1000-1200 lumen rechargeable project light for 50$

[Klaus]

500 lumen rechargeable project light for 50$

How to do this ? Thanks to CPF quite easily - read on .... ;-)

First of all thanks to PeLu, Phantomas2002, kb0rrg, Gman (and others I might have forgotten) for inspirations, ideas, hints and tricks on this subject.

Basic idea was to have a BRIGHT xD-Mag style flashlight using the 12V MR16-type halogen bulbs

In the end I used a Brinkmann Promax 2D as it was cheaper than the 2D Mag I had on hand and would have been unhappy to use just for the nice aluminium body - the really inferior quality of the made in china Brinkmann wasn´t a point as these parts were not used anyway - just needed the aluminium housing anyway. (Sidenote: If you can´t decide flashlight-wise between Mag xD and Brinkmann xD - thrust me - the Promax ist a cheap product - so as often you get what you pay for.) The nice benefit was that the ProMax head fitted the 51mm MR16 reflector perfectly with no need for further parts - the Mag would have needed some additional fitting work as described in other threads.

I decided for the 2D version as I wasn´t into the 3/4/5/6D´s as much because of size and weight considerations - using this type of modification can be used as well for the larger ones with benefits in runtime and/or max putput.

The bulb/reflector assembly used was the OSRAM DECOSTAR IRC 51 20W MR16 combo at a claimed 6000 cd for the 10° version - the 35W bulb (no reflector) has 900lm so I guess I can safely calculate the projector based 20W part at 450 to 500 lumen based on this data (no data for the 20W bulb alone available) - these OSRAM IRC bulbs/reflectors are using power saving techniques like reflecting heat back onto the filament therefor reducing the needed current to heat it and OSRAM claims that the 20W part acts like a standard 35W part in light output.

I´m using 12 AA 900ma NiCd cells - 12 cells were used as all my trials showed the 20W bulb drawing around 1.7A and pulling the voltage down to around 12.x V quite fast in this configuration after some initial higher voltage with freshly loaded cells. 10 cells which nominally should give 12V under load weren´t sufficient for the high load - they quickly dropped to 10.xV under load. The initial higher voltage of around 14V I think shouldn´t harm too much as it is only for a short time and getting some less life out of a bulb rated for 4000h wasn´t considered to be a major concern.

I tested 1600ma NiMh Sanyo Twicells (my favorite cells for general use), 600ma NiMh el cheappos, pretty much same behaviour besides obviuosly runtime differences.

I ended up with the 900ma NiCd on two reasons - firstly the NiCd´s are much easier to charge - you will understand this better reading the part on how I load the beast - and they were on sale in my local electronic shop at around 0.90 USD pcs - so I thought better get twelve of these for the project (with soldering and fitting into the body still to come) than to wreck twelve of the much more expensive Sanyos.

Where to fit 12 AA´s was the next issue - as we know 4AA can replace a D type diameter-wise while being a bit shorter - (AA approx 5cm - D approx 6cm) - therefore you can calculate

8AA - 2D plus spacer
12AA - 3D plus spacer
16AA - 4D plus spacer
20AA - 4D
24AA - 5D
28AA - 6D

The 4D-6D´s were too bulky for me (while 5D would fit nice for the project with 2 X the 12AA setup at double capacity - maybe the 20AA in the 4D would be sufficient as well with the lower relative load and possibly less voltage drop - actually Phantomas was using 20AA / 4D with a 75W bulb in one of the projects inspiring me at a **somehwat** reduced runtime) - so I tried to fit 12 AA into a 2D to get the smaller formfactor. As the original bulb holder / reflector wasn´t necessary anyway I took apart a mag 2D - enough space with no problems with the bulb/holder plus switch assembly removed - just a switch would be needed anyway.

This was the point where I decided to give it a try (after on the fly trials with all the parts until here) and got the Brinkmann to fit everything into. Soldering the 12AA´s together was OK (just barely OK to be honest - working but looking ugly) - fitting them into the body was a challenge - scratched some of the cells and had some shorts while working it in - finally had it fit into the body with no further issues - just very careful .......

The switch so far is ugly but functional (maybe have to work on this a little more) - fits into the original switch "hole" of the body but protrudes a bit as the cells are inside at that point of the body - maybe I will modify it with a very tiny switch fitting the "hole" better but unable to handle the current which just switches a larger one inside the housing near the top handling the load. Using a 3D would mean being able to use the original switch spacewise (should be OK to handle 20W I think, hope, ..... but don´t know) but is larger - and to be honest it was just the 2D Brinkmann on sale ..... might have used the 3D if that one would have been on sale though ..... maybe - just like the smaller size as well.

