Static from hand killed my expensive mouse, can it be repaired?

[Pcmaker]

I have a Steelseries mouse from 2008 that is no longer being manufactured. It's the only mouse that doesn't make my hand hurt when I use it. A used one from eBay goes for around $450 which is a ton of money.

This morning, I touched the mouse and received a static shock. Usually, it stops working for 2 seconds, but it starts working again. Now it's dead for good. I put it in a different USB port, no luck. I tried a different mouse and that worked, but when I plugged in the dead mouse, the new mouse wouldn't work until I reset the PC.

I've been nursing this mouse since 2008. I've only replaced the switches and encoder, but I don't know what to look for in repairing what the static killed on the  little board.

Anything I should look for on the board after I open it up?

[fzabkar]

Look for ESD protection components near the USB electronics.

Measure the resistance between the +5V and Ground pins of the USB connector.

[Pcmaker]

Look for ESD protection components near the USB electronics.

Measure the resistance between the +5V and Ground pins of the USB connector.

On the mouse itself?

[fzabkar]

Look for ESD protection components near the USB electronics.

Measure the resistance between the +5V and Ground pins of the USB connector.

On the mouse itself?

Yes. Also tests for shorts between ground and each of Data + and Data -.

[EHT]

I have a Steelseries mouse from 2008 that is no longer being manufactured. It's the only mouse that doesn't make my hand hurt when I use it. A used one from eBay goes for around $450 which is a ton of money.

What is this thing? When i search for Steelseries there are new mice around £50 price point.

[DaJMasta]

It's usually a single, small IC package that is located near the connector/cable.

That said, I really doubt static killed your mouse, and I also doubt that there has been no mice made in the last 14 year that would suit your hands, but attempting a repair is a fine way to proceed.  I would check to make sure all the cord connections make good contact through the cable, as generally the most common point of failure is going to be the interface between the cable and the board, or in the cable/connector itself.  Beyond that, verifying the voltage on the USB power line when it's plugged in (to see if something is shorted and dragging it down) and inspecting for any grime that got in through small gaps that could have pushed on something or shorted something on the board itself.

Also worth checking what the OS reports when it's connected - whether there is a device identified or if it just acts like nothing has been present - and this isn't always apparent in notifications.

[Pcmaker]

I have a Steelseries mouse from 2008 that is no longer being manufactured. It's the only mouse that doesn't make my hand hurt when I use it. A used one from eBay goes for around $450 which is a ton of money.

What is this thing? When i search for Steelseries there are new mice around £50 price point.

It's the Steelseries Cataclysm mouse. They stopped making them back in 2008

[Fflint]

It is not a solution you're looking for, but if you don't manage to fix it you could get a cheap travel laser mouse(tiny) and have a custom enclosure 3d printed to match the shape of the old one.

Perhaps you own a 3d printer, or you know someone with one? Also if you could do the CAD by yourself, that can make the whole thing a lot more cost effective. There is a free(open source) powerful CAD software called FreeCAD. Version 19 is actually usable as a proper CAD if you can get through learning the interface without pulling your hair out. If you do go down this route I'll send you some YouTube video links to good tutorials.

It is just a thought. Another one may be to convince someone to do the CAD for you in hope of selling the model(or the print) in future.

[abdulbadii]

I touched the mouse and received a static shock. Usually, it stops working for 2 seconds

so weird super hi, expensiive quality product not even esd proof, should ensure it really is.
is its body conductor ?

[Pcmaker]

I took it apart and this is the PCB. I don't even know where to start looking as to what is broken. can these boards be bought by themselves with all the parts installed or are they made specifically for Steelseries?

[Pcmaker]

I touched the mouse and received a static shock. Usually, it stops working for 2 seconds

so weird super hi, expensiive quality product not even esd proof, should ensure it really is.
is its body conductor ?

They sold for $99 each when they came out around 2008. But the problem is they don't make these anymore and the used ones on eBay are going for $350 to $600 and from Bulgaria

[T3sl4co1l]

Yep, no ESD components, there's even positions for them right by the connector, see the USB pair coming off it, CDM/CDP.

