Corridor Emergency Egress Lighting via GTD20

[birtclp]

I have a corridor application where emergency egress lighting is required. Is there any code-related reason why I cannot wire all of the lights in the corridor to turn on upon loss of normal power to the space using a 20A generator transfer device? The GTD20 will be located above the lay-in ceiling at one end of the corridor and will be fed by 2 separate circuits from two separate panels, one of the panels being an emergency panel fed via the building ATS and the other a normal power panel.

The AHJ is rejecting this design, but I think he is confusing the UL 1008 GTD20 with a UL 924 bypass device, in which case he would be correct. Failure of either breaker will not leave the space dark as the GTD20 will switch if the normal power circuit fails. Failure of any lighting element will not leave the space dark as there are 20 separate light fixtures in the corridor wired via the GTD20. The GTD20 will bypass switching upon loss of normal power so switch position will not impact emergency egress lighting upon loss of normal power to the space. The two circuits feeding the GTD20 are in separate conduits coming from separate panels. The AHJ wants multiple circuits run out to the light fixtures, either alternating fixtures or dual switched fixtures or something similar.

I believe that the two circuits to the GTD20 located in the space meets requirements of NEC 700.17. AHJ disagrees but has indicated he is willing to discuss.

Before I pursue this I wanted to run it by a few others to get opinions/thoughts.

 

[dkidd]

I have done it that way a number of times.

Here's a link http://www.bodine.com/downloads/specs/GTD.spec.L4100004.pdf

 

[birtclp]

I have done it that way a number of times.

Ironically I have also, and with this same AHJ providing plan review.

I'm hoping he is just confusing UL 924 devices with UL 1008 devices. In the case of a UL 924 device (switch bypass relay) I would agree with him.

 

[texie]

I have a corridor application where emergency egress lighting is required. Is there any code-related reason why I cannot wire all of the lights in the corridor to turn on upon loss of normal power to the space using a 20A generator transfer device? The GTD20 will be located above the lay-in ceiling at one end of the corridor and will be fed by 2 separate circuits from two separate panels, one of the panels being an emergency panel fed via the building ATS and the other a normal power panel.

The AHJ is rejecting this design, but I think he is confusing the UL 1008 GTD20 with a UL 924 bypass device, in which case he would be correct. Failure of either breaker will not leave the space dark as the GTD20 will switch if the normal power circuit fails. Failure of any lighting element will not leave the space dark as there are 20 separate light fixtures in the corridor wired via the GTD20. The GTD20 will bypass switching upon loss of normal power so switch position will not impact emergency egress lighting upon loss of normal power to the space. The two circuits feeding the GTD20 are in separate conduits coming from separate panels. The AHJ wants multiple circuits run out to the light fixtures, either alternating fixtures or dual switched fixtures or something similar.

I believe that the two circuits to the GTD20 located in the space meets requirements of NEC 700.17. AHJ disagrees but has indicated he is willing to discuss.

Before I pursue this I wanted to run it by a few others to get opinions/thoughts.

I think the AHJ is correct. Not all of Bodine's stuff is listed for use in an emergency system. The GTD20 is listed only for Article 702 installations. Their bypass relays are listed as UL 924 but you are not just bypassing a switch in your application. UL 1008 listing does not in itself mean a transfer device is acceptable for use in an emergency system.

 

[birtclp]

I think the AHJ is correct. Not all of Bodine's stuff is listed for use in an emergency system. The GTD20 is listed only for Article 702 installations. Their bypass relays are listed as UL 924 but you are not just bypassing a switch in your application. UL 1008 listing does not in itself mean a transfer device is acceptable for use in an emergency system.

and there it is, right on their cut sheet, stating

The GTD20A has been tested by Underwriters Laboratories in accordance with the standards set forth in UL 1008 for Automatic Transfer Switches for Optional Standby Systems. The GTD20A has also been tested by Underwriters Laboratories in accordance with standards
set forth in CSA C22.2 No. 178 for Automatic Transfer Switches. The GTD20A is UL Listed for field installation.

They do push it as being usable for emergency lighting. Irony here is this is not the reason the AHJ is disallowing this, so he may be accidentally right, but right is right. I'm gonna still look further to see if there is another device available via a different mfr. Thankfully there is still time to rework the design.

Appreciate quick responses so far.

 

[texie]

and there it is, right on their cut sheet, stating



They do push it as being usable for emergency lighting. Irony here is this is not the reason the AHJ is disallowing this, so he may be accidentally right, but right is right. I'm gonna still look further to see if there is another device available via a different mfr. Thankfully there is still time to rework the design.

Appreciate quick responses so far.

The allowance for a switch bypass has been in Art 700 for a while now and as you point out requires a listing to UL 924. The transfer device for branch circuits has never been allowed in Art 700 and I think that is why their listing shows only use on 702 systems.
Of note however is that branch circuit transfer devices are now recognized in the 2017 edition at 700.25. Unless you are on the 2017 edition, I don't think there is a compliant way to transfer a branch circuit.

 

[birtclp]

The allowance for a switch bypass has been in Art 700 for a while now and as you point out requires a listing to UL 924. The transfer device for branch circuits has never been allowed in Art 700 and I think that is why their listing shows only use on 702 systems.
Of note however is that branch circuit transfer devices are now recognized in the 2017 edition at 700.25. Unless you are on the 2017 edition, I don't think there is a compliant way to transfer a branch circuit.