And I will add one of this molded sleeve parts over the body anyway as it gets real hot when running continously - so the neopren will make this more comfortable and also hide the ugly switch. So maybe the switch will stay - don´t know yet .....

Loading the part was the other issue - having soldered the AA cells to make them fit the body and with the troubles getting them in - loading the AA cells outside in an standard loader obviuosly wasn´t an option. So I had to find a way to load the 12-cell pack inside the light. Most loaders aren´t able to handle 12 cells - and are expensive when smart - after some looking around I decided to just use some reasonable priced 18V power supply I thought might fit and use NiCds to work around the troubles when fast-charging NiMh´s - and finally found some old power supplies I had around (yes - I´m too coming from the IT world - and this was a former US robotics modem power supply - no more need after moving to ISDN and DSL now) - problem was it was 20V AC - needed 18V DC - one 1USD part got rid of that problem .....

So I put a socket into the light where normally the spring resides - to load I just unscrew the tail-end and plug the loader in - one hour+ later (loading at around 900ma) - voila ! I might make the socket fit into the body´s tailend directly to get rid of that ...... maybe ...

So after all of that you might ask yourself - so what ?

its fu**ing BRIGHT BRIGHT BRIGHT

it runs around 30 minutes continously

it is CHEAP

it is rechargeable

Part count & cost:

1 x OSRAM OSRAM DECOSTAR IRC 51 20W MR16 approx 7-8 USD
1 x Brinkmann Promax 2D or equivalent approx 10-15 USD
12 x 900ma NiCd´s approx 10-12 USD
1 x 18V DC 1A power supply approx 10-15 USD
Various cheap parts like switch, socket approx 1-2 USD

So we´re talking around 35 to 50 USD for a rechargeable 30min runtime 500 lumen light - variations in power / runtime are easily done using 35W / 50W MR16 bulbs (I really recommend these IRC OSRAMs) for higher output (900 lm / 1200 lm) or higher capacity cells like 1800ma NiMh´s for a 1 hour runtime - larger bodies like 3-6D´s can be used for less work in the switch area and/or longer runtime. For me this was a fun project and while I honestly wouldn´t have killed for a high-output flashlight and never considered paying a couple of hundred bucks for a light in this class but with the new "toy" on hand I wonder how I have lived without one )

Klaus

 

[Gman]

Hey, nice job Klaus! I know those Osrams, they are a good lamp. What kinda beam quality you get from that setup?

Yeah, seems the higher ah NIMH or Nicads would improve the run time some. Seems odd you went to 900 ah, but I understand the logic.

LOL... I know a guy in the UK who'll be watching this project closely!

Keep us informed.

 

[yclo]

Nice work!

Any heat problems?

YC

 

[JoeyL]

This sounds cool.
I'd love to see pictures of this.

I have an attache case I'd like to place a few sealed lead acid batteries with a recharger into with one of those hand held 500,000 candlepower searchlights. It would be heavy, but not as bad as a car battery which is what I drag along on long camping trips for reliable lighting.

Does it get hot when you run it for a continuous 30 minutes?

 

[kb0rrg]

Wow, great work. I have wanted to build this light myself, but moving cross-country and job hunting has distracted me. I love those 12v lamps! In fact, I have found the 50W on my bike helmet to be way overkill for 90% of my riding needs, thus I have added a 10W to my helmet. This is the perfect output for most riding. I switch to the 50W at high speeds and difficult terrain. The beam quality on these is amazing.

Some thoughts and questions.

I have been estimating that my bike helmet light that uses a 50W bulb is in the 800 lum range. (this is based upon the M6...500lum for 30watts). By using this method that would make a 20 watt lamp about 330 lum. I have tried to look up data on these lamps, but can not find any. What is your source? If you are correct, than I have been underrating my helmet light

<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>Originally posted by Klaus:
...actually Phantomas was using 20AA / 4D with a 75W bulb in one of the projects inspiring me at a **somehwat** reduced runtime) -
<HR></BLOCKQUOTE>

I am really interested in reading about this thread. How did I miss it?? When/where was it?

<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>Originally posted by Klaus:
[QB

And I will add one of this molded sleeve parts over the body anyway as it gets real hot when running continuously - so the neoprene will make this more comfortable and also hide the ugly switch. So maybe the switch will stay - don't know yet .....
[/QB]<HR></BLOCKQUOTE>

Just a thought...the neoprene may insulate the batteries even more, thus making the internal temp very high.

