Life vs. number of starts for CFL UV lamp.

[Bart Z. Lederman]

I realize this can be a somewhat "hot" topic, but I'll ask
the question anyway.

Honeywell has come out with an easy to install (one hole)
Ultraviolet light that can be added to forced air vents. The
cost seems reasonable, so I decided to add one to my home
heating / cooling system. It uses a CFL without the "F": a
lamp about 6 inches long, two parallel tubes, for pin in-line
base, but clear (no phosphor). It also comes with a lot of
warnings not to look directly at the lamp, and has warning
stickers to put on the HVAC equipment. There is a plastic
indicator that glows to show the lamp is on.

There is no switch on the device: the lamp is on all of the
time. My thinking is that it's not going to be doing much good
while the air isn't circulating past it. (The instructions
point out you can position the lamp so it will irradiate the
air-conditioning coil, if you have one, which would be good for
keeping mold from growing on any condensed moisture. But the
way my unit is set up there isn't a good place to put the lamp
to do that.) So I'm thinking of hooking it up to a circuit
where it will get power only while the circulating fan is
operating.

This, of course, leads to the hot topic: can I expect
better lamp life if the lamp is on all of the time, or will it
be better if the lamp isn't operating while the fan is off, but
will have to start more often.

I realize the answer is going to depend a lot on the duty
cycle of the fan, and I can probably take some measurements in
the future to find out what that is: but if anyone has any
experience in this area and would like to express an opinion,
I'm interested.

(I've taken lighting design courses, but this topic wasn't
covered: and CFLs were developed long after I took the course.)

Thanks.

 

[Victor Roberts]

I realize this can be a somewhat "hot" topic, but I'll ask
the question anyway.

Honeywell has come out with an easy to install (one hole)
Ultraviolet light that can be added to forced air vents. The
cost seems reasonable, so I decided to add one to my home
heating / cooling system. It uses a CFL without the "F": a
lamp about 6 inches long, two parallel tubes, for pin in-line
base, but clear (no phosphor). It also comes with a lot of
warnings not to look directly at the lamp, and has warning
stickers to put on the HVAC equipment. There is a plastic
indicator that glows to show the lamp is on.

There is no switch on the device: the lamp is on all of the
time. My thinking is that it's not going to be doing much good
while the air isn't circulating past it. (The instructions
point out you can position the lamp so it will irradiate the
air-conditioning coil, if you have one, which would be good for
keeping mold from growing on any condensed moisture. But the
way my unit is set up there isn't a good place to put the lamp
to do that.) So I'm thinking of hooking it up to a circuit
where it will get power only while the circulating fan is
operating.

This, of course, leads to the hot topic: can I expect
better lamp life if the lamp is on all of the time, or will it
be better if the lamp isn't operating while the fan is off, but
will have to start more often.

I realize the answer is going to depend a lot on the duty
cycle of the fan, and I can probably take some measurements in
the future to find out what that is: but if anyone has any
experience in this area and would like to express an opinion,
I'm interested.

(I've taken lighting design courses, but this topic wasn't
covered: and CFLs were developed long after I took the course.)

Thanks.

You are right that switching the lamp will reduce the number
of OPERATING HOURS, but not necessarily the number of
elapsed hours between lamp changes; and it will save energy.

You are also correct that we can't begin to answer this
question until we know the number of minutes or hours per
start and the number of starts per hour or hours per start.
For a furnace application I suspect the time per start is
well less than an hour and there is more than one start per
hour.

Fluorescent lamps are life rated on a 3-hour per start
cycle. However, the exact relationship between time per
start and lamp life depends upon the lamp design and the
ballast design. If operated continuously they may operate
20% to 100% longer. If operated at less than 3 hours per
start their life will be reduced.

Life is a function of both operating hours and number of
starts. CFLs tend to be rated for 10,000 hours at 3 hours
per start which is only 3333 starts. If you operate them at
30 minutes per start, then you will reach 3333 starts after
only 1667 hours of operation. Of course the lamp will
normally last longer than 1667 hours at 30 minutes per start
but perhaps not more than 3000 operating hours if started
every 30 minutes.

