Maker Pro
Maker Pro

Running mains fluorescents from inverter

T

The Other Mike

Currently got a remote observation site (wildlife) with no grid fed
power nor any prospect of it.

Half the site has a few modified 4ft T5's (36W) retrofitted with 12v
IOTA Ballasts (2D12-1-32) fed from a lead acid battery charged by a
solar panel. The other half of the site has 5ft T5's with magnetic
ballasts fed by a Honda EU20i generator which despite being a quiet
suitcase model and loads of additional soundproofing is still way too
noisy. Near silent operation is essential. Hauling fuel is also a
PITA as its a long way from the road.

So I need a way of powering the 5ft T5's (58W) from a low voltage DC
supply. IOTA only make ballasts up to 40W and they need a circa 50v
supply, realistically I need to keep to 12v to keep the solar array
price down.

So thoughts turned to an inverter fed from an uprated solar array and
battery.

A cheap modified sine wave inverter (circa 500W capacity) on a 100Ah
brand new battery fails to even kick even one 5ft tube into life. The
manufacturer says these inverters are not compatible with fluorescent
tubes but doesn't elaborate any further.

Does anyone have any ideas on how to get these lights working off
grid?

A change of ballast to an electronic type? (all indications are this
could won't work?)

Moving to a pure sine wave inverter (extremely expensive) ?

A different inverter supplier rather than 'one hung lo china inc' ?'

A ballast supplier that offers 12v ballasts that will drive a 58W
tube?

A homebrew 12V fluorescent inverter, running at high frequency that
will drive 5ft tubes and costs not a lot? :)


--
 
G

Gazz

The Other Mike said:
Currently got a remote observation site (wildlife) with no grid fed
power nor any prospect of it.

Half the site has a few modified 4ft T5's (36W) retrofitted with 12v
IOTA Ballasts (2D12-1-32) fed from a lead acid battery charged by a
solar panel. The other half of the site has 5ft T5's with magnetic
ballasts

So I need a way of powering the 5ft T5's (58W) from a low voltage DC
supply.
So thoughts turned to an inverter fed from an uprated solar array and
battery.

A cheap modified sine wave inverter (circa 500W capacity) on a 100Ah
brand new battery fails to even kick even one 5ft tube into life. The
manufacturer says these inverters are not compatible with fluorescent
tubes but doesn't elaborate any further.

Does anyone have any ideas on how to get these lights working off
grid?

prolly cheapest and easiest to change the 5 foot fittings to 4 footers with
the 12 volt ballasts you know work.

you'll need a pure sine wave inverter to run a ferromagnetic ballast, bit of
a waste of money just to run a few lights, if you wanted to run laser
printers, charge electric toothbrushes etc at the site, then it may be
worthwhile.
 
M

Mho

Did you try rubbing the tubes to get them started, especially with a static
generating material like nylon or wool?

----------

"The Other Mike" wrote in message

Currently got a remote observation site (wildlife) with no grid fed
power nor any prospect of it.

Half the site has a few modified 4ft T5's (36W) retrofitted with 12v
IOTA Ballasts (2D12-1-32) fed from a lead acid battery charged by a
solar panel. The other half of the site has 5ft T5's with magnetic
ballasts fed by a Honda EU20i generator which despite being a quiet
suitcase model and loads of additional soundproofing is still way too
noisy. Near silent operation is essential. Hauling fuel is also a
PITA as its a long way from the road.

So I need a way of powering the 5ft T5's (58W) from a low voltage DC
supply. IOTA only make ballasts up to 40W and they need a circa 50v
supply, realistically I need to keep to 12v to keep the solar array
price down.

So thoughts turned to an inverter fed from an uprated solar array and
battery.

A cheap modified sine wave inverter (circa 500W capacity) on a 100Ah
brand new battery fails to even kick even one 5ft tube into life. The
manufacturer says these inverters are not compatible with fluorescent
tubes but doesn't elaborate any further.

Does anyone have any ideas on how to get these lights working off
grid?

A change of ballast to an electronic type? (all indications are this
could won't work?)

Moving to a pure sine wave inverter (extremely expensive) ?

A different inverter supplier rather than 'one hung lo china inc' ?'

A ballast supplier that offers 12v ballasts that will drive a 58W
tube?