Are you using the body of the light as you ground? If so, how did you connect the wire to it?

How did you secure the batteries in the tube?

Does the lamp/reflector assembly you are using have an integrated glass lends?

I have not had any heat problems with my 50W light. But I have the batteries in a separate location and the back of the reflector is open to the air.

All in all great work. I am jealous

 

[Klaus]

@ Gman - thanks

>> What kinda beam quality you get from that setup?

beam quality - ompf - best I ever saw - flawless and even - need to try to get some beam pics - the OSRAM´s really are great little bulbs/reflectors (no I´m not a nationalist - grin)

>> Seems odd you went to 900 ah, but I understand the logic

cells - I was afraid to kill the NiMhs too fast due to the charger issue and I feared the soldering/fitting job - as it went OK and the higher capacity would be nice to have ..... maybe on the next one.

re UK-guy - guess who emailed me

@ yclo

heat problems - at least you have warm fingers in the winter - on my runtime tests (and while free-air-testing the parts) the cells got quite warm when approaching the discharged state - possibly around 2/3 to 3/4 of the discharge curve when running continously - anyone know the chemical reason for that ? - this is why I think about adding the sleeve - also when loading at around 1C the cells get warm when reaching the full state - this is by design and I use this as an additional "sensor".

@JoeyL - pics will/should follow - did you thought about small solar panels to load a smaller battery to save weight ? When you use a charger you need an AC power outlet.
And yes - it does double-duty as a heater on continuous rundown around 20min (its OFF after 30min) - but we are approaching winter anyway ....

@kb0rrg - thanks

>> What is your source?

Check out Osram IRC on page 8

Most manufacturer should give you ratings of their bulbs - I found lumens for the bulbs alone and candela ratings for the bulb/projector combo´s at the different spreads and tried to correlate those.

>> Phantomas2002 Mag4D 20AA / 75W "burner"

This is from Phantomas2002 hidden deep down in the thread
CPF thread "42LED"

To the 75 watts bulb: It is a usual 75W bulb with reflector (50mm diameter) from General Electrics. In Germany available from Conrad Electronics. And WITHOUT safety glass in front. This kind of lamp fits EXACTLY in the front of a Mag. I changed the plastic lens against a thermoresistant glass (at first I used usual window glass...2minites burntime...biiing... ). I also used a Power-MOSFET to swith the current (6 Amps are a little bit to much for the Mag-switch...). And theres only ONE way the 20 AA cells fit into the Mag: four parallel and five of these packs in a row. connect always 2 cells parallel and then these twopacks in a row. if you use 1000mAh cells (preferably from SANYO) this makes a 12V/2000mAh Accupack. I soldered on a connector, which is hidden behind the tailcap for charging. And voila you got something to outshine a car headlight by far.... *evilgrin*

Always remember: THE MORE YOU PUT IN THE MORE YOU GET OUT

My friends always say "The new Binford 2000 Power Flashlight"....

>> neoprene may insulate the batteries even more

Good point - might try anyway - the sleeve is on sale right now - really need to hide that ugly switch ..... (don´t tell anyone)

>> body of the light as your ground ?

Nope - but the cells do touch the body (tight fit) and it seems the heat gets dissippated quite nicely to the metal

>> How did you secure the batteries in the tube

Its quite a tight fit anyway - in the front the bulb/reflector combo / wire assembly and in the tail end the socket for the loader squeeze the cells from both sides in addition - so as there was no rattle or movement I didn´t lose any sleep on this.

>> Does the lamp/reflector assembly you are using have an integrated glass lends?

Yes - as the standard mag lens might have melted I thought why not just use one of the bulb/reflector combo´s WITH lens and don´t worry about it - nice fit - no problem.

>> I have not had any heat problems

Plus all those holes you drilled into that poor little Mag-Head - wait ´til the mag-revenger gets you

Thanks to all for comments & congrats - whoever needs more details shouldn´t hesitate to give me an email

Regards

Klaus

 

[pwell]

Hi Klaus

It was really interesting to read your post. I'm interested in seeing photos of this too. I'm also wondering where I can find more info on the OSRAM 20W MR16 bulb. And I have a question. Does this buld do most of the beam shaping for you, or does the Brinkman reflector do it? (or a combination?) I'm not familiar with the Brinkman.