That being said, there are special ballasts, known as
"programmed rapid start" or sometimes "programmed start",
that are designed to start the lamp very gently. These can
be used to get long lamp life in applications where the
lamps will be started frequently.

Is there a part number of your lamp or on the ballast? Does
the system use a glow switch? (A glow switch will lead to
poor lamp life if started frequently.) At least your lamp
has 4 pins, which means there is some hope of finding a good
programmed rapid start or programmed start ballast to
operate it.

--
Vic Roberts
http://www.RobertsResearchInc.com
To reply via e-mail:
replace xxx with vdr in the Reply to: address
or use e-mail address listed at the Web site.

This information is provided for educational purposes only.
It may not be used in any publication or posted on any Web
site without written permission.

 

[Bart Z. Lederman]

You are right that switching the lamp will reduce the number
of OPERATING HOURS, but not necessarily the number of
elapsed hours between lamp changes; and it will save energy.

You are also correct that we can't begin to answer this
question until we know the number of minutes or hours per
start and the number of starts per hour or hours per start.
For a furnace application I suspect the time per start is
well less than an hour and there is more than one start per
hour.

The way the system runs now, at least in the summer during
the cooling period, is that the time per start is considerably
more than one hour. With the weather we've been having, it
will start in the afternoon, and run continuously into the
evening. Eventually, the system will go off, and will
generally remain off until the next afternoon. I also have
a four time period thermostat, so even in the winter, the
system is 'off' for extended periods of time, and it is my
recollection that the on-off cycles tend to be fairly long.
That's why I'm thinking that having the lamp on only when
the fans run will be worth while.

That being said, there are special ballasts, known as
"programmed rapid start" or sometimes "programmed start",
that are designed to start the lamp very gently. These can
be used to get long lamp life in applications where the
lamps will be started frequently.

Is there a part number of your lamp or on the ballast? Does
the system use a glow switch? (A glow switch will lead to
poor lamp life if started frequently.) At least your lamp
has 4 pins, which means there is some hope of finding a good
programmed rapid start or programmed start ballast to
operate it.

There is, unfortunately, relatively little information.
The ballast is "buried" within the unit. The lamp mounts
in a large twist base unit that can be thought of as a sort
of 'plug' which fits into the unit with the lamp sticking
through the hole in the duct. To remove the lamp you twist
this unit counterclockwise to unlock it, which also shuts off
power to the lamp so you can't see it when it's on (a necessary
safety feature). The unit is quite well designed, but unless
I dismantle it I don't think I can get any information on the
ballast. (I'm fairly willing to do that, though, but a quick
look at the unit when installing it makes me think that
dismantling it may be a one-way operation, and at the moment
I'd really rather not sacrifice it.) Since I don't hear any
hum, and the unit isn't too heavy, I think it's a safe bet
that it's an electronic (rather than magnetic) ballast of some
kind.

I don't recall any markings on the lamp itself, but I'll
double check. I'd like to know more about it, because at
the moment the only replacements are to buy the Honeywell
units at the same store. I can make measurements of the
lamp if someone wants to match it up. I would expect it to
be some reasonably mass-produced lamp: Honywell wouldn't
launch a product like this without having a reasonable
supply of lamps, and I don't think they make them themselves.
I suppose they could custom order a lamp that only they would
have, but my past experiences with them make me think they
usually don't do "tricks" like that. The lamp is definately
of the "CFL" form factor, two parallel tubes connected at the
far end, with a white plastic base. It is not one of the
conventional single tubes with pins on the end lamps, and it
certainly does not use glow starters.

I had been thinking for some time of installing some sort
of germicidal lamps, using conventional straight tubes and
ballasts. However, this unit solves all of the problems of
obtaining the necessary lamps (they don't appear to be
standard off-the shelf items in any local store), wiring
up a ballast, finding a place to hook it in, cut openings
in the duct, rig up a door with a safety switch, install
some sort of safe peephole, and all of the other necessary
work. This unit installed quickly in one hole (plus two
small pilot holes for the fastening screws), ready to plug
in. It is tough to beat that for a home installation.

