Amps, lamps and milliamps

I have a pair of VU meters which use the LM081CP IC as the audio amp. I use a Radio Shack 12 VAC, (6.3-0-6.3) 1.2 Amp transformer to power a simple bipolar power supply with a pair of 7909 voltage regulators and a couple of filter caps. The meters have holes on each side for mini lamps. They make the meters look really cool.

Now here’s my problem. The only mini lamps that I can buy a Radio Shack are 12 volt at 25ma. These lamps work for a few weeks and then start to burn out. Not all at once though. First one, then another, etc., until finally they are all dead. The lamps are hooked up in parallel across the transformer secondary. Obviously they are not rated right; either that or they are just poorly made.

Any ideas? I can post some photos and a drawing of the schematic if that would help.

Do they have adequate air flow around them? Thinking maybe there is excessive heat?

Terry part of the problem is ''china'' . You might be pushing the bulbs limit . Have you tried placing a resistor in there to lower the volts 1 or 2 ? that might also add a softer start on them .
I dont think heat is a problem ? i have seen hundreds of these tiny bulbs jammed in gromets and wrapped up in aluminum in stuff over the years .

Why dont you use LED ?

What type bulbs are they? dimension, base,or type number?

Use diodes and run them on DC, that will make them last longer too. Running them on half wave will reduce the current but also reduce the light.

http://www.radioshack.com/product/index.jsp?productId=2102815

LEDs are an interesting idea, but I'm not sure how they would look. Plus wouldn't I have to get some DC for them?

I've been up to my armpits inside my mixer today trying to fix a bad pot. (I know, if I put my pot in a pipe instead of a mixer... )

Later on I'll post some photos of everything so you can get a better idea about what's going on.

I'm getting ready to re-assemble my mixer- Wish me luck.

Watch those fingers !

The control works now but I have no idea why. I did crimp the clips a little, and of course re-soldered it in. However in my haste I apparently didn't center it exactly right on the board. The shaft is off center in the hole. I can get the knob on so I can live with it. I''m not about to tear the thing apart again. The knob will be a little stiff. Heck, maybe that's why it works, who knows.

I did a little casting about and found these.

http://www.ebay.com/itm/370709567539?var=640050473716&ssPageName=STRK:MEWAX:IT&_trksid=p3984.m1423.l2649

LED would give off a more white light or you could use colored ones .
For DC just find the spot that feeds the bulbs and add a diode and a drop resistor

Ken g wrote:LED would give off a more white light or you could use colored ones .
For DC just find the spot that feeds the bulbs and add a diode and a drop resistor

White is too bright.



I've cut this box in half so that only the VU meters remain and I've replaced the headphone jack with an on/off switch. I've adjusted them so they mirror the bar output meters on my mixer. This way I can see from across the room if the signal is overmodulating. Of course I can only do this if they are illuminated.
Aren't these cool meters?

Are there ventilation holes in the top? Don't see any on the side.

try putting a 1N4004 diode in series. even though you are using DC for the lamps the extra voltage drop from the diodes should help by dropping the voltage by 6/10ths of a volt.

Resistance is Futile wrote:try putting a 1N4004 diode in series. even though you are using DC for the lamps the extra voltage drop from the diodes should help by dropping the voltage by 6/10ths of a volt.

The lamps are across the transformer output, i.e. 12VAC @ 1.2 A

I'm trying to get some photos uploaded, but I'm also busy trying to teach my dog to give himself a bath. If he could grow a thumb he could do the dishes and take out the garbage too.

Photos-

These are the back of the VU meter where the lamps go, the meter amp, the power supply, and the lamps.

Resistance is Futile wrote:try putting a 1N4004 diode in series. even though you are using DC for the lamps the extra voltage drop from the diodes should help by dropping the voltage by 6/10ths of a volt.

Hey that's what I said....only I didn't say it as eloquently as Cliff!

Cliff you are much better at expressing yourself than I am.

Ok then put more resistance on the led to dim it down , put in a diode .....etc etc... how can we help you ??

I stuck a white diode hooked to a battery into one of the meters. It looked weird. These meters need the warmer glow of an incandescent to compliment the cream colored faceplate. I like the limiting resister idea. What value? Also, would it be too much drain on the 7909's to use 9 volt lamps?

I've been checking the data sheet for the TL081CP. It can use a supply voltage from 7VDC to 36VDC. I could build a new supply using an 18 volt doorbell transformer and 78/79-12's. Using ohms law that means that if the resistance remains the same and the voltage goes up the current goes down. Does that sound right?

