Do Sealed Endcaps On Vho Cause Fires?

[John Kelly]

I thought this would be better in its own thread. These are two recent posts I made in the thread http://archive.reefcentral.com/forum...hreadid=250465
WOW!!! My endcap just caught on fire

This is what I believe causes most of the VHO endcap meltdowns and fires....look at the pictures:



and this:



Do you see the difference?
I recently took out 2 URI 50/50 bulbs that have been in for 4 months. When I originally put them in I made sure the bulb pins were NOT LOOSE and that the endcaps were sealed around the bulbs good. During that time I have not taken off or moved the canopy or messed with the bulbs. The first picture shows no darkening of the tube. There is no corrosion and the bulb pins are still tight. The second picture shows obvious darkening of the tube (same tube) on one side only and on this same side the bulb pin is EXTREMELY LOOSE. I can pull it out at least an 1/8 of an inch and wiggle it at least a 1/4 inch or more up and down. It was NOT this way when I first put the bulb in and I did NOT use any force whatsoever to put the bulb in. I believe that these bulbs are cheaply made and it has come loose due to heat or excessive resistance through that particular pin inside of the bulb (note: frick-n-frags started this thread because his 6 MONTH OLD bulbs nearly burned his house down). I skimmed through this entire thread and found at least 6 different problems of one sort or another that people have had with their URI BULBS and most, if not all, were less than a year old. The problems didn't seem to be confined to any certain type (50/50, actinic, aquasun, etc...), but they were all URI. I don't recall seeing any other brand name mentioned. I don't know whether URI pins are soldered on or crimped, but whatever it is it doesn't hold for very long.
I just purchased two new URI super actinics, but this really makes me want to return them and buy some metal halides instead. I was planning to keep using the 50/50's until I found this.



UPDATE, I believe the problem is here:

Without much force at all, I pulled the bulb pin out. The only thing that holds the pin in place is that stiff cardboard stuff. The wire that leads into the pin is connected to the pin by a single stamp-like crimp (see picture). I noticed that the cardboard stuff around the pin hole was very brittle as if it had been charred a little. I really don't see much of a problem with the crimp connection as long as the bulb isn't constantly giggling around while it is on. Anyway, the problem didn't appear to be at the pin, but affected the pin so that it would end up getting hotter than the other one. I took the bulb out to the trash can and broke it (see 2nd picture). The second picture is looking at the inside of the bulb end. THE PROBLEM IS HERE and it is a combination of things that leads to Fire.
First of all, I don't think these bulbs were ever meant or designed to have a sealed endcap placed on them. Notice the dark ring on the glass inside of the bulb. This is the dark burn spot that I saw on the outside of the bulb. I placed my waterproof endcap on the bulb and the dark ring on the inside of the bulb glass is EXACTLY where the endcap seal contacts with the bulb. Now, look at the bulb filament and the thin oval metal housing around it. The filament is not straight, but has a bend in it. See where the filament and the metal oval housing are closest to each other? This spot gets the hottest. Now look at the bulb glass and notice where the burn mark is largest in diameter. Here is my theory as to what is happening:

1. The bulb is placed in a sealed waterproof endcap.
2. In some bulbs, the filament (which is not straight, but curved) and the oval metal housing on the inside of the bulb are slightly closer together than in other bulbs.
3. The filament heats up in this area more so than it normally would because of the extra heat from the use of sealed endcaps.
4. The wire that leads from that side of the filament to the bulb pin is now substantially hotter. Look at the picture...the wire is black on that side at the point where it goes out to the bulb pin.
5. Because the wire is burning hotter, the bulb pin is hotter inside of the endcap.
6. The bulb pin becomes so hot that it begins to melt down the endcap.
7. POOF, everything goes up in flames.






What I am saying is that SEALED ENDCAPS ARE, IN SOME PART, RESPONSIBLE FOR MELTDOWNS AND FIRES WHEN USED ON VHO BULBS.

Symptoms of possible endcap meltdown occurring:
1. Obvious dark blotches or spots on the outside of the bulb near the ends.
2. Loose bulb pins. The looseness is caused by the excessive heat on the stiff cardboard that holds the bulb pins in place.

[Crawdad1]

1. Obvious dark blotches or spots on the outside of the bulb near the ends.

What made you come up with this conclusion?

