Discharge data for Phillips 5761 on AW's new C Li-ion cells

With a little encouragement from LumenHound, I have tested AW's new C size Li-ion cells in a 2C mag running the Phillips Focusline 5761 bulb. My light has had extensive modification to reduce resistance. This includes a magcharger bi-pin slug soldered directly to solder braid which is then soldered to upper switch contact and a strip of solder braid that connects the focus wheel to the body, bypassing the setscrew using a "Jimmy strap" from JimmyM. It also has the tailcap spring fixed for low resistance using more soldered solder braid.

Anyway, I started the test with both cells at 4.20 V at rest and measured bulb voltage at the bulb pins. A fan was blowing on the flashlight head throughout the test, and no reflector or lens was installed. And now for the data:




I terminated the test at 6V, which took 28 minutes. The bulb was still fairly white even at the end. The head of the light was warm after 5 minutes and hot after 10 minutes. The battery tube was warm at the end, but not hot.

Immediately after the run I measured the cell temperatures by wrapping the cells together in a towel and placing a thermocouple between them. By holding the thermocouple tight against the cells and keeping the whole thing wrapped in a towel, the temperature came to equilibrium in about 1 minute at 120F. Since Li-ion chemistry should be kept below 140F and it is reasonable to assume that the inside of the battery would be warmer than the outside, this is close to the safe operating limit of the cells.

After a 30 minute rest, the batteries had open circuit voltages of 3.78V! I am at a loss to explain this degree of voltage rebound.:thinking: Does anyone have an idea why they measured so high?

So, in closing, I really like this set up and I hope this data encourages more people to try it. I can melt plastic trash can liners with this thing and it's only a 2C light! It's like a mag85 with a larger hotspot at a fraction of the size. Gotta love it!

C.J.

Thanks for that test, Mud. I would love to see how the 1185 fares. The 5761 takes a lot of current -- 5.5 amps or something like that -- compared to 3.3 or so for the 1185. So for not a lot of difference in lumens, the 1185 should push the runtime out to nearly a hour and be a lot easier on the cells, too. Didn't AW say the cells shouldn't be pushed past 5 amps? But it's nice to know that these cells can hold up to the heavy draw.

Hello C.J.,

At higher current draws, the internal resistance of the battery lowers the voltage. When you take the load off, the voltage springs back up. The RC people use a similar low voltage cut off and expect their cells to rebound to 3.7 volts per cell. If it does not rebound that high, they estimate that some damage has been done.

I believe the 140 F high temperature reference is to the actual outside temperature of the cell. When the outside temperature of the battery exceeds 140 F, the inside temperature is too hot.

Tom

C.J., thanks for putting the time and effort into setting it up and charting your results with those new cells. :thumbsup:
That first twelve minutes (above 6.7V) looks killer. You must have an ear to ear grin right about now. :grin2:

Thank you for this valuable information. I inferred from things that I had read that the 5761 was intolerant of direct drive and flashed over 7-7.1 volts. Thanks again.

John

Thanks for the thread, Mudman cj. I have been running the 5761 on two 2600 mAh batteryspace unprotected 18650s for several months and this is an improvement. The 5761 when the voltage is kept up towards 7.2 volts is brighter than the 1185, plenty has been posted on that. The whole point of my build was to get these C cells. But it takes 5 clicks to fire them up in my light and I have more than one set of these new Cs so it is the light. I too have a system for wiring a ceramic socket I make into a low resistance build. I am inclined to do away with the battery internal protection (boy this should fire our friend, LuxLuthor, up as he detests this idea) my light needs to fire up when I call on it and I do not need to be clicking away when my attention should be elsewhere. That will eliminate the spacer too. Otherwise maybe a soft start, a NTC or a driver would get around this. Without the protection I am at risk to damage from low voltage damage to the battery. My chargers should take care of the overcharge protection.
I need to get around the multi clicking, that is for sure. Ideas?

I was going to ask you for more information about your idea to incorporate an NTC to provide soft start, but then I discovered that my cells fire up the 5761 on one click every time. I have been thinking that I just got lucky and my cells are "special", but maybe it is my light after all. I am very surprised to hear that you need 5 clicks. :awman: Maybe we can notice a pattern regarding how many clicks people need to fire up the 5761 vs. characteristics of their lights such as resistance.

AW also has a single click light! I am going to take the circuit apart and analyze every component from socket to set screw, there is gotta be a way, and I will post it, I bet it will be something simple and overlooked. But when it lights up it is brighter on charged batteries than my 5761 on 12 Sanyo 2700 AA's.

