You are at The Krib ->Lighting [E-mail]



  1. [M] Lighting: IceCap Ballast EMI Problem
    by (Gary Dudley) (20 May 92)
  2. IceCap wiring info...
    by Mike Bateman <> (Thu, 31 Aug 1995)
  3. more about e-ballast
    by Augustine J Rodriguez <> (Augustine Rodriguez) (Sat, 27 Jul 1996)
  4. Help, Ballast Blew! :(
    by (Craig Bingman) (Tue, 29 Oct 1996)

[M] Lighting: IceCap Ballast EMI Problem

by (Gary Dudley)
Date: 20 May 92

I had gotten an Icecap 450 a couple months back, checked that it basically 
worked, then put it away until I get to the project that will use it.  This 
thread prompted me to pull it out and check it more carefully.  This is all 
based on my single sample, driving two rather old 48" VHO lamps, over just 
a few days (no lifetime testing).

First, on the subject of TV interference.  Mine doesn't affect the TV, but 
the TV is on cable, and so there is no easy path from the Icecap to the TV 
by radiated interference, and conducted interference doesn't seem to get 
into the TV picture either.

I poked around with an oscilloscope.  The ballast label has a wiring 
diagram, where the first fluorescent tube is connected by two red wires 
at one end, two yellow wires at the other end, for the second tube its 
orange and blue.  The red and orange pairs have the high voltage, and 
are driven with the same voltage, while the yellow and blue pairs are 
near ground.  All pairs have a low voltage for the lamp filaments.  You 
people having TV interference problems may want to try connecting the 
wires so the red and orange pairs are at opposite ends of the fixture.  
This way the radiated currents from each lamp would be in opposite 
directions and would mostly cancel out (not perfectly, but better than 
having them add together).  Also, the wiring itself can radiate, and you 
should keep the wiring as short as possible, and near to the lamps as 
much as possible, to help keep all the fields canceling out.  Keeping 
the wires bundled or twisted toegther may help too.


       DO THIS                                 NOT THIS

   red==####lamp1####==yellow              red==####lamp1####==yellow

  blue==####lamp2####==orange           orange==####lamp2####==blue

The device operates in the 30 to 40 KHz frequency range, as a switching 
supply, so it generates quite a bit of higher frequencies.  I can see 
quite a bit of 30KHz in the AC line current.  I'm a little surprised to 
not see any FCC approval on the ballast, I thought this device would 
fall within their jurisdiction.  Maybe this is part of the "problem", 
it should be covered but wasn't designed to meet the requirements.  BTW, 
the FCC is taking a pretty dim view of marketing unapproved devices, 
but correcting interference from approved devices is considered the 
user's problem (and it sound like the level of interference people are 
seeing MAY be within limits).

Other issues I note:

1.  As in most switchers, it only draws current at the peak of the AC 
       voltage.  Unfortunately, it draws a LOT of current when it is 
       first powered up (I measured single half cycle pulses between 
       30 and 40 amps).  I would expect this to wear out or fuse the 
       contacts in a timer used to switch it on/off automatically, 
       since most timers are rated for 10 to 15 amps.  Anyone seen 
       this problem?

2.  The ballast has "smart" "protective" circuitry inside, which 
       includes shutdown if it thinks something is faulty.  If the AC 
       power goes out for about 1 to 5 seconds, the lamps don't ever 
       relight.  Shorter or longer outages don't cause a problem.  
       Around here, 1 to 2 second outages are relatively common, so I 
       guess I can't use this ballast for any critical lighting.

3.  General specsmanship:  Here I'm comparing the same lamps, with the 
       ICECAP ballast versus a conventional transformer ballast (a 20 
       pound brick ;-), each configuration operating for 15 minutes 
       to allow things to stabilize.  I make no great claims of 
       precision, but I tried my best, if anyone wants all the gory 
       details of equipment and measurements, email me.

       Icecap claim            measured (Icecap vs conventional)
       40% brighter            90% (ie 10% DIMMER)
       62% less electricity    80% (ie 20% less electricity)
       6X lamp life            How?  The filaments are always powered, 
                               the lamps get about the same power, 
                               everything should wear out just as fast.

Overall, I'm not real impressed.  In its favor, it IS pretty efficient, 
and I'm waiting for the programmable dimmer to become available (any 
news here?  Couple months ago it was "RSN").  All the little things I 
infer from its operation say its an ok, but not great, design, probably 
not ready for mass production.  IMHO.  But it looks an awful lot like 
the "original" "smartballast".  Yeah, I bought one of those too, it died 
(not under warantee :-( ), so I took it apart, thought "gee, I could do 
a lot better, if I had the time...", replaced the (improper) fuse in it 
with what it should have, it now works.  I had a talk with the "chief 
engineer" of smartballast at the MACNA3 conference, wasn't real impressed.

Just to try to spawn a discussion, how do people feel about the effects 
of fluorescent light flicker?  Conventionally ballasted lamps flicker 
120 times a second, Icecap ballasted lamps flicker in the 60KHz range.  
Julian Sprung thinks algaes and zooxanthellae benefit by the short dark 
periods from fluorescents, but presummably the electronic ballast flicker 
is much too fast to be of benefit.  OTOH, I'll conjecture, fish visual 
systems may be able to sense 120 Hz flicker, and this may contribute to 
stress, or problems raising fry, etc.  And/or the electric/magnetic fields 
(so which is more important?) from the lamps may upset the lateral line, 
again presummably the lowwer frequency can excite the nervous system more 
than the higher frequency.  I don't really want to open up the "ground the 
tank" debate again, lets just keep it to frequency related effects.