This project reminds me of my first high power project: a 50W car cigarette lighter spotlight and a 6.5 AH sealed lead acid batt. I carried the battery in a belt pack. It went for about 40 minutes and was MEGA bright. We used it for bush bashing and the neighbour always called me when he wanted to go rat shooting.

Before sealed lead acid came out, a friend of ours used to lug around a spotlight and car battery. The car bat took up a whole backpack.

 

[Klaus]

Hi to Korea pwell,

working on the photos

I posted the URL for the OSRAM PDF file on the IRC bulbs/combos above in the earlier post - pls find it again here.OSRAM IRC INFO
If you need/want more info you might give the OSRAM web site a try at OSRAM WEB SITE and surf around there - quite some info if you dig deep enough.

The OSRAM part I am using is not just the bulb but a complete set of bulb/reflector - I think this setup is what makes the whole thing an "MR16" - the reflector is very nice and multi-faceted - I guess this makes the nice beam. I´m not using the Brinkmann reflector at all - it is one of the spare parts now - I´m basically just using the aluminium housing of the original light.
Due to this I can´t focus the beam though.

What you describe from your former high power lights is the old dilemma of high input = high output - but where to get the high input without too much weight

Actually this is why I decided to go with the 2D and sacrify the switch assy - just to get the smaller form factor (if you can name a 2D "small" anyway)

But possibly the next one will be either a 5D using 24AA´s or a 6D using 10 1/2D´s - will do some free-air-testing soon - and then decide to go for more power or more runtime

Or maybe just use my 3D and swap the three 7500 ma NiMh´s against 5 1/2D´s and see how to optimally fit some 10W mag-charger type bulb at 6V ? Would still have the normal reflector though ... Unfortunately there is no OSRAM IRC bulb/reflector at 6V - just few 6V bulbs from 5 to 50W - no reflector setup.

We´ll see

Klaus

 

[Klaus]

Update

modified the light with the grip shield and lite-ring from Nite Ize which were on sale recently - covers the ugly switch and gives a nice feel and you can run it continuously without getting too warm (hands). I also modified the charger with a switchable resistor to select a lower 100ma charge current for a c/10 style charging alternative (switchable from 1C to C/10 now).

I´m also looking in some cheapo IC to have deltaV voltage charge "intelligence" added - maybe for the next one where I think about using the 35W IRC Osram (even better efficiency - so possibly around 900-1000lumen) driven from some 20-24 Sanyo 1600ma NiMh cells - just free-air testing if I would need 24 or if 20 would suffice. And then need to get a cheap or used (just housing anyway) 4D or 5D Mag or similar - theoretically this should give me around 1 hour runtime at possibly 1000 lumens but with the quite large form factor. Maybe I even could squeeze the 20 in a 3D - hhmmmhhhhhh ....... just want to use up that other power supply at 1.2A 18V´s ,,,,,

----- testing done

I have looked into the 35W version of this Osram IRC reflector/bulb combo - on paper it gives you more than double the CP at "only" 75% more power. (20W ~ 6000CP/10° vs 35W ~ 12500CP/10° [old datasheet even reads 14000CP] - the 35W bulb alone is rated at 900 lumen)

I have free-air-tested the setup to run nicely using 24 NiMh 1600ma cells - using 10 sucked the voltage down to below 10V, even with 20 only 11V - using 12 cells (as in the 2D 20W one) also was no good as the load at approx 3A which is 2C even for the nice 1600ma Sanyos was too high for the NiMhs - so finally testing with 24 of these cells was nice and BRIGHT - initially overdriving the part to around 44W at 13,2V and 3330mA.

So I finally found why NiCds have a better name than NiMhs for high drain applications - in the 20W light the 12 900ma NiCds behaved much better at 2C load than 12 of the NiMhs at the 2C load of the 35W - but 24 of the NiMh´s (which is around 1C load then) should give around 1 hour runtime at 900-1000 lumens - sounds nice to me.

More to come I hope - just don´t want to buy one of the needed 5D Mags new right now at the exorbitant pricing they have in europe

Klaus

 

[Klaus]

Update II

yup - just had to try that 35W OSRAM IRC bulb of ~ 900-1000 lumens in the modified 2D brinkmann housing with the 900ma NiCds - WOW - and we need to come up with some special "blinded by the light" smiley

This thing was so unbelievable bright I just had to leave it that way - (as I broke the 20W bulb combo anyway

) - so for now the 24 x 1600ma NiMh project for that bulb combo is delayed until I find some el cheapo 5D housing and/or a real need for 1 hour runtime for that beast.