Northern Lights said:

AW also has a single click light! I am going to take the circuit apart and analyze every component from socket to set screw, there is gotta be a way, and I will post it, I bet it will be something simple and overlooked. But when it lights up it is brighter on charged batteries than my 5761 on 12 Sanyo 2700 AA's.

Click to expand...

I found it. First realize there are manufacturing tolerances in batteries and bulbs and switches so my findings of combinations will not necessarily be constant.
I took the light apart and anyalyzed every part of the circuit. I found over all circuit resistance to include the tail cap which is modded too to be 0.6 ohms on my meter.

I used several bulbs, 5761, and both sets of batteries. On jumpers the batteries and bulbs showed these characteristics too. On 0.6 ohms or less circuit resistance the batteries would not click over, look like a short to the protection circut. By incrementally increasing resistance they work at 0.8 or above. By removing the jumper from the tail cap and going to the spring allowed the light to work at 3-5 clicks. Adding resistance into the cap circuit it got better. Single click with enough resistance. How many lumens is 0.1 ohm in this set up? Well, I will plug that into some calculations later. The lights that work on single and doubel clicks have enough resistance so as not to trip the circuits in those particular batteries, because of tolerance and difference it is not accurate to say which light has the lowest internal resistance, that can only be fairly measured by using a standard meter between all of the lights. Each case is therefore unique.
edit: I usually run a fuse in the tail cap, have not done that yet so I will check on that, I have no idea what that adds to the resistance and also it would be useful to see what is the voltage at the bulb at the different resistance settings. I will do that too.
edit2: Inrush current Limiting NTC is the way to go for me. I found a source with free samples. I will take those last readings and then determine the start up resistance necessary to get a single click by increasing the resistance on the jumper/battery set up and determine the difference between the bench model and the actual light and then order some samples. Looks like the running resistance will be near zero, I know my mod has a lot less resistance to begin with than many other solutions represented. For instance I use copper pin sockets, even gold plating, aluminum or brass has a higher resistance than pure copper; see the thread on that issue. So I can afford a little residual thyristor resistance in the mod. Darn, this shoots down my half brained excuse to take the protection circuit out. Am I accepted back into the fold now?

Warning! Math content below may not interest most readers. You may want to just skip down to the summary.

Light output is proportional to the power of 3.1 to 3.4. If we use 3.4 to get the maximum difference and use a resistance of 0.1 Ohms, the only thing we then need is the current. Knowing that the 5761 pulls 5.5 Amps with fresh cells helps, and gives us what we need to see the effect of 0.1 Ohms of resistance on fresh cells.

At 5.5 Amps of current, a 0.1 Ohm resistance will drop 5.5*0.1=0.55 Volts for an output ratio of [(7.2-.55)/7.2]^3.4=0.763. So 0.1 Ohms of circuit resistance will only give about 76% of the light output when the cells are holding 7.2 Volts. Assuming 900 lumens at 7.2 Volts gives 684 lumens with the added 0.1 Ohms of resistance. For a 0.08 Ohm resistance the ratio is about [(7.2-.44)/7.2]^3.4=80.7% and for a 0.06 Ohm resistance the ratio is about [(7.2-.33)/7.2]^3.4=85.3%. Going from 0.1 Ohms to 0.06 Ohms would give a muliplier of [(7.2-.33)/(7.2-.55)]^3.4=1.12 or an increase of about 12% in light output, or 108 lumens.

Let's also consider once the light has been running a while, and bulb current has dropped and so has the voltage dropped across the 0.1 Ohms of circuit resistance. Since power consumption equals current times resistance and power consumption is proportional to the voltage to the 1.6 power, it follows that V*I*C=V^1.6 where C is a constant, and I*C=V^0.6. So, I1/I2=(V1/V2)^0.6 and from this we can calculate that when voltage has dropped to 6.6 Volts, for example (about 2/3rds the way through the run shown in the first post), the bulb current has dropped to 5.5*(6.6/7.2)^0.6=5.22 Amps which gives a voltage drop of 0.52 Volts and an output ratio of [(6.6-.52)/6.6]^3.4=75.6% of the light output, which is now only about 783 lumens (900*(6.6/7.2)^1.6=783 lumens) with no resistance drop and 783*.756=592 lumens with the added 0.1 Ohms.

In summary, reducing resistance has a significant effect on light output. The 5761 can give an extra 200 lumens of estimated nominal output by getting rid of a 0.1 Ohm circuit resistance completely. If the circuit resistance were reduced from 0.1 Ohms to 0.06 Ohms, the gain in light output would be about 100 lumens.

