External Power supply current limit set to MAX amps to find SHORTS on a PCB

[Billy Mayo]

External Power supply current limit set to MAX amps to find SHORTS on a PCB

The Techs that sit next to me at work , use a technique of finding shorts

When there is a SHORT from VCC to a component or a SHORT to ground to a component or a SHORT from VCC to ground

The use an external power supply and set the current limit to MAX 3 to 5 amps , but the voltage is at 5 volts or less

They turn OFF the circuit , And Pulse or trigger the Power supply rails with an external power supply and set the current limit to MAX 3 to 5 amps , but the voltage is at 5 volts or less

The Shorted component or what is causing the short, will cause a component to get HOT or explode the component, plus it leaves damage to the blown traces on the PCB board

How can you do this technique without cause blown traces? they have the current limit set to high? and would be a good current limited setting?

Also , Is this technique done for only a SHORT that is for VCC to ground? without damaging other components?

Cause I would think if you had A SHORT that is a SHORT from VCC to a component or a SHORT to ground to a component , and use an external power supply and set the current limit to MAX 3 to 5 amps , but the voltage is at 5 volts or less , that it will damage the other components or STRESS them out

I think this technique is only used when you have a SHORT from VCC to ground right? without cause STRESS damage to other components?

 

[ronv]

Not a very good way to do it if you want to save the board, but I guess it works if you don't want to spend the time to find the problem.


**broken link removed**

 

[KeepItSimpleStupid]

Thermal camera, anyone? But keep the current below rated current.

You should have asked for one for XMAS.

 

[Mike odom]

The voltage is set below 5V so that the other components are not damaged. Even though they allow a max of 5A to flow into the short, it will not flow into other components.

I would not set the current limit for max, set it for 1A, then use a thermal (phone) camera to look to see which component is heating up, or if you have a temperature probe, this works as well. Most traces/vias will handle 1A without damage. You still might damage the component pads if you over heat it.

At my last tech job, we had a row of tantalum capacitors on our 12V bus, one of which would short out. We basically did the same thing, set our power supply for max Amp out and bypass the fuse and protection circuitry and feed the 12V bus directly. The offending cap would start smoking and sometimes blow out. At first we used a thermal camera to find them with a 1A source, but this was a hassle going to get the camera everyday, so we used the 5A supply and identified the offending part w/o damaging the boards (Flight Data Recorders).

 

[shortbus=]

Here we go again! The big shake-up here did not change anything, but loose good people.

 

[Billy Mayo]

then use a thermal (phone) camera to look to see which component is heating up,

But a direct short doesn't cause heat or have voltage across it

We basically did the same thing, set our power supply for max Amp out and bypass the fuse and protection circuitry and feed the 12V bus directly.

You just said not to sure the power supply for max amp out, you said it would damage the traces , pads, and other components

What did u set the voltage on the power supply at to feed the BUSS rails?

 

[Mike odom]

We set it at 12V since it was a 12V bus, but when the power supply goes into current limit that voltage would drop to about 3V.

A short is usually smaller than the traces on the pcb, so it would exhibit more resistance and would heat up more at the same current flow than the surrounding traces. If it is a component that is shorted, they tend to heat up as well.

Yes, I said not to set it at full amperage supply, then I explained we did it for a very specific short test, where we knew we had a shorted tantalum cap and that it would blow out rather quickly when we hit it with 5A. At 3A, it wouldn't show itself.

 

[large_ghostman]

Funny thing is, in a old post of yours about trouble shooting, I mentioned using current to trace a short, and now guess what your work place uses this!
How come in your other fault finding shorts thread you never mentioned this, just meters that were used

 

[large_ghostman]

So my other post for apologies to body certain body orifices, save me having to actually say sorry twice, I did it the first time under duress, the second time would be over my dead body, so just assume the other one applies here as well, it dosnt but for the sake of not getting shouted at by dad it does.