Lesson learned is definetively that NiCds do have a much lower voltage drop on high load than NiMhs - MUCH MUCH lower - compare my two setups:

To achieve a satisfying voltage for the 35W bulb I needed 24 1600ma NiMhs (and these are the Sanyos - named to be the best cells you can buy) - resulting in a 1C load on them. Going higher like to a 2C load (12 pcs) had the voltage drop unacceptably even with NiMhs being rated for 2C loads.

In comparison the NiCd´s were behaving perfectly fine at 2C as in the original setup using the 20W bulb - and as well doing fine with the 3,6C the 35W bulb put on them which consumed 3300ma at around 13,2V - but then I became curious and put BOTH the 20W and 35W bulb on the 12 cells - still doing fine - and this was at around 5C without major voltage drop - means I could even happily run the 50W/1260 lumen OSRAM bulb combo.

But as the 35W bulb is the most efficient one according to the paper specs I will let things stay as is right now - with the voltage and current I measured the setup initially is at around 45W and 1200 lumes anyway - for maybe 15 to 20 minutes or so going gradually down to somewhere between 900-1000 lumens which should still be fine - while no run-down test done so far.

And yuck - according to Brocks site and the SF sites data I´m even beating the M6 and 10X Dominator in lumens/runtime hands down - and all this for 50 bucks - project closed

(for now at least)

Klaus

 

[patg]

Would this project be much easier if you use 6 volt reflector bulbs? The MR11 20W version apparrently has a 400 lumen output with a 35mm reflector diameter. 6V would only use 1/2 the cells as 12V.

This looks like a neat project! I'd like to make it with less batteries (like maybe just 4C alkalines. What do you think?

I found the reflector bulbs here: http://www.reflectalite.com/halogenpage.html

Also, Willie Hunt's PWM voltage regulator looks perfect for this application: http://www.cs.indiana.edu/hyplan/willie/lvr.html#anchor0001

Let me know what you think.

Pat

 

[pwell]

<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>Originally posted by Klaus:
Update II

...but then I became curious and put BOTH the 20W and 35W bulb on the 12 cells - still doing fine - and this was at around 5C without major voltage drop - means I could even happily run the 50W/1260 lumen OSRAM bulb combo.

Klaus<HR></BLOCKQUOTE>

This is great news! I've been looking for the 20W and 35W 10 degree bulbs everywhere, but cant find them in Korea. They only have the 38 degreee ones. What they do have is the 50W 10 degree bulbs (model : 44870 SP). They are only about $2.50 here, so I picked up 5.

I did a quick test of the 50W, 10 degree bulb on a 4 Ah SLA and BOY DOES IT ROCK!

I was hoping to get the 35W 10 degree bulb too, since it looks like the most efficient. With Klaus's setup (12 NiCads in series) I guess we could get a runtime of 20 minutes.?

Since I have some spare 50W bulbs, would anyone be interested in doing a swap for 20W and 35W 10 degree ones? I suppose I can order them in, but I'm leaving Korea soon to go back to Australia. I expect it will be harder to find them there, but fingers are crossed.

It would be great to make a detachable bezel so you could easily change between the 20W and 50 heads, depending on how much light you need. I envy those with a lathe!

Please keep us updated Klaus! I cant wait to see pics if you have any.

Regards
Sean

 

[PeLu]

<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>Originally posted by Pat G:
Would this project be much easier if you use 6 volt reflector bulbs?
<HR></BLOCKQUOTE>

The Osram bulbs are not available for 6V and Klaus wants to use them for a good reason.

<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>
Also, Willie Hunt's PWM voltage regulator looks perfect for this application:
<HR></BLOCKQUOTE>

He's well aware of the Willie LVR.

 

[Klaus]

PeLu said it pretty much already (thx Peter) - but let me detail things out a bit more - ouch - this is getting longer by the minute I type .....

@ Pat G

>> Would this project be much easier if you use 6 volt reflector bulbs? The MR11 20W version apparrently has a 400 lumen output with a 35mm reflector diameter. 6V would only use 1/2 the cells as 12V.

6V wouldn´t need as much cells - agreed - but for the same wattage you would need the double current then because W = V * A - in the case of a 20W bulb at 6V this would be 3,3A - this would bring your 4C alkaline cells to their knees right away.

