MAX757 Run Time Test

[ElektroLumens]

CPFer's,

(NOTE: I posted this to another thread, and it was suggested to me to start a new thread with it, so here it is.)

I did manage to do a run time test with the MAX757 circuit, and the Luxeon Star. After I started it, I realized I was using an older circuit board, which is only at 77% efficiency. I decided to just continue with the test anyway, because I wanted to start with completely fresh Energizer AA alkaline batteries. I set the output voltage at about 3 volts, and 300mA. Here are the results. I didn't check for input current as I only have one .1 ohm resistor, and it is a hassle to switch it back and forth from input to output.

TIME Vin Vout mA
9:45pm 2.885 3.015 300mA
9:51 2.701 3.003 280mA
9:58 2.597 2.996 270mA
10.05 2.509 2.989 270mA
10:14 2.424 2.986 260mA
10:24 2.348 2.979 260mA
10:46 2.211 2.969 250mA
10:58 2.142 2.965 240mA
11:15 2.073 2.960 240mA
11:35 1.988 2.955 230mA
12:04 1.755 2.939 210mA
12:15 .573 2.637 100mA
12:15 Shutdown.

Ran for over 2 hours at full brightness or near full brightness. Even at 230mA, the Luxeon is still fairly bright.

I will do the test with 'C' batteries (gotta start earlier!!). I can also do it with 'D' cells. I do not have any rechargables, and it is not a good idea to run them down too far anyway.

I suppose that the 'C' batteries would last 6 hours, and 'D' cells should go for 12 hours. Perhaps longer because it is easier for the larger batteries to supply the heavy current. Also, I need to use the 83% higher efficiency MAX757 circuit next time.

Hey, this can get expensive, doing these tests!


It seems like the regulator leveled out for a bit at around 270mA to 240mA. I guess we could toss this around and find a optimum setting for long battery life. I think the rules change with nimh. I think the higher efficiency regulator shoud make a difference, possiby 6% better. After 2 hours and 20 minutes, it was still pretty bright. It faded out real quick in the last 10 - 15 minutes or so. Add 8 minutes to the run time (6% of 140). If we set the regulator down to 270mA, we might attain 3 hours. ??


Wayne www.elektrolumens.com

 

[lambda]

Be waiting to see the rest; good job.

 

[**DONOTDELETE**]

<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>Originally posted by ElektroLumens:
I do not have any rechargables, and it is not a good idea to run them down too far anyway.

Hey, this can get expensive, doing these tests!<HR></BLOCKQUOTE>

Save some money and improve efficiency by buying some NiMH's and a good charger, one that will keep the batteries charged at all times but not overcharged.

Stop your tests when cell voltage falls to 0.9 volt and you won't damage them.

 

[ElektroLumens]

<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>Originally posted by Duggg:
Save some money and improve efficiency by buying some NiMH's and a good charger, one that will keep the batteries charged at all times but not overcharged.

Stop your tests when cell voltage falls to 0.9 volt and you won't damage them.<HR></BLOCKQUOTE>


Thanks Duggg,

Good suggestion. I've been thinking about it for a while. I haven't felt the need, except for these tests. A set of AA alkalines last forever in my Garrity/Luxeon/MAX757 mod. When used intermittently, they last a lot longer than in a continuous run test, as you already know.

I should, however, break down and purchase the nimh and charger.


Wayne www.elektrolumens.com

 

[**DONOTDELETE**]

Hey the times for the 2AA alkaline tests are almost exactly what I predicted (for the 1 volt cutoff anyway)

Since the regulator is able to suck every little bit of power from the batteries, here are some revised projections:

2AA lithiums - 4:40
4AA alkalines - 6:40
6AA alkalines - 12:00 (!!)
2C alkalines - 5:20

Lets see how close (or not) I am. Actually since its a more efficient circuit, it could run for slightly longer.

"Perhaps longer because it is easier for the larger batteries to supply the heavy current." ... maybe, but just a little better. All alkalines start to strain when you want lots of current from them.

