New Latitude 5470

I'm on BIOS 1.17.3, but I'm not sure I've used it on battery enough since upgrading to see if there's a difference.

I'm also on the latest SSD firmware according to the Sandisk tool; when I last checked, with an aftermarket Sandisk drive the Dell firmware tool did not work, but that was over a year ago when the machine was new.

Worth checking on the Sandisk firmware and trying the machine out on battery to see if the BIOS fixed it without my knowing it.

 

I have did some thorough testing, here are my findings posted on dell forum.
http://en.community.dell.com/support-forums/laptop/f/3518/p/20022231/21026554#21026554

I am planning to get NVMe SSD and 66whr 6Cell battery which is 11V. Already got the M.2 SSD bracket. I can get NVMe SSD, but don't know where to find 66WHr battery, asked dell person on forums he told me i need to chat with Dell sales rep and ask about SKU 451-BBUE. Which I did and got told they are out of stock for those, and offered me same 62whr 7.4V battery I have right now. I don't think there is something wrong with this battery, but probably this one should not be used with HQ processor, need stronger battery for that. If nothing I will get 48whr, it is also 11V. It will last less, but at least no surprises.

 

Here is about battery and 2.5" inch SSD on page 10 here
http://topics-cdn.dell.com/pdf/latitude-e5470-laptop_owner's manual_en-us.pdf

 

Thanks for the part number on the 11V 66Whr battery! Hope those come back into stock soon.

 

I wonder whether this one for the E5480 would fit, and what the voltage is? I don't think we have an E5480 thread, but will look.

 

I was thinking the same. We need to look all available batteries for e5480 and if they are all 11V then that means Dell found the issue and adressed it with battery change. The e5470 has 3 official batteries, 47whr, 62whr and 66whr. Only this one that we have, 62whr , is 7.4V. Other two are 11V. The 52Whr 11V battery I mentioned earlier is actually for 5450 and 5550, but seems to also fit 5470. Dell person told me it is not for 5470 but I see on partsPeople and eBay that it is listed as compatible.

 

My config is the same as I stated in my original post. For the battery drain test I also had the following extras plugged in: a USB hub (4 ports), wired USB keyboard, wired optical mouse, HDMI monitor. I turned the laptop screen on, I usually have it off as I use the laptop as portable desktop, meaning it is always on AC (that's what I also set in BIOS).

This was probably the first time I had it on battery for longer than 5 minutes so I am sure the final outcome is skewed (new-like battery, 95% of factory capacity). In short: I let it drain the battery till 29% and no shutdown. I know I could have waited longer, but I preferred not to stress it too much as I was already pleased with the result. As I said, this test was nothing close to how I use my laptop.

I wrote a simple C program to take 100% CPU power and ran it on 2 cores (two instances). So the CPU had 2 cores busy 100% each plus I was doing some PDF reading. I stopped at 29% battery left and I had been checking the voltage level - it was constant for 10-15 minutes at 6.97V and did not want to move lower (anticipated shutdown right below 7V, but it did not happen).

Here is the battery info just before I stopped the test:

native-path: BAT0
vendor: Samsung SDI
model: DELL TXF9M64
power supply: yes
updated: Sun 22 Oct 2017 01:04:23 PM EDT (10 seconds ago)
has history: yes
has statistics: yes
battery
present: yes
rechargeable: yes
state: discharging
warning-level: none
energy: 18.1564 Wh
energy-empty: 0 Wh
energy-full: 61.9932 Wh
energy-full-design: 61.9932 Wh
energy-rate: 33.858 W
voltage: 6.969 V
time to empty: 32.2 minutes
percentage: 29%
capacity: 100%
technology: lithium-polymer
icon-name: 'battery-low-symbolic'


BTW, I am still on BIOS ver. 1.15.4

 

So did you try to test it without any drive in? One thing I've observed is my SSD can get quite hot, easily peaking at 63-65C during normal work, while its critical temp is 70C. M.2 brackets from Ebay have some sort of heat dissipation mechanism built-in, I wonder if normal size SSDs should have something similar added.

 

Thanks mintuser.

I think your battery might be too new and undamaged, and that your premature shutdown might happen closer to 10%. Not sure how much your program stresses the battery, I used the command

stress -c 4 -m 96 -i 3 -d 2

You have to install stress utility first.

Tell me do you get in dmsg something like this
ahci 0000:00:17.0: port does not support device sleep

Especially when closing the lid on laptop and opening again on battery.

I have not tested with mechanical HDD yet, I can't take apart my Thinkpad X220T at the moment.

 

Yep, my battery looks new and that's why I'd rather not wear it out running such artificial tests. My new laptop arrived with one dead pixel and instead of returning it I opted for the screen replacement via next day Dell service. The technician who arrived was super friendly, he offered to replace the CPU thermal paste free of charge, and we chatted for half an hour after he was done. I remember him specifically mentioning that batteries on Latitudes do not last long and wear our quickly. In general he was super knowledgeable about my Dell model, he deals with hundreds of them each month, replacing the screen took him no more than 5 minutes while still chatting with me So if you're serious about pursuing the shutdown issue I suggest get in touch with one of those "field guys", I bet they outclass many Dell remote support specialists.

Still, my recommendation would be to take out the drive and boot off a USB stick (or a RAM disk - even better). No-drive test would be your baseline result, if your laptop gets shutdown with no drive, it will do so with any drive on board. Having said that, your stress command looks very demanding, how likely is it that in an everyday situation you're gonna be maxing out all 4 cores + memory + i/o? I am not saying Dell is not at fault here, they ship HQ models with a more capable power brick, so increasing the battery potential would have made sense as well. Since these are business models Dell may listen and pressing them may work.

As for your kernel message, nothing similar on my system but that may depend on the kernel version. I did observe various lid/suspend/battery related quirks, a few of them. For instance, I normally wake up my laptop by either USB keyboard or USB mouse and that works well. But when I unplug the power cord even for one second and plug it back in, then waking up by USB device does not work anymore and I have to lift the lid up and press the power button (discovered while untangling computer cables). Still, as I am 100% on AC power + an external LCD, issues like that do not bother me at all.