PVC parts/MR11 bulb...Help?

[Nochrome]

Ok, there I was in Home Depot and I saw this MR 11 12v Halogen bulb (20w Phillips) and one thing led to another and I started thinking I could make a light with that bulb and some PVC plumbing parts.
I was pleased to find a fitting that fits the bulb perfectly and even has a screw down "bezel". Looks pretty cool with the bulb in it actually. I'm guessing heat will be the reason this is never gonna work but its more of an exercise to learn how to wire it up. So here's the questions.
1. I figure I can have mostly any battery configuration cause I can make the handle as long or short as I want (Or bigger or smaller diameter for that matter) so what battery setup would power the MR11 brightly and for a reasonable period of time? What kinda power can this bulb handle?
2. The light is plastic of course so how am I going to get power from + and - to the switch and the bulb.
3. Any ideas on a switch? (would kinda like a momentary function)
4. Any other thoughts ideas comments?
It might be apparent from my questions that I'm a rank beginner so thanks for any help. I just want to see if I can build something fun for just a few dollars and with basic find em any where parts. If nothing else I just want to turn it on and watch it melt. :thinking:

 

[mdocod]

it's a fun project... I made one a few years ago thinking the same thing... I never use it, but it's neat none-the-less...

I used an MR16 and 15 AA NIMH IIRC, but it has been awhile...

I have driven 12V long-life MR-16 halogens with as many as 18AA cells in a 6D mag before, but I always rested the cells after charging to prevent instaflash.


Somewhere around 15-18 NIMH or NiCD cells is the ballpark you want to be in. Pretty much any NIMH/NiCD cells around or larger than AA cells will work fine. The bulb will draw ~2.5A when overdriven like this, You can pretty much decide what size flashlight size and what runtime you want by selecting different cell options...

Alkaline cells won't work. Well.... they might for a few minutes, but it would be a serious waste of ~$10-20 worth of alkaline cells. Even D size alkaline cells into this sort of load would just totally fall flat on their face, delivering less than 10% of the capacity they are "rated" for with such a high drain rate. If you must keep the project as cheap as possible, get Tenergy brand NiCD AA cells from batteryjunction for $0.85 each, that's less than a name brand AA alkaline at the supermarket so that should keep you in budget and you'll be able to recharge them and have them around to experiment with other projects. If you want to splurge a little and make an investment in good cells now, get some eneloops or other LSD NIMH cells. If you want something in-between price wise, then I suggest the Titanium 2200mAH cells, they should be reasonably robust. The titanium brand of cells from battery-junction is easily the best bargain in batteries anywhere. Buying into lower capacity cells (like 2200 vs 2700) is actually not a bad thing, as lower capacity cells almost always have better overall behavior characteristics, less self-discharge, and better voltage maintenance under a load.

Moving up to larger pipe size and various adapters you could easily do C cells, or D cells. Again, NICD cells are on the cheap, but still in the $2.5-5 per cell range in this larger size, so this may be more than you wanted to spend....., or you can spend some big bucks on NIMH cells, but I wouldn't recommend that unless you think you have some other use for large NIMH cells around the house that would offset their high cost (you'd be looking over $100 in batteries to do this)...

Since you are just slapping cells end-to-end in a tube, make sure whatever cells you select have button tops on them. Many NICD cells these days are sold with flat tops as they are more commonly found in soldered/welded installations in rechargeable devices.

Alternatively, you could probably do 5 3.7V protected 18650 li-ion cells

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

I used a wire and a spring to make contact, but this was pretty dumb.... If I had to do it over again, and cared more, I think I would have rolled up some copper sheet (see hobby lobby, comes in rolls of thin gage if you want something locally)... And put the copper sheet roll up inside the PVC...

As you can see I drilled a hole and slapped a toggle switch on it, kind of neat... use whatever switch you like, just needs to be rated for ~24V ~5A or better.

As for melting... if you only run it for a few minutes at a time it will be fine. I played around with mine a bit and didn't have problems..

I should point out that I made a similar little "mr-16 housing" and installed it on the back of my SUV to use as an auxiliary light for backing up and other stuff. I accidentally left it on one time, and after driving all the way home with it, it was completely melted into a pile of pvc goo and deformed MR-16 on my bumper So they will melt if left on too long...

That's all I can think of for now to say...

