Update on BatteryGeek

That was a bit much to ask but I said I'd be willing to re-run my tests and amend my reviews with the latest information.

With that in mind I began a series of reasonably careful tests to see both how much time I could squeeze out of the pack and how little. At each stage I let Sean know how the test had fared and solicited his feedback.

My general usage on my laptop involves email, text file editing and web browsing. There isn't a lot of disk access and I rarely use the CD-ROM drive (IE, almost never). In other words, my power needs are minimal. On the other hand my Averatec C3500 uses an AMD Athlon XM-M processor which runs quite hot and that's why I use a coolpad. In general I keep power consumption as low as possible even when I'm running off wall power just to keep the machine running cooler.

The one test I ran that was atypical was test 3 where there was very regular Ethernet network traffic going while I was running a program that was in constant communication with another PC on my local network. I also had the screen brightness turned up to 66% which is more than three times higher than I normally leave it.

Test 1

Started 7:53am, full charge, laptop battery removed, screen brightness at 12% (the controls have 15 steps above "off" but I'm using % for clarity), coolpad connected, 802.11b radio off, Ethernet connection on. Audio turned off. Screen set to turn off after 1 minute. Speedswitch set for max battery. Computer died ~1:00pm - 5hrs 10 minutes runtime.

This test wasn't much fun since there was no battery backup but Sean had asked if I could try running without the internal battery and so I did. I suspect that he's had problems with people running their computers off the pack with a dead internal battery and complaining about short runtimes on the pack. This makes sense as with the computer pulling power from the pack to run the CPU as well as charge the internal battery you'd see quite shorter runtimes. I always use the pack with a fully charged internal battery for just this reason.

Performance wasn't great since the machine was running at 400MHz the entire time but since I was doing text editing and browsing during the testing period, it wasn't that bad either.

Note that I had stepped away from my desk when the computer died so I've given the pack the benefit of the doubt and estimated the runtime on the high end.

Test 2

Basically the same as test 1 except that for this test the 802.11 radio was used instead of the 10/100 Ethernet connection, the internal battery was hooked up and the screen was set slightly brighter. This time the computer ran for 5hrs before switching over to the internal battery.

Test 3

Test started 11:10am Sept 23, 802 off, Ethernet on, usb mouse, coolpad, screen brightness at 66%, dynamic switching, audio off, screen off in 1 minute, regular network traffic on the Ethernet line
Battery died at 3:30pm - 4hrs 20 minutes.

Test 4

No USB connections, 802 on, 100bt off, screen brightness at 20%, internal battery connected, speedswitch at max battery, audio off, screen off in 1 minute. This is basically the same test as #2 except that the coolpad isn't connected. Started at 10:10pm Sept 24 battery died at 2:40am Sep 25 - 4 hrs 30 minutes

This result was interesting. It's clear that running with the coolpad increases runtime by allowing the computer to run cooler.

Summary

By turning everything down as low as possible I can squeeze five hours out of the battery pack. It's at the low end of the 5-7 hour estimate given by Sean when I placed my order so I'm not thrilled by that. Especially since Sean's original estimates were based on my purchasing the 15-21-118 battery pack which is rated at 118WH rather than 130WH for the model 130 which is what I actually received.

During this testing I've been regularly sending the results back to Sean at BatteryGeek and he still hasn't come out and said why his original estimate was so much higher than the actual runtimes. The best he's done is say "My theory is that 5 hours is all that you will be able to get out of this battery. Based on your test the reason why you are getting less than 5 hours is obviously because of your USB mouse & coolpad running along with your screen set at the brighter level. Also, I know that having the Palm connected does also reduce your run time however minimally."

To me, those statements aren't much different than his saying "I expect you'll get the runtime you're getting". This is the rough equivalent of buying a car that's advertised to get 35mpg, actually getting 30mpg and having the salesman respond with "yep, that's the best I'd expect." Add to that the fact that running with the coolpad actually extends by runtime and it seems like his response was simply off the cuff.

With an estimate of 5-7 hours I expected that in power saving mode I'd get at least something close to the middle of that range rather than the low end. In an ideal world I'd expect that I'd get 7 hours in power saving mode and 5 when running in "who cares about conservation!" mode.

So, would I recommend BatteryGeek? I'd have to say I would with reservations. In my opinion their customer support went from abominable to barely acceptable with Sean's recent attention but given that he still hasn't responded to the differences in estimated vs actual runtime (and their web site claims 4-10+ hours of additional runtime with this battery) I'd recommend that if you're going to buy a pack from BatteryGeek, ignore their estimates and get the biggest pack you can afford. If you're lucky you'll get something over twice what you're used to.

Originally posted 2005-09-28 13:00

Updated: 2005-10-01 05:05

Two days before this update was posted I sent an email to Sean at BatteryGeek with a copy of the update attached and let him know when the update was scheduled to go live.

I didn't receive a reply.

A day after the update went live I received an email from Sean merely asking that I remove the batteryGeek email addresses from the original review.

Amazing.

Updated: 2006-01-22 12:10pm

Apparently BatteryGeek has also been operating under the name New Laptop Batteries for some time now. Beware. They're the same people, same products and I'm sure exhibit the same behavior.