how to test ballasts??

[treesprite]

Is there an easy way to test ballasts? I have 2 fixtures that both have one set of lights out, and I don't know if it's the ballasts or if there is something wrong with the wires or switches (everything looks fine to my naked untrained eye). One is a PC and the other is a T-5.

[Sikryd]

Is there an easy way to test ballasts? I have 2 fixtures that both have one set of lights out, and I don't know if it's the ballasts or if there is something wrong with the wires or switches (everything looks fine to my naked untrained eye). One is a PC and the other is a T-5.

 

Depends on who you ask......I use my tongue

 

A multi-meter is the only way I know of.

[treesprite]

So is it just a matter of testing current? If no current is going through then it's bad?

[Jon Lazar]

Is there an easy way to test ballasts? I have 2 fixtures that both have one set of lights out, and I don't know if it's the ballasts or if there is something wrong with the wires or switches (everything looks fine to my naked untrained eye). One is a PC and the other is a T-5.

treesprite,

 

I would put a multimeter on the ballast and see if there's electricity going in and coming out of it. I don't know how much current your ballast is supposed to put out, but no electricity means the ballast isn't working.

 

Also, PC lights are electronically the same as T5s...the PC manufacturers just take a T5 bulb and bend it into a U.

 

Jon

[Coral Hind]

You should be checking for continuity or voltage not current.

 

Did you swap the bulbs around and you are still getting the same side that is out?

Have you made sure that the ballast wires are secured into the bulb holders?

Is there just one switch for both bulbs or does each bulb have its own switch?

Is this a fixture that is in use over a tank or can you bring it to frag fest?

[Origami]

You should be checking for continuity or voltage not current.

 

First try swapping bulbs. Then try swapping ballasts. See if the problem follows either. If it follows the bulb, that's the likely problem. If it follows the ballast, that's the likely problem. If it follows neither, then it's likely a wiring problem. In that case, try a continuity check using a multimeter on the input and on the output wiring. If you find a problem, fix it. If not, it's then that I would put the multimeter in voltage mode and trace the input AC voltage from the fixture input to the ballast input. If you have AC on the input side of the ballast and the wiring is good to the light, I would suspect the ballast is bad and would replace it (provided that the problem is not the bulb).

 

Who makes the lights, Forrest? Some manufacturers do a shoddy job of wiring and it could be as simple as a loose connection inside the housing. I had a light that had that problem once.

[treesprite]

These are both Current USA fixtures. The PC one is not in use. The T-5 one is the one over my 65.

 

The PC fixture is not in use right now, but I would use it if both bulbs worked. Nothing happens when I switch out bulbs. I don't see anything wrong with wiring inside that would be obvious to an untrained eye. That one I could bring with me to the frag fest if anyone is willing to have a look at it.

 

The T-5 fixture is over my 65. One of the light sockets melted because of either water or salt creep getting to it (or so I believe), so I thought there was a wiring problem. I took apart the fixture the other day and didn't see anything obviously wrong with the wires themselves, and switching bulbs makes no difference. These bulbs have been out for like half a year without the corals being affected, so I wasn't terribly worried about it, but I want to use the ballasts in a retrofit hood as long as they still work.

[cabrego]

These are both Current USA fixtures. The PC one is not in use. The T-5 one is the one over my 65.

 

The PC fixture is not in use right now, but I would use it if both bulbs worked. Nothing happens when I switch out bulbs. I don't see anything wrong with wiring inside that would be obvious to an untrained eye. That one I could bring with me to the frag fest if anyone is willing to have a look at it.

 

The T-5 fixture is over my 65. One of the light sockets melted because of either water or salt creep getting to it (or so I believe), so I thought there was a wiring problem. I took apart the fixture the other day and didn't see anything obviously wrong with the wires themselves, and switching bulbs makes no difference. These bulbs have been out for like half a year without the corals being affected, so I wasn't terribly worried about it, but I want to use the ballasts in a retrofit hood as long as they still work.

 

 

Melted socket is likely a sign of corrosion of the socket causing the ballast to burn out. I had a recent experience with this. It looks appears the only solution to my problem is to replace the ballast and melted sockets.

[Coral Hind]

What wattage are the bulbs? I might have the ballast to replace them with?

 

A lot of fixtures with PC bulbs use individual ballasts? When you had the PC fixture open did you see two small ballasts?

[treesprite]

What wattage are the bulbs? I might have the ballast to replace them with?

 

A lot of fixtures with PC bulbs use individual ballasts? When you had the PC fixture open did you see two small ballasts?

The T-5 fixture has 2 of 39 watt bulbs per ballast, and the PC has 1 of 40 watt bulb per ballast.

