Wierd grounding problem with laptop

[SentinelAeon]

I have a wierd problem with a laptop - when charger is connected and you hold finger on a touchpad, the cursor will sporadically start jumping around. I did all the usual things, connect usb mouse (no problem there), remove battery (problem persists), reinstall drivers, different drivers, safe mode, different socket, 6 different chargers. I tested 100's of laptops using the very same power socket and never had a problem. Even the last time i was testing it i dont remember having any problems.

It appears to be a grounding problem. If i touch the metal case of the laptop with my left hand, the problem goes away. What else can i do, test ? Is it possible that some ground connection somewhere moved out of place, like from touchpad or around the power jack and thats the problem ? I would appreciate more info on what more i can test. The DC jack on laptop is standard asus 2 pin (5.5mm 2.5mm), charger uses standard 3 pin AC cable, the sockets i used (230V) only have 2 pins, i could try on the one that also has ground pins. Never had a problem till now. Tnx for your help.

 

[rjenkinsgb]

It is most likely just the lack of a mains ground. There will be a small amount of capacitive leakage through the power supply, causing some AC voltage on the computer internal "ground", but at high impedance.

The touchpad is capacitive, so the AC leakage can confuse the circuitry with false signals.

 

[aliarifat794]

Inspect the DC jack for any visible damage or loose connections.

 

[SentinelAeon]

Inspect the DC jack for any visible damage or loose connections.

Ok, i will take it apart and check the DC port.

I did some measuring with multimeter from the outside and for some reason measuring resistance between heatsink and positive on DC jack shows they are connected. I am pretty sure this is not supposed to be. Disconecting most things, i measure 125k resistance between positive on MB and metal casing of the laptop. Can someone comment on that ?

 

[JimB]

It is most likely just the lack of a mains ground. There will be a small amount of capacitive leakage through the power supply, causing some AC voltage on the computer internal "ground", but at high impedance.

The touchpad is capacitive, so the AC leakage can confuse the circuitry with false signals.

I completely agree with this.

Inspect the DC jack for any visible damage or loose connections.

This is most unlikely.

JimB

 

[SentinelAeon]

Btw, i tested on a socket with mains ground. The problem remains unchanged. I tested like 100 laptops in this apartment and never had a problem.

What else can i do and test ?

 

[SentinelAeon]

I put the laptop apart and couldn't find any faults. What i did find is that the original asus charger for a similar laptop will work ok and the touchpad wont malfunction, i am guessing the original asus charger fixes the ground thing. I still find it funny that this is the first laptop that has this issue with the exact same charger and socket.

Since my understanding of ground loop is not very good, i would just like to ask this. Touchpad is connected with aluminum foil to the case of the laptop - to ground. Would it make any difference if i removed this foil and tried without it ?

 

[rjenkinsgb]

Touchpad is connected with aluminum foil to the case of the laptop - to ground

That's screening to minimise the effect of or emission of noise - removing it would likely make things worse.

 

[SentinelAeon]

That's screening to minimise the effect of or emission of noise - removing it would likely make things worse.

Ok, i understand. What if i kept this screening but removed just it's connection to case of the laptop, to be disconected from ground ?

 

[Tony Stewart]

The charger has excessive CM noise leakage pF across the isolation flyback transformer which to a mismatched AC impedance on DC source upsets the touchscreen threshold.

Either shunt the Vcm to PE. Connect VGA up or wire case gnd to PE gnd using R~<10k or C>= 10nF from VGA Gnd or foil under laptop or replace charger or add 1mH CM choke to charger with Y caps to PE or ....get rid of charger.

It probably exceeds SMPS freq 1mA AC leakage current to PE so your finger impedance creates a high interference voltage. Either suppress it to PE with low impedance or raise series impedance with a CM Choke or both. Often designers add a few pF from primary to secondary to try to cancel from opposite phase and may not have tested it for that in production.

This is a common problem with "isolated" secondaries in cheap flyback chargers lacking good specs and testing. It can also destroy USB communication with grounded towers.

 

[Diver300]

Is is possible to swap live and neutral on the mains power into the laptop?

If Wikipedia is to be believed, the mains sockets in Slovenia are Schuko so can be turned around to swap live and neutral. Also, if the laptop charger is fed with a C7 connector, that could be swapped instead.

If would be worth finding out if there is a difference if live and neutral are swapped.