"If we set the regulator down to 270mA, we might attain 3 hours. ??" ... thats why everyone's searching for the all great magical number... ther perfect compromise between runtime and brightness.

Can't wait to see the results of the next test.

 

[JollyRoger]

Wayne, I'm interested in how the more efficient regulator does.
I'm hoping that 3 hours can be attained without reducing the brightness...hmmm...further improvements in efficiency? I know, always bugging you for more efficiency.

BTW, how is the ls heatsinked during this test? I'm sure the heatsinking also has a significant amount to do with brightness...

Are you using one of you signature cut-up aluminum discs?

 

[ElektroLumens]

<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>Originally posted by roger:
Wayne, I'm interested in how the more efficient regulator does.
I'm hoping that 3 hours can be attained without reducing the brightness...hmmm...further improvements in efficiency? I know, always bugging you for more efficiency.

BTW, how is the ls heatsinked during this test? I'm sure the heatsinking also has a significant amount to do with brightness...

Are you using one of you signature cut-up aluminum discs?<HR></BLOCKQUOTE>


Heat sink? Let me put it to you this way. It is really hard to solder the leads on the Luxeon, when it is connected to the heat sink. I'll take a photo of the monster and post it. I is multiple 2" X 4" aluminum plates, finned, with copper plates in between. I was using it to run 2 Luxeons, at 500mA. The plates barely even get warm, even at the high current. I designed this heat sink with a bike light in mind, using 8 'C' or 8 'D' cells as a power source, and a step down regulator. Never really finished, because I canablized the parts for other projects before I finished.


Heat sink? That's not a heat sink! This is a heat sink!! Crocodile Dundee said that, didn't he?

Wayne

 

[ElektroLumens]

<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>Originally posted by Someguy:
Hey the times for the 2AA alkaline tests are almost exactly what I predicted (for the 1 volt cutoff anyway)

Since the regulator is able to suck every little bit of power from the batteries, here are some revised projections:

2AA lithiums - 4:40
4AA alkalines - 6:40
6AA alkalines - 12:00 (!!)
2C alkalines - 5:20

Lets see how close (or not) I am. Actually since its a more efficient circuit, it could run for slightly longer.

"Perhaps longer because it is easier for the larger batteries to supply the heavy current." ... maybe, but just a little better. All alkalines start to strain when you want lots of current from them.

"If we set the regulator down to 270mA, we might attain 3 hours. ??" ... thats why everyone's searching for the all great magical number... ther perfect compromise between runtime and brightness.

Can't wait to see the results of the next test.

<HR></BLOCKQUOTE>


I'll pick up a few 'C' batteries, and run a test with them tonight, with a higher efficiency regulator.

I would like to run 4 AA's in 2 sets, parallel. I might or might not use a diode, as suggested by the lonely flashaholic in Spain. I don't think the Double Barrel flashlights use diodes in their design. The diode would waste some energy.


Wayne

 

[JollyRoger]

Hmm....4AA's will be interesting...but if you are going to put 4AA's in a flashlight, wouldn't the step down regulator you make be more efficient?

 

[ElektroLumens]

<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>Originally posted by roger:
Hmm....4AA's will be interesting...but if you are going to put 4AA's in a flashlight, wouldn't the step down regulator you make be more efficient?<HR></BLOCKQUOTE>

Roger,

That's a good question. I wonder how the AA's will perform in 2 sets parallel?

I have designed a simple step down circuit using the LT1073 chip. I used larger components, (through hole, etc.) If it performs well, it could be easily redesigned to surface mount components, and within the 1" diameter size constraint.

I am going to run another run time test tonight with 2 'C' batteries. I also have some Eveready E2 AA's, I'd like to test. I think I only have time for the 'C' test tonight, as it might go all night long?

I guess I should run it at 300mA. Perhaps I'll do another test at 270mA tomorrow.