 

Melted socket is likely a sign of corrosion of the socket causing the ballast to burn out. I had a recent experience with this. It looks appears the only solution to my problem is to replace the ballast and melted sockets.

So even though the wires don't look bad that are connected to the socket, this could still have caused damage to the ballast? The other ballast I replaced the switch for about a year ago because it was melted , but the ballast was not affected nor anything else, just the switch.... there was a thick coat of salt creep in that corner, so I'm assuming I somehow splashed water in there.

[Jon Lazar]

The melting could have damaged the endcap so that there's an opening in the circuit.

 

The T-5 fixture has 2 of 39 watt bulbs per ballast, and the PC has 1 of 40 watt bulb per ballast.

 

 

So even though the wires don't look bad that are connected to the socket, this could still have caused damage to the ballast? The other ballast I replaced the switch for about a year ago because it was melted , but the ballast was not affected nor anything else, just the switch.... there was a thick coat of salt creep in that corner, so I'm assuming I somehow splashed water in there.

[cabrego]

The T-5 fixture has 2 of 39 watt bulbs per ballast, and the PC has 1 of 40 watt bulb per ballast.

 

 

So even though the wires don't look bad that are connected to the socket, this could still have caused damage to the ballast? The other ballast I replaced the switch for about a year ago because it was melted , but the ballast was not affected nor anything else, just the switch.... there was a thick coat of salt creep in that corner, so I'm assuming I somehow splashed water in there.

 

 

Looking at the wires only gives you a piece of the puzzle, most end caps will actually will open with gentle pry of a screw driver. And you can see the inner workings of it which are really quite simple typically consisting of conducting material such as copper. If I had to guess I would say your conducing material began coroding away causing a decrease in electrical resistance and and a large current spikeand fried the ballast.

 

But, the only way to know for sure is to open it up and inspect it.

[Coral Hind]

cabrego, The corrosion would increase resistance which would create the heat that would melt an endcap.

 

Forrest, Bring the PC fixture to frag fest and we can take a look at it. If your tank can do without the T5 fixture bring it also.

[cabrego]

cabrego, The corrosion would increase resistance which would create the heat that would melt an endcap.

 

Forrest, Bring the PC fixture to frag fest and we can take a look at it. If your tank can do without the T5 fixture bring it also.

 

 

I think there are several ways to form an argument for what happend, mine is based on simple equations;

 

Power=current*voltage

or , P=IV

 

 

and V=IR (ohms law)

 

or I=V/R

 

 

 

I assume a large amount of power went through the circuit causing the heat. Given a constant voltage, only increasing the current will do this; it follows that the resistance must have decreased. As copper corrodes away copper material is literally lost, and since resistance is directly proportional to the cross sectional area of the conducting material the reistance must go down.

 

Thats my basic logic at least for now..

[Jon Lazar]

I think it melted because it got too hot.

[treesprite]

I think it melted because it got too hot.

But "why" it got too hot gives more indication as to possible state of the ballast... I guess.

 

I'll bring the PC fixture to the frag fest with me and leave the T-5 one at home.

Edited June 25, 2009 by treesprite

[Origami]

... since resistance is directly proportional to the cross sectional area of the conducting material the reistance must go down.

 

Material resistance is inversely proportional to material cross section. If what you were saying above is true, then high tension wiring would benefit from being hair-thin or thinner.

 

(Resistance = (material resistivity * length) / cross-sectional area)

 

Corrosion (metal oxides) have lower conductivity (read that as higher resistance) than metals. This is because their "free" electrons are bound up in the covalent bond of the oxide. If these oxides begin to interfere with the movement of electrons (that's resistance) heat goes up because you've put a resistance in where once there was virtually none.

 

Since Forrest has a melted socket, it's clear that heat caused it. Some possible failure scenarios that support this observation include:

 

1) Corrosion in the contact.) Since ballasts are current-regulated devices, corrosion in the contact would cause excess heat.

2) Corrosion between contacts. A path between contacts would sap current and put heat where it didn't belong.

3) Failed ballast. A ballast that has lost it's ability to regulate current could put excess heat into it's load.

 

Forrest, your ballast may or may not be working. I'd first focus on replacing that melted socket (I'm assuming it's really melted and not just brown from heat). It sounds like you need to do that anyway. As for whether or not the ballast is also bad: If you can open up your light and swap ballasts inside the fixture (actually, just remove the suspect ballast and replace it with the known-good ballast - there's no sense in putting the good ballast on a bad load), you can probably definitively determine if you've got a ballast failure. If you haven't, replacing the socket should fix you up. If the suspect ballast will not fire the light on the known-good circuit, you'll probably have to replace it also.