 

[Tony Stewart]

I would hope it is a grounded plug (F)?

 

[Diver300]

I would hope it is a grounded plug (F)?

A lot of laptop power supplies are not grounded and only have 2 pin connectors for the mains.

Most mainland European sockets are non-polarised, so live and neutral can be swapped, even if grounded.

 

[Nigel Goodwin]

I would hope it is a grounded plug (F)?

Pretty well all laptops are class II, so there's no earth connection anyway - so no point in a grounded plug.

 

[Tony Stewart]

Pretty well all laptops are class II, so there's no earth connection anyway - so no point in a grounded plug.

Class II does not control or guarantee RF isolation from the CM noise in a SMPS.

For those who do not understand...
In Class II or 2 appliances, the user is protected by at least two layers of insulation. For this reason, Class 2 appliances are also known as Double Insulated. Class 2 or II items do not require an earth connection for safety.

Yet, some need protective earth (PE) ground for functional noise diversion to prevent susceptibility to EMI issues.

This laptop battery charger, for example. The laptop is Class 2, yet the charger is class 1 such as this HP charger.

All the symbols are for electrocution safety, not RF interference. Notice that it is missing the double insulated safety symbol.

Isolation or Electrical "Insulation" is only tested at line frequency by shorting the AC pair and apply several kV to measure secondary leakage to PE ground greater than say 100uA for some products and 1 mA or more for others depending on the continent or country or class of product specifications. But they do not measure the impact of CM noise injected between the case 0V and one's finger on a touchscreen.

Either exchange or toss the charger with a better one or shunt the 0Vdc to PE gnd with a 10 nF or so cap in series. The wire inductance and series C form an LCR or RC series filter, so this may require some skill , I think. A hardwired ground requires some faith that your charger will never see a direct lightning strike or otherwise have insulation failure. This may not be wise if your power grid is not buried.

Often 2 stages of primary CM filters are used in better ACDC supplies as the magnetic useful impedance for attenuation often spans less than 2 decades and the harmonic content spans more than the CM choke. If I were to guess, this is the problem with your charger.

 

[Tony Stewart]

I have a wierd problem with a laptop - when charger is connected and you hold finger on a touchpad, the cursor will sporadically start jumping around. ,,... I tested 100's of laptops using the very same power socket and never had a problem. Even the last time i was testing it i dont remember having any problems.

Can you do us a favour and share the brand / model of Laptop and Charger. Most laptop chargers are universal 19.5V +/- 1 to 20%.

Does the problem move with the charger or laptop?

 

[Nigel Goodwin]

Class II does not control or guarantee RF isolation from the CM noise in a SMPS.

For those who do not understand...
In Class II or 2 appliances, the user is protected by at least two layers of insulation. For this reason, Class 2 appliances are also known as Double Insulated. Class 2 or II items do not require an earth connection for safety.

Yet, some need protective earth (PE) ground for functional noise diversion to prevent susceptibility to EMI issues.

Very often such chargers are only two core mains leads, the secondary is connected back to live or neutral, using special isolation components.

I suspect the issue here is just down to a crappy third party charger, and probably best to source an original one?.

 

[rjenkinsgb]

Designs also vary tremendously from make to make.
My original Panasonic Toughbook PSUs are three pin, with "Must be earthed" on the label and appear to have a 1M bleed resistor between mains earth and the PSU negative out.

Those are 15-16V ones for three cell batteries. Machines with four cell batteries use 19-20V.

At the other extreme, one old USB PSU I had was so bad with leakage, you could not use the touchscreen on a phone or ipad while connected to it. With those it had a kind of locking out effect, possibly due to it sensing inputs or noise over too large an area??

 

[Tony Stewart]

I have never seen a laptop with am (ungrounded) Class II charger and this unknown charger is without a doubt incompatible with it's lappie touchscreen due to insufficient CM noise filtering or suppression of interference.

. Keep in mind differential noise is not a factor as that DC voltage is regulated down to the motherboard logic level Vdd.

 

[Nigel Goodwin]

I have never seen a laptop with am (ungrounded) Class II charger

???

I've PAT tested hundreds of laptops and PSU's, and without exception they are all class II devices. 'Some' use a three core lead (all plugs in the UK are three core anyway), but the earth isn't connected to the output. As they don't have metal cases there's nothing to earth, so you can't test than as class I.

But as you say, it's presumably just a crap third party PSU.