Elite 1700 and Eneloop Data for Hotwire Modders

Greetings Everyone

There are a number of choices of batteries for hotwire mods, but when it comes to the AA NiMH variety, it seems to me that only two make sense: Elite 1700s and Eneloops. The Elites have great voltage retention at high loads, and the Eneloops...well, they are just Eneloops.

Over the last few weeks I have been experimenting with several hotwire configurations using both Eneloops and Elite 1700s. I find myself frequently going back to Silverfox's NiMH shoot out data to check out voltage at different discharge currents.

I thought that it might be helpful if I did some additional testing of two of the more popular hotwire NiMH cells, focusing on the discharge rates of interest. I also thought it would be useful to take a closer look at the initial voltage during the first 5 seconds of discharge as this is when instaflashing of the bulb is a possibility without a soft start on the hotwire mod.

My purpose for posting these results is not necessarily to compare the Elite 1700s to the Eneloops, or to say that one is better than the other. Rather, it is to augment Silverfox's test results and provide a reference for those wanting to use these cells in hotwire mods. Anyone with a flashlight that has a regulated power supply that applies a constant power load on the cells would be better off looking at my Ragone Plot for AA cells here.

So below are the results of a large number of constant current discharge tests of both Eneloop and Elite 1700 AA cells.

Here are the cycling conditions for both cells:

charge: 2000 mA to a -delta V of 10 mV (per Eneloop datasheet)
rest: 30 minutes
discharge at specified current to 0.8 V

All of this was done on a Maccor cell cycler. Nickel tabs were spot welded to the end caps of the cells to minimize contact resistance.

The Elite 1700s were discharged from 1A to 10A in 1A increments. The Eneloops were discharged from 1A to 5A in 0.5A increments. (I actually discharged the Elite 1700s in 0.5A increments, but the performance difference is so small with this change that it makes the plots too difficult to read).

I realize that a real hotwire discharge is not constant current. However, as someone pointed out to me when I first started posting to CPF, it is not constant resistance either since the resistance of the filament changes with applied voltage. But I think that constant current is a reasonable representation of how the cells will perform in a hotwire application.

So here are the plots. I scaled the voltage axis from 1.4V to 1V instead of 0.8V in order to expand the axis and better differentiate the discharge curves. Each curve is labelled with a number that represents the discharge current.

The Elite 1700 complete discharge curves:



The Elite 1700 voltage profiles for the first 5 seconds of discharge:



The Eneloop complete discharge curves:



The Eneloop voltage profiles for the first 5 seconds of discharge:



And by request, here are the discharge curves for both Eneloops and Elite 1700s overlaid for 1A-5A discharge:



Also by request, here are the discharge curves for both Eneloops and Elite 1700s overlaid for 6A-10A discharge:



Suggestions for how I can make this data more useful are welcome, as are suggestions for additional testing. For example, I suspect that some of you might want to see higher discharge currents for both the Elite 1700s and the Eneloops. Also, I was thinking that the addition of a high capacity 2500 or 2700 mAh NiMH cell might be useful, but I don't know which is most popular among hotwire modders, or even if hotwire modders use high energy NiMH cells.

I hope you find this this useful. Looking forward to comments.

Cheers,
BG

Thanks! I've been looking for precisely thess curves on both cells.

I've been running flat-top elites in custom high current 4aa > D holders I bought five years ago thru cpf that accomodate flat top anodes. (I've never had much luck soldering blobs to act as anode buttons) Also, my hobby chargers won't charge fewer than 3@ NimH or NiCad cells in series.

lemlux said:

Thanks! I've been looking for precisely thess curves on both cells.

Click to expand...

I have been using the CBP 1650 curves in Silverfox's Shoot Out study, assuming that they were similar to the Elite 1700s. Boy, was I wrong! The Elite 1700s are substantially better. I was amazed that they maintain >1.2V for 30% of a 10 amp discharge.

Cheers,
BG

Hello Battery Guy,

Keep in mind that while the CBP 1650 cells were impressive during normal testing, the greatly improved when used hot. I didn't test those cells hot at 10 amps, but if you review the 15 amp test you will find that the hot cells came in roughly 0.2 volts higher than the normal test.

In order to make the most of the CBP 1650 cells, you needed to charge at a minimum of 2C using a -dV of 10 mV and then you needed to use the cell before it cooled down. If the cell temperature was in the 110 - 140 F range, you were good to go. Needless to say, this is hard on cells and cycle life is reduced.