A homebrew 12V fluorescent inverter, running at high frequency that
will drive 5ft tubes and costs not a lot? :)


--
 
V

vaughn

Gazz said:
you'll need a pure sine wave inverter to run a ferromagnetic ballast,

But what about swapping to a newer electronic ballast? I would expect them to
have a pulse-mode input which should even work off a cheap square wave inverter.
(But I haven't tried) Besides, they are more efficient.

Anyhow, thanks to this thread, I now know how to find 12 volt ballasts without
breaking the bank.

Thanks!
Vaughn
 
F

F Murtz

The said:
Currently got a remote observation site (wildlife) with no grid fed
power nor any prospect of it.

Half the site has a few modified 4ft T5's (36W) retrofitted with 12v
IOTA Ballasts (2D12-1-32) fed from a lead acid battery charged by a
solar panel. The other half of the site has 5ft T5's with magnetic
ballasts fed by a Honda EU20i generator which despite being a quiet
suitcase model and loads of additional soundproofing is still way too
noisy. Near silent operation is essential. Hauling fuel is also a
PITA as its a long way from the road.

So I need a way of powering the 5ft T5's (58W) from a low voltage DC
supply. IOTA only make ballasts up to 40W and they need a circa 50v
supply, realistically I need to keep to 12v to keep the solar array
price down.

So thoughts turned to an inverter fed from an uprated solar array and
battery.

A cheap modified sine wave inverter (circa 500W capacity) on a 100Ah
brand new battery fails to even kick even one 5ft tube into life. The
manufacturer says these inverters are not compatible with fluorescent
tubes but doesn't elaborate any further.

Does anyone have any ideas on how to get these lights working off
grid?

A change of ballast to an electronic type? (all indications are this
could won't work?)

Moving to a pure sine wave inverter (extremely expensive) ?

A different inverter supplier rather than 'one hung lo china inc' ?'

A ballast supplier that offers 12v ballasts that will drive a 58W
tube?

A homebrew 12V fluorescent inverter, running at high frequency that
will drive 5ft tubes and costs not a lot? :)


Try this, and oatleyelectronics have lots more goodies.

http://secure.oatleyelectronics.com...d=229&osCsid=e1382db5b7e8301a03011e405258f2ca
 
A

Andrew Gabriel

But what about swapping to a newer electronic ballast? I would expect them to
have a pulse-mode input which should even work off a cheap square wave inverter.

They also have power factor correction inputs (usually achieving 0.95 - 0.98)
and these might not like square wave. Many nowadays are rated 50-60Hz or DC,
and the DC voltage can be lower than mains in many cases, 180V is not
uncommon, and I have one here which says 160V.

One thing you do have to be careful about when running off batteries is not
to allow the ballasts to run below their minimum rated voltages, as that
causes excessive current in the transformer primary, and quick burnout.
Some of the better ones include protection circuitry to switch-off below
the min rated voltage.
 
T

The Other Mike

They also have power factor correction inputs (usually achieving 0.95 - 0.98)
and these might not like square wave. Many nowadays are rated 50-60Hz or DC,
and the DC voltage can be lower than mains in many cases, 180V is not
uncommon, and I have one here which says 160V.

One thing you do have to be careful about when running off batteries is not
to allow the ballasts to run below their minimum rated voltages, as that
causes excessive current in the transformer primary, and quick burnout.
Some of the better ones include protection circuitry to switch-off below
the min rated voltage.

So presumably I could rectify and smooth and regulate the near sine
wave inverter output of circa 340v pp within the lamp housing and feed
an electronic ballast with say 200v?

Any indication of suitable makes and models?


--
 
T

The Other Mike

I think you'd benefit greatly from zooming out for a rethink. You'er
asking for low voltage panels to charge a 12v battery, which will then
get upconverted to 240v which must be done the expensive sine way,
which will then be downconverted to 35v to run the tubes. Its madness.
Far more practical wuold be warm white LEDs that will run off 12v via
a resistor or constant current regulator. Take a look at
rapidonline.co.uk, if you can wade through the long list of led
options you'll find some very satisfactory deals there. No need for a
240v invertor anywhere, or problematic ballasts, no compatibility
issues, and LEDs are far longer lived, far more robust and much more
versatile than fl tubes, and trivial to dim to save power.

I can't see the economics working, yes its wasteful with multiple
stages of conversion but the fluorescents are already there, also
LED's seem way more expensive because of the number of light sources
required.

