I mean when turn on the power of UV transilluminator, the light-on lamp of 302 nm or 365 nm cannot be seen. Normally, I will see the UV lamp when light on for most of the UV transilluminators, including the old one in our lab.
There is a picture I found on their website. You can check it.

I used a photospectrometer to check the spectrum at UV lamp and black glass positions, and found there is visible light emitted from the lamps. When the lights go through the black glass filter, there is nearly only UV rays transmitted. I guess it’s why the image of DNA bands is with a better contrast. Other UV transilluminators may have a black glass that allows partial visible light to go through. Therefore we can see the lamps when power on. Your opinion about safety is helpful to me. Thank you. I will suggest our personnels to watch out the power button and obey an operation instruction. There is a metal frame with UltraSafe UV shield above the black glass. I think it is important for gel cutting.

If you look at EtBr’s excitation and emission spectrum, it is found the absorption is strongest near 300 nm. It is why most people use 302 nm UV to illuminate EtBr and cut bands. But you can see there is also absorption at 470~500 nm of EtBr which means if the intensity is high enough, the fluorescence is acceptable. 470~500 nm is blue spectrum, and many manufacturers use blue LEDs to make LED transilluminators to excite safe dyes. I used some of the LED transilluminators and the result is as you said so so for EtBr but wonderful for safe dyes. I tried UltraBright LED transilluminator and found the sensitivity for EtBr is a little bit better but acceptable. Maybe sensitivity 2~5 ng is enough for EtBr since only rare people need to observe less than 10 ng.

There are not so many LED transilluminators in market. I think LED visible light is much safer than UV. This is a video I found yesterday about [some] LED transilluminator.
http://www.youtube.com/watch?v=mjqBlxcQBds

Many researchers told me in US this instrument of UltraBright LED transilluminator is quite different from others like SafeImager or DarkReader and they don’t like to wear amber glasses to cut gel. Canalon said the difference is minimal. I am interested in these different opinions.

From the attached EtBr spectrum, it is possible to excite EtBr by VIS light, expecially blue wavelength. Band image relates to contrast. If intensity of blue light is high enough, the contrast of band image can be acceptable. We cannot say ONLY UV can be used for EtBr.

---

Attachments:

If you highly doubt about excitation at 500 nm and emission at 600 nm with a filter to cut 500 nm background to see the bands, you can simply check the mechanism of Sybr dyes. The difference between excitation and emission wavelength of SyBrSafe is even smaller. So it is possible for EtBr to excite with storng 500 nm light and observe the emission of 600 nm with an amber filter to cut the 500 nm background.

It is hard to say. From the spectrum of EtBr, it shows the response of 302 nm is approximately 5 times higher than that of 500 nm. For consideration of intensity of lights, UV lamptube is with diverse light but LED is with a more intense and focusing light. It means the convergence of light cone angle is smaller for LEDs. LED’s light flux is much higher than UV lamps. It is also hard to measure the luminance with a fair condition. The UV luminance meter is with a sensor compatible for UV and blue light needs the luminance meter with sensor suitable for visible light. So, I think if we want to judge the feasibility of 500 nm, just consider the contrast of bands and sensitivity of DNA gel. If the sensitivity is acceptable, it is not needed to use UV. The diffuculties of current LED transilluminators are, first, their LED intensity is not strong. Second, their amber filter is not optimized to cut the blue background.

Blue light can excite EtBr absolutely. Many people knows. All the issues are ONLY about contrast.