The Elite 1700 cells seem to perform great under more normal conditions, and they also seem to hold up better, making them a better cell, at least in my book.

Tom

I know when I talked to Mike at CBP's, who had very aggressive performance charts posted for various cells, including the Elite 1700, he tested them as hot off the charger as he could, soldered test leads directly to NiMH cells, and used the CE-35.

One thing I think would be useful is to overlay the two brands, using different colors for the cell type--but only for 3, 4, 5 Amp (& with new, higher Amp plots with Eneloops, since I believe Tom had pushed them up to 10A). If it is also possible to repeat the voltage display on the right side of the table, because it is hard to see the faint gray horizontal lines as you move accross. If not, c'est la vie.

Personally, I wouldn't use Elites for under 5A. Above 5A is where you really see the differences, and Elite advantage.

5 second plots. SWEEEEEET!

Thanks Battery Guy and a huge thank you to SilverFox from a new user.lovecpf

SilverFox said:

Hello Battery Guy,

Keep in mind that while the CBP 1650 cells were impressive during normal testing, the greatly improved when used hot. I didn't test those cells hot at 10 amps, but if you review the 15 amp test you will find that the hot cells came in roughly 0.2 volts higher than the normal test.

In order to make the most of the CBP 1650 cells, you needed to charge at a minimum of 2C using a -dV of 10 mV and then you needed to use the cell before it cooled down. If the cell temperature was in the 110 - 140 F range, you were good to go. Needless to say, this is hard on cells and cycle life is reduced.

Click to expand...

I had heard this, and I did see your discharge results for the CBP 1650 hot off the charger. It is nice to see the Elite 1700s performing very well without this limitation.

SilverFox said:

The Elite 1700 cells seem to perform great under more normal conditions, and they also seem to hold up better, making them a better cell, at least in my book.

Tom

Click to expand...

They definitely seem to perform well, at least when new. I don't have any cycling data on them yet, but I have high hopes for them.

Cheers,
BG

LuxLuthor said:

I know when I talked to Mike at CBP's, who had very aggressive performance charts posted for various cells, including the Elite 1700, he tested them as hot off the charger as he could, soldered test leads directly to NiMH cells, and used the CE-35.

Click to expand...

I have never been able to locate the Elite 1700 discharge curves on CBP's website.

I agree with the idea of soldering or spot welding leads to the cell. Contact resistance is one of those experimental variables that is hard to measure and rarely reproducible, so I am of a mind that it should be eliminated if at all possible. However, I would rather not test the cells hot off the charger. I gave a lot of thought to how long I should let the cells rest after charging. In the end, I put a thermocouple on an eneloop and monitored the temperature. The cell was within 2 degrees C of room temperature in less than 15 minutes, so I doubled that and used a rest time of 30 minutes. I think that testing the cells at or near room temperature is more representative of how people will actually use them.

LuxLuthor said:

One thing I think would be useful is to overlay the two brands, using different colors for the cell type--but only for 3, 4, 5 Amp (& with new, higher Amp plots with Eneloops, since I believe Tom had pushed them up to 10A). If it is also possible to repeat the voltage display on the right side of the table, because it is hard to see the faint gray horizontal lines as you move accross. If not, c'est la vie.

Click to expand...

So, increase the eneloop discharge current to 10 A, and make a plot with the eneloop and Elite 1700 data overlaid? I can do that!

Not sure I can do much to darken the horizontal gridlines, but I will give it a try. I am using the Maccor software to make these plots, and it is somewhat limited in its capabilities. To make the new plots with the overlaid eneloop and Elite data, I may need to import the data into Excel, in which case making the gridlines darker will not be a problem.

This wil take me a few days. I won't be able to get the eneloops back on test until next week.

LuxLuthor said:

Personally, I wouldn't use Elites for under 5A. Above 5A is where you really see the differences, and Elite advantage.

Click to expand...

One thing that I noticed from these curves is the difference between the initial voltage under load and the voltage "plateau". I think that having a small voltage difference here is advantageous because it allows you to have the highest voltage plateau as possible without instaflashing the bulb in hotwires without soft starts. Ideally, if there was no voltage difference, instaflashing would not be a problem. You can see from these plots that this difference is significantly smaller for the Elite 1700 cells compared to the eneloops.

What this means is that for a given hotwire setup with a critical instaflash voltage, you can have a higher operating voltage during discharge if you use the Elite compared to the eneloops. Perhaps everyone already figured this out, but it has not become apparent to me until recently.

