Replacing Neon Lamp with LED

[BrokenYugo]

I've found running two LEDs in parallel, with the polarity flipped, provides less flicker and each one protects the other from excessive reverse voltage.

Pretty sure I've seen LEDs with two die that do this internally

Interesting.
I've never seen such LEDs myself, but often optocouplers. Do you have lniks/references? Thanks.

obsolete but; https://www.elfadistrelec.dk/Web/Downloads/86/20/Lx-bipolar_eng_tds.pdf

if you can live with the mixed color there's crap tons of dual color leds that'll work

They are not entirely obsolete, e.g. https://www.mouser.com/c/optoelectronics/led-lighting/led-emitters/standard-leds-through-hole/?q=bipolar%20LED&instock=y

I've not tested it but I would think such a bipolar LED running straight off 240VAC with just a resistor would flicker less than a neon, as the LED conducts and makes some light nearly to the zero crossing, whereas the neon cuts out entirely by something like 40-50V, EDIT: and doesn't light again until the voltage raises to the strike voltage of something like 80-90V.

[rfclown]

People don't have the same eyes.
Tip: try holding the LEDs in your peripheral vision instead of looking directly at them. Odds are you'll see the flickering. There are also people that see 100 or 120 Hz flickering directly in their field of vision.

I would think it would be highly dependent on the LED. The ones I'm using emit some light with very little current. I put 3V DC across the circuit and I still see some light. I have LEDs in my junk bin at home from when I was a teen in the 1970's which would probably have barely perceivable light with 1mA and would certainly flicker. There's no reason to use LEDs like that anymore. I'll try looking at it in the dark to see if I can notice flickering. I'm actually at work now and I had it laying on a workbench in a brightly lit room.

[floobydust]

A neon lamp on mains does flash, each electrode (cathode) lit on alternating 1/2 cycles. youtube vid high speed camera 1,000fps in slow motion showing that.

LED nightlight, typically 100R 1/4W surge resistor with 1uF and 120VAC, commercial products use carbon resistors but I wasn't comfortable with their flammability.
I changed to 1/4W fusible and it works... until... if you switch on at a mains peak it fused. Such a hassle.

[james_s]

Neons do flicker, but to my eye it is not nearly as visible as it is with LEDs, probably down to the fact that most are single die and used without a bridge rectifier. Either way it doesn't take much capacitance to fix that.

Carbon resistors will typically burn out without actually setting anything on fire. Especially if they are heavily overloaded they just pop.

Side note about neons, if you run them from DC only the cathode will illuminate, and lifespan is significantly reduced.

[paul_g_787]

People don't have the same eyes.
Tip: try holding the LEDs in your peripheral vision instead of looking directly at them. Odds are you'll see the flickering. There are also people that see 100 or 120 Hz flickering directly in their field of vision.

I definitely can see it (100Hz flicker). I have to look really hard in the daytime to see it but I can just about. However in the dark I can see it flicker clear as anything!

[james_s]

100 Hz flicker is very, very obvious to me, it's something that really jumped out when I visited England. Part of that is the lower frequency and part of that is probably that I'm used to 120 Hz flicker, it got less noticeable after a day or two but motors and transformers still all sounded weird to me. My friend from there is over here somewhat regularly for business and has said the same thing about the motors and transformers here sounding weird.

I can see flicker at much higher frequencies than that though, 300-400 Hz PWM backlight flicker gives me a headache and I can see it in my peripheral vision, especially if I'm tired. The 1 kHz flicker of Philips Hue lamps is visible any time I move my head or eyes, that took some getting used to at first, now I can mostly tune it out. The PWM flicker of many car tail lights is very visible to me too, again largely down to motion. When people say that the human eye cannot perceive flicker above x frequency they are ignoring the fact that a person is rarely sitting still just staring into a light source. The moment you move your eyeballs to look at something you turn a flickering light into a dotted line and that strobe effect is readily visible well into the kHz range.

[StillTrying]

Some of us must have fast eyes,   I can easily see 50Hz & 100Hz flicker, and the 25 frames per second of a TV picture even though the backlight doesn't flicker
I sometimes wonder how lip readers manage off a TV screen.

[james_s]

Many modern TVs do have backlights that flicker, it is synchronized to the video frame rate.