I am very impressed with your work Northern Lights, and I think this is a big enough difference to make it worth some effort. Please provide more details on incorporating the NTC into the light.

What a great post. I love the numbers, I have been aware of this for ever but if I posted that over the course of a run in a 2c-cell I averaged 6xx lumens someone would tell me that an ROP does that not realizing or applying this principle which is a universal fact!
I have worked hard to get the resistance down in a couple of 2c-cells I was working on for the 5761. I wired directly the bi-bin pr socket that I built to the positive and grounds plus the socket uses copper connections. My dmm says my entire system, tail cap and bulb too, has a total of 0.55 ohms. The bulb is by far the majority of the resistance. I cannot get AW cells that I have to work on "million" click starts. There could be variances in the protection pcbs. I did take AW cells apart and removed the protection and added a single board for two cells from BatterySpace, it has a common tap so it balances cells, a series charger can be used. But the point was the first two pcbs I got worked, rated, 7.4 volts, 5-7a protection. I just built another pack tonight, tried two of the same boards bought at a different time. They will not start either! I may get the next higher amperage one to solve that. They must have a wide variation of tolerance because I used the pack in the same light that worked beautifully before. I put an NTC into the circuit to fire it up and once on removed the NTC from the circuit and I measured 5.2 amps on the running light.

I have tried two models of NTC surge protection thyristors from AMETHERM.
http://www.ametherm.com/Inrush_Surge_Limiters.htm
I do not have the numbers in front of me but here is what I found. Extrapolating the values on the different models always came up with a residual resistance of .8 to 1 ohm when operating at 5 amps. The NTCs are rated by amps and the specs show how much resistance at 100% and 50% is residual in the thyristor, so you can take the 5 amps as a ratio of the total maximum amperage of the NTC and extrapolate it to figure how much residual resistance is when it is heated up. The NTCs have too much resistance left in the system for a 7.2 volts system to get any benefit from them, it will allow the light to start up on a single click but the end result was about a 0.5 volt drop!! And you can see by the above math that is a significant loss of lumens.

For those systems that will fire up the 5761 with one click on the AW battery; if you have the identical PCB protection, that is to say if you tested it and it had the same performance values as the ones I got then you are only getting them to fire on one click because you have resistance higher than my light in the circuit and that translates to a loss of lumens. But realize the likely hood is the circuit differences probably lie mostly in the range of tolerances in the battery's pcb.

This two C-cell light even with a diminished return on the 5761 still is probably he brightest initial combination for a 2C-cell size. The size of the light is important to me so I will work with it. The AW cells are probably the best solution for this combo but the balance is critical to get them to fire on one click and keep the lumens. The 18650, 2600 mAh from battery space are unprotected, because of that they worked ok and getting the most from them was sure easier than getting the AW cells working for me.

My current plan is to get the correct pcb again from BS and put it lengthwise on the cells, the pack. I wired the taps to the pcb using two strips of circuit board repair tape ontop of a clear tape that insulates the batter for each tap. The PCB will be in a plastic housing and the pack shrink wrapped. The pcb will fit into the tail cap. That will allow me to use the space of the AW spacer to put a jack behind the switch for charging. Current configuration with the pcb flat on the tail of the pack makes the whole thing as big as two of the original batteries and therefore it fits, but I have no room for a charging jack.

Very interesting thread. First of all....just wait until your father gets home and finds out that his boys are playing with unprotected Lithium cells, or worse yet taking the PCB out of individual cells. You both will be grounded for a long time! LOL! Seriously, you guys are at the high end of the Bell curve in terms of understanding and using Li-Ions. My main concern was for all the noobs who didn't take the time to read or understand their safety....especially when charging.

Back to the thread....I'll use one of FM's 1.5D lights with a HD & tailcap spring mod with 2 AW-C and see how what measurements I get once my 5761's get here.

Then I'll figure out which host to use 2x EMoly's 3000mAh (same 26mm diameter as AW C's, but 70mm long) and see how that works.

To update you, LL, I wish I had not started this projects, AW's coming driver would be the best solution!!! But I started this before his announcement. I have assembled and disassembled the pcb on the batteries. It works but is off by about 0.5 amps, that is it takes about .6 more ohms in the circuit for it to allow the system to work, same as the original protection! BatterySpace realized that two of their PCBs were rated too high for 18650s since my first experiments and lowered one board's threshold and discontinued the other board. Now this is getting a little beyond my ability so I contacted an inovative genius I know and as he reversed engineered the boards has pointed out I need to increase one resistor and that will place the threshold higher. So much to do so little time and I still need him or someone else to tell me how to solder to one of those itty bitty boards!
Those BatterySpace PCBs on a sidexside pair of 18650, 2600 mAh BatterySpace cells and two packs of these in a sidexside D cell would make a screaming great 5761. A 3D or modified 2.25 D would be such geat hosts for this battery.
Another combination in a tri-bore D is using 6 of AW 17500s. Just remember the two-C rule.

