Twin Tube HF fluorescent lighting for the workshop

[John HaineParticipant @johnhaine32865]

Adam, is the earth connected to any of the lights? If it isn't the only way there can be an imbalance between L and N is via an earth fault in your wiring. If the lights do have an earth then sometimes a filtering capacitor can create enough of an imbalance to trip an rcd. So if there isn't an earth check your wiring carefully as it might be dangerous.

[Adam HarrisParticipant @adamharris13683]

Hi John, yes all lights connected to L,N and E. I am going to connect each of the 5 to their own on/off switch and will revert with the result for those interested. Makes sense to do that anyway I suppose as it means I don't have to waste energy on having all twin lights on at the same time. Furthermore they are also VERY bright and really too bright for all 5 on at the same time ( but correct level of brightness per machine/work area over which they are suspended).

[Andrew TinsleyParticipant @andrewtinsley63637]

HF ballasts operate in the kilohertz regions. They are usually a multivibrator circuit. This means that the resulting hash can go up into the medium and even VHF wavebands. This is why I don't like them. I can't listen to my radio because of the interference.

Andrew

[MuzzerParticipant @muzzer]

If your HF ballasts screw up your radio it suggests you should have bought them from a more reputable outlet. It's illegal to sell stuff that doesn't meet the EMC requirements (look at EN55011). Are you using Chinesium crap?

You will also note that despite what you say, the colour temperature is usually specified for fluorescent tubes.

And as mentioned on various previous occasions, LED lights are generally about the same as fluorescent lights in terms of efficiency.

The of the key benefits of HF ballasts is that they turn on almost instantaneously and can coax a little more life out of the tubes before they give up the ghost. As a tube gets older, the voltage rises until it can't be kept running.

For me, one key benefit of LED lights in the workshop is the lack of glass tubes above head height. Belting one of those or dropping one when changing it on top of a ladder could be pretty messy.

Some HF ballasts (and many LED drivers) have very little inherent energy storage, so although it's true that they switch at several kHz, they often have a significant 100Hz component.

Murray

[HughEParticipant @hughe]

I agree with Muzzer. If the correct spec'ed device is used then there should be no interference to other electronic equipment. Note that there are two version one for commercial and one for residential use. The commercial can give problems in a residential environment as the EMI protection is not as good. EMI is both conducted and radiated so if you have a problem the you to check which is causing the problem. Reading the tech advice from GE suggests you should keep electronic equipment as far away as possible, which may not be possible in a home workshop.

Also the earthing is important. If you have as high impedance earth then the light fitting could be acting as an aerial, if metal.

Hugh

Edited By HughE on 12/10/2018 21:26:35

Edited By HughE on 12/10/2018 21:27:04

[SamsarandaParticipant @samsaranda]

Muzzer, your comment about “Chinesium Crap” is a bit harsh as nowadays it is almost impossible to buy equipment that is either not a Chinese product or contains some of its basic elements that have been produced in the orient. Can you name any fluorescent light units that do not contain anything produced in the Far East?

Dave W

[Andrew TinsleyParticipant @andrewtinsley63637]

hello Dave,

yes I should have said 100hz flicker for magnetic ballasts. This frequency is way above what the average human eye can see. I don't have information to hand about the subliminal effects, so can't comment.

I have examined HF ballasts which operate from as low as 4kHz to near 50 kHz. The basic problem is that the circuits are mainly multivibrator. This leads to a sharp edge on the cut off which produce harmonic oscillations into the mHz region, thus producing copious interference in the RF spectrum. Few HF ballasts have sufficient RF suppression. As Muzzer said, they are cheap Chinesium crap, as opposed to the expensive Chinese stuff manufactured for reputable European majors

Interesting that you sited the GE blurb. I had a hand in writing it!

Andrew.

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Andrew Tinsley on 13/10/2018 13:18:05:

hello Dave,

yes I should have said 100hz flicker for magnetic ballasts. This frequency is way above what the average human eye can see. I don't have information to hand about the subliminal effects, so can't comment.

[ … ]

.

The ergonomic problem [as distinct from the strobe-effect on rotating machinery] is not so much related to what we can see, but to how the eye/brain reacts to off-axis stimulation.