The first time I ever saw a PAL-50 CRT TV was about 20 years ago when I was involved in the original Xbox game console. We had a TV shipped over from Germany to test something and the flicker looked so bad I couldn't see how anyone could watch it. I suppose you get used to it after a while and don't even notice anymore.

[rfclown]

So I compared a neon with the LEDs that I wired with a resistor. The neon had a 68k soldered to the lead. My back-to-back LEDs have 47k x 2=96k. I looked at them in the dark. LEDs are WAY brighter than neon. My 60 year old eyes don't see flicker. I tried looking in my peripheral vision. Don't see flicker. If I move me head back and forth I can't see flicker. If I wave them back and forth rapidly I can distinctly see the stobe effect. The on/off duty cycle of the neon and LEDs seem similar. So for the OP's problem, I don't see anything more necessary than back-to-back LEDs with a resistor. If some some people notice the flicker (which is obviously true based on this post) then I wouldn't think it needs be any worse with LEDs over neon (as long as appropriate LEDs are chosen). With a large series resistor there isn't any of the problems associated with capacitor/recitifier method. I wondered if the LEDs would excite glow-in-the-dark phosphor, so I put them behind a glow-in-the-dark star that was on the ceiling in the kids room. No luck. I thought that might me a good way to smooth out flicker. Maybe with the right color LED and the right phosphor? One of my LEDs is white and the other is yellowish.

[Vovk_Z]

So I compared a neon with the LEDs that I wired with a resistor. The neon had a 68k soldered to the lead. My back-to-back LEDs have 47k x 2=96k. I looked at them in the dark. LEDs are WAY brighter than neon. My 60 year old eyes don't see flicker.

I told about this in Reply #41.

[Benta]

So for the OP's problem, I don't see anything more necessary than back-to-back LEDs with a resistor.

The OP doesn't have a problem, it's just about playing around and learning things.

I wondered if the LEDs would excite glow-in-the-dark phosphor, so I put them behind a glow-in-the-dark star that was on the ceiling in the kids room. No luck. I thought that might me a good way to smooth out flicker. Maybe with the right color LED and the right phosphor? One of my LEDs is white and the other is yellowish.

White LEDs already have a phosphor layer. The chip is mostly blue in cheap devices and excites a yellow(ish) semi-transparent phosphor. To do this, the exciter wavelenght has to be shorter than the phosphor wavelength (=higher energy), so your experiments were futile.

More advanced LED technologies use UV chips, which gives more possibilities on phosphor mix and thus better color rendering.

[tooki]

Many modern TVs do have backlights that flicker, it is synchronized to the video frame rate.

The first time I ever saw a PAL-50 CRT TV was about 20 years ago when I was involved in the original Xbox game console. We had a TV shipped over from Germany to test something and the flicker looked so bad I couldn't see how anyone could watch it. I suppose you get used to it after a while and don't even notice anymore.

It was a big enough problem for enough people that high-end PAL TVs doubled the frame rate to 100Hz, presumably using some form of buffers to show each frame twice. That must have been quite a technical achievement before the days of digital video! Later ones used digital buffers, and Philips even made some that were actually 75Hz TVs using motion interpolation similar to what modern TVs use.

[tooki]

More advanced LED technologies use UV chips, which gives more possibilities on phosphor mix and thus better color rendering.

Do you know of any examples of UV-based LEDs in the wild? I’ve heard of them as a theoretical, but have never found one actually on sale. And given how inefficient UV LEDs still are, it comes as no surprise that blue is the foundation for modern LED illumination. Certainly the high-CRI white LEDs are not UV-based; you can get high-CRI remote phosphor panels, and they’re designed for blue LEDs.

[Zero999]

I wondered if the LEDs would excite glow-in-the-dark phosphor, so I put them behind a glow-in-the-dark star that was on the ceiling in the kids room. No luck. I thought that might me a good way to smooth out flicker. Maybe with the right color LED and the right phosphor? One of my LEDs is white and the other is yellowish.

White LEDs already have a phosphor layer. The chip is mostly blue in cheap devices and excites a yellow(ish) semi-transparent phosphor. To do this, the exciter wavelenght has to be shorter than the phosphor wavelength (=higher energy), so your experiments were futile.

