a bit, compared to tungsten illumination.

[3roomlab]

Recently i start feeling pain in my eyes specially when i sit on computer and work, I recently moved to a new house where i have energy savers installed,  it looks me like the reason is energy savers, some of my friends says they are not good in such environment i also saw some debates on internet regarding LED lights and energy savers.

Previously i had tube lights in my lab and i never had any issue, I'm thinking to change lights in my lab but before getting started i thought lets discuss with you guys' experience. 

what do you use for light in your lab, any good ideas to share, thanks.

are they point source or difused?
I think point sourced are generally problematic for eyes and looking at things long term.

[Ian.M]

Mains voltage LED lighting and most fluorescents are generally problematic in an electronics lab, because of the amount of RFI they emit.   Cheap and nasty LED lights may actually be quieter than expensive ones with good CRI.

Fortunately magnetic ballasts for fluorescent tubes are still available, which emit very little RFI.    Long tube 'daylight' fluorescents and lots of them is the way to go for general lab lighting.  The length of the tube reduces sharp shadows laterally, and if you have several tubes well spaced out side by side, you can also minimise front to back shadows. 

Add task lighting, chosen for its low RFI and good CRI on angle arms or goosenecks, or as adjustable spotlights under the front edge of the lowest shelf over your bench, and you'll be far more likely to be happy with your bench lighting.   

If you are doing sensitive  RF or audio stuff, and need a really low RFI environment, you may need to resort to halogen lighting, without dimmers, powered by a linear transformer or even 12V batteries, though 12V LED strip lights with integrated resistive droppers are also worth considering.

[3roomlab]

Fortunately magnetic ballasts for fluorescent tubes are still available, which emit very little RFI.    Long tube 'daylight' fluorescents and lots of them is the way to go for general lab lighting.  The length of the tube reduces sharp shadows laterally, and if you have several tubes well spaced out side by side, you can also minimise front to back shadows. 

I agree
T5 4 foot is over 2500 lumens per tube @ 40w.
a good LED setup could do that for maybe <20w, but I have not seen any commercial product do this well out of box and well diffused.

[wilfred]

There is an old teardown Dave did here https://www.eevblog.com/forum/blog/eevblog-533-led-fluoro-tube-teardown/

This thread prompted me to go back and watch the video again because I use one of the exact same tubes and it produces an annoying whine. Not near 100Hz and I definitely cannot hear 16KHz as  I saw in the teardown.

I also note someone else reported the noise too.

I just attached a cord directly to the end of the tube so no fitting or old ballast. The noise comes from the tube. So that is something to consider.

I mostly use two DIY lights made from 5 and 4 compact fluorescent bulbs.

It is not to be underestimated having super bright even light on your bench.

[David Cutcher CEG]

I'm surprised that nobody (so far) has mentioned color temperature.
https://en.wikipedia.org/wiki/Color_temperature
 And do you have an outside window (sunlight) in your new space? Focus appears to be on the background hum of the ballasts.
As a shop teacher, I insist on quality full spectrum fluorescent lights, if only to read the resistors. I have no problems, and neither do my students. They do comment about lighting in other classrooms. I have as close to full spectrum LEDs in my home office. Only the best. No headaches.
David Cutcher "Certified Evil Genius"

[Bud]

I settled on 4000K for the shack. However i found that the light quality also depends on the color the walls are painted. My  shack is painted in sunny yellow, so lighting it with a 4000K LED fixture produces comfortable environment. I tried the same lamp in the hall painted in very light cyan and Gosh,  it was terrible. There something like 2700 ...3000K would work better.

[sairfan1]

do you have an outside window (sunlight) in your new space?

Yes i have a reasonable size window, and it really helps me while day time if its sunny my eyes feel really good, but i mostly work in the night that's why its a big issue for me.

However i found that the light quality also depends on the color the walls are painted.

I was conscious about it, that's why specially i choose white color for my labs.

[soldar]

I always had a couple pantograph lamps I could move around and place as needed. They get a lot of use and movement.

I used incandescent but in hot weather they give off too much heat.

Compact fluorescents were OK.

Finally I have transitioned to LEDs and I am extremely happy with the light intensity and color and they give off little heat.

[dnwheeler]

DIY Perks (YouTube channel) has a project making LED "studio" lights (but generally useful, too). They have a high CRI and adjustable color temperature, and can be made fairly inexpensively.

[3roomlab]

DIY Perks (YouTube channel) has a project making LED "studio" lights (but generally useful, too). They have a high CRI and adjustable color temperature, and can be made fairly inexpensively.

youtube.com/watch?v=DhbMnQt14_o]https://www.youtube.com/watch?v=DhbMnQt14_o[/url]

*rephrased
I think the video is misleading
@ 5:15, the video shows what appears to be a very very very high end looking LED spectrum, sun like wavelength density.
and then at that moment he seems to suggest that is the LED sold on the ebay listing
a spectrum of such density is likely CRI98, but I think it is impossible for such a LED to be 120-140lm/W claimed by the ebay seller
if he claimed 95+ CRI without showing the spectrum, the china LED specs of 120-140lm/W might be plausible
but CRI 95+ in order to solve what he calls a "green" face, may or may not be the final outcome.
to shift into rich red spectrum, the led must have red peak at 630-650nm. not all CRI95+ have this peak. and the ebay seller does not display the spectrum.

