Lower wattage of an IR bulb permanently?

[Emil Frolund]

Hi!

I have two 90w/12v infrared IR bulbs which is put into a r7s sockets. From thes socket goes in total four wires which is connected to our power supply. We have set two of the wires in serial so they pull 45 watts each, in total 90 watts.

The bulb is 90w / 12v

Our power supply is 12v / 8.3 amps

It should be able to power the bulbs since the power supply is 99.6 w and the bulbs are 90w in total.

But they dont, they cannot power the two bulbs.

However, if we put a bigger power supply which is 10a / 12v then it works.

If we then, after having the bigger power supply connected, switch to the smaller 99,6w power supply then it works as well.

It is only if we try to power the two bulbs directly from small one that it doesnt work.

My questions are

why can't the 99.6w power supply power the 90w IR bulbs? (They are in serial so they use 45w each)

is there a way we can lower the wattage of the bulbs? Maybe a resistance? Something else? I just need to lower them by 5 wattage and I'm sure it'll work.

Thanks in advance, and thanks for reading my long post!

 

[Skogsgurra]

Hej Emil

It is all about the cold resistance of the bulbs. When the filament is cold, the resistance is around ten times lower than the operating (hot) resistance. That is not a problem for a car accumulator or a transformer - both can deliver the inrush current without any problem.

An electronic power supply does not have that capability. There is a current limit which is activated within milliseconds if current gets above the set limit. So, you need a PSU with more punch in it to be able to start the bulb.

The fact that you can start the bulb on a heftier power supply unit is proof. And if you transfer to the smaller PSU, you can continue operating from that. But, if you switch the bulb off, you have to "kick start" it again from the more powerful PSU.



Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[Emil Frolund]

Hi Gunnar,

Thanks for your execllent reply. This tells me all I need.

There's one thing though, that we need to know about: Is it possible to lower this current limit that will kick the bulbs? For instance with a resistance?

Thanks in advance.

 

[Skogsgurra]

I think that you can. If you put a resistance in series (to avoid current limit activation), you may heat the filament enough and then short-circuit the resistor to get full power. You may not need more than 50 - 100 milliohms. Which, in my world, is a piece of Kanthal wire. Or even a piece of fence wire. Even barbed fence wire. You find such things everywhere.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[Emil Frolund]

Incredible, we will test this out and see if it works.

A question. On another forum a user told me that we might "start a fire" from the setup that we have now.

we have 2x 90w / 12v / 7.5a infrared bulbs in series

We have a 99.6w / 12v / 8.3a power supply.

I don't see anywhere where anything is being used at overcapacity and thus will be damaged or being overheated?

If we cannot get the 99.6w power supply to work, we might order 120w supplies because we know those work with the bulbs.

Do you see anything where a 120w / 12v / 10a power supply might damage anything, so that we potentially could start a fire from the setup we will be building?

Personally I have a hard time seeing it, but I am not very talented within electronics.

Thanks in advance, this is extremely valuable for us!

 

[Skogsgurra]

A much simpler and less vulnerable way of doing it would be to use a simple 230/12 V transformer with 250 VA rating. That will give you lots of head-room and costs less. If you need to vary output, you can use a dimmer that can handle inductive loads. Quite often, a simple dimmer for incandescent lamps works well if you put a snubber across transformer primary (to get up to the triac holding current before the gating pulse ends).

Re, starting fires. I don't see why that would be a problem. Unless you direct the lamp towards something combustible. Those are things that people like to say. Sometimes. I wouldn't pay much attention to it.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[zeusfaber]

Emil,

One other thing to watch:

They are in serial so they use 45w each

If they were simple fixed resistances, they would actually only use 22.5w each (as well as each seeing only half the voltage, two loads in series doubles the overall resistance, so the current is also halved)

Lamp loads aren't quite the same - put 6V across a 12V-rated lamp and it will stabilise at something more than half the current and a quarter of the power each (but still less than all the current and half the power each). I can't find a proper I-V curve, so can't put a proper figure to it. (anyone else?)

Why might this matter?

1. The lamps will end up drawing less current than you expect (once you get over the initial surge).

2. Together, the lamps will provide less IR than a single 90w lamp would on its own.

3. The IR output from the lamps is likely to have a quite different spectrum from what's advertised.

Maybe none of these things is a problem - but it's worth asking the question early.

A.

 

[Emil Frolund]

Thanks for your replies,

Gunnar:

Do you have a link to some specific resistors you would recommend we try?

I heard that we might need to use power resistors, in that case do you have a recommendation?

Thanks in advance

 

[waross]

You could try staggering the lamps;
Energize only one and then the second lamp.
No guarantees, we don't know at what current your current limit becomes active, but worth a try.
The 1/10 figure given by Gunnar is the accepted ratio for the initial current of an incandescent lamp.
But these are IR lamps. They may be running cooler than normal lamps and so will be lower on the initial current curve.
The ratio may lower, 1/9 or 1/8. That means that it is more likely that staggering the lamps will work than if they were normal lamps.
Still no guarantee.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Emil Frolund]

Waross

Thanks for your reply.

How would we stagger the lamps?

Thanks in advance!

 

[waross]

Turn one lamp on. Then turn the second lamp on.
No guarantees.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[crshears]

OP says the lamps are in series; it's not clear to me how you envision turning the lamps on one at a time...

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]

 

[waross]

Sorry. I became confused.
I like Gunnar's idea of using a transformer.
If you are running the lamps in series, the voltage is probably not important and the AC ripple should have no noticeable effect.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[cswilson]

What people are advocating here is a form of "soft start" circuitry to limit the inrush current.

The simplest soft start circuit is a "negative temperature coefficient" resistor in series. This will have high resistance at start up, then will reduce as current starts flowing through. It could be a good balance for the positive temperature coefficient of the filaments.

I don't have time to look for an NTC resistor that would have the specs you need, but the search shouldn't be too difficult.

 

[Emil Frolund]

@cswilson

Thanks for your input, this sounds like a great idea. if you ever get the time to find the specs of the Thermistor you would suggest we try out, we would love to hear what you think! It is super valuable for us.

After our multimeter tests it shows that the bulbs draw 10.8 amps from the beginning and then go down to 5 amps after they have started up. Which would you suggest we try?