Retrofitting T8 tube fixtures to LED

[Fly-Guy5]

I will be retrofitting existing T8 fixtures to new LED tubes and pulled the following up during my research. Have any of you dealt with this issue?

"There is a deeper and more dangerous issue here that few people are aware of. My comments here specifically apply to tubes which are intended to be directly connected to a 120vac source (or branch circuit in UL terms). The tombstone or lampholder MUST BE RATED FOR DIRECT CONNECTION TO A BRANCH CIRCUIT. MOST ARE NOT!!! As of this writing, I don’t believe Leviton has a single holder that is rated to connect to a branch circuit directly. IT IS NOT SUFFICIENT TO SAY THAT BECAUSE A HOLDER IS NON-SHUNTED AND RATED AT SAY 600v AND 600w, it can be used. I have found only 1 manufacturer of holders(Stucchi 3000 series of Italy) that has tested and, in writing, certifies it can be used for DIRECT CONNECTION TO A BRANCH CIRCUIT(although I’m sure there are others since folks are selling retrofit ‘kits’ that contain the tube, holders, and a label) Look at the installation sheet of ANY Leviton tombstone, and the 2nd bullet in the upper left hand corner states it’s not intended for connection to a branch circuit. [The technical reason is that an arcing and sparking hazard exists when the tube fails in a way that produces a partial for full short circuit and high currents are drawn through the tombstone contact where it touches the pin. There are also issues with the physical installation of the tube and exposure of ‘live’ contacts to the installer or the luminaire]Keep this in mind if you choose to do a retrofit using direct connect tubes, or if you hire a contractor to do it for you! You don’t want an accident in your facility number one, and secondly don’t want to find out that your insurer won’t cover you because you did not have the proper tombstone!"

 

[al hildenbrand]

May I suggest googling "ballast bypass t8 LED", or "direct wire t8 LED", or "ballast free t8 LED" and explore the returns.

There are a lot of non-shunt tombstones, of varying heights, available.

FYI: Many direct-wire T8 LEDs are sold as kits with tombstones included, but they are generally very short and may be unsuitable for your specific fixture.

 

[ksmith846]

There is T-8 LED lamp that works with the ballast still wired......no retro fitting work required......

here's a link

http://gc-lighting.com/categories/tube/

 

[mtfallsmikey]

Where did this article come from?

 

[al hildenbrand]

Where did this article come from?

It appears to be a comment on 3/22/17 3:39 PM near the bottom of this page.. At least that's what a simple phrase search resulted. There is still no definitive "provenance" for the ideas.

 

[lakecitieselectric]

There are 3 different types of retrofits A,B,and C.

A- uses the existing ballast - existing ballast fails and you have compatibility issues.
B- direct wire - safety issues, and I have some jurisdictions not allow them.
C- new led driver, lamps, and sockets - best install but expensive upfront, but less problems over the life of the product.

Sometimes its easier to sell a new fixture. The customer gets a new lens, and all new components.

 

[lakecitieselectric]

There is T-8 LED lamp that works with the ballast still wired......no retro fitting work required......

here's a link

http://gc-lighting.com/categories/tube/[Look at the website, even they admit they compatibility issues and you may be required to change the sockets.

 

[Telesis]

I know this post is a few months old, but I am the person who posted the text quoted by the OP on another site. I'll comment further here.

To date, Leviton does not have a tombstone that can be used with self ballasted LED tubes intended for direct connection to branch circuits. If you look at their installation sheets, one of the bolded warnings is 'NOT INTENDED FOR DIRECT CONNECTION TO THE BRANCH CIRCUIT.'

In my research, I also discovered a couple letters from Leviton's Director, LED Business Development, Lampholder Product Management stating that they have noted 'significant safety hazards' and go on to outline what those are. The following is quoted from a letter from that Director "To Whom It May Concern" It's only a couple sentences long, but the first is worth quoting here.

"Leviton has notified UL and NEMA that it currently does not support the use of its standard G13 and G5 base linear fluorescent sockets in LED "tube" retrofit applications."
(source: Leviton letter dated 1/3/12. If you Google "Leviton January 3, 2012" you can see the letter for yourself)

I contacted Leviton again just before writing this post and they confirmed they still do NOT have a product to be used in these applications.

I contacted Etlin-Daniels in Canada and the technical person I spoke to said their tombstones weren't to be direct connected either, but I didn't see it in writing.