Wayne www.elektrolumens.com

 

[ElektroLumens]

Further tests results:


Last night I did the run test again, using 2 Energizer e2 batteries. I got almost 3 straight hours of burn time. Here are the results:


MAX757 STEP UP REGULATOR, DRIVING A LUXEON STAR

TWO ENERGIZER e2 BATTERIES:

TIME_____Vin______Vout_____Amp
8:51pm____2.867____3.015_____290mA
8:54pm____2.788____3.010_____270mA
9:02pm____2.640____3.004_____290mA
9:11pm____2.495____3.005_____290mA
9:28pm____2.342____2.994_____270mA
9:38pm____2.270____2.992_____270mA
10:09pm___2.083____2.957_____270mA
10:18pm___2.053____2.984_____250mA
10:30pm___2.000____2.980_____260mA
10:35pm___1.967____2.980_____260mA
10:48pm___1.888____2.978_____250mA
10:53pm___1.848____2.978_____260mA
10:58pm___1.797____2.972_____260mA
11:03pm___1.748____2.969_____260mA
11:08pm___1.699____2.967_____250mA
11:17pm___1.606____2.958_____240mA
11:25pm___1.285____2.916_____200mA
11:29pm___1.252____2.907_____170mA
11:33pm___1.178____2.880_____160mA
11:38pm___1.050____2.848_____130mA
11:48pm___0.600____2.651_____below 100mA, and cannot get accurate reading.
11:48pm___shutdown.

I hooked the battery up a few minutes later, after it cooled down, and it powered the light brightly. Showing that if the light is used intermittently, much more than 3 hours use would be realized.

Wayne

 

[JollyRoger]

Great test, Wayne! What was the initial setting? It wasn't 300ma, right? Did you turn it down a bit?

Can't wait for the step-down test....

 

[remuen]

Hi Wayne

Interesting run time test! About 2 and a half hour a reasonable flat output current curve and then a slow dimming during the last 30 minutes.

Did you take your new MAX757 design with the high efficiency for this test?

It would be interesting also to see the efficiency in the table. You should invest all you money in an additional 0.1 ohm resistor to measure the input current

.

Duggg explained in another thread that the batteries are able to supply more energy (in watt-hours) when you draw less current out of it. See Duggg's post here Regulation question - why not *flat* output?. I think exactly this happens here with the lower current of only 300mA. Can you give me an idea about how less bright the LS is if you drive it with 300mA instead with 350mA?

Btw, the table is now much easier to read!

 

[ElektroLumens]

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

Interesting run time test! About 2 and a half hour a reasonable flat output current curve and then a slow dimming during the last 30 minutes.

Did you take your new MAX757 design with the high efficiency for this test?

It would be interesting also to see the efficiency in the table. You should invest all you money in an additional 0.1 ohm resistor to measure the input current

.

Duggg explained in another thread that the batteries are able to supply more energy (in watt-hours) when you draw less current out of it. See Duggg's post here Regulation question - why not *flat* output?. I think exactly this happens here with the lower current of only 300mA. Can you give me an idea about how less bright the LS is if you drive it with 300mA instead with 350mA?

Btw, the table is now much easier to read!

<HR></BLOCKQUOTE>

Hello Rene,

Several things were diferent in this test. One is that I used Energizer e2 batteries, which are supposed to have more amp hours in them.

I also used the MAX757 circuit with the lower DSR. It uses a 22uH inductor from Coiltronics. Better quality, and lower resistance, but also lower inductance. I noticed that at the same input voltage, there was about 10mA differnce in current output. The net effect however is that efficiency is better. So I started the test at 3 volts in, and 290mA.

Yes, I would like to run 4 DVM's (Vin, Vout, Amp in, Amp out), and use 2 .1 ohm resistors. It would make it so much easier to do these tests.

I can't remember exactly how to explain it, but because of the way our eyes perceive light, we only notice an increase or decrease in light if it is double. So unless the light decreases a lot, we cannot perceive the decrease or increase. I cannot tell much difference in the amount of light, from 290 to 300mA. There is a perceivable difference between 300mA and 350mA, and between 350mA, and 400mA. There is a huge difference in light output between 300mA and 400mA.

Oh, the table. I did the first table in Word, and when I copied and pasted it to this forum, it came out wierd. So I decided to try it the way I did, and it is easier to read.

Wayne