Dimming isn't a requirement. Its for work lighting as its a near
windowless building, the requirement is for an even spread of light
such as provided by a fluorescent tube (operating microscopes and
laptops in the same room is very common occurrence)

T5 fluorescents are somewhere around 70-100 lumens per watt. Most
LED's at appear to be much lower efficiency with a holey spectrum and
providing a very directional light source. but I'll ignore those
deficiencies for now.

Currently there are I guess somewhere around 17000 lumens lighting an
area of about 20 x 8ft so make that around 1000 lux, but we might get
away with a little bit less.

At about the typical "90 lumens" per white LED package that would
require about 180 separate light sources costing anything from GBP 2 -
4 a piece. All requiring heatsinking and being around 1/3 the
efficiency of the fluorescents, almost certainly requiring a bigger
solar array and a bigger battery.

So before those LED's are mounted and wired that's GBP 500 - 700 for
the equivalent of the fluorescents that are already in situ which cost
significantly less than a 10th of that.

GBP 500-700 buys a very nice pure sine inverter.


--
 
T

The Other Mike

Do you need fluorescent tubes or would LED bulbs be a more efficient
choice for solar powered illumination?

The local home/garden center has 120 volt 7.5 watt LED bulbs (lumens
equivalent to 40 watt incandescent) for $10US each. These bulbs work
with a dimmer, so they are not picky about the power source.

Two of these LED bulbs should provide more light than a 5 foot
fluorescent tube plus provides power savings.

Two?

Really?

USD 10?

I thought this was the only one worth bothering with?

http://www.gelighting.com/na/energysmartLED/home.html

450 lumens output

USD 50 a piece

To give me 17000 lumens equivalent (100 lumens /watt * 58W * 3
fittings) that requires 37 lamps costing USD1800

Giving a total load of more than 300W compared to circa 170W

You might be able to see why I keep coming back to conventional
fluorecents every time :)


--
 
T

The Other Mike

Anyhow, thanks to this thread, I now know how to find 12 volt ballasts without
breaking the bank.

The IOTA ones are a bit crude (check the wiring layout!) They don't
like cold temperatures and light output is also down quite a bit.


--
 
T

The Other Mike

Not cheap, but would these kits, designed to turn an ordinary
fluorescent into an emergency light, do the trick?

http://www.tlc-direct.co.uk/Main_In...ng_Emergency_Index/Conversion_Kits/index.html


I looked at them a long time ago but it's been suggested that they are
not ideal for external DC supplies as their heatsinking isn't up to
the job. The supplied battery packs have a limited run time 0of a
couple of hours and this isn't enough to 'break' them.

A bit like a computer UPS. Rated at say 1000VA for 10 minutes to get
though short breaks in supply or to enable an orderly shutdown but no
good for long term use 'on battery' Feed them from an external
battery with a higher capacity and they can go into meltdown without
extensive cooling modifications.

--
 
D

Dave Liquorice

At about the typical "90 lumens" per white LED package that would
require about 180 separate light sources costing anything from GBP 2 -
4 a piece.

LED lighting is evolving very quickly, you can get strips these days
with LEDs mounted every few inches, rather individual "bulbs".

I wouldn't like to say how their efficiency compares to florry
though...

Or there are GU10 LED or R50 sized reflectors that only take a watt
or two at mains voltages. The aldli/lidl 3W Warm White 3000K LED R50
ES "reflector" I got to try is a pretty close match to a 60W R50 ES
tungsten.
"Livarno EdiLight".

Directional of course not spray light everywhere like a florry.
 
D

Don Klipstein

Currently got a remote observation site (wildlife) with no grid fed
power nor any prospect of it.

Half the site has a few modified 4ft T5's (36W) retrofitted with 12v
IOTA Ballasts (2D12-1-32) fed from a lead acid battery charged by a
solar panel. The other half of the site has 5ft T5's with magnetic
ballasts fed by a Honda EU20i generator which despite being a quiet
suitcase model and loads of additional soundproofing is still way too
noisy. Near silent operation is essential. Hauling fuel is also a
PITA as its a long way from the road.

So I need a way of powering the 5ft T5's (58W) from a low voltage DC
supply. IOTA only make ballasts up to 40W and they need a circa 50v
supply, realistically I need to keep to 12v to keep the solar array
price down.

So thoughts turned to an inverter fed from an uprated solar array and
battery.