I share this to make folks aware that simply switching to non-shunted tombstones from a reputable manufacturer like Leviton can be putting your customer or company at risk especially when they are on record as prohibiting their use in these applications. And yet I see myriad suppliers selling Leviton non-shunted sockets for use in retrofits with direct connect LED tubes.

To date, as I stated in the quote from the OP, I have only found one manufacturer that has put in writing that their sockets can be directly connected to branch circuits or 'mains' and that's Stucchi of Italy.

This is very clearly a foggy issue!

 

[Electric-Light]

15% and 30% tint grey sunglasses

15% and 30% tint grey sunglasses

Ballast bypass is frowned upon in the US, because they're dangerous to anyone who come upon re-lamping at some point as well as compromising the reliability of branch circuit. Any incident at socket will damage the socket and trip the branch and every fixture on the same switch has to be turned off to repair the damage. Notice how products that claim to be UL listed have a list of specs that are incoherent such as ballast gutted LED lamp chassis sold as chassis but include specs such as number of LED elements in place. These were likely submitted for approval with lamps already attached as non-user replaceable or part # specific only but not as an open ended application general purpose integral ballast LED lamp bulb fixture.

Many are only UL listed as a component, not a general purpose end user products. The power cord on vacuum cleaners are like 40-50ft long and they're 17 AWG and push 12A. Well, cords like these are approved as a component but attaching a UL listed two prong receptacle do not make the end product acceptable. The cord is not listed as general purpose extension cord use.

You just have to read what they have to say to realize it takes some vivid creativity to make numbers work out with LEDs. Commercial application integral ballast LED lamps with DLC listings aren't exactly cheap. I don't really know the motivation for fitting TLEDs but many claims made by the people who sell LEDs can be misleading or completely irrational so you have to do your own fact checking.

A significant energy savings realized from retrofitting 32W 741 lamps come from shaving down light output by about 1/4. The fixture efficiency recovery from wiping down dust and brand new LEDs that have not accumulated any lamp degradation helps with impression of non-compromise but think about the long run.

FO32/741 lamps pull 29.5W/lamp on ordinary ballast driving them at 88% and become stabilized at about 2400 lumens output per lamp by the time they're more than half way past the rated life. This gets to 2,100 lumens or less with years of dead bugs and dust accumulation. So when LED lamps that make 2,050 lumens when they're fresh can match the output of existing system in the conditions they were found while cutting the wattage by almost 50%. The performance can even improve over existing if the existing fixtures are in terrible shape and repairs are made such as replacing yellowed lens. But initial lumens are only good for the first impression, which is not the goal of lighting energy savings. LEDs have the highest lumen depreciation aside from metal halide so the initial lumen is just a flare up. Whenever you have an LED retrofit, my suggestion is that you make visual observation of "after" wearing a 15% tint sunglasses to get a feel for the lamp performance at about half way into lifetime, and a 30% tint sunglasses for how it will feel near the end of lamp life.

Except for L90 rated fixtures or active degradation compensation like N80 fixtures LED retrofit will always appear brighter than HPT8 when they're both designed to the same mid-life lumens, because LEDs have substantially more degradation which means a significantly more initial over lighting is required to ensure desired lighting level down the road.


Without making output compromise, ballast-lamp system efficacies:
standard electronic ballast + 741 is 80 LPW (100% kWh)
HPT8 93.4 lm/W (85% kWh)
high cost LED 108.9 (73% kWh)

drastic savings aren't possible without shaving output or use wattage for existing system created by LED sales rep instead of lamp-ballast data.

http://gc-lighting.com/wp-content/uploads/16T84FBYP-Tube-T8-16W-BYP.pdf

"16W REPLACES 32W 60% Energy Savings" meaning that the type of lamp it purportedly replace uses 40W. I can't think of what wild imagination was used to assume a fluorescent equivalent of wide beam 2,050 lm initial lamp uses FORTY watts per lamp.

This product they sell are LEDs rated in hours until 30% accumulation of permanent LED degradation, but I don't know the "design lumen" to use for fixed output LED systems. Anybody know authoritative figure for what to use as design lumen for LEDs?
The 2,050 lm is initial lumens.

I will be fair and use a more appropriate comparison. Two lamp is the most common ballast configuration for 32W T8. Philips
IOPA-2P32-LW-N runs a pair of 2XL 25W 48" rated at 2,305 design lumens at 77% output for 38W(Despite the double life, the less expensive standard XL lamps with lower upfront cost are more popular)

The LED sales team at GC claims savings are based on some whacked imaginary numbers that are so far off base it's not even usable.

