What's new

Battery Drain While Plugged in

Votality

Member
Hi,

I was using my sp4 (i5 8g, 256gb ssd) this evening and noticed that while using it with ac power connected my battery was depleting while using a VM guest. My real concern is... is it possible the battery charger actually is insufficient to charge the sp4 when its under full load, or do you think this is some strange bug? (Read below for a description of the problem)

The battery report looks like as follows:

21:30:49 Active Battery 96 % 36,525 mWh
21:42:09 Active AC 88 % 33,337 mWh
23:18:35 Active Battery 72 % 27,285 mWh
23:18:36 Active AC 72 % 27,277 mWh
23:18:54 Active Battery 72 % 27,270 mWh
23:19:00 Active AC 72 % 27,247 mWh
23:19:11 Active Battery 72 % 27,240 mWh
23:19:17 Active AC 72 % 27,217 mWh
23:19:51 Active Battery 72 % 27,210 mWh
23:20:29 Active AC 71 % 27,045 mWh
23:26:34 Active Battery 71 % 26,985 mWh
23:26:55 Active AC 71 % 26,887 mWh
23:46:00 Report generated AC 79 % 29,970 mWh


The initial drop from 96% to 88% was valid as I knocked the power connector out and didn't notice. But after plugging it back in and using my machine the battery continued to deplete. The switching between battery and ac, was when I was plugging the charger in/out to see what was happening. The depletion happened from 21:42:09 to 23:18:35 and the connector was in and lit the whole time. The connector and charger are all like new (same as the charge port and contacts)

All that was connected to the sp4 at the time was a usb3 hub and us3 lan adapter. I was watching Netflix in edge and using a VMware guest.

Now the most interesting part of this story. As soon as I suspended my VMware guest (running under VMware workstation 12), guess what....the battery started charging. (i.e at 23:26:55 I suspended the VM and it charged to 79% when I generated the report. It continued to charge to 100% without the VM running). My sp4 is all up to date.

I read this: Surface battery won’t charge | Battery not detected | Troubleshoot battery

While my sp4 does turn on, its suggestion to uninstall the driver to control the battery may be worth a try. Too late here to muck around tonight but I will tomorrow. I will watch what happens with the guest running again also.

As I asked above... is it possible the battery charger actually is insufficient to charge the sp4 when its under full load, or do you think this is some strange bug?

EDIT: After reading other threads it indeed looks like it may just be the charger is insufficient. Do you think I have just cause to demand a higher rated power charger. The hub and network adapter I think, would have negligible power usage (few watts at 5v max)
 
Last edited:
The charger that comes with the SP4 is 36 W. The Core i5-6300U has an official TDP of 15 W but a boosted TDP of 25 W (Intel® Core™ i5-6300U Processor (3M Cache, up to 3.00 GHz) Specifications). Throw in SSD activity, the USB hub, LAN adaptor, and the integrated GPU (probably a big draw given that videos in Edge might not be that efficient), I wouldn't be surprised if you handily exceeded the 36 W charger. For me, playing a video game can cause a slow drain on the battery even while plugged in if it's a particularly demanding game.
 
Yea I agree with you. I talked to Ms surface support, they are getting me to do the usual reinstall the power driver etc, But honestly I don't think that has anything to do with it. As soon as the vm stopped. So did the battery drain. (I know because as soon as I stopped it the power indicator changed from not charging, to charging)

I will try the heavy load again soon without the hub and after I change the power settings back. For the moment I changed the vm guest from high performance power to the battery saver power profile and I changed the maximum performance limit on the sp4 itself to 99% so it wont use the turbo boost. (although I'm not sure if the cpu load of guests combined obey the cpu limit in windows)

I want to have my machine doing some computationally heavy work overnight when I'm not using it so I would like it to use the turbo boost.

My question is: If the adapter is insufficient to power the device under full load do I have just cause to demand a 60w charger for free. I don't believe that it is adequate to ship a device that is incapable of powering itself under full load. But I must first establish that is without doubt the case.
 
Last edited:
What is full load to Microsoft. It sounds like you have the Surface hub or docking station. I thought that had a much heavier power supply than the Surface.
 
How you use your machine is your choice...IMHO no, you should not be given a free 65W charger.

Are you joking.

IF this is the case (the charger is insufficient, which i still haven't determined conclusively yet) there should be no usage scenario in which the battery is depleted when connected to mains. Nothing I'm doing is not normal usage. I.e tweaking the system to overclock etc. What it means is the stock power adapter chosen was inadequate. It should have ample capacity to keep the system powered under full cpu and gpu load and supply the full power on the usb3 port as allowed in its spec.

Lets imagine if every laptop did the same. i.e Would only run under the balanced power profile or not full cpu load otherwise it discharged the battery and required you to buy a new power adapter to run under full load. The laptop makers would have a class action law suit. But all of this conjecture is jumping the gun. I still have to put it under full load again and see if the issue is reproducible.
 
Last edited:
Are you joking.

IF this is the case (the charger is insufficient, which i still haven't determined conclusively yet) there should be no usage scenario in which the battery is depleted when connected to mains. Nothing I'm doing is not normal usage. I.e tweaking the system to overclock etc. What it means is the stock power adapter chosen was inadequate. It should have ample capacity to keep the system powered under full cpu and gpu load and supply the full power on the usb3 port as allowed in its spec.

Lets imagine if every laptop did the same. i.e Would only run under the balanced power profile or not full cpu load otherwise it discharged the battery and required you to buy a new power adapter to run under full load. The laptop makers would have a class action law suit. But all of this conjecture is jumping the gun. I still have to put it under full load again and see if the issue is reproducible.
Well gee. I bought a new car knowing I got the v4 and not the v6. Now I find it can't tow my pontoon boat. Should I get a new motor for free?
 
All I can add is that there are plenty of products (laptops, tablets, phones) from plenty of the biggest tech companies in the world which discharge under heavy load while plugged in.. I've personally had it occur on an iPad, a nexus, the surface 3, an HTC phone. Can be a touch annoying, but that's just life.

It may be poor design or implentation etc, but there is absolutely nothing about it which would hold up in court.

Also, anandtech tested the 65w on the sp4 and I believe they found that it didn't supply any more power to the tablet, which to me sounds like the surface internals are only designed to handle 45w, maybe for safety reasons given its dimensions (just a random guess.. I'm no engineer, hardware software or otherwise)
 
I'm not sure where it's stated anywhere that the included charger would be able to handle a full load and continue to charge the device. It would be nice, I guess, if it could charge under full load but I don't think you have any grounds to argue that there has been any kind of false advertising. The charger charges the device, even under full load but not enough to prevent battery drain with heavy usage.
 
I'm not sure where it's stated anywhere that the included charger would be able to handle a full load and continue to charge the device. It would be nice, I guess, if it could charge under full load but I don't think you have any grounds to argue that there has been any kind of false advertising. The charger charges the device, even under full load but not enough to prevent battery drain with heavy usage.

Like i said i haven't ascertained yet if this is the case. Ive been too busy to run any tests. When it depleted it was a vm running a vpn and Netflix on the host (hardly the most taxing thing that you could do) but i certainly will be running stuff on the device that will use full cpu in the future hence the concern.
 
Back
Top