ESD causes quite severe currents through whatever it's sunk to, very briefly.  On the order of 15A peak, with a half-height duration of 50ns or thereabouts.  It's not much energy, at least to something macroscopic like us (it hurts enough to be annoying..), but the peak power is incredible, and for the tiny (micron-sized) structures on a chip, it's easily fatal.

And something like a mouse, ought to be designed to be fully insulated (but there's enough things on this one, I'm not sure, maybe there's exposed metal parts near some of those buttons?).  But it seems you exceeded that insulation level.  It must've been a stonking great spark indeed.  It might've exceeded the voltage it was rated to (typically these things are tested up to 15kV).  Which doesn't bode well.

The main chip seems to be a Freescale (now NXP) MC9S08JM16 microcontroller:
https://www.nxp.com/docs/en/data-sheet/MC9S08JM16.pdf
It's only rated to 2kV ESD, and definitely needs external components to provide protection to this kind of level.  (Probably they figured the enclosure would give enough insulation, so left off anything extra.)  Very likely it's shit the bed, at least the USB pair if not additional internal damage (like shorting the supply pins together).  Can go around the device with a multimeter on "diode test" and see which pins are diode-like (probably good) and shorted (or low resistance, and how low at that).

If someone's dumped its firmware, it's possible to get a replacement part, program it, and keep on using the same rest of the mouse.  This has some caveats:

1. They probably read-protected the chip, so it will require some hacking to dump it.  Often an exploit is required.  This is non-trivial, and likely requires clever programming (maybe the USB port, programming port, or other pins are vulnerable to unexpected codes) or special hardware (such as a tool to momentarily drop power to the MCU, as it's executing a critical section, confusing it and causing it to pass the check).
2. Programming it, requires some hardware and software to connect to the chip itself.  Typically a programmer adapter, drivers for it, and a software tool to send the file to it.  Sometimes very simple hardware is required (some chips can be programmed with a generic serial adapter).  They were nice enough to provide the manufacturer's recommended programming port, by the looks of it (J2, see datasheet page 329).
3. New from scratch: it's always a possibility, if the original firmware isn't available, to write it new.  Knowledge of the pin functions, opto sensor, and device hardware (well, that's one of two datasheets down at least) is required, and simply a lot of time to bring it all together (100hr+?).  It looks like there are one or a few open-source mouse designs, maybe one can be ported to the MCU, that would save a few hours digging through hardware specs.  In any case, needless to say this would be prohibitively expensive, unless you happen to have the exact skill set to do it yourself -- in which case, still, you're far better off just getting the replacement unit.  Alas.

And, it's still possible that your chip has its memory intact, but the physical issues caused by ESD damage may prevent accessing it, besides the above reasons.

Oh hah, the sensor actually exists, brand name Avago even:
http://www1.futureelectronics.com/doc/AVAGO%20TECHNOLOGIES/ADNS-9500.pdf
so that's the datasheets.  And the other chip in the corner is just an ordinary EEPROM, nothing special there.

Tim

[wraper]

First check continuity of the cable wires. Static discharge may be a coincidence.

[Pcmaker]

Besides checking the continuity of the USB wire, this may be out of my expertise. Is it worth taking to an electronics repair shop? Buying a used one on eBay will cost me around $450 plus shipping and taking a chance on it being still operational.

[ledtester]

Besides checking the continuity of the USB wire, this may be out of my expertise. Is it worth taking to an electronics repair shop? Buying a used one on eBay will cost me around $450 plus shipping and taking a chance on it being still operational.

Not likely. They can't replace the microcontroller because they don't have the firmware. They could replace the 9500 but you don't know if that's the problem. If the mouse doesn't even produce a USB connection event I would suspect there's something wrong with the MCU.