[cabrego]

Material resistance is inversely proportional to material cross section. If what you were saying above is true, then high tension wiring would benefit from being hair-thin or thinner.

 

(Resistance = (material resistivity * length) / cross-sectional area)

 

Corrosion (metal oxides) have lower conductivity (read that as higher resistance) than metals. This is because their "free" electrons are bound up in the covalent bond of the oxide. If these oxides begin to interfere with the movement of electrons (that's resistance) heat goes up because you've put a resistance in where once there was virtually none.

 

Since Forrest has a melted socket, it's clear that heat caused it. Some possible failure scenarios that support this observation include:

 

1) Corrosion in the contact.) Since ballasts are current-regulated devices, corrosion in the contact would cause excess heat.

2) Corrosion between contacts. A path between contacts would sap current and put heat where it didn't belong.

3) Failed ballast. A ballast that has lost it's ability to regulate current could put excess heat into it's load.

 

Forrest, your ballast may or may not be working. I'd first focus on replacing that melted socket (I'm assuming it's really melted and not just brown from heat). It sounds like you need to do that anyway. As for whether or not the ballast is also bad: If you can open up your light and swap ballasts inside the fixture (actually, just remove the suspect ballast and replace it with the known-good ballast - there's no sense in putting the good ballast on a bad load), you can probably definitively determine if you've got a ballast failure. If you haven't, replacing the socket should fix you up. If the suspect ballast will not fire the light on the known-good circuit, you'll probably have to replace it also.

 

Yes, I meant to say resistivity in that previous statement.

 

Forrest, what kind of light are you working with?

[Origami]

(actually, just remove the suspect ballast and replace it with the known-good ballast - there's no sense in putting the good ballast on a bad load)

 

^^^

That reads wrong. Sorry. It should say, remove the known good ballast and replace it with the suspect ballast. If it starts the lamp, the suspect ballast is good. If it doesn't, it's likely bad. Don't attach the known-good ballast to the fried socket.

[treesprite]

 

 

I have been searching for a replacement endcap for the Nova Extreme for a few months and still haven't found one, so I have given up on that.

 

I switched around wires in the T-5 fixture and the bulb did not light, but I may have don't something incorrectly. I still can't figure out a way to remove the ballasts without cutting up the metal fixture housing.

Edited June 26, 2009 by treesprite

[cabrego]

I have been searching for a replacement endcap for the Nova Extreme for a few months and still haven't found one, so I have given up on that.

 

I switched around wires in the T-5 fixture and the bulb did not light, but I may have don't something incorrectly. I still can't figure out a way to remove the ballasts without cutting up the metal fixture housing.

 

 

I know I have seen replacement socket sets of 4 for the nova extreme on websites. I have seen them sold in sets of 4 with the metal bracket. You can also buy the DIY end caps and extract the sock, they are essentially the same size but, do not fit on the nova extreme bracket but are useful for testing.

[treesprite]

I know I have seen replacement socket sets of 4 for the nova extreme on websites. I have seen them sold in sets of 4 with the metal bracket. You can also buy the DIY end caps and extract the sock, they are essentially the same size but, do not fit on the nova extreme bracket but are useful for testing.

That's interesting, because I have searched and searched, yet all I can find anyone selling JUST the endcaps. I don't want to buy an endcap just to take it apart. Eventually I will be taking the working ballast from the fixture to use in a canopy.... what I want is for the currently questionable ballast to be ok so that I don't have to buy a new one when I make that canopy. Not having the 2 bulbs working is not affecting my corals, so as far as the fixture itself goes, I don't really mind the bulbs being out.

[cabrego]

That's interesting, because I have searched and searched, yet all I can find anyone selling JUST the endcaps. I don't want to buy an endcap just to take it apart. Eventually I will be taking the working ballast from the fixture to use in a canopy.... what I want is for the currently questionable ballast to be ok so that I don't have to buy a new one when I make that canopy. Not having the 2 bulbs working is not affecting my corals, so as far as the fixture itself goes, I don't really mind the bulbs being out.

 

 

Have you tried contacting current direclty? I am sure they can help finde replacment parts.

[treesprite]

Have you tried contacting current direclty? I am sure they can help finde replacment parts.

I really just want the ballast to work so I can use it when I build the canopy - the SPS have been growing ok without the 2 missing bulbs, so I'm not worried about repairing the T-5 fixture itself anymore.

[Coral Hind]

These are the end caps I use for my DIY T5 setups. I have never had any issues with them.

 

http://www.businesslights.com/challenger-t5-fluorescent-twist-and-lock-lampholder.html