You could use 5 * 1/2D rechargeable cells (available at upto 3700mA) in a Mag 3D without major trouble to get 6V at 3,7A resulting in around one hour+ runtime - but you would run in two issues:
- the 35mm MR11 (1,37") wouldn´t fit the head as nicely as the 51mm MR16 (2") does (and the 35mm ones are 10-15% less efficient)
- and as PeLu mentioned I´m using the Osram IRCs for a reason - and this is to get any possible photon for the electrons available - and there is no OSRAM IRC at 6V.
Look at the Watt/Lumen for your proposed bulb - 400lumen for 20W ~ 20lm/W - the 20W Osram MR16 I used initially was around 500 for the same wattage ~ 25lm/W - a nice and cheap >20% gain.

I am actually looking into the 6V issue for a 3D conversion using the above mentioned 5 1/2D cells - where using a MR16 bulb combo could potentially solve the problem of the not-to-nice Mag reflector but at the cost of losing the adjustability of the beam - so if I would go from 3 X 7500ma D-cell to 5 x 3700ma 1/2D-cell (not shure on that yet as I would lose 20% capacity) to get 6V / 3,7A there would be some choices:

- using a 6V MR16 for which I´m still searching, only saw 6V MR11s so far
- just use a halogen PR bulb which are available in different flavors around 6V at 60 to 100 lumens and keep the adjustability instead of the better reflector (still looking into how to keep the cam-feature AND get a better reflector).
The efficiency of the best PR base halogens in the 6V area is pretty much the same as standard 6V MR16 ones at ~ 20lm/Watt

And still need to decide if its better to keep the 4V .85A bulb with 3Dcells at almost 9 hours runtime with 60 lumens or to go to one of the 6V setups at upto 100 lumens but only 3,5 hours.

Might just keep the curremt 3D as is - modified too much flashlights lately for higher brightness at shorter runtime and might keep that one for that 60 lumens / 9 hours setup.

And for 6V/12V in general - you can also use a 5D with rechargeable Dcells for 6V or a 6D with 10 1/2D cells for even 12V and higher drain capability compared to alkalines - while I didn´t tested these configurations for their behaviour under the load of the high wattage bulbs - (the Mag charger uses a 5 1/2D pack with 2200mA NiCd and a 10 or 11W halogen bulb) - but I still think my 2D equivalent is just a too nice form factor for 1000/1200 lumens ....

>> Also, Willie Hunt's PWM voltage regulator looks perfect for this application:

First of all let me say that I very much like Willie´s LVRs without ever having used them because they solve at least one of the incandescent bulbs inherent main problems (simplifying a bit you can say incandescents need a stable voltage for good efficiency [ok ok - the LVRs solve even some more]) but I seem to have solved that by experimenting a bit and by using 12 NiCd cells with my operating voltage going from 14V to 12V - which is right in the perfect range for these type of bulbs - so you can name me either
- just lucky having found it

- a typical german for my fanatism in details

- again just lucky running into it "by design"

(12 fitted as well as 10)

Also Willie himself gave me some evidence so-to-say afterwards when looking into whether to add an LVR or not - on the page you linked he writes: QUOTE

I want to run a 50 watt 12 volt bulb off of a 12 volt 6.5 AH SLA.
Will your LVR help?

The problem with using a "12" volt SLA to run a "12" volt bulb is that the 12V SLA (sealed lead acid battery) voltage is 14.0 charged and around 10 discharged, and most 12 volt bulbs are "cold" at 12.0 volts because they are designed to run for 3000 hours. They are much more efficient at 14.0 volt and still will last 50 to 100 hours. My LVR's can only regulate voltage down, not up, so you need to use a lower voltage bulb, or a higher voltage battery to make a "good" system.

The other problem with such highly loaded SLA is the capacity will be small relative to available capacity. With a 50 W bulb on a 12V 6.5AH SLA, your run time will be about 30 minutes. However, the SLA has about 70 Watt Hours of energy, so it "should" be able to run the 50 Watt bulb for about 1.4 hours. NiCad or NiMH should be seriously considered for discharge time of less than 4 hours. SLA really do best above 8 hour discharge times.

So how should I run my "12" volt bulb?

I recommend a 12 cell NiCad pack if the bulb is really 12.0 volts and a 16 cell NiCad pack if the bulb is 14.0 with a LVR3E custom adjusted to your battery voltage for warning flashing and cutoff. Alternatively, a 24 volt SLA (two 12 volt SLA in series) can be use with a LVR3 model 24V12. ENDQUOTE

So it seems I found one of the solutions Willie also recommends - a pack of 12 NiCds - while also for the sake of the project "500 lumens for 50 bucks" (which now even became the maybe even better 1000-1200 lumens for 50 bucks project) I hesitated to add another 50 bucks just for the LVR alone - while certainly having the LVR even just for nicety & professionality & a better switch & the "dimming and such" functions would be nice though - possibly Willie could even help to fine-tune a LVR for this project but as far as I can tell unknownwise he´s a very busy man these days and I didn´t wanted to "harm" him with this pet-project of mine not intended for mass-production and therefor certainly unprofitable for him - and BTW - does someone know what he´s doing at SF ? wouldn´t a SF with LVR be really nice ?