Use a diode, not an LED. Look back at the posts from me and better yet, from Cliff. Cliff took the time to list what diode you should use. with your incandecent bulb. If you use one diode in series with your bulb it will act as a half wave rectifier.

Without beating the idea to death, I wonder about the thermal specifications of the existing bulbs and if there is adequate ventilation?

The dropping resistor will depend on the current draw of the lamps. Ohm's law.

OK still try a diode as you would get half the wave and also cut the current so it should still help, or try series two of the bulbs so the only run on half the voltage prolonging the life. You would get a warm light rather than bright white. Another Idea is to use a cap in series since its A.C. ,to create reactance, experiment with the values.

Okay- on the agenda for today is a trip to radio Shack. I'm going to load up on spare bulbs and diodes. I'm not going to build a new power supply. This one works so why fix it? I might get some 9 volt bulbs and try coming off of the back, (DC), side of the power supply.

And N7ZAL- Every idea is appreciated and I've been beat to death by professionals, like my landlady. so don't worry. However the heat these lamps put out is so low, and the meter case so large, that I really doubt it is a factor. But you never know and further experimentation may prove that it is.

I personally have come to think that it has to do with resistances in parallel. Ohms law for parallel resistance, (8+8=4 Ohms, etc.), would make the combined resistance very low and thus current draw very high.

We'll see. Gee, looks like I'm gonna have a really fun afternoon.

you are right about ohms law. However I still am in a quandry as to why the bulbs are connected to the circuit card? The card you show has two regulators on it and they require DC to function. So if you say they are (the bulbs) running on AC, its confusing(to me anyway)
you have three wires coming out of the board on one end red, green, and black. Are those DC?
On the other end you have two yellow wires that have the bulbs spliced into them, so I am assuming they are 12 AC, right? If so you have all those bulbs in parallel? If so then your total current just for the bulbs is only 100ma. and the transformer is rated at 1.2 amps?
So that presents no problem if you are using only 1.1 amp to run your amp.

Your bulbs are using 25ma, So using ohms law R= E/I 12v/.025amps= 480 ohms so 480ohms/4 bulbs=120 Ohms.
without seeing the circuit of your power supply is it possible that the regulators are trying to stabilize the voltage when the lamps are first lit which is almost a direct short and the regulator is trying to compensate by upping the voltage till the filament heat up and have more resistance? And send a pulse back through the diodes? I am not for sure on that, but just a guess. Or there is a spike put out by the transformer on first turn on.
I'm just trying to sort through possible problems other than the obvious crappy bulbs scenario.

Well Terry, you are there and if they don't run hot...forget about my suggestion of heat being a factor.

Keep us posted...curious minds want to know.

Resistance is Futile wrote:you are right about ohms law. However I still am in a quandry as to why the bulbs are connected to the circuit card? The card you show has two regulators on it and they require DC to function. So if you say they are (the bulbs) running on AC, its confusing(to me anyway)
you have three wires coming out of the board on one end red, green, and black. Are those DC?
On the other end you have two yellow wires that have the bulbs spliced into them, so I am assuming they are 12 AC, right? If so you have all those bulbs in parallel? If so then your total current just for the bulbs is only 100ma. and the transformer is rated at 1.2 amps?
So that presents no problem if you are using only 1.1 amp to run your amp.

Your bulbs are using 25ma, So using ohms law R= E/I 12v/.025amps= 480 ohms so 480ohms/4 bulbs=120 Ohms.
without seeing the circuit of your power supply is it possible that the regulators are trying to stabilize the voltage when the lamps are first lit which is almost a direct short and the regulator is trying to compensate by upping the voltage till the filament heat up and have more resistance? And send a pulse back through the diodes? I am not for sure on that, but just a guess. Or there is a spike put out by the transformer on first turn on.
I'm just trying to sort through possible problems other than the obvious crappy bulbs scenario.

Wow! I love someone who knows what they're talking about.
To answer the first question- The power supply is as simple as it gets, two diodes, a couple of caps and the regulators. These IC's are just buffer amps for the meters and don't need much filtering.
And yes, I have the bulbs on the secondary of the transformer, before the diodes.

The trip to Radio Shack was a waste because they only had two bulbs. But-I have a universal power diode, (black with a band, I don't know what number), and I'll try bench testing those two with it and see what happens. Is there some test, voltage or current, that I should try?

P.S. This thread has become a lot of fun and I'm learning a lot. I like that.