To add to the previous post and bump it, I just recieved another 2c cell 5761 with a kiu socket and another set of AW C cells. These C cells start on a single click in this light and also in my existing light where as another set I have with the previous light will not start in either light. That confirms it. The variance is in the battery pcbs. I like this new set, because I now have them in a light with a focusing head, no spacer because I used my easy make modified tail cap http://candlepowerforums.com/vb/showthread.php?t=162767 pict at the link. And the ability to flood 12 feet in front of me is paramont along with a spot and size is a major issue. Now when I get the new driver for the other bodies...

Northern Lights said:

To add to the previous post and bump it, I just recieved another 2c cell 5761 with a kiu socket and another set of AW C cells. These C cells start on a single click in this light and also in my existing light where as another set I have with the previous light will not start in either light. That confirms it. The variance is in the battery pcbs. I like this new set, because I now have them in a light with a focusing head, no spacer because I used my easy make modified tail cap http://candlepowerforums.com/vb/showthread.php?t=162767 pict at the link. And the ability to flood 12 feet in front of me is paramont along with a spot and size is a major issue. Now when I get the new driver for the other bodies...

Click to expand...

That is interesting. I also have a feeling that AW will be swamped for a while doing the C soft starters that are selling like hotcakes.

Laugh with me. I am back to square one. Who is it that keeps posting that "resistance is you friend" about th 5761 because he keeps insta-flashing? He is correct. Now this new set of cells that works single click flashes bulbs. I had a very expensive day; even rested which btw has little affect on lions. The other cells since they must soft start by multiple clicking or NTC or added resistance do not. I remember what anticipation is now, my drivers... please hurry...

icantsee has used it as I recall and sad to say resistance is still my friend. Flashed couple 5761's without AW's driver lately after senting driver in and :poof::poof:

Started with stock mag 2c with AW's driver in it, did one resistance fix at a time including, tailcap spring, JimmyM's neg strap fix, soldered both sides stock switch contact ring, pro=golded everything and no :poof:

With a nicely done smo reflector to match it turned into my new wow light.

Jim

Northern Lights said:

Laugh with me. I am back to square one. Who is it that keeps posting that "resistance is you friend" about th 5761 because he keeps insta-flashing? He is correct. Now this new set of cells that works single click flashes bulbs. I had a very expensive day; even rested which btw has little affect on lions. The other cells since they must soft start by multiple clicking or NTC or added resistance do not. I remember what anticipation is now, my drivers... please hurry...

Click to expand...

I used to have a double click light, I recently lubricated the o-rings and threads and may have got some on the spring or other contacts. Now I have a 15 click light, did I lower the resistance? How can I fix this?

nightstalker101 said:

I used to have a double click light, I recently lubricated the o-rings and threads and may have got some on the spring or other contacts. Now I have a 15 click light, did I lower the resistance? How can I fix this?

Click to expand...

What type of bulb and batteries? I presume that it is AW cells and a 5761. Then you can fix it for $1.20.
cures insta flash too:
http://candlepowerforums.com/vb/showthread.php?t=170972
I have used it here:
http://candlepowerforums.com/vb/showthread.php?t=172284]
and here:
http://www.cpfmarketplace.com/mp/showthread.php?t=168555

This is the solution to the problem of my post #16
http://www.ametherm.com/Data%20Sheets/SL12%201R010.pdf

Northern Lights said:

What type of bulb and batteries? I presume that it is AW cells and a 5761. Then you can fix it for $1.20.
cures insta flash too:
http://candlepowerforums.com/vb/showthread.php?t=170972
I have used it here:
http://candlepowerforums.com/vb/showthread.php?t=172284]
and here:
http://www.cpfmarketplace.com/mp/showthread.php?t=168555

This is the solution to the problem of my post #16
[URL="http://www.ametherm.com/Data%20Sheets/SL12%201R010.pdf"]http://www.ametherm.com/Data%20Sheets/SL12%201R010.pdf[/URL]

Click to expand...

Thank you very much, this looks like a great solution, I'll give this a try. Thanks again