Thanks. In SC we are under 2014 and probably will be until 2019 at the rate we update.

Hopefully Bodine and others will go through testing, etc. required to get these units classified under 700.25 of 2017 NEC.

 

[birtclp]

Guess this is what the listing will be going forward...

http://productspec.ul.com/details.php?ccn=WPWR

 

[birtclp]

...and I found at least one device from ETC.

https://www.etcconnect.com/Products...ing-Systems/Branch-Circuit-ELTS/Features.aspx

 

[dkidd]

Here is the proposal that resulted in 700.25.

Note that is from ETC.

 

[birtclp]

So prior to 2014 NEC many engineers (me included) were using load control relays and branch circuit transfer devices for transferring 20A branch circuit emergency loads, then the 2014 NEC clarified that this was not acceptable, only it only addressed load control relays used as transfer devices and ignored branch circuit transfer devices. So the 2017 NEC has now addressed the latter, identifying them as Branch Circuit Emergency Lighting Transfer Switch (BCELTS).

If I'm off on any of this please correct.

So going forward in order to design per my original description I will need a BCELTS that is UL 1008 listed to the proper category, and that will only suffice if the project is being designed under the 2017 NEC (unless I can get something in writing from the AHJ permitting this prior to adoption of the 2017 NEC).

I am appreciative of all the input, very helpful.

 

[texie]

So prior to 2014 NEC many engineers (me included) were using load control relays and branch circuit transfer devices for transferring 20A branch circuit emergency loads, then the 2014 NEC clarified that this was not acceptable, only it only addressed load control relays used as transfer devices and ignored branch circuit transfer devices. So the 2017 NEC has now addressed the latter, identifying them as Branch Circuit Emergency Lighting Transfer Switch (BCELTS).

If I'm off on any of this please correct.

So going forward in order to design per my original description I will need a BCELTS that is UL 1008 listed to the proper category, and that will only suffice if the project is being designed under the 2017 NEC (unless I can get something in writing from the AHJ permitting this prior to adoption of the 2017 NEC).

I am appreciative of all the input, very helpful.

I think that about sums it up. I know from experience that the whole issue of the proper use of load load control relays as permitted in the 2014 NEC is wildly misunderstood in the field by many.
Help me understand something though... while the need for a load control relay in this day and age is obvious, what is the real advantage to using a BCELTS as allowed in 700.25 in the 2017 NEC? You have to wire back to the emergency panel anyway, so why not just power the lights from there in the first place? What am I missing here?

 

[birtclp]

You can switch the entire circuit, for example all of the lights in an existing corridor can be wired via a single BCELTS, then you only bring a normal circuit and an emergency circuit to the BCELTS.

 

[texie]

You can switch the entire circuit, for example all of the lights in an existing corridor can be wired via a single BCELTS, then you only bring a normal circuit and an emergency circuit to the BCELTS.

Yes, I get that but you can do this with just an ALCR to bypass local switching/controls and avoid having to wire to the normal panel at all. You just put the lights on the emergency circuit. I must be missing something here.

 

[birtclp]

Aren't 2 separate power sources to the space required so tripping a single breaker will not leave space in dark? The control relay only senses normal power. If the emergency panel breaker is tripped all of the lights in the space turn off regardless of the status of normal power.

 

[texie]

Aren't 2 separate power sources to the space required so tripping a single breaker will not leave space in dark? The control relay only senses normal power. If the emergency panel breaker is tripped all of the lights in the space turn off regardless of the status of normal power.

OK, now I see where you are coming from. I was thinking of, say, a typical arrangement where you have both normal and emergency fixtures in the same area and you don't have to meet this requirement. You're right, if all the lights in an area were on 1 branch circuit then a BCELTS could be an option.

 

[Electric-Light]

How do you positively prevent accidental 3-way switch situation? If emergency one has tripped for some reason and forgotten about and it is turned off at normal panel for routine service work, how do you ensure circuit being worked on won't get energized when someone turns on the breaker in the emergency side?

 

[steve66]

How do you positively prevent accidental 3-way switch situation? If emergency one has tripped for some reason and forgotten about and it is turned off at normal panel for routine service work, how do you ensure circuit being worked on won't get energized when someone turns on the breaker in the emergency side?

As soon as they turned off the normal power, the emergency relay should switch over to the "emergency side". I would assume the emergency side would be connected on the load side of a emergency transfer switch.

So the light fixture would remain on, even after the default source of normal power was turned off.

 

[steve66]

I would have two issues with your use of the GTD device to connect all lights in a corridor.

1. That doesn't provide the 2 required separate lighting systems - a normal system, and an emergency system. At least that's what I remember the code requiring. By putting all the lights on the emergency you have eliminated any "normal" system.
2. Only necessary, required egress lights can go on the emergency system. You can't just put all the lights on the emergency system. That gives a higher chance a ballast will short, or something will go wrong that will trip a branch breaker and shut off all the emergency lighting in one area.

Now, if you pick every 3rd light and put it on the GTD, I would be good with that.

 

[birtclp]

There is normal and emergency to the GTD20, which is located above the lay-in ceiling in the corridor. If a properly listed BCELTS (Branch Circuit Emergency Lighting Transfer Switch) were used then I believe this would meet intent of the requirement. If the normal power breaker is shut off then the BCELTS transfers to the emergency panel circuit, which is still energized.