[John Kelly]

When you have VHO bulbs that are less than 6 months old and dark interior burn type marks (not uniform, but more like spots/ blotches) start appearing on the outside of the bulb, you can figure that something just isn't right. Then, when the bulb pin on that side of the bulb is so loose it is about to fall out and you know it was originally tight, you can figure that something just isn't right. After you realize that the burn mark is in the shape of an endcap seal and directly underneath where the endcap seal touches the bulb, you think to yourself that something just ain't right. Finally, after reading posts about how endcaps melt from within and some are mysteriously burned around the seal, you draw the conclusion that "Obvious dark blotches or spots on the outside of the bulb near the ends" may be an indication that there is abnormal heat being produced inside, which may cause problems outside of the bulb; such as endcaps melting and fires starting.

HTH

[Frick-n-Frags]

That makes infinitely more sense than the "moisture" arguement which I think is totally off base. So I'll bump this one up. IMO, the endcaps have to go. ( but I'll still use the bulbs)

[SPasse]

Hi All,

As a EE (hardware engineer), VHO end caps in salt spray environments have been a concern of mine since I first saw them in use.

One of my earlier jobs as an engineer was to develop electronic equipment that would pass the infamous UL (Underwriters Laboratories) salt spray test. This was an interesting engineering exercise, even for pressure sensors, etc that were only limited current (sensor & non-heating) devices.

But VHO end caps have additional factors to deal with that makes designing them even more of a challenge.

1 Even properly functioning VHO lamps get hot and generate UV light.

1A This combination of heat and UV is very hard on the end cap gasket and socket insulator materials.

2 The daily heating & cooling cycle cause condensation and hydroscopic “wicking� of the condensation into the contact area.

3 The contact themselves are also subjected to this daily heating & cooling cycles. Even without the salt spray issues, this is hard on contact interfaces.

Based on the above conditions, you should consider the following when dealing with these devices:

1 Use only end caps that are rated for VHO service in wet environments.

2 Inspect them regularly

3 Replace them as at some point before they fail.

I suspect that some of the current VHO end caps that are being used, even if they were designed for VHO use, were not also designed for the additional salt spray environment issues.

If there is an issue with the filament/pin/crimp interface, that would cause a problem even with none of the above factors coming into play.

Regards

Scott

[Frick-n-Frags]

So what exactly would you think some moisture inside the cap would do if there is no corrosion to cause a failing contact WRT to causing ignition???

The only reason I ask(and I can only go from my scenario) is because I found no traces of corrosion in any of my other 15 "good" endcaps or the bulbs or bulb pins.

I have to stick with either an open in the contacts causing an arc across the growing gap or the heat from a filament too close to the bulb side igniting the cap through the glass(which is what reefartist has come up with). But this was just an examination of my situation, which I examined pretty closely. Why would wicking moisture melt the rubber seals to the bulb surface? Furthermore, my bulbs were in open air and the ends were not even over the prop tub. Even exposed copper in wire nuts right there wasn't tarnished and has not tarnished to this day.

Not a big deal, as I determined that the endcaps had to go period, but just looking for more perspective.

[John Kelly]

(posted to the other thread also)



I fished the other end of the same URI bulb out of the trash can. This end had no visible outer darkening on the glass where the endcap seals were holding it (see previous post pictures). See how nice and straight the filament is compared to the curved one (see previous post pictures). Also, notice how the oval metal housing isn’t separated and the wire that leads to the pin isn’t blackened at the base. I don’t know if the bulb is designed this way or if the curved filament in the other end should actually be straight; either way, as it currently stands in my opinion, sealed plastic end caps and VHO bulbs (URI in this instance) makes for a potentially bad situation. I have called URI to find out more details about the design of their bulb, but am now waiting for a return call.

[marineaddict]

Just joining the thread as I have VHO's with the german style waterproof endcaps.

[SPasse]

Frick-n-Frags,

I was not meaning to suggest there was not a pin crimp/open contact issue with your lamps, if true, this would defiantly cause a meltdown. I was just making some general observations of what I have observed with electrically related failures in wet environments.

But if a seal on an end cap does fail, a little bit of SW will very quickly destroy the brass/copper alloys used in contact material.

Post meltdown forensic analysis of a fried end cap is tricky, a rather complex chicken/egg story. (Been there done that)

Regards,

Scott

[SPasse]

ReefArtist,

Have you contacted URI about this?

Most electrical manufactures have a failure analysis engineer. I am betting that they would love to get a hold of these specimens to determine if there is a QA issue with their manufacturing process.

