4xIMR26500's Driving: Osram 64623 or Osram 64458?

[My3kidsfather]

Which one bulb do you prefer for fun purposes? Assume using 4 x AW's IMR 26500's; Borofloat glass; alum. reflectors, AW's soft-start switch mod on a standard 4C cell mag.

LuxLuthor's spreadsheets say the 458 has lower amp usage of 8.8 amps (vs. 10.1 amps for the 623) at 14.8-16.8 volts. And the 458 has a vertical element that is said to be great at throw focusing in contrast to the 623's horizontal element.

The 623 has greater lumage at 15 volts of 6100 lumens vs. the 458's lower 3931 lumen at about 15 volts. Which one do you find has better color? useful runtime? greater wow factor? Other attributes?

 

[fivemega]

Which one bulb do you prefer for fun purposes? Assume using 4 x AW's IMR 26500's; Borofloat glass; alum. reflectors, AW's soft-start switch mod on a standard 4C cell mag.

LuxLuthor's spreadsheets say the 458 has lower amp usage of 8.8 amps (vs. 10.1 amps for the 623) at 14.8-16.8 volts. And the 458 has a vertical element that is said to be great at throw focusing in contrast to the 623's horizontal element.

The 623 has greater lumage at 15 volts of 6100 lumens vs. the 458's lower 3931 lumen at about 15 volts. Which one do you find has better color? useful runtime? greater wow factor? Other attributes?

For your case, 64623 or 64633 will work much better than 64458

64458 really needs 5x26500.

 

[My3kidsfather]

Thanks FiveMega, I checked out the 64633 and it looks like it can take higher volts than the 64623 before flashing- maybe the 633 is a bit more reliable? I had not read anything on the 633 but will be on the look out for threads. Pretty efficient bulb? or just not known yet?

Which brings to me to another question. What sort of problems with heat dissipation will one encounter with 4 and 5 battery setups?

 

[LuxLuthor]

Thanks FiveMega, I checked out the 64633 and it looks like it can take higher volts than the 64623 before flashing- maybe the 633 is a bit more reliable? I had not read anything on the 633 but will be on the look out for threads. Pretty efficient bulb? or just not known yet?

Which brings to me to another question. What sort of problems with heat dissipation will one encounter with 4 and 5 battery setups?

Remember to always take voltage drop from overall resistance of setup into account, which with these 3 bulbs running 9-11 Amps will likely be a decrease of 3 volts (Ohm's Law -- 10A x 300 mΩ--AW has said his D driver has a total resistance alone of 225mΩ), so 4 x IMR26500 gives measured starting Vbat of 16.4V, then from resistance, bulb can end up only receiving 13.4Volts.

When you pick bulbs from my destructive testing, I intentionally used a setup with only 19mΩ, so your setup's resistance must be included, depending on your setup. The efficiency of a bulb (more of a LED conecpt) can be seen with Ictorana's graphs in this thread (6th one on right for each category).

How "hearty" a bulb will be is best determined by the default manufacturer "Life," and the % of overdrive you push it. The 64623 and 64458 are both 4,000Hr, while 64633 is only 50Hr Life rating.

With 4 cells, I think you will get a more optimzed result with the 64623, given your resistance.

The 64633 is a 150W bulb, and has a narrow range of performance & tolerance...ideally 16-17V delivered to the bulb (after resistance taken into account). It would need 5 x IMR26500 cells. This is an ideal candidate for AlanB or JimmyM's regulated drivers, set to deliver 16.8V so it has a decent life.

Your other question on heat...I think the best guideline is to let the heat in your hand be your guide. If the head and/or body is getting uncomfortably hot (vs. warm to mildly hot), then you should stop and let the light cool down. The IMR26500's are rated up to 60° C (140° F), so you have some leeway with them technically.

 

[My3kidsfather]

Remember to always take voltage drop from overall resistance of setup into account, which with these 3 bulbs running 9-11 Amps will likely be a decrease of 3 volts (Ohm's Law -- 10A x 300 mΩ--AW has said his D driver has a total resistance alone of 225mΩ), so 4 x IMR26500 gives measured starting Vbat of 16.4V, then from resistance, bulb can end up only receiving 13.4Volts.

When you pick bulbs from my destructive testing, I intentionally used a setup with only 19mΩ, so your setup's resistance must be included, depending on your setup. The efficiency of a bulb (more of a LED conecpt) can be seen with Ictorana's graphs in this thread (6th one on right for each category).

How "hearty" a bulb will be is best determined by the default manufacturer "Life," and the % of overdrive you push it. The 64623 and 64458 are both 4,000Hr, while 64633 is only 50Hr Life rating.

With 4 cells, I think you will get a more optimzed result with the 64623, given your resistance.

The 64633 is a 150W bulb, and has a narrow range of performance & tolerance...ideally 16-17V delivered to the bulb (after resistance taken into account). It would need 5 x IMR26500 cells. This is an ideal candidate for AlanB or JimmyM's regulated drivers, set to deliver 16.8V so it has a decent life.