Common 'good practice' in large offices was to install fluorescent fittings alternately 'end for end' to smooth-out the intensity fluctuations. … Another solution [useful when appropriate] is to use 'circular fluorescents' which, of course, put the two ends of the same tube nearly coincident.

MichaelG.

[Andrew TinsleyParticipant @andrewtinsley63637]

Hello Michael,

I can't see what point you are trying to make. The distribution of fluorescent lights in offices are indeed as you describe. This is simply to even out illumination and attempt to produce shadow free lighting. It has nothing to do with flicker. Are you saying that such a layout minimises any subliminal effects?

I would be interested in further information if you are claiming that it reduces subliminal effects. I am always willing to learn. I have never come across such information, but that only means I have probably missed it.

Andrew.

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Andrew Tinsley on 13/10/2018 14:05:45:

Are you saying that such a layout minimises any subliminal effects?

.

In a word … Yes

The subliminal effects are reduced, simply because the ON end of one tube is adjacent to the OFF end of the other tube so the 'microclimate' [for want of a better term] is illuminated by the average of the two.

MichaelG.

.

Reference source, for the present, is only my personal recollection from BSc Ergonomics.

… but I will try to locate something more authoritative over the next few days.

.

Edit: Hopefully, Dave [S.O.D.] will find two photocells, and a dual trace 'scope … to demonstrate that the two ends of the tube flicker in anti-phase.

Edited By Michael Gilligan on 13/10/2018 14:49:43

[Andrew TinsleyParticipant @andrewtinsley63637]

Thanks for that Michael,

You may well have something. But I am still very dubious. The 100Hz phases for two lamps are always in phase, they have to be because the same 50Hz voltage is driving both of them, For what you are claiming, the voltage needs to be 90 degrees out of phase, between the two lamps.

Maybe I am missing something? I am not trying to cause dissension, just curious. There is too much arguing on forums, but not too bad here!

Andrew.

[Neil WyattModerator @neilwyatt]

Posted by Andrew Tinsley on 13/10/2018 16:06:57:

Thanks for that Michael,

You may well have something. But I am still very dubious. The 100Hz phases for two lamps are always in phase, they have to be because the same 50Hz voltage is driving both of them, For what you are claiming, the voltage needs to be 90 degrees out of phase, between the two lamps.

Maybe I am missing something? I am not trying to cause dissension, just curious. There is too much arguing on forums, but not too bad here!

Andrew.

I think Michael is suggesting that the lamp is brighter at one end than the other depending on the direction of electron flow. By placing tube L:N then the flow in adjacent tubes will be in opposite directions, 180 degrees out of phase.

I looked to see if there were any slow-motion videos but they aren't slow enough.

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Andrew Tinsley on 13/10/2018 16:06:57:

Thanks for that Michael,

You may well have something. But I am still very dubious. The 100Hz phases for two lamps are always in phase, they have to be because the same 50Hz voltage is driving both of them …

.

Andrew,

I think what you missed might be that I carefully said that the fittings were swapped end-for-end.

… this 'mechanically' introduces a phase shift.

Like Neil, I have thus far failed to find video or similar evidence to support my case.

MichaelG.

[Andrew TinsleyParticipant @andrewtinsley63637]

Hello Neil,

Now I am really confused! As soon as current flows (electrons one way and ions the other!). The lamp emits light from end to end. The idea that one end of the tube is brighter than the other is incorrect. So I don't think that this is Michael's point (?). I understood him to infer that the 100Hz flicker would be out of phase at adjacent ends of tubes in a row. Thus cancelling out the 100Hz flicker. The adjacent ends of the tubes are in fact in phase with each other so I cannot understand how this would reduce the 100Hz flicker.

As I said before, I could well be missing the point.

Andrew.

[Frances IoMParticipant @francesiom58905]

I suspect in a large office the lamps must be fed from all three phases to even out the load which might even out flickers.

Many years ago I wrote the software to allow control of each individual light fitting by dialling a specific number from a specific phone that had been associated with that fitting – saved a lot of money (peak rate electricity to take away heat of lamps) + office occupants were very happy as I suspect the susceptibility to seeing flicker is more common than suspected

Edited By Frances IoM on 13/10/2018 17:39:41

[Michael GilliganParticipant @michaelgilligan61133]

I don't think this either supports or refutes my proposition, but it's interesting: **LINK**

http://www.giangrandi.ch/electronics/fluorescenttubes/fluorescenttubes.shtml

MichaelG.