A cheap modified sine wave inverter (circa 500W capacity) on a 100Ah
brand new battery fails to even kick even one 5ft tube into life. The
manufacturer says these inverters are not compatible with fluorescent
tubes but doesn't elaborate any further.

Does anyone have any ideas on how to get these lights working off
grid?

A change of ballast to an electronic type? (all indications are this
could won't work?)

Moving to a pure sine wave inverter (extremely expensive) ?

A different inverter supplier rather than 'one hung lo china inc' ?'

A ballast supplier that offers 12v ballasts that will drive a 58W
tube?

A homebrew 12V fluorescent inverter, running at high frequency that
will drive 5ft tubes and costs not a lot? :)

I would replace the 5-footers with 4-footers. More 4-footers are made
than all other sizes of fluoros combined. Going with the flow will
greatly improve your ballast choices.
 
M

m II

"Man at B&Q" wrote in message

Since you appear to be in the UK, the first product source below won't
be much help, but the concept might be an alternative.

Do you need fluorescent tubes or would LED bulbs be a more efficient
choice for solar powered illumination?

The local home/garden center has 120 volt 7.5 watt LED bulbs (lumens
equivalent to 40 watt incandescent) for $10US each.

More like 30W or less equivalent in a real world comparison.
Two of these LED bulbs should provide more light than a 5 foot
fluorescent tube plus provides power savings.

15W LED v. 58W flourescent, not a hope.

MBQ


-----------------------

and very bad light colour.


mike
 
A

Andrew Gabriel

I would replace the 5-footers with 4-footers. More 4-footers are made
than all other sizes of fluoros combined. Going with the flow will
greatly improve your ballast choices.

Whilst that is likely in the US (I'm always somewhat surprised to see
large US supermarket lit with rows and rows of 4' T12 tubes), it's not true
here. 4' was a popular size in the home in the 1960's, but in supermarkets
and other similar installations, they tend to be 5', 6' or 8' tubes (8' is
becoming rarer now), but never 4', possibly because of their low loading
compared with the longer tubes.
In offices, you can find 4' tubes in 1200x600mm modular ceiling lights,
but 2' tubes in 600x600mm modular ceiling lights are much more common.
(These are giving way to externally ballasted compact fluorescents now.)
 
D

Dave Plowman (News)

The other half of the site has 5ft T5's with magnetic
ballasts fed by a Honda EU20i generator which despite being a quiet
suitcase model and loads of additional soundproofing is still way too
noisy. Near silent operation is essential. Hauling fuel is also a
PITA as its a long way from the road.
So I need a way of powering the 5ft T5's (58W) from a low voltage DC
supply. IOTA only make ballasts up to 40W and they need a circa 50v
supply, realistically I need to keep to 12v to keep the solar array
price down.

My feeling is that magnetic ballast mains florries fed via an inverter
would be very little more efficient than 12 volt halogen lighting.
 
"Man at B&Q" wrote in message



More like 30W or less equivalent in a real world comparison.


15W LED v. 58W flourescent, not a hope.

MBQ


-----------------------

and very bad light colour.


mike
Depends entirely where you NEED the light. The LED can provide as much
light in a restricted area as the flourescent does - but will NOT
light as large an area to that brightness. So the question is - how
much light do you need and where???
If the light scattered all over by the flourescent is needed - use
flourescent. If it is just wasted (not needed anyway) try the LED
solution.
 
electronic invertors get over 90% efficiency, the difference between
the 2 is huge
I agree - if the inverter is run in it's "sweet spot" - but if,
say, a 1200 watt inverter is used to power a 13 watt load, the
efficiency drops WAY DOWN because the quiescent load (what it takes to
run the inverter with no load) becomes a sizeable percentage of the
total power consumed. - and just how efficient IS a magnetic ballast??
Likely something closer to 80 than 90 - so combine that (let's say
80) with the (possibly) 70% efficiency of a lightly loaded inverter,
and you are down to about 56% efficiency - where, particularly for
task lighting, a low voltage halogen is DEFINITELY worth considering.
(and an LED positively SHINES.) (pun intended)
 
D

Dave Plowman (News)

electronic invertors get over 90% efficiency, the difference between
the 2 is huge

That may well be their peak efficiency. But driving an inductive load
which requires a decent sine wave? And the efficiency of the fittings is
likely poorer too if not getting the correct waveform.

It would be very interesting to know the actual efficiency of this setup
in practice. I'd say you'd be surprised.
 
Top