Savings per fixture based on $0.11 / kw energy cost, 12 hrs / day lamp usage, 2x36W fixture / $10 fluorescent tube, $45
ballast with 15,000 hr lifetime, $35 LED tube with 50,000 hr lifetime

WTF $45 15,000 hr ballast and $10 15,000 hr lamp are they talking about? Call up your supply house and get quotes per lamp for box quantities of 25W XLL, 25W 2XL and IOPA-2P32-LW-N. Think about the time it takes to replace and rewire two sockets with non-shunted sockets to power vs 1 for 1 ballast swap out.

So you've got lamps starting out at 1,850 and settles down at 1,775 lm until they go poof some 50,000 hours later (60,000 hrs at 3hrs with instant start B50 life) at 19W per lamp.

In comparison, you've got 16W per LED integral ballast T8 that starts off at 2,050 lm and fades off to 1,435 lm after 50,000 and that's if the integral ballast doesn't fail during this time. Since I don't know the appropriate industry practice for what to use as design lumen, I'm gonna take the middle of degradation allowance value and use 85% of 1,740 lm per 16W ballast bypass lamp.

Using actual data based calculation, I come up with 19W per lamp vs 16W per lamp which means the energy saving 15% at 3W per lamp. Not even close to 60%.

1,775 lm stabilized value with negligible further decay per 19W or 93.4 mean lumens per watt. (HPT8)
1,740 lm per lamp at 15% decay.. and further depleting to 1435 lm per lamp at 50,000 hrs rated life. per watt fading down to 1,435 lm at 50,000 hrs. 108.9 mean lumen with 89.7 terminal lm/W. (LED)

This assumes that further degradation of LEDs down to 1,435 lm/lamp do not cause it to fall below acceptable light level. If it does, then LEDs need 21.4% greater installed capacity which raises the wattage which puts energy saving over fluoro to zero, or place them on a different useful life schedule and depreciate the LED system using L85 life instead of L70 life.

 

[James L]

I'm curious why all the mention of tombstones being "non shunted" ??

I know the difference, but I'm wondering does that mean everyone is using tubes which are fed from a single end?

Dual-end tubes don't require non-shunted tombstones.

Also curious if others label their fixtures as having been modified?

 

[al hildenbrand]

I'm curious why all the mention of tombstones being "non shunted" ??

One of the ballast bypass common methods is to connect a hot and neutral 120 Volt supply to a pair of LED tube pins, but only on one end. If the tombstone slots are shunted together, either internally or externally, when the 120 V supply is turned on, there will be a direct short.

I know the difference, but I'm wondering does that mean everyone is using tubes which are fed from a single end?

That has been, until quite recently, the UL listed manner to configure a line voltage ballast bypass LED tube. The other end's pair of pins have no electrical purpose, the pins only help mount the tube securely in the fixture.

Also curious if others label their fixtures as having been modified?

It depends upon the manufacturer's instructions and / or UL listing requirements. If labeling is required to help those who eventually have to replace lamps, this informs them of a bit of what is going on electrically.

 

[James L]

One of the ballast bypass common methods is to connect a hot and neutral 120 Volt supply to a pair of LED tube pins, but only on one end. If the tombstone slots are shunted together, either internally or externally, when the 120 V supply is turned on, there will be a direct short.


That has been, until quite recently, the UL listed manner to configure a line voltage ballast bypass LED tube. The other end's pair of pins have no electrical purpose, the pins only help mount the tube securely in the fixture.

I get the concept. I've installed l.e.d. tubes with both shunted and non-shunted. I was just curious if it's just kind of a running assumption that single-end tubes are used

It depends upon the manufacturer's instructions and / or UL listing requirements. If labeling is required to help those who eventually have to replace lamps, this informs them of a bit of what is going on electrically.

I was curious about labeling because I do. That pic I posted is a label I have made, and I stick two on each fixture - whether anyone requires them or not

 

[al hildenbrand]

There is a UL type C that is evidently a double end feed 120 V ballast bypass. Phillips is selling them.

My point is that this LED retrofit market seems to be developing new products and is not settled yet.
Tapatalk

 

[Electric-Light]

There is a UL type C that is evidently a double end feed 120 V ballast bypass. Phillips is selling them.