Your other question on heat...I think the best guideline is to let the heat in your hand be your guide. If the head and/or body is getting uncomfortably hot (vs. warm to mildly hot), then you should stop and let the light cool down. The IMR26500's are rated up to 60° C (140° F), so you have some leeway with them technically.

LUxLuthor- Might as well start purchasing the parts for the 64458 and let her rip. First, a 4D MagLite. Probably AW's SoftStart for D's too as it will handle the Osram's bi-pin width I think. More lenses, and more reflectors. (I think AlanB's or JimmyM's regulated drivers are a bit above me right now. Got to learn a few more things first.) Already got 5xIMR's here.

Thanks for the info on resistance- it's not futile. So as I see it now, the 623 for 4xIMR and I think it sounds like the 458 is all around best for 5xIMR's. Gotta try it. Have you tried working with heatsink compounds with your bulbs? It's gotta help at least a bit in runtime before dangerous heat buildup. I am thinking it might work with my new potted ROP's, even with bi-pins too between the adaptor and the bulb-cup? I'm gonna try as soon as I decide which combos will stay in certain hosts.

 

[LuxLuthor]

LUxLuthor- Might as well start purchasing the parts for the 64458 and let her rip. First, a 4D MagLite. Probably AW's SoftStart for D's too as it will handle the Osram's bi-pin width I think. More lenses, and more reflectors. (I think AlanB's or JimmyM's regulated drivers are a bit above me right now. Got to learn a few more things first.) Already got 5xIMR's here.

Thanks for the info on resistance- it's not futile. So as I see it now, the 623 for 4xIMR and I think it sounds like the 458 is all around best for 5xIMR's. Gotta try it. Have you tried working with heatsink compounds with your bulbs? It's gotta help at least a bit in runtime before dangerous heat buildup. I am thinking it might work with my new potted ROP's, even with bi-pins too between the adaptor and the bulb-cup? I'm gonna try as soon as I decide which combos will stay in certain hosts.

Heatsink compounds are not needed for Incands like they are for LED's. The IR output of Incands transfers very well to the head/body. Trust me on that.

The more important issue is to pay attention to the temps as I said (even if "by the seat of your pants" manner). Those who insist on running their high watt bulbs for long runs, are the ones who have ruined reflectors (the metalized reflector surface can peel/bubble away from the aluminum--depending on quality, of course). You may have some plastic parts in switch that are also subject to heat damage.

Finally, the stock Mag switch contacts are really not designed to hold up to the higher currents over time. They should be replaced with a Judco switch (like Mac used in his original Torch 14.4V 64623), or using Jimmy's regulated driver setup which takes all the brunt of the watt stress, leaving the mag switch unaffected.

 

[Chodes]

Once you have a 64623/64458 built you will see why there is no need for heat transfer compound.

If you put your hand near the front of the light for more than a second or so , you will have a serious burn. Hold some newspaper 2 inches in front of the lens and flames should occur in seconds.

It's entirely different from a LED where you need to try to get the heat away from the LED. With one of these big bulbs , anything close to the bulb automatically gets very hot very fast.

Ballpark figures - you will be able to run for around 2 minutes from cold.
Around 1 minute once light is already warm.
Big variations depending on ambient temperature , using your hand to absorb some heat and your own decisions on how hot is safe.

With either of these bulbs , variations of 0.5 volts at the bulb can make a significant difference with heat too , so different lights / cells / switch combinations will have an effect.

 

[My3kidsfather]

FiveMega, LuxLuthor, Chodes- Thankyou for your responses. I can see from your texts that these higher-lumage lights need to be carefully handled. Lesson heard and noted. Stronger switches needed for longer-term reliability.

 

[LuxLuthor]

FiveMega, LuxLuthor, Chodes- Thankyou for your responses. I can see from your texts that these higher-lumage lights need to be carefully handled. Lesson heard and noted. Stronger switches needed for longer-term reliability.

The most cost effective recommendation is to use your existing Mag Switch, and add Jimmy's soft-starting, regulated driver under a KIU aluminum pedestal/bulb holder.

A slicker step up, allowing drop in replacement is to go with AlanB's regulated driver.

I have inserted a Judco 12A switch into the Mag switch, but it gets tricky for grounding and setscrew.

 

[sylathnie]

There is also the slightly older but still extremely functional (and economical) Mosfet switch. An excellent how-to is located here.

Please also take into account some sort of lockout plan for the light. (Remove batteries, unscrew tailcap, keyswitch, etc) These people aren't joking around when they say it will product flames within seconds. Safety first!

You may want to play for a bit with each of the bulb setups and find what feels the best to you. Pure white light from an over-driven 62623 is beatiful but the tight hotspot and nice clean beam shape from an axial filament also has it's draw.

