UK Vintage Radio Repair and Restoration Discussion Forum - View Single Post - Workshop: LED replacements for flourescent tubes.

mhennessy · 18th Nov 2017, 2:55 pm

I agree - I'd stick with what you have.

I'm not convinced that LEDs have better CRI than good florescent tubes. Maybe some do, but certainly not all. The CRI of a florescent tube is indicated in the spec - for example, an "840" tube is 8 on a scale of 0 to 9 (and the 40 means 4000 Kelvin). The cheapest tubes from Screwfix, etc, are 8. The tubes fitted originally might well be worse - they might not even have that rating scheme if they're really old, in which case, who knows what they are? Meanwhile, I can't recall seeing equivalent numbers on LED lamps until really very recently - they usually quote the CRI on a scale of 0 to 100 (if at all), and 80 should be the minimum to go for...

As your existing units use conventional ballasts, they're about as quiet as it gets in EMI terms. The only downside of conventional florescent lamps is the 100Hz flicker, which is very subjective.

Here, I have a number of conventional types, plus a HF ballast driving a pair of 80W T5 tubes immediately above the bench. I have to turn that off when lining up radios, or indeed anything that is sensitive. It runs at 100kHz approx (depends on the brightness setting), and the waveform applied to the tubes is several hundred volts. It's no wonder it causes problems.

I also have an LED strip under a shelf. This is fed from a DIY current source, using a conventional mains transformer and a linear regulator. It doesn't chuck out masses of light, but provides a reasonably OK background light and generates no noise at all.

With LED lights, it seems to be a binary choice between 100Hz flicker or RFI. I'm very sensitive to flicker, and I'd rather not have the RFI either. It wouldn't be so bad if you at least knew which of the two evils you were going to get, but there's no way of telling from the description or packaging - and that's a big problem. Lots of people are affected by flicker, even if they don't realise it - as such, there should be some way of identifying these types. Hopefully one day that will happen. In the meantime, your approach of buying a sample before buying a batch seems to the only option we have...

18th Nov 2017, 2:55 pm	#6
mhennessy Dekatron Join Date: Jul 2007 Location: Evesham, Worcestershire, UK. Posts: 4,244	Re: Workshop: LED replacements for flourescent tubes. I agree - I'd stick with what you have. I'm not convinced that LEDs have better CRI than good florescent tubes. Maybe some do, but certainly not all. The CRI of a florescent tube is indicated in the spec - for example, an "840" tube is 8 on a scale of 0 to 9 (and the 40 means 4000 Kelvin). The cheapest tubes from Screwfix, etc, are 8. The tubes fitted originally might well be worse - they might not even have that rating scheme if they're really old, in which case, who knows what they are? Meanwhile, I can't recall seeing equivalent numbers on LED lamps until really very recently - they usually quote the CRI on a scale of 0 to 100 (if at all), and 80 should be the minimum to go for... As your existing units use conventional ballasts, they're about as quiet as it gets in EMI terms. The only downside of conventional florescent lamps is the 100Hz flicker, which is very subjective. Here, I have a number of conventional types, plus a HF ballast driving a pair of 80W T5 tubes immediately above the bench. I have to turn that off when lining up radios, or indeed anything that is sensitive. It runs at 100kHz approx (depends on the brightness setting), and the waveform applied to the tubes is several hundred volts. It's no wonder it causes problems. I also have an LED strip under a shelf. This is fed from a DIY current source, using a conventional mains transformer and a linear regulator. It doesn't chuck out masses of light, but provides a reasonably OK background light and generates no noise at all. With LED lights, it seems to be a binary choice between 100Hz flicker or RFI. I'm very sensitive to flicker, and I'd rather not have the RFI either. It wouldn't be so bad if you at least knew which of the two evils you were going to get, but there's no way of telling from the description or packaging - and that's a big problem. Lots of people are affected by flicker, even if they don't realise it - as such, there should be some way of identifying these types. Hopefully one day that will happen. In the meantime, your approach of buying a sample before buying a batch seems to the only option we have...