There are people building their own mice... here's a recent hackaday article:

https://hackaday.com/2020/03/12/build-your-own-mouse-for-high-performance/

And a design with schematics and firmware:

https://www.overclock.net/threads/progress-on-a-fully-custom-mouse.1724838/post-28267334

and some modules for motion sensors:

https://www.tindie.com/products/jkicklighter/pmw3360-motion-sensor/

https://www.tindie.com/products/jkicklighter/adns-9800-laser-motion-sensor/

[Fraser]

Time to stop and consider EXACTLY what happened.

1. You touched the mouse and witnessed an ESD event…. Fine
2. What material is the mouse casing made from ? Plastic, Metal of Plastic with a sprayed conductive layer inside (black sooty or metallic in appearance. Is there a connection of some sort from the PCB ground plane to the mouse casing ?
3. Where did you touch the mouse for the discharge to occur? On a button, scroll wheel or just the casing.
4. You say this happened before and the mouse returned to life after a few minutes. Did you have to disconnect it from the host for this to happen and where did you likely touch the mouse to cause the ESD discharge ?
5. Is your mouse connected to an earthed appliance or an isolated computer such as a laptop that is without an Earth connection ?

Why do I ask all these questions ? Well to understand a failure and assist in identifying the location of likely damage, you need to look at the case scenario in detail. Static electricity can definitely damage sensitive electronic components and may disrupt their normal operation even if not directly contacted by the discharge. The high voltage needs to find its way to a very different potential for a discharge event to occur. In domestic situations this is commonly you carrying the high voltage charge and the charge causing an ESD event to ground, or a component at a significantly different charge. Potential. So in the case of your mouse, the charge travelled down your arm, onto the mouse casing or it’s buttons and found a path to the mouse 0V rail or USB cable screen. If we consider this to be a fact for the moment, you have to wonder how the charge damaged an IC or other component that was not directly in the path of discharge. For example, a metal cased mouse with its case DIRECTLY connected to the screen of an earthed computer would see a faraday cage effect where the charge goes around the PCB but not through it or it’s components. If the case is plastic, without an earthed spay conductive screen, the charge can travel to the plastic and potentially find a lower charge potential component in close proximity to the case mounts or even through the case, a short distance to a component on the other side. Much depends upon the insulating qualities of the plastic used. If the case is earthed via connections to the PCB before reaching the USB earth connection, there can be collateral damage issues as well.

Let us consider for a moment the potential earthing or lower charge paths within the mouse. Any of the USB connections to the host computer is a potential charge discharge path and it is interesting that your mouse effectively stops your other mouse from working. As has been stated, it is common to use ESD protection TVS diodes on all USB pins as these limit over-voltage events by taking the voltage to the USB chassis connection and/or the 0V rail. In your mouse I see no such protection so the manufacturer was relying upon the ESD protection built into the USB bridge within the processor IC. Not great practice but that is another story.

Let us consider failure modes here for a discharge via a USB connection.

The USB connections are +5V, 0V, D+, D- and screen (0V or Earth)
The mouse has low voltage Multi Layer Capacitors (MLC) on some pins but not all. The CDN (D-) and CDP(D+) capacitors are not fitted. Such capacities are more for data line noise suppression than any sort of ESD protection. MLC capacitors are not known for their tolerance of physical or electrical abuse and are regularly seen to go short circuit in laptop deployments. It would be possible for an ESD event to cause damage to an MLC capacitor that causes a short circuit or low impedance. In your situation I would remove the MLC capacitors connected to the USB connections in case one has failed. You can refit them if this does not remove the fault. It I’d does remove the fault, either test the capacitors with 5V across them or fit new ones.

That is enough from me for now.

Fraser

[Fraser]

Regarding component markings on the PCB….