Bart.

 

[Victor Roberts]

The way the system runs now, at least in the summer during
the cooling period, is that the time per start is considerably
more than one hour. With the weather we've been having, it
will start in the afternoon, and run continuously into the
evening. Eventually, the system will go off, and will
generally remain off until the next afternoon. I also have
a four time period thermostat, so even in the winter, the
system is 'off' for extended periods of time, and it is my
recollection that the on-off cycles tend to be fairly long.
That's why I'm thinking that having the lamp on only when
the fans run will be worth while.



There is, unfortunately, relatively little information.
The ballast is "buried" within the unit. The lamp mounts
in a large twist base unit that can be thought of as a sort
of 'plug' which fits into the unit with the lamp sticking
through the hole in the duct. To remove the lamp you twist
this unit counterclockwise to unlock it, which also shuts off
power to the lamp so you can't see it when it's on (a necessary
safety feature). The unit is quite well designed, but unless
I dismantle it I don't think I can get any information on the
ballast. (I'm fairly willing to do that, though, but a quick
look at the unit when installing it makes me think that
dismantling it may be a one-way operation, and at the moment
I'd really rather not sacrifice it.) Since I don't hear any
hum, and the unit isn't too heavy, I think it's a safe bet
that it's an electronic (rather than magnetic) ballast of some
kind.

I don't recall any markings on the lamp itself, but I'll
double check. I'd like to know more about it, because at
the moment the only replacements are to buy the Honeywell
units at the same store. I can make measurements of the
lamp if someone wants to match it up. I would expect it to
be some reasonably mass-produced lamp: Honywell wouldn't
launch a product like this without having a reasonable
supply of lamps, and I don't think they make them themselves.
I suppose they could custom order a lamp that only they would
have, but my past experiences with them make me think they
usually don't do "tricks" like that. The lamp is definately
of the "CFL" form factor, two parallel tubes connected at the
far end, with a white plastic base. It is not one of the
conventional single tubes with pins on the end lamps, and it
certainly does not use glow starters.

I had been thinking for some time of installing some sort
of germicidal lamps, using conventional straight tubes and
ballasts. However, this unit solves all of the problems of
obtaining the necessary lamps (they don't appear to be
standard off-the shelf items in any local store), wiring
up a ballast, finding a place to hook it in, cut openings
in the duct, rig up a door with a safety switch, install
some sort of safe peephole, and all of the other necessary
work. This unit installed quickly in one hole (plus two
small pilot holes for the fastening screws), ready to plug
in. It is tough to beat that for a home installation.

Bart.

If you have a 4-pin lamp it is not instant start, and if you
think it has an electronic ballast then it is probably not
switch start. That leaves rapid start or one of its
variations. If you can't or don't want to rip it apart,
which I fully understand, then your best option seems to be
to switch it on and off with the blower motor and see how
long the lamps last.

--
Vic Roberts
http://www.RobertsResearchInc.com
To reply via e-mail:
replace xxx with vdr in the Reply to: address
or use e-mail address listed at the Web site.

This information is provided for educational purposes only.
It may not be used in any publication or posted on any Web
site without written permission.

 

[TKM]

If you have a 4-pin lamp it is not instant start, and if you
think it has an electronic ballast then it is probably not
switch start. That leaves rapid start or one of its
variations. If you can't or don't want to rip it apart,
which I fully understand, then your best option seems to be
to switch it on and off with the blower motor and see how
long the lamps last.

I agree. It sounds like the circuit is a rapid start. Another thought then
is to connect the lamp to a delay circuit so that once it
starts (triggered by the fan switch), it stays on for a minimum time. There
are such circuits (simple adjustable thermal types) in garage door openers
that keep the light on for a few minutes after the door is operated.

But as Ioannis suggests, the lamp life in this case is not determined by
when the lamp fails to light; it's determined by when the UV output drops to
an ineffective level. You need the UV depreciation curve for the lamp as
well as the lamp life rating.

Terry McGowan