@ Sean

the 44870SP is the right one - you will be able to run it with 12 NiCd - at somewhat reduced runtime but BRIGHT - look for the highest capacity NiCs you can get / afford - you might also consider using a larger housing to fit 24 pcs (5D) for double the runtime - while the 50W one isn´t as efficient as the 35W - according to the Osram specs the 50W is 300cd/Watt while the 35W is 357cd/Watt. You might also check if you can get the Philips Masterline ES MR16s in Korea or Australia which on paper seem to be equivalent to the OSRAM IRCs - they even come in a 8° version which I think would be even better beamwise and at even more wattages (20,30,35,45) which might come in handy when finding the right brightness/runtime mix. Look at

Philips MR 16 Halogen Masterline ES "Energy-Saver"

Oh and it seems also with these Philips parts the most efficient (at least on paper) is the 35W version at 14000cd / 8° - I think these are pretty similar to the OSRAM IRC parts - at the same ° the cds are identical - while the 8° version might be nicer though than the 10° Osram if you want to throw a good beam - maybe need to try to get one of those to check it out.

>> It would be great to make a detachable bezel so you could easily change between the 20W and 50 heads, depending on how much light you need. I envy those with a lathe!

Hmmh, you would need multiple Mag/Brinkmann heads - I think doing them yourself is too much of a hassle - but if you do it right its a quick job to chance the bulb combo anyway - I right now use a (o sh** - what´s Lüsterklemme in [american] englisch ?) screw connection but recently bought a halogen bulb base (not installed yet) - using which it would be very fast & easy to change the bulb combo.

While having true detachable bezels would certainly be nice - possibly waterproof ones (using the Brinkmann head the 51mm MR16 combo fits nice but not 100% waterproof - the Mag-head fit is a bit looser still) - and my (ugly but hidden) switch also isn´t as waterproof as I would like - so still plenty of room for improvement like in using the discussed WH LVR as the switch just using a tiny button fitted nicely through the aluminium housing, having the head waterproof and easily swappable and having the loader be "intelligent" - oh my - I think I leave this one as is right now and wait for Santa Claus to drop a used/cheap 5D on me for that 1hour 1000/1200 lumens 24 X 1600NiMh 35W light-saber I start to drool about ...

- oh wait - and what about using the 24 pcs in series instead of 2x12 in series/parallel - having therefor some ~24V and being able to nicely use a WH LVR ??? Maybe even a 4D with 20pcs would be sufficient then and be smaller & nicer though - oh sh** - someone made me to start to THINK again ----

Klaus

 

[patg]

Klaus,
Thanks very much for your considerate explanation.
Your project sounds great. I wish you good luck and look forward to reading further news about it.

Peter, No thanks to you for your curt reply. It was of absolutely no help at all to me or anyone else.

Pat

 

[Klaus]

High Load NiCd / NiMh difference

Again with the help of fellow CPF posters (thx PeLu and others) the different behaviour of NiCd compared to NiMh became clearer to me - for a nice explanation go to NiCd vs NiMh and read especially starting at page 17 in this document Substitution of Rechargeable NiCd batteries

What seems to be the case in the newer breed of higher-cap NiCds (available now at upto 1200ma) now is that also these are using the foam or spongy electrods (used in NiMH) instead of the sinter ones to get the higher capacity.

This results in my warning for people trying to use these high-cap NiCd for projects like the above - they need to use either the "standard" cap NiCds (at around upto 900ma) able to handle the high load - or can go with the even higher-cap NiMhs anyway as the NiMhs disadvantage of the inferior behaviour under high loads (>1C) is shared by these higher-cap NiCds as well.

There seems to be a new development of "high-power" NiMhs as opposed to the traditional high-cap NiMhs which seem to go back to using Sinter for the Ni - these will be able to have a good high-load behaviour at a somewhat reduced cap.

I hope this is clear enough

Klaus

 

[brightnorm]

For Klaus and all the others in this thread,

I had just gotten my hard-earned M6 and was drooling over it when I started reading about your creative experiments. It became clear that you guys are the ones on the cutting edge, not Surefire.