Most glow in the dark phosphors are green and can be excited with blue light. I've not had any problems with glow in the dark stuff not glowing, when exposed to LED lights. I discovered this back in the 90s, when I got hold of some of blue LEDs, when they first became available, albeit very expensive.

More advanced LED technologies use UV chips, which gives more possibilities on phosphor mix and thus better color rendering.

Do you know of any examples of UV-based LEDs in the wild? I’ve heard of them as a theoretical, but have never found one actually on sale. And given how inefficient UV LEDs still are, it comes as no surprise that blue is the foundation for modern LED illumination. Certainly the high-CRI white LEDs are not UV-based; you can get high-CRI remote phosphor panels, and they’re designed for blue LEDs.

I don't know of any, but it's often not specified. I suppose the advantage of using UV is the phosphor's emission becomes more dominant.

Modern deep violet/near UV LEDs are very efficient, perhaps often more so than blue. This data sheet gives an efficiency of 61.7% for 385nm and 395nm LEDs, vs 57% for blue 450nm.
https://www.mouser.co.uk/datasheet/2/810/NewEnergy_StarBoard_Horticulture_Luminus_DataSheet-2509530.pdf

On the other hand, the higher the energy of the radiation exciting the phosphor, the greater the Stokes shift losses, so UV would be a net loss, efficiency-wise.

[tooki]

More advanced LED technologies use UV chips, which gives more possibilities on phosphor mix and thus better color rendering.

Do you know of any examples of UV-based LEDs in the wild? I’ve heard of them as a theoretical, but have never found one actually on sale. And given how inefficient UV LEDs still are, it comes as no surprise that blue is the foundation for modern LED illumination. Certainly the high-CRI white LEDs are not UV-based; you can get high-CRI remote phosphor panels, and they’re designed for blue LEDs.

I don't know of any, but it's often not specified.

No, but the forward voltage is specified, and it seems to always match that of blue LEDs. The Vf of UV LEDs rises dramatically with wavelength. A true UV-A emitter from a respectable source (e.g. Osram) has a single-die Vf of 4V (±0.5V), and UV-B and -C LEDs from 5V to upward of 7V!

Even just the ~400nm near-UV LEDs have noticeably higher Vfs than blue LEDs.

I just took a look on Digi-Key, and there is exactly one line of white LEDs containing models whose Vf is high enough (and the drawing explicitly showing it having just a single die, not multiple in series) that I thought it had to be real UV: the Lumex Titanbrite line, whose 5W (SML-LXL99UWC) and 3W (SML-LXL8047) models have a Vf of 4.5-4.6V. The 405nm near-UV version of the SML-LXL8047 line has a Vf of just 3.6V. But the blue (454-465nm) versions of both models also have exactly the same 4.5-4.6V Vf, so I guess they, too, are blue-based.

Hence my skepticism that any UV-based phosphor-converted LEDs are actually on the market.

[james_s]

I'm reminded of a curious effect I noticed some time back when I first built a violet laser pointer after HD-DVD lost the format war and pickup units became available cheaply. In my house the cover plates on the outlets and switches are made of a hard white plastic and when excited by the violet laser it fluoresces light greenish. It even has some persistence, if I sweep the beam across a switch plate it will leave a green trail that fades out over a second or so. I have no idea why those plates fluoresce, they are not glow in the dark. Maybe it's part of the dye that makes them white? I don't know.

[rfclown]

I wondered if the LEDs would excite glow-in-the-dark phosphor, so I put them behind a glow-in-the-dark star that was on the ceiling in the kids room. No luck. I thought that might me a good way to smooth out flicker. Maybe with the right color LED and the right phosphor? One of my LEDs is white and the other is yellowish.

White LEDs already have a phosphor layer. The chip is mostly blue in cheap devices and excites a yellow(ish) semi-transparent phosphor. To do this, the exciter wavelenght has to be shorter than the phosphor wavelength (=higher energy), so your experiments were futile.

Most glow in the dark phosphors are green and can be excited with blue light. I've not had any problems with glow in the dark stuff not glowing, when exposed to LED lights. I discovered this back in the 90s, when I got hold of some of blue LEDs, when they first became available, albeit very expensive.

Put a blue LED behind a glow in the dark star (greenish) and sure enough there was some persistance.