Cheers,
BG

I meant to inquire if anyone has seen or conducted self-discharge tests on the Elite 1700 cells. Personal experience or anecdotal observations would be useful, but obviously a well conducted experiment with replicates over the course of several months to a year would be great. Unlikely, but great!

Thanks!
BG

Battery Guy said:

I have never been able to locate the Elite 1700 discharge curves on CBP's website.

Click to expand...

This is what Mike had sent me earlier, and which came from the factory. I never saw one that he independently verified.

Tom's Eneloop graph is here.

Battery Guy said:

However, I would rather not test the cells hot off the charger. I gave a lot of thought to how long I should let the cells rest after charging. .... I think that testing the cells at or near room temperature is more representative of how people will actually use them.

Click to expand...

I agree. Just wanted to explain some of the more aggressive discharge graphs I had seen earlier at CBP.

Battery Guy said:

So, increase the eneloop discharge current to 10 A, and make a plot with the eneloop and Elite 1700 data overlaid? I can do that!

Click to expand...

Probably be good to do an 8A of Eneloop which Tom had done also.

Battery Guy said:

One thing that I noticed from these curves is the difference between the initial voltage under load and the voltage "plateau". I think that having a small voltage difference here is advantageous because it allows you to have the highest voltage plateau as possible without instaflashing the bulb in hotwires without soft starts. Ideally, if there was no voltage difference, instaflashing would not be a problem. You can see from these plots that this difference is significantly smaller for the Elite 1700 cells compared to the eneloops.

What this means is that for a given hotwire setup with a critical instaflash voltage, you can have a higher operating voltage during discharge if you use the Elite compared to the eneloops. Perhaps everyone already figured this out, but it has not become apparent to me until recently.

Click to expand...

Well that is true that the Elite has a higher plateau for better voltage in a Hotwire if you are direct driving. Then there's that other pesky self-discharge issue with those Elites!

LuxLuthor said:

Probably be good to do an 8A of Eneloop which Tom had done also.

Click to expand...

I am planning on taking the eneloops up to 10 A at 0.5 A increments to complete the curves. I will then make some plots with Elite and eneloop curves overlaid.

I noticed that the curves Mike made went up to 30 A. Do you think that there would be an interest to discharge the Elites higher than 10 A?

LuxLuthor said:

Then there's that other pesky self-discharge issue with those Elites!

Click to expand...

I have been wondering about this, but wasn't able to find any information on it. So, how bad is the self discharge of the Elites?

Cheers
BG

Battery Guy said:

I am planning on taking the eneloops up to 10 A at 0.5 A increments to complete the curves. I will then make some plots with Elite and eneloop curves overlaid.

I noticed that the curves Mike made went up to 30 A. Do you think that there would be an interest to discharge the Elites higher than 10 A?

Click to expand...

Most people will start looking at SubC or Lithium Manganese/Lithium Iron Oxide for high current applications. However, if you look at some of the higher watt Osram bulbs in my signature testing link, a number of us have driven those in Mags or Elephants using the 1700's. I made packs of 7s that fit in an Elephant, as shown with the Gold model in first post here.

Battery Guy said:

I have been wondering about this, but wasn't able to find any information on it. So, how bad is the self discharge of the Elites?

Click to expand...

The 1700 are much better made and hold their charge longer than the 2/3A which are popular for tri-bore mags to get the voltage up high in 2D size (i.e. for the Mac Torch fire starting Osram 64623 type setups).

I don't know how to answer in terms of "how bad," except that I developed a habit of freshly charging every time before I use a light like that, even if I charged it the day before. Once you weed out substandard cells (which are more common among the Elite/IB 2/3A cells), I seem to remember they will hold 1.2V on a full charge by DMM measurement (which is not all that useful vs. measurement under load) for at least 4-6 months.

LuxLuthor said:

One thing I think would be useful is to overlay the two brands, using different colors for the cell type--but only for 3, 4, 5 Amp (& with new, higher Amp plots with Eneloops, since I believe Tom had pushed them up to 10A).

Click to expand...

Lux

I added a plot that has both Eneloop and Elite discharge curves for 1-5A. It will take me a few days to collect the 6-10A data on the Eneloops.

Cheers,
BG

LuxLuthor said:

......Personally, I wouldn't use Elites for under 5A. Above 5A is where you really see the differences, and Elite advantage.

Click to expand...

......Well that is true that the Elite has a higher plateau for better voltage in a Hotwire if you are direct driving......

Click to expand...