.

More good stuff here: http://edisontechcenter.org/Fluorescent.html

… but still no simple confirmation 

Edited By Michael Gilligan on 13/10/2018 18:16:42

[Neil WyattModerator @neilwyatt]

Posted by Andrew Tinsley on 13/10/2018 17:36:01:

Hello Neil,

Now I am really confused! As soon as current flows (electrons one way and ions the other!).

The current flow is almost all electrons from cathode to anode. Presumably they need some distance to build up speed and the anode end is brighter.

[Andrew TinsleyParticipant @andrewtinsley63637]

The use of three phase in an office would go someway to reducing 100Hz flicker. Good point Frances! However I think the flicker most if not all people see with choke fed fluorescents is NOT 100Hz, but flicker due to near end of life effects.

Andrew.

[Neil WyattModerator @neilwyatt]

No, I've found it on Wikipedia.

Under Electrical Ballast:

"In ballasts that control two or more lamps, line-frequency ballasts commonly use different phase relationships between the multiple lamps. This not only mitigates the flicker of the individual lamps, it also helps maintain a high power factor. These ballasts are often called lead-lag ballasts because the current in one lamp leads the mains phase and the current in the other lamp lags the mains phase. "

Under Fluorescent Lamp:

"Even among persons not sensitive to light flicker, a stroboscopic effect can be noticed, where something spinning at just the right speed may appear stationary if illuminated solely by a single fluorescent lamp. This effect is eliminated by paired lamps operating on a lead-lag ballast. "

[EmgeeParticipant @emgee]

As Neil says Thorn lighting used to sell matched lead/lag flurouscent lights to not only improve PF but also reduce strobe effect.

In the 1970' and 80's most electrical specifications for factories where machinery was to be be used stated the need to connect lighting in large areas using all 3 phases to reduce strobe effect.

Emgee 

Edited By Emgee on 13/10/2018 19:03:27

[Frances IoMParticipant @francesiom58905]

in the very large open plan offices in the late 70s and 80s tubes were changed on a regular schedule irrespective of use but designed so that all tubes were replaced well before end of life – one feature I was due to add to control software was to report on actual usage of each tube but I left the consultancy before it was approved. I suspect that the cost of tubes was small wrt labour cost to change them and that a fixed sweep through a floor was cheaper to administer. In those days complete floors of dealers would be switched around between European, Japanese + American business hours over a night with partitions etc also moved – all our testing was done between 1am and 3am.

Edited By Frances IoM on 13/10/2018 20:01:50

[Adam HarrisParticipant @adamharris13683]

For those interested in the tripping of the 30mA RCD problem, I have now independently switched all 5 twin HF 70W "Viper" fluorescents rather than wiring in parallel and if switched on one after the other, they work absolutely fine with no RCD tripping. Thus it must be the start up surge that trips the RCD when more than one is started at the same moment. There is no evidence of running earth leakage from 700w of this fluoresecent lighting which was my other concern.

[Adam HarrisParticipant @adamharris13683]

They are extremely good and cheap lights

[JonParticipant @jon]

Glad your sorted Adam.

For some time, decades have run cheap fluorescents likes of Wickes from a 4 bank switch.
Rear of partition of shop on 1 switch
Right side a single on 1 switch
Left side now a twin on 1 switch
Left further down on another switch.
Loft is linked in parralel off one of the feeds via a pull cord.
Never ever tripped a 6A MCB but the invertor and compressor did multiple times an hour on different circuits.

Only time notice any flickering whatso ever is when a tubes on its way out.
Also i dont buy rotational argument with what appears stationary, got a piccy some where from 16 yrs ago chuck spinning 1000rpm, quite blatant.
24 yrs ago an 100hz telly new out then you could detect strobing by waving your hand across. 50hz real easy to see.

Have seen some sellers stating the K rating for fluorescents, lower the softer whereas high 6000K its whiter and 'seemingly' brighter along hurtful eyes and shadowing with led! When a tube goes i just pop down to local hardware and buy one there and then paying over the odds around £4.

[Author]

Posts