help designing a pcb to be broken

Thread Starter

jpcrowd

Joined Mar 1, 2022
10
Hello,

I'm a graduate student at university studying Civil Engineering. However, my research has more of a mechanical engineering emphasis to it. A significant part of my research is looking into the impact resistance of PCBs and their components. To gather data and validate the models I've created, I need to design and test/break some very simple PCB panels. Needless to say, my electrical knowledge is pretty ends at V=IR. So I could use some help if y'all don't mind.

I need to design a simple square PCB panel (about 5"x5") with a row of chips running down the center, connected to wiring that runs out to the edges, and a way to measure the voltage on either end. The test consists of taking the PCB, bending it a predetermined amount, and checking to see if the current runs through either end. I've attached sketches of what I am trying to do.
Panel_Top_View.PNG PanelTest_Side_view.PNG
The main thing I need help with is picking out a chip and a part to measure the voltage. Also, given a voltage in one end, what should it be coming out the other side of the chip? I've downloaded the program Eagle to create a drawing to send off to manufacturing.

Thanks for the help. Please let me know if I need to clarify anything else, as I know so little I don't even know what I need to be considering.

Thanks,
John
 

MrChips

Joined Oct 2, 2009
30,714
My recommendation would be to measure resistance, not voltage.

You should be looking at degradation of solder joints manifested as increasing resistance.
The chip I would select would be an SMD resistor with the lowest value that your instruments can measure with sufficient resolution. Accuracy is not important. You are trying to measure change in resistance. Thus if your resistor value is 1Ω you want a resolution of 0.001Ω or 1mΩ.

SMD resistors of size 0805 are commonly used.
 

MrChips

Joined Oct 2, 2009
30,714
Hi MrChip, working with so small values, changes in temperature wouldn't play a role here as well?
Temperature will play a role regardless of the resistance value.

I am taking into consideration the resistance value of the PCB copper tracks and solder joints.
We want the SMD chip resistance to be as low as possible, comparable to the resistance of the tracks and joints.

Edit: Better still, use a 0Ω SMD chip resistor.
 

MrChips

Joined Oct 2, 2009
30,714
What are the failure modes?

  1. stretched copper tracks
  2. broken track
  3. broken chip
  4. broken solder joint

Edit: The failure mode will most likely be abrupt, i.e. from 0Ω to infinity instantly and not gradual. Hence high resolution is not necessary, in which case measuring voltage will be adequate.
 

MrSalts

Joined Apr 2, 2020
2,767
I think you are on the wrong thread @MrChips . The thread about what comes first, voltage or current.
If the OP has more comfort and tools to measure voltage, the other thread pretty much demonstrated that ohms law is really a scientific law - the OP,can use voltage, current or resistance as he wants to design his experiment.
 

MrChips

Joined Oct 2, 2009
30,714
I think you are on the wrong thread @MrChips . The thread about what comes first, voltage or current.
If the OP has more comfort and tools to measure voltage, the other thread pretty much demonstrated that ohms law is really a scientific law - the OP,can use voltage, current or resistance as he wants to design his experiment.
That was really unnecessary!:(
And should be deleted.
 

Thread Starter

jpcrowd

Joined Mar 1, 2022
10
What are the failure modes?

  1. stretched copper tracks
  2. broken track
  3. broken chip
  4. broken solder joint

Edit: The failure mode will most likely be abrupt, i.e. from 0Ω to infinity instantly and not gradual. Hence high resolution is not necessary, in which case measuring voltage will be adequate.
The models I'm currently looking at have the solder joint breaking first. I haven't considered stretched copper tracks though
 

MrChips

Joined Oct 2, 2009
30,714
So to was redesigning the OP's project and changing his question. He had a perfectly valid plan to use voltage.
Maybe. OP did not understand what comes out at the other end of the wire.
I gave a reasonable explanation why resistance measurement is the way to go.
Then I reconsidered and explained why voltage measurement will also be adequate.

What's your problem?
 

Jon Chandler

Joined Jun 12, 2008
1,029
Consider using 100mm × 100mm boards, as these can be made by many Chinese fabs at $5/10 boards.

Resistors seem like a good choice, as it's easy to measure resistance on the passive devices. Consider using a variety of physical sizes (i.e., 0402, 0603, 0805, 1210), oriented both perpendicular and parallel to the flex direction.
 

Thread Starter

jpcrowd

Joined Mar 1, 2022
10
Hi MrChip, working with so small values, changes in temperature wouldn't play a role here as well?
Even if temperature played a major role I don't think I would mind too much. Just need to make sure the circuit hasn't been broken. I'm looking at this problem more from an mechanics prospective than and electrical one.
 

Thread Starter

jpcrowd

Joined Mar 1, 2022
10
Consider using 100mm × 100mm boards, as these can be made by many Chinese fabs at $5/10 boards.

Resistors seem like a good choice, as it's easy to measure resistance on the passive devices. Consider using a variety of physical sizes (i.e., 0402, 0603, 0805, 1210), oriented both perpendicular and parallel to the flex direction.
Thanks for the info about the board sizes
 

Thread Starter

jpcrowd

Joined Mar 1, 2022
10
My recommendation would be to measure resistance, not voltage.

You should be looking at degradation of solder joints manifested as increasing resistance.
The chip I would select would be an SMD resistor with the lowest value that your instruments can measure with sufficient resolution. Accuracy is not important. You are trying to measure change in resistance. Thus if your resistor value is 1Ω you want a resolution of 0.001Ω or 1mΩ.

SMD resistors of size 0805 are commonly used.
Hey MrChips,

After looking at the SMD resistors, I'm worried I won't be able to get a reasonable stress/strain due to the small size.

what chip can I use to give me a reasonable signal (voltage, current or resistance) to send a signal in one side of the board and output on the other side - given that I need an SMD chip that is at about12mm long to get reasonable stress/strain on the chip. Please reply with a specific chip under $4 each

Thanks
 

Ya’akov

Joined Jan 27, 2019
9,071
Hey MrChips,

After looking at the SMD resistors, I'm worried I won't be able to get a reasonable stress/strain due to the small size.

what chip can I use to give me a reasonable signal (voltage, current or resistance) to send a signal in one side of the board and output on the other side - given that I need an SMD chip that is at about12mm long to get reasonable stress/strain on the chip. Please reply with a specific chip under $4 each

Thanks
That‘s a really large package. I would just look for a multi lead package that has GND pins on at least two sides, then connect to those on each side.
 
Top