Just a quick comment. The only Hotwire I have that uses AA cells, is a fully resistance tricked (complete switch fix and tailspring) 6 AA ROP. While this light only draws about 4.3A with fresh cells, I have to say the difference between eneloop's and Elite 1700's is quite noticeable. While the 3854H isn't known for being a "white" lamp, it is much whiter with Elite's than with eneloops. That 0.5 Volt or so difference halfway through the discharge, does make a difference, even at ~4 Amps.

I'm sure there is a trade off in runtime, but the eneloops look so yellow, I've never bothered to check. On a side note, I've never had instaflash problems with the Elite's. I do always let the cells rest a half hour or so before firing up the light, however.

Dave

LuxLuthor said:

Once you weed out substandard cells (which are more common among the Elite/IB 2/3A cells), I seem to remember they will hold 1.2V on a full charge by DMM measurement (which is not all that useful vs. measurement under load) for at least 4-6 months.

Click to expand...

Curious as to how you weed them out. I just purchased a relatively large number of Elite 1700s. They have all been numbered and dated. I was considering using the break-in mode on my C9000 on all of them, logging the capacity, letting them sit for a week and then discharging them to determine if any have an unusually high self-discharge rate.

The only problem with doing this is that it will take a LONG time. Is this overkill, or highly recommended for the Elite 1700s?

Cheers!
Battery Guy

Great thread! Needs to be a sticky :wave:

Eneloops and elite 1700s are 2 of my favorite batterys for hot wires. I'm running 16ea elite 1700s in my m*g458 and they deliver 10.5a. I have found the elite 1700s don't self discharge nearly as much as 2700 mah AAs do. Eneloop AAs are about all I put in my m*g85s anymore. After 2700 mah AAs sit for a month they don't have as much power left in them as the eneloops and they are not as bright to start with.
Billy

Battery Guy said:

I added a plot that has both Eneloop and Elite discharge curves for 1-5A.

Click to expand...

That really shows side by side, especially with the horizontal lines at 0.05 increments only--very easy to follow now.

I would keep the 6-10A graph separate, as too many lines becomes chaotic, and don't believe you need .5A increments. We can interpolate with whole numbers.

Ahhh...the cell weeding out process. There was no practical way to do individual cell discharges. I needed to develop a relative easy measurements using a DMM. This was much more important with the 2/3A Elite cells. It was an arduous and repeated charge voltage & self-discharge rate evaluation based on DMM arrival voltage, charge, and discharge voltage sortings. If I got new cells that arrived with <0.9V, they were rejected. 0.9 to 1.1V questionable. >1.1V were put in initially good category, charged and voltage measured 1 hour afterwards. Sort was done on this good category of 1 hour post charge voltages.

Almost all of the high V on arrival and after charge turned out to be ideal, but would be discharged anyway at 8 or 10 Amps and recharged to verify. All categories of first sorted cells were left sitting after charging for at least 15 to 30 days to check for comparitive amount of self-discharge within their categories--resulting in more sorting. Those that kept higher voltage after 15-30 days were discharged and any with abnormally low voltage went into gray zone category. Good ones on discharge were recharged and again observed for peak charge voltage and self-discharge rate over another 10-15 days.

I would say that out of 1,000 cells, I got 800 that were great. 100 that worked if made into a pack with each other, 50 that were clear rejects and returned for replacement, and another 50 that remained in a substandard gray zone that I used in my own pack lights. Once I started doing this weeding out process, as far as I know, never got a request to replace a pack with a failed cell. Did some eventually go bad? Sure, but there was no way to know after 6-12 months what had happened to them, so I only had a warranty period of 3-4 weeks after arrival.

I started doing this because the guy who used to make all the cell packs (Mad Maxabeam--well actually his nephew made them for him) didn't do any quality cell screening, and I had a 75% pack failure rate from him. You know when it happens, as the cell feels like a ladyfinger firecracker went off inside your Maglite--and then the light goes dead. In any case, if people didn't keep them charged, overdischarge them beyond noticeable dimming, subjected them to the highest current demands of various hotwires in long runs without allowing heat to dissipate, the cells eventually deteriorate no matter what you do.

Having said all of that, I never found the same failure rate with the Elite 1700's. There was I think an 1600 or 1800mAh high current Titanium brand that looked like this that had high failure rate and electrolyte leakage, so I got away from those. Elite 1700 are great cells.

By request, a plot showing both Eneloop and Elite 1700 discharge performance at 6, 7, 8, 9 and 10 A discharge has been added to the first post.

Cheers,
BG

Those last two show it all.