I am extremely frustrated when I read about all your experiments, based on what seems to be comprehensive knowledge and insight into the electronic esoterica of lights.

I have abosolutey no skills or knowledge in this area, but I greatly appreciate the power and beauty of well designed flashlights that push the boundaries of current technology.

When you have finally refined one of your projects down to a functional working prototype that is easy to handle and doesn't look too ugly, would you ever consider selling it?

I am genuinely interested in purchasing one of these CPW-produced lights, if in fact it really can out perform/outrun a light like the M6, as long as it's in an easily portable form.

I'd love to get a reply from one or all of you working in this area.

Best regards,
Brightnorm

 

[lightuser]

All the big companies are good at making viable products; an outfit like SF produces many designs and actually reponds to the real needs of the end user. Many companies produce fine designs; and I think it's fair to say, Mr. Gransee may agree, that bringing any flashlight to the market is much like reinventing the wheel. There are multitudes of design flaws, adjustments and challenges. In contrast we hobby types are just playing and enjoying. The value of it is that CPF has a freewheeling aspect to it that proprietary concerns cannot really afford to indulge in. There is a lot of good exchange of ideas, unencumbered by the competitional aspects. It has a role to play in countering the limiting effect of competition on the exchange of ideas. For an engineering student it can be an educational supplement. An amazing array members post here from common users to highly specialized engineers. It's a nice common ground. Maybe there will evolve a library of "CPF specials" that are great DIY designs and plans derived from the free exchange of ideas. Already there are a number of great conversions available here like the DB37 and the RayoVac LS, and the new ArcLS is partly a CPF inspiration. As high tech parts are newly available more possibilities open up and get discussed. My own area of interest is the small HID flashlight because it has such high efficiency and impressive candlepower. As far as I know there are yet no small handheld HID designs in production. The technology is there yet no company has made it a production item. A light the size of the SF M3 can be made that puts out 1500 lumens. It's bound to happen soon...

 

[vcal]

<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>Originally posted by Klaus:

for a nice explanation go to Substitution of Rechargeable NiCd batteries
I hope this is clear enough

Klaus<HR></BLOCKQUOTE>

Klaus, would you happen to have an ident. URL for that page containing those pdf files? (I can't get into 'em otherwise with this old browser

) Thanks.

 

[Klaus]

@vcal

The Web-site hosting the thing is : http://europa.eu.int/comm/environment/waste/nicd.htm
The main document directly is at http://europa.eu.int/comm/environment/waste/nicd.pdf

Hope this works out for you

@brightnorm

nonono - don´t get frustrated with your beautiful SF M6 - I would almost kill to get one

- or an E2 or whatever SF

My joking statement about beating the SFs was just that - a joke - you can´t really compare a production design to a DIY project.
Lightuser is exactly on topic - there are some nice CPF specials (and this one is hopefully also considered one) - but IMO all of those aren´t production type ones.

For your request to get one of those - I "closed" the project (for now) as IMO its no more ugly

and pretty functional as it is right now - the reason I updated the thread continuosly was to inform potential "followers" about the progress I had through the "design, test & built" phase.

As it is right now I consider it a worthwile DIY project not requiring too much skills - I personally wouldn´t want to go into even limited "production" as I should be doing other things

and also lack the soldering skills to not be ashamed of it.

Also I guess you are from the US and I´m in germany - so this would be a nightmare to coordinate - I think there had been others on the forum (like Gadget - seems to be busy these days) having the time and skills to help others out having less hands-on skills. Lets see if someone else responds.

Just in case - this project always was and is "free" for everybody to work on it, use it, commercialize it, whatever - it was even originally inspired from CPF anyway - I think I made this clear from post one. (The only thing I would like is that others would share any experience & enhancements with CPF as well)

@lightuser

good points. On HID - you´re absolutely right - lets see what comes out of the austrian mountains on that - I would expect "real products" to be maybe another 1-2 years away - IMO the HID-bulbs are still too expensive and the "niche" outside diving & biking possibly too small - and for the "tactical" use the limitations are maybe too critical preventing the use for now.

I would love to see such a 2D / 8NiMh design discussed in the other thread (PeLu seems to be too busy these days) - I guestimate it to have around 2-3 times the runtime of the original 20W / 500lm / 2D design discussed in this thread - but at possibly 3-4 times the price as just the bulb is around 100-150 USD vs 5-10 USD - hopefully the HID designs will come down in price significantly as volumes go up.

Keep us posted on yours

Klaus