(this is the spectrum capture of my cfl light, a common cfl. nothing special)
disclaimer : I am only a noob at lighting, but I cannot un-see this spectrum he showed

[floobydust]

CRI is just a number, a bunch of math that can be fudged. It is not great with spikey (LED) spectrum. High-CRI does not work well with film photography but the number says it should.

The DIY fixture is two sets of three LED strips with different colours (ice blue, warm white, daylight pure white), each with a PWM dimmer. Not very bright, and better for background lighting, not task lighting.

[3roomlab]

CRI is just a number, a bunch of math that can be fudged. It is not great with spikey (LED) spectrum. High-CRI does not work well with film photography but the number says it should.

The DIY fixture is two sets of three LED strips with different colours (ice blue, warm white, daylight pure white), each with a PWM dimmer. Not very bright, and better for background lighting, not task lighting.

I remember reading that the standards body has defined a new color standard to replace the CRI, the TM-30-15.
TM-30-15 depends on a 99 color spots, CRI is up to 16.
manufacturers refused to use the new rating
I think it is because with CRI, it is easy to "cheat" buyers. the new rating will also cause many of the existing products to fall into mediocre low range.
this in turn can be seen in the actual pdf at say digikey, I have seen not one using TM-30-15
there are only so few pdf that will show you actual spectrum, and there are always the same few manufacturers that show intentional vague data of the spectrum (thick unrealistic plots)

[David Hess]

Long tube 'daylight' fluorescents and lots of them is the way to go for general lab lighting.  The length of the tube reduces sharp shadows laterally, and if you have several tubes well spaced out side by side, you can also minimize front to back shadows.

I agree; I much prefer linear fluorescent tubes to any other form of lighting for the reasons you give.  In addition, desk mount fixtures which hold two 2 foot tubes for close in lighting are available and tubes are available with whatever color rendering you prefer.  I find that a mixture of cool and warm tubes works best but there are lots of options.

Power line EMI has become so strong that I do not notice any addition from electronic ballasts for linear fluorescent tubes.  The designs I have inspected use soft switching topologies like resonate Royer converters so maybe that is why.

[HB9EVI]

for years I used a lamp with G23 socket and fluorescent bulb what often required to be turned off when it interfered too much with the scope; then I found led bulbs matching for the socket - what a surprise: EMI was even worse than with the fluorescence bulb. just recently I decommissioned it and got me a led lamp with integrated magnifying glass - which is free of EMI and does a good light (day light 6600K)

[schmitt trigger]

Back in the days, when all ballasts were magnetic, the thumb rule was to always use a dual-tube ballasts.

The reason was that fluorescent tubes flickered tremendously, which could strain your eyes.
On a dual-tube ballast, one of the tube currents was phase sifted with the aid of series caps, such that as one tube was going down in brightness, the other was reaching peak luminosity.

This was eliminated with high frequency ballasts in the 90s. However the newer cheap designs do away with most if not all the bulk powerline filtering, and therefore the high frequency carrier is modulated with powerline frequencies.

[akowalczyk]

Cheap LED strips on Ebay can work great, they aren't a point source so you have less shadows. Figuring out a fixture to hold them can be a bit of a problem though, my roomate and I came up with a PVC frame held up by thick magnet wire and wall hooks. Only fell down a few times...and fortunately never when we were soldering

[sairfan1]

can you please share some link what kind of those one, (As there are a lot of led strips now a days I'm not taking risk that i buy some different with poor result)
do they get hot with time?

thanks,

[Psi]

i have 2x 50W  95%CRI YUJI leds in my smallish lab.

Plenty of light and good quality light too.


If you think 100W of light is overkill for a small lab, you're wrong.
This would be overkill
https://store.yujiintl.com/collections/frontpage/products/yujileds-bc-series-high-cri-cob-led-900h-1500w-pack-1pcs

[floobydust]

i have 2x 50W  95%CRI YUJI leds in my smallish lab.

What format bulb are you using? Are you using just their high CRI emitters on your own heatsink etc.? The Yuji LED's phosphors appear much better.
I'm a bit tired of pissing around with 10W-50W LED's mounting them on a heatsink, adding lens etc. and trying to beat an Ikea desk lamp 
Yuji wants $499 USD for a desk lamp.

I've been trying GU10's with E17 adapters and such. But the GU10 is so narrow beam, it's like a flashlight.
Something that could mount on 2020 hardware would be great for making LED lighting.

[Psi]

i have 2x 50W  95%CRI YUJI leds in my smallish lab.

What format bulb are you using? Are you using just their high CRI emitters on your own heatsink etc.? The Yuji LED's phosphors appear much better.
I'm a bit tired of pissing around with 10W-50W LED's mounting them on a heatsink, adding lens etc. and trying to beat an Ikea desk lamp 
Yuji wants $499 USD for a desk lamp.

I've been trying GU10's with E17 adapters and such. But the GU10 is so narrow beam, it's like a flashlight.
Something that could mount on 2020 hardware would be great for making LED lighting.

yep, just two 50W cob emitter on a large heatsink and slow speed fan. I have a 30cm x 30cm diffuser sheet mounted over it to make it look a bit better and prevent it casting hard shadows.

[Housedad]

My lab is in my basement.  There already were six fixtures with (4) cool white t-12 bulbs each, and I added six 4 ft daylight t-8 2 bulb shop lights over the benches. 

I really like the effect even though my wife says that if I add any more the house will start to fly from the photon pressures. 😎