Smart     :  When you only believe half of what you hear...
Brilliant :  When you know which half to believe!!

IceCap wiring info...

by Mike Bateman <>
Date: Thu, 31 Aug 1995

> From: Earle Hamilton <>
> Regarding unsightly wires with the Ice Cap ballasts - - I would
> caution against braiding the lines together.  If I understand
> correctly the solid state ballast first converts 60 cycle to 30,000
> cycles per second.  This gives a very steady light that is with no
> flicker.  At any rate, perhaps a more recent engineering grad out
> there can comment on the effect of high frequency current on impedence
> (resistance). Since impedence acts like resistance and is a function
> of frequency - you may be asking for trouble by braiding the wires
> together.  This is my semi-educated guess and I will yield to somebody
> with better credentials.  BTW, you could call customer service at ICe
> Cap and ask them.  If there is a problem with braiding the wires then
> they should caution folks about the problem.
> Any takers out there who would like to follow up on this and give us
> an informed answer instead of an opionion?
I spoke to Chris at IceCap Industries and he said there is no problem
with braiding of the wires.  In fact he said it could reduce reduse any
noise that might be coming from the cables themselves.  Chris also noted
that they had tried this with a 150ft harness with no side effects.
I'm in the process of installing one of these myself and I'm very
interested in doing a tidy wiring job.
Of possible interest to others working with the IceCap...
After getting my system all wired for testing I powered the system up
and was standing in awe over the pretty bright lights when my dearest
sweetheart hollared across the room "What the -bleep- are you doing over
there?! You've killed the TV picture!?!"  Sure enough, it was a winter
wonderland on the TV.  Snow blizzard on CBS.
I was bummed.  I called IceCap and they informed me that this can occur
with TV antenna reception.  Cable is apparently immune to the effects of
the IceCap.  However, there is a fix.
Chris told me to ground my reflector.  I'm not sure how it works but
this is the fix.  Any kind of sheetmetal or even wire mesh like they use
on screen doors will be enough to drain off the interference.  It only
has to cover the entire length and width of the bulbs.
So far I am quite happy with the IceCap/VHO system.  I'm watching
closely to see if the system lives up to its claims.  So far the whole
system runs VERY cool.  The bulbs (4 36" 95W AquaSuns) get no warmer
than standard flourescent tubes and the ballast remains cool to the
The claim of 2.5x brighter than standard flourescents remains to be
seen.  Maybe 2.5 brighter than a Sears shoplight.  Ultra TriLux bulbs
are damn bright and although these bulbs appear brighter, a meter would
be required to determine how much.
// Michael A. Bateman                            
// Independent Networking Specialist   
// St. Louis, MO
// 314.282.3883

more about e-ballast

by Augustine J Rodriguez <> (Augustine Rodriguez)
Date: Sat, 27 Jul 1996

>>I installed the new dimmer designed by Niche Engineering.
>Is this the dimmer rig that Champion sells?

Yes.  It is the non-programable one.  It sells for about $130.  The fully
programable one sells for about $350.  The non-programable one requires an
external timer.
BTW - Both Champion and Niche do a good job of walking you through the
installation, which is fairly straight forward -  as long as you accept the
fact that no fluorescent lamp will be linear during dim-up and dim-down
like  incandescent lamps.  The ground plane (a 1-1/2" wide piece of Al foil
running the length of the tube and wired to the main ground) is essential
in reducing the impedance and thus getting many small increments of light
intensity changes during the ramp-up/down cycles.  Also the bulbs need to
be "broken in" for 50 hrs or so before they dim properly.  The Hg is not
properly dispersed throughout the bulb until this time.

Augustine Rodriguez
Elk Mound, WI, USA

Help, Ballast Blew! :(

by (Craig Bingman)
Date: Tue, 29 Oct 1996
Newsgroup: rec.aquaria.marine.reefs,rec.aquaria.misc

In article <>, Mazer  <> wrote:

>Two reputable companies for transformers that i can recommend are advance and

Yes, they make fine ballasts.

>> What does everyone recommend for VHO?

>Again, i think the MH is the better route. there'll be other opinions, 
>of course.
>I have heard (archives) that Icecap ballasts do not live up to their marketing
>claims. that was about 1 year ago, so it may have changed.

Actually, from a somewhat limited data set, I'd say that the Icecap is 
actually a pretty good ballast.  there were major reliability problems 
with the product Several years ago, but these seem to have been more or 
less worked out now.

Perry Tishgart and I used to go around and around about this stuff.  
Mainly about internal reflector lamps, but that is another story.  And it 
was a long time ago.

The Icecaps work well.  It is now a stable product, as nearly as I can 
tell.  I can't remember the last time I heard a complaint about them, and 
it seems to be time to just let go of the old horror stories.  

VHO is a feasible way of keeping many types of corals in a reef tank.
So is metal halide.

The problem with all this is that there just isn'a a unique solution to 
lighing problems, skimmers, anything else.  There is always more than one 
way to skin a catfish.


Up to Lighting <- The Krib This page was last updated 29 October 1998