Damage on uC from continuity test ??

[Mat]

Hi,

I just had to replace the uC (dsPIC30f6010A) from this prototype board I'm working on because of a fried GPIO output buffer...

Is it possible that the damage was done by probing the pins for continuity against ground or other adjacent pins with a Fluke 23 III multimeter (unpowered circuit of course)?

And while I'm at it : what's the difference between the 23 III and the 70-series multimeters ?

Thanks in advance guys,
M.

[Psi]

Good question, I'm interested to know what the official answer is

Many years back i had a plasma display driver IC stop working immediately after I probed pins with DMM on continuity range.
I still don't know for sure that's what caused it to fail but i've always suspected it was damage from the DMM.

[kizzap]

Not 100% sure on this, but depending on the uC and the DMM, is it possible that the continuity setting put out a voltage that was higher then the allowable voltage on the pin?

-kizzap

[Dave]

Try measuring the voltage on the open contacts of the multimeter on continuity test. Post the result.

[mikes]

Manual for 23-III shows open circuit continuity test voltage <3.1 V and a short circuit current ~400 uA. That's unlikely to have caused a problem.

[KJDS]

I've fried stiff before when probing just by clumsily bridging a port to ground or supply line.

[Neilm]

I have seen this happen as well. Damage can happen as you are effectively taking the GPIO port above the power rail (as it is turned off at the time) 400uA isn't much but it can be enough to damage the protection circuit in this case. Generally, what is in chips is designed to stop very small ESD events and not much more.

Neil

[FrankBuss]

I don't think that the low current of the multimeter can destroy it. But the datasheet of the dsPIC30f6010A says: "The analog inputs have diodes to VDD and VSS as ESD protection.". Does this mean, that the digital inputs don't have ESD protection? Then this could be the reason, when touching a contact with a finger.

I've not used dsPICs so far, but the PIC 16F and 18F series and never had such a problem, even when I bridged incidentally a GPIO to VDD or VSS.

[Neilm]

ESD damage doesn't always result in immediate failure. It can just reduce the life of a component so it will fail later.

I've not used dsPICs so far, but the PIC 16F and 18F series and never had such a problem, even when I bridged incidentally a GPIO to VDD or VSS.

But did you take them above VDD? That is when damage can happen.

Neil

[FrankBuss]

But did you take them above VDD? That is when damage can happen.

I've just tried it the hard way: connecting GND to one of my circuits with a PIC18F252, then applying 3.1V with a 1k series resistor to each pin for some seconds each. The circuit is still working. There were other chips connected as well, but the voltage at the pins was still between 1V and 2V, so I guess not much ESD protection, otherwise one of the ESD protection diode would have reduced the voltage to 0.7V or below (but I don't know much about the analog side of chips). So I would say it is no problem. But to be sure that it doesn't reduce the life time of the chip, someone could ask the Microchip technical support, usually they are very helpful.

[SeanB]

Many IC's only have limited ESD protection, small feature sizes limit the amount of energy they can handle. You might have just had enough static charge on you dischargeds through the junction to kill it.

[jimmc]

Looking at p177 of the dsPIC30F6010A data sheet (http://ww1.microchip.com/downloads/en/DeviceDoc/70150E.pdf)
under Absolute Maximum Ratings it gives

Input clamp current, I_IK(V_I< 0 or V_I> V_DD) .........±20 mA
Output clamp current, I_OK(V_O< 0 or V_O> V_DD) ...±20 mA

So the 400uA from the meter was well within permissible limits; as has been said ESD is the most likely culprit.

Jim

[Mat]

Thanks a lot everyone!

Yeah, I'm leaning towards ESD too now that I think about it. I'm 99% sure I was wearing the ESD bracelet thing but could have forgotten....

M.

[Baliszoft]

dsPIC30F withstands a lot more abuse than the specified 20mA input clamp. I have tested up to 800mA without damaging the pin itself.