 

[Nochrome]

First of all, thank you for a very helpful and detailed response.
The copper lining is a great idea. I was trying to think of a way to line the tube. I actually started with the idea to use copper plumbing parts but found that idea to be too expensive.
I also had the idea that I could use a rheostat type switch to vary the output?
Not that I undertstand such things but wouldn't that be a like a soft start feature?
I have some of the eneloop batteries (recent costco deal) so I could try those but I would like to keep the handle short.
Is there anything wrong with stacking the 3.7v type batteries such as the cr123's or 18650's? Is there any danger to stacking say 4 or 5 or 6 (end to end )? Incidentally how bright might a light like this be?

 

[mdocod]

First of all, thank you for a very helpful and detailed response.
The copper lining is a great idea. I was trying to think of a way to line the tube. I actually started with the idea to use copper plumbing parts but found that idea to be too expensive.
I also had the idea that I could use a rheostat type switch to vary the output?
Not that I undertstand such things but wouldn't that be a like a soft start feature?
I have some of the eneloop batteries (recent costco deal) so I could try those but I would like to keep the handle short.
Is there anything wrong with stacking the 3.7v type batteries such as the cr123's or 18650's? Is there any danger to stacking say 4 or 5 or 6 (end to end )? Incidentally how bright might a light like this be?

Within reason, there isn't anything wrong with stacking 3.7V batteries. There are battery packs out there for tools that are cramming enough cells for 28V and 36V and probably even higher out there somewhere. Higher voltage systems with many more cells in series are in various applications, so this is not really an issue within the range you are dealing with.

However, there is the issue of maximum safe discharge rates.

Most RCR123 cells on the market are lithium cobalt chemistry, with a true capacity of ~500-600mA, and a maximum manufacture suggested discharge rate of 2C, or about 1.1-1.2A. Your bulb is going to be pulling ~2-2.5+A (varies depending on the specific bulb), so if you are going to use LiCo cells, you need to have around ~1.25AH of true cell capacity. The best option here would probably be AW brand protected 18500 cells, you could probably get away with 5 of them before the bulb blows. 6 cells would almost certainly insta-flash it, and 4 would not be driving the bulb hard enough to be worth it.

5 18500 cells end to end would be ~10 inches long (batteries only), and run for ~30-35 minutes on that bulb. Not too bad I suppose... Keep in mind that when overdriven like this, it's no longer a 20W bulb. Check out this link:
http://i61.photobucket.com/albums/h75/pike444/bulbs/W-4743.jpg

that's a fairly typical 12V long life halogen bulb being overdriven. Complements of LuxLuthor's destructive bulb testing. Find the full thread in the threads of interest....

At ~18.5V it will run ~37W, and be producing roughly 1091 bulb lumen. Your bulb may vary.

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

Alternatively, you could use IMR16340 cells from AW, which are lithium manganese chemistry, and can handle safe discharge rates of up to 8C, they are ~550mAH rated cells, and into this load they will deliver ~450mAH of capacity, for a runtime of somewhere around 10-12 minutes. Total length a little under 7 inches for the cells.

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

You *could* use something like a rheostat, or "dimmer" style switch, but it would need to be rated to handle the power it would have to dissipate, it's not going to be some little miniature thing, you would need something that could really dissipate a LOT of heat. Resistance just chews into efficiency anyways, I wouldn't bother personally. Just get a good high current switch.

Eric

 

[Nochrome]

Wow! once again thank you. Now I can get started. It sounds like it will produce enough light to be a worthwhile project. Very useful info on the batteries. I will do some reading on the AW's you mention and I'll try to post some pictures if can get it to work.

 

[fivemega]

Similar thing with 12AA in M*g host first offered by Ginseng years ago here called Polaris. Then I offered 12AA M*g which is basicaly similar but better quality battery holder and different bulb. To use with MR11 or MR16 this type socket is easier to use.
Better choice is this kind of extension with 5C M*g using 5 AW's li-ion C cells.
Finally best choice is using 6x17670 serial battery holder in 2.5D M*g with soft start which can even overdrive 12 volt, 35 watt, 4000 hours life MR16 similar to this.
BTW why MR11 with 1.4" diameter? MR16 is 2" diameter which will fit in M*glite head.

 

[Nochrome]

I must confess I have no good reason to use the Mr11. I'm just trying to throw something cheap together so I can learn which direction I want to go before I build a real light.
Also I kinda like putting things together from junk and spare parts. Thanks for the battery suggestions.
I have the head of the light assembled uisng one of the bi-pin sockets you get at the hardware store. I lit it up using a car battery . It is certainly bright. I just have to figure out how to power it without spending a fortune on batteries and thus defeating the purpose of it being cheap. I guess I could carry the car battery around.... :thinking:. Thanks for taking the time to respond.