R= Resistor
C= Capacitor
V=Voltage
D=Data
P= Positive
N= Negative

So a component marked “CV3” would ve a capacitor on a supply rail. Likely a decoupling (Bypass) capacitor.
One marked “CD3” would cue a capacitor on a data line. Likely RFI and noise suppression.
The “CDP” is a Capacior connected the plosive data line. Likely fir RFI and noise suppression.
The Resistors are similarly identified. So “RDP” is a resistor involved with the Positive Data line and “RDN” is involved with the Negative Data line

Good thinking by the designer 

[Fraser]

If wishing to move onto actual diagnostics, you could check the supply rails of the mouse to ensure they are present and correct. A shorted MLC will likely drag a supply rail down and may even activate the host computers over-current protection device. These are commonly self resetting fuses and they can take a minute or two to reset…. Sound familiar ? The host PC could have been experiencing temporary over current situations in the past and triggered the auto resetting USB 5V supply fuse. Such a protection circuit might also disable other devices on the same USB hub supply rail.

Fraser

[Fraser]

My thoughts on the auto restring fuse are possibly wrong as they reset quicker than I said ! It could be a LDO regulator in the mouse in distress causing an issue though.

Fraser

[Pcmaker]

If I buy another used one, is there a way to add ESD protection to it in case it happens again?

[Fraser]

ESD power and data path protection is easily added as there are several TVS diode array case formats to suit different applications. The challenge is in deciding where ESD protection is needed. If the mouse has a plastic case, I would spray conductive paint on the inside and use spring contacts to connect to the paint layer. The contacts should be connected to the USB screen connection.

Fraser

[nvmR]

As user Fraser discussed, it is important to consider failure mechanisms before the mouse itself.
One of these is the outlet you are using for example. It can be checked using an outlet checker to see if GND is preset and ok (as in image attached) .

Regarding fixing the mouse:
The fact that the usb port doesn't work after plugging in points to a short which trips the e-fuse in the usb port. If you have a power supply with current limiting capability, we can understand how much the mouse is taking from the usb-port. In any case, checking for resistance between 5V and GND when the device is off, and looking for capacitors which may be shorted is critical.
This is again quite similar to Fraser's tips

[Pcmaker]

ESD power and data path protection is easily added as there are several TVS diode array case formats to suit different applications. The challenge is in deciding where ESD protection is needed. If the mouse has a plastic case, I would spray conductive paint on the inside and use spring contacts to connect to the paint layer. The contacts should be connected to the USB screen connection.

Fraser

The housing is plastic. So I should just spray the entire PCB with a special spray and it's good?

[james_s]

I took it apart and this is the PCB. I don't even know where to start looking as to what is broken. can these boards be bought by themselves with all the parts installed or are they made specifically for Steelseries?

It's a fully custom PCB designed from the ground up for that mouse. It probably has a lot in common with other mice made by the same company in the same era but there is no way you will be able to buy just a replacement board off the shelf. Most of the components on it are standard off the shelf parts but as someone else mentioned already, the firmware is the secret sauce that makes it all work and you probably won't be able to get your hands on that.

You might be able to hack up a similar mouse and stuff the guts inside the case from this one but that's likely going to be some work.

[Ian.M]

Another approach, once its been proven the TQFP-48 custom programmed MCU is faulty (see  Fraser's comments above for why it might not be), and if you cant source a board from the same model mouse that still enumerates when its USB is plugged in (possibly from a mechanically knackered mouse) to transplant a good MCU from, would be to remove the MCU and 'dead bug' a QFN package ATmega32U4 within its footprint, patching it to the TQFP footprint as required to pick up signal and supply voltages required.  I suspect you'd also need to change the crystal as an Arduino Leonardo uses a 16MHz crystal and I *think* that one's 12MHz.  Then you could write your own firmware for it as-if it was an Arduino.

Unlike Fraser, although I would suggest the same nickel screening spray, (on the inside of all plastic case parts except for the sensor window/lens, *NOT*on the PCB),  I would connect to the nickel screen once the paint is fully cured with copper tape with conductive adhesive, then solder grounding wires to the copper tape, as I think grounding springs would be difficult to mount reliably.  Ideally, overspray the edges of the copper tape with more  nickel paint after soldering the tape end of the grounding wires.