 

[My3kidsfather]

sylathnie- Another option for switches is a good thing. I was wondering how to idiot proof the 4D mag after it is turned into a hot light. Maybe just remove one battery until needed? Gotta think that thru.

Thanks for your comment. Having trouble buying Osrams and shipping them here to Alberta. Not many options.

 

[LumenHound]

Having trouble buying Osrams and shipping them here to Alberta. Not many options.

Know what you mean.
I might be able to help you with the 64458 and 64623.
Check your PM.

 

[LuxLuthor]

sylathnie- Another option for switches is a good thing. I was wondering how to idiot proof the 4D mag after it is turned into a hot light. Maybe just remove one battery until needed? Gotta think that thru.

Thanks for your comment. Having trouble buying Osrams and shipping them here to Alberta. Not many options.

None of the places I have linked on my Hotrater-Lux.xls file that is linked in topics of interest ship to Alberta?????

 

[LumenHound]

Lux, many US companies use UPS to ship to Canada. Our UPS is a mere shadow of your UPS. I'm afraid they are known as the carrier of last resort north of the border. A shame if you ask me.
The minimum brokerage fees UPS-Canada charge us for stuff valued in the $10.00-$50.00 range can sometimes raise the final cost of a low value item up to 5 times or more of the original purchase price.

 

[LuxLuthor]

Lux, many US companies use UPS to ship to Canada. Our UPS is a mere shadow of your UPS. I'm afraid they are known as the carrier of last resort north of the border. A shame if you ask me.
The minimum brokerage fees UPS-Canada charge us for stuff valued in the $10.00-$50.00 range can sometimes raise the final cost of a low value item up to 5 times or more of the original purchase price.

Wow! That truly sucks. I wonder why the mail/delivery system in Canada seems to lag behind otherwise advanced manufacturing/retail output.

 

[LumenHound]

Off topic but good info for Canucks wanting extreme hotwire bulbs...
It's got nothing to due with the mail. It's just that USP isn't a viable option in Canada for low value packages from the US.
The mail is amazing between Canada and the States.
Canada Post has a very unique reciprical agreement with USPS when it comes to goods moving across the border in both directions. If it's shipped USPS we pay a very modest brokerage fee that tops out at 5 bucks CDN for stuff with a value of $100.00 or less. Most times we don't even get charged that if we haven't gotten more than $100.00 declared worth of stuff from the US in the last 3 months. No taxes either. The mailman just drops it off. It's terrific if you ask me.

 

[My3kidsfather]

Off topic but good info for Canucks wanting extreme hotwire bulbs...
It's got nothing to due with the mail. It's just that USP isn't a viable option in Canada for low value packages from the US.
The mail is amazing between Canada and the States.
Canada Post has a very unique reciprical agreement with USPS when it comes to goods moving across the border in both directions. If it's shipped USPS we pay a very modest brokerage fee that tops out at 5 bucks CDN for stuff with a value of $100.00 or less. Most times we don't even get charged that if we haven't gotten more than $100.00 declared worth of stuff from the US in the last 3 months. No taxes either. The mailman just drops it off. It's terrific if you ask me.

I have changed all shipments to USPS as well. Slower but much less costly and reliable too. UPS really stuck me the last time i shipped by ground with about $35.00 in fees above the shipping costs for inspections and customs papers. If an e-store requires UPS shipping the deal is now canceled at my end.

 

[LuxLuthor]

Sorry to continue off topic, just this last question, then I'm done.

When you say "amazing" what do you mean in terms of days? I have ordered things on occasion by regular post FROM Canada (that my USPS carrier delivers), it seems to take a fairly long time, relative to USPS within USA.

Most of my First Class letters to other states arrive in 1-3 days (depending on distance), First class packages are more like 3-5 days. Priority Mail packages are 1-3 days.

 

[xeonsaga88]

Remember to always take voltage drop from overall resistance of setup into account, which with these 3 bulbs running 9-11 Amps will likely be a decrease of 3 volts (Ohm's Law -- 10A x 300 mΩ--AW has said his D driver has a total resistance alone of 225mΩ), so 4 x IMR26500 gives measured starting Vbat of 16.4V, then from resistance, bulb can end up only receiving 13.4Volts.

Shouldn't the total internal resistance be < 30mOhm??. As stated in the AW incan D thread (http://www.candlepowerforums.com/vb/showthread.php?t=177316) . Or am i missing something out

 

[LuxLuthor]

Shouldn't the total internal resistance be < 30mOhm??. As stated in the AW incan D thread (http://www.candlepowerforums.com/vb/showthread.php?t=177316) . Or am i missing something out

Here was my talking about that, and AW responding below mine. There was another post or PM where AW just told me 220mΩ. I think it is wise to take the higher end of the resistance range. Then you have to add the internal resistance of the batteries, contact spring, etc. If you did spring resistance mod, you could probably get it down to 250mΩ