My point is that this LED retrofit market seems to be developing new products and is not settled yet.
Tapatalk

or do they? I see a link for it but when it says not found when I click on it. The background picture on Canadian Philips shows L and N pins. Interesting. http://www.lighting.philips.ca/products/product-highlights/instantfit

ANSI specification defines single capped FL lamp as something like 26W Dulux while normal linear lamps are defined as double capped. There's really no defined standard for feeding direct line to fluorescent lamp sockets. The single end feed is only to make installation faster.

Some drop-in LED double ended lamps are internally shunted on each end so it can feed from either pin. This type of lamp will cause a short circuit when inserted into a single end direct wire conversion. I have seen generic ballast bypass lamp that is single end fed with line voltage pass through across the length of the lamp to allow another lamp to piggy back off of it to make 8' conversion faster. This type of lamp will cause a short circuit if connected to a "double end fed" direct wire conversion. We know mismatching lamps of all colors end up in the ceiling. It's unrealistic to expect T8 drop-in LED, T8 120v only LED, T8 single end fed LED, T8 double end fed LED to not get mixed up. http://www.birddogdistributing.com/2-and-3-foot-LED-T8-Tube-Light-Ballast-Bypass.html This suggests possibility of direct 347v exposure from light fixtures.

Some integral ballast LED lamps are made of glass. As a general expectation, a directly wired 277v and 347v hot is not supposed to be available at fluorescent lamp sockets. You can expect component ruptures or vaporized internal leads with some chance of capacitors or transistors shooting through glass envelope. If the light was switched off first, the breaker will trip upon turning it back on. 3-4 incompatible lamps(as discussed above) already seated to short across the line will probably not clear on their own. Closing a 347v breaker into a short can create a very dangerous situation. Tripping something upstream and causing a significant disruption is expected. Just one of 101 reasons ballast bypass is incredibly stupid.

Remember. As I said before, a vacuum cleaner cord is a UL listed component. Simply joining it with a UL listed two prong molded female end do not make the whole thing proper. The specifications detailing exact details about the LED lamp and its internal element count on Tamlight advertised as "LED lamp ready" hints that it might have been UL listed as an LED luminaire assembled with off the shelf TLED listed as an appliance component.

 

[Bartstop]

I know this is an old thread but are there any updates? Is anyone making tombstones rated for connection to the branch circuit? I see Leviton still has the warning in their instruction sheets.

 

[garbo]

I will be retrofitting existing T8 fixtures to new LED tubes and pulled the following up during my research. Have any of you dealt with this issue?

"There is a deeper and more dangerous issue here that few people are aware of. My comments here specifically apply to tubes which are intended to be directly connected to a 120vac source (or branch circuit in UL terms). The tombstone or lampholder MUST BE RATED FOR DIRECT CONNECTION TO A BRANCH CIRCUIT. MOST ARE NOT!!! As of this writing, I don’t believe Leviton has a single holder that is rated to connect to a branch circuit directly. IT IS NOT SUFFICIENT TO SAY THAT BECAUSE A HOLDER IS NON-SHUNTED AND RATED AT SAY 600v AND 600w, it can be used. I have found only 1 manufacturer of holders(Stucchi 3000 series of Italy) that has tested and, in writing, certifies it can be used for DIRECT CONNECTION TO A BRANCH CIRCUIT(although I’m sure there are others since folks are selling retrofit ‘kits’ that contain the tube, holders, and a label) Look at the installation sheet of ANY Leviton tombstone, and the 2nd bullet in the upper left hand corner states it’s not intended for connection to a branch circuit. [The technical reason is that an arcing and sparking hazard exists when the tube fails in a way that produces a partial for full short circuit and high currents are drawn through the tombstone contact where it touches the pin. There are also issues with the physical installation of the tube and exposure of ‘live’ contacts to the installer or the luminaire]Keep this in mind if you choose to do a retrofit using direct connect tubes, or if you hire a contractor to do it for you! You don’t want an accident in your facility number one, and secondly don’t want to find out that your insurer won’t cover you because you did not have the proper tombstone!"

I have retrofitted one 2 lamp ( remember tubes go into the ground or so my inspector said ) in my house. They came with 4 new lamp holders rated at 600 volts and forget the wattage. I marked on each tombstone where I landed the hot wire. I plan on a installing a fuse holder with a 1 amp non time delay fuse. I would not trust converting a 277 volt luminare. Saw hundreds of cracked tombstone some with a piece missing due to ruff lamp replacements over my 50 years as a sparky.