 

[audioguru]

But a direct short doesn't cause heat or have voltage across it

Again, you do not understand.
It is true than an IC that is a direct short will not have voltage across it and will not get hot. But the traces on the pcb that feeds the short have voltage across them and high current in them and will probably BURN OUT!

 

[large_ghostman]

Again, you do not understand.
It is true than an IC that is a direct short will not have voltage across it and will not get hot. But the traces on the pcb that feeds the short have voltage across them and high current in them and will probably BURN OUT!

So if they burn out will they get hot???

There you go Billy I saved you the trouble of asking

 

[Billy Mayo]

When an IC or component gets very hot, is this a short or what? I always thought the compont internally is shorted inside causing it to get hot , so what is it than if its not a short ? i

 

[kubeek]

When an IC or component gets very hot, is this a short or what?

It is dissipating a lot of power. For example a light bulb or electric oven is very hot and you can burn your fingers, but still it is not shorted and is operating perfectly well.
So the first question you need to ask is how hot is said component supposed to be under normal operation. Second thing, if the component being hot is caused by the component itself, or some other factor (for example a hot zener diode caused by overvoltage or too low value of the series resistor - some resistors can degrade to having different than original value, by quite a large amount)

Some components really do fail shorted, power transistors and mos-based technologies being the most notable example. Some components like high speed opamps can get hot if they are left to oscillate wildly, but this doesn´t instantly mean the component is damaged, in this case it is rather working perfectly well.

 

[audioguru]

Billy,
Don't you know how to calculate heating? It is the voltage across it times the current in it.
If it is a dead short then it has no voltage across it and 5A in it then its heating is 0V x 5A= 0W. The pcb traces will heat with 5V x 5A=25W.
If it is an extremely high resistance and has 5V across it and no current in it then its heating is 5V x 0A= 0W.
If it has a partial short then it has maybe 4.5V across it and maybe 4.5A in it then its heating is 4.5V x 4.5A= 20.25W. The pcb traces will have heating of 0.5V x 4.5A= 2.25W.

 

[Billy Mayo]

If it is a dead short then it has no voltage across it

How can a dead short have no voltage ? if you have a +10 VCC to ground short

The SHORT will have +10 volts across it going to ground

Some components like high speed opamps can get hot if they are left to oscillate wildly, but this doesn´t instantly mean the component is damaged, in this case it is rather working perfectly well.

No, I'm talking about when an IC chip is extreme HOT , not normal operating hot , I mean you put your finger on the IC chip and it will hurt burn it

Is the SHORT internally in the IC chip or you're saying it's overvoltage or undervoltage that is cause the IC chip to get extreme HOT?

 

[kubeek]

Dead short means 0 ohms, so there cannot by any voltage across it. Basically the short has much lower resistance than the connecting pcb traces, so all voltage drop will be across those traces and not across the shorted part.

 

[Ian Rogers]

I think you are confused by a component going low resistance.... Electricity will always find the shortest path to ground.... So you can't have a dead short from 12v to 0v, the tracks and components a very low resistance causing FAR too much current to flow.... Remember if you stick a nail across + and - the nail becomes a path for electricity to flow and becomes a consumer of the current... It will allow the maximum current to flow... This isn't a dead short, its consuming current so its a consumer with a low resistance, no different that putting a 2 ohm resister across 5 v rail...

 

[Billy Mayo]

oh ok so when an IC is shorted internally, it is a very low resistance which has a voltage drop, and the current is VERY high? which causes it to be very very hot?

 

[audioguru]

On the other website forum he is asking how to "repair" a pcb that is burned into charcoal.

 

[Billy Mayo]

Dead short means 0 ohms, so there cannot by any voltage across it.

When is a component or something a dead short? and how do you find a dead short when there is no voltage drop or heat? no voltage drop is like an OPEN or BREAK , cut trace

Cause an OPEN = Zero volts
Dead short = Zero volts

So what's the difference?