Regards,

Scott

[SPasse]

ReefArtist,

I took the liberty of drawing some comment graphics on your picture:



#1 The uneven coloration of the starter filament is interesting

#2 The apparent deformation of the metal ring surrounding the filament is interesting

#3 The apparent darkening of the glass surrounding the top lead is interesting.

Possibilities:

#1 Bad crimp of the filament on “top� side crimp?

#2 Pin/lead wire being heated from bad connection at the socket/pin interface?
IMO, not as likely as “A�

As I said in my previous post, I bet URI would like to see those lamp ends.

Regards,

Scott

[John Kelly]

SPasse,

Thanks for the comment graphics. In that same picture, the dark ring on the outer bulb sleeve is exactly where the endcap seal was touching the bulb. What I believe is happening is that if the bulb end was NOT sealed, then the crooked filament may cause an increase in temperature and even a little darkening over time at the bulb end, but not enough to cause the bulb pin to melt an endcap. When the endcap is sealed, this is trapping additional heat and therefore causing the crooked filament to become even hotter, which then causes the bulb pin to be hotter, which then breaks the endcap melting barrier.
The socket/ pin interface was originally fine. I think it became deteriorated by the filament/metal housing heat and not the other way around (filament/metal housing being deteriorated by socket/pin interface).

[SPasse]

ReefArtist,

I am thinking…



Regards,

Scott

[kastanzaman]

SPasse,
Do you know of any particular endcaps that are VHO and wet environment rated? Are these the ones usually sold by reef lighting vendors. Is there a way to tell by looking at the endcap?

[SPasse]

Kastanzaman,

Do you know of any particular endcaps that are VHO and wet environment rated?

I knew somebody was going to ask that… The last time that I looked at that issue was several years ago, I need to re-visit that.

Are these the ones usually sold by reef lighting vendors.

Well hopefully… I am kind of a stickler for being able to talk to the manufacturer of any given component. (When in doubt, talk to the designer )

Is there a way to tell by looking at the endcap?

Want list:

1 Plastic able to deal with heat/UV

2 Compression seals able to deal with heat/UV

3 Well designed contacts able to deal with heating/cooling i.e. expansion/contraction cycles.

4 Gastight contact interfaces.

Regards,

Scott

[pickle311]

I'm tagging this one, it has got me very concerned. I'm running VHOs with the sealed german style endcaps.

[SPasse]

ReefArtist,

Can you get me a super close-up of this area?



Regards,

Scott

[SPasse]

Hi All,

I stuck this thread for awhile.

Regards,

Scott

[Frick-n-Frags]

thanx for the feedback Scott.

[SPasse]

ReefArtist,

I took the liberty of changing the thread title.

I don’t want to draw any conclusions – yet.

Regards,

Scott

[SciGuy2]

I've noticed that nearly every old VHO lamp I've used has had some degree of blackening on at least one end near the endcap seal area. I've seen it so often that I almost thought that it was normal.

[SPasse]

SciGuy2,

I've noticed that nearly every old VHO lamp I've used has had some degree of blackening on at least one end near the endcap seal area. I've seen it so often that I almost thought that it was normal.

It is, the metals that make up the filament are “boiled/blasted� off and deposited on the glass.

But blacking in and isolated spot can be indicative of something else besides normal processes.

Regards,

Scott

[Crawdad1]

ReefArtist I must admit I have not read the original thread from start to finish because I get real perturbed with read a post that states (just tagging along) and I know there will be quite a few. In the original thread how many people (head count please) had an actual problem with the end caps catching fire? Was it 1,2 6 12? I'm just curious at to how wide spread this problem may be.

[thedude]

You can see that same blackening on many older normal dry environment flourescent bulbs as well. When I replace them here in my office (ceiling types) many times they are black at one end. That usually only happens once the bulb starts to flicker or dim.

It sounds to me like the saltwater environment is just eating these things up. It is somehow still getting inside the bulbs and corroding things. You just don't see this kind of failure rate on bulbs in a dry environment.

Something interesting to try would be to seal the insertion point of the endcap (where it meets glass) with silicone, and then also seal any other screw on areas or where the endcap meets the hood. Just goop the crap out of it and see if that helps the lifetime of the bulb.

If it does, then our endcaps maybe splash proof, but they aren't airtight.

Just a thought

I haven't experienced a fire, but I have had smoke and a melted endcap. Now my t5's only run for 2hrs a day, 1hr in AM, 1hr at PM. I don't trust them

[SciGuy2]

No fire here either. However, I had the internal part of the endcap melt and smell like burnt plastic.