How to Test Diode & transistor leakage

[danny davis]

How do you check and test Diode Leakage? in circuit test? and out of circuit test?

How do you check and test transistor Leakage? in circuit test? and out of circuit test?

 

[Audioguru]

Re: Testing Diode leakage & transistor leakage

I have seen tens of thousands of circuits and have never seen any leakage in a diode or transistor.
Why do you want to waste time trying to measure something that probably does not exist?

 

[danny davis]

Re: Testing Diode leakage & transistor leakage

My Service Manuals at work , in the back of the service manuals there is 30 pages of troubleshoot charts for each boards in the circuit and it says that to change a diode or transistor because of leakage

It has lists and lists for diodes and transistors have will cause a problem on the circuit board

It must exist if I have a bunch of service manuals , they are written in the 60's, 70's for sure

The problem I'm having is how to you test the leakage of a diode or transistor?

The manual will say D109 test the leakage if there is a problem on circuit board XXX because it will causes this problem XXX
The Manual will say Q12 transistor test the leakage if there is a problem on circuit board XXX because it causes this problem XXX

But I don't know how to test the leakage in circuit or how?

- - - Updated - - -

The circuit boards have only METAL CAN transistors and IC op amps

When I get RMA's circuit boards that have been out in the field , they come back with intermittent problems

When I do a freeze test or a Hot air test on each METAL CAN transistor, some of the transistor will be faulty and flaky

But then the transistors will fix themselves after and will never cause intermittent problems again

Does the freezing and hotting them up fix them? like it fixes the doping or functions molecules internally again

It something to do with chemicals

Because I have noticed with Silicon transistors that are those ( black plastic packaging types NOT the metal cans)

Those METAL CAN transistor I'm not sure if they are the same Silicon transistor like the black plastic package types are , they do react different with intermittent problems

The Silicon Transistor black package types, when they are intermittent , you freeze or heat them up and they stay like that, they don't fix themselves like the METAL CAN transistor do

Are the Metal Can transistors the same type of silicon and are the same as the black plastic types?

 

[Audioguru]

Re: Testing Diode leakage & transistor leakage

What do you think leaks out of a diode or transistor? Beer?
When a diode or transistor leaks then it messes up the circuit and the problems obviously show that leakage caused it.

You are silly when you compare a metal can IC or transistor with one in a black epoxy package, because you never saw the different maximum allowed limitations on the datasheet.

 

[FvM]

Re: Testing Diode leakage & transistor leakage

May be leakage current measurements have been performed routinely with germanium transistors in the fifties and sixties?

 

[betwixt]

Re: Testing Diode leakage & transistor leakage

I was thinking the same, leakage virtually disappeared as a problem when silicon became the prevalent semiconductor material. Perhaps these boards are so old they are using germanium transistors and diodes.

Danny, or whatever your name you are using today, give us a clue about what these boards are from or what they do. No need for specifics, just give us an idea what kind of product they are used in. If you are working with products designed and built 50 years ago it would explain some of your less coherent postings.

Brian.

 

[Audioguru]

Re: Testing Diode leakage & transistor leakage

On the other website terms like "Gyro", "Rudd", "Flaps" and other aero things led us to believe that this guy who knows nothing about electronics is "repairing" (destroying) these circuit boards for passenger airplanes. Unbelievable. We will never fly again!
He said he has been to MANY job interviews.
I would not let this guy fix a toaster.

 

[danny davis]

Re: Testing Diode leakage & transistor leakage

If you are working with products designed and built 50 years ago it would explain some of your less coherent postings.

Aerospace, 50's and 60's circuit boards

They use METAL CAN transistors and METAL CAN IC CHIPS , not the dual in line packages

Since they are METAL CAN transistors , do they have leakage current tests? because it's a METAL can and not a silicon black plastic package type transistor?

I think the silicon Metal can transistors are different than the silicon black plastic package type transistor

How are they different?

 

[FvM]

Re: Testing Diode leakage & transistor leakage

Generally speaking metal can transistors aren't different from plastic case types. In detail, you'll want to refer to specific types.

First commercial silicon transistors appeared 1958 and became commonly used til mid sixties. Before only gemanium was in use.

 

[danny davis]

Re: Testing Diode leakage & transistor leakage

So why does my service manual from the 60's say to test the Leakage of silicon diodes and transistor " IN CIRCUIT TEST" but how do i do this please? Do I need a transistor tester or semiconductor tester?

Will a freeze test or heat test , checks the leakage of a silicon diode or transistor?

 

[FvM]

Re: Testing Diode leakage & transistor leakage

I don't know why. As other contributors I thinks that it's more or less useless. But if the test is of any essential importance, I'm sure that the service manual will describe the procedure and required instruments.

 

[danny davis]

Re: Testing Diode leakage & transistor leakage

the service manual will describe the procedure and required instruments.

The service manual doesn't , the tech should know how to do this, which I don't know how to , this is why I'm asking you guys, can you please help me out

What instrument would you use or how would u do this please?

 

[FvM]

Re: Testing Diode leakage & transistor leakage

There will be at least specifications of acceptable leakage current at voltage x.

 

[danny davis]

Re: Testing Diode leakage & transistor leakage

But how do you test the diodes and transistors in circuit for the leakage? how would u do this please?

Do I keep the power on the circuit, or turn the power off?

Do I set my DVM meter to current and put it in series with the circuit? on the base, emitter or collect?

 

[Audioguru]

Re: Testing Diode leakage & transistor leakage

But how do you test the diodes and transistors in circuit for the leakage? how would u do this please?

Do I keep the power on the circuit, or turn the power off?

Do I set my DVM meter to current and put it in series with the circuit? on the base, emitter or collect?

How do you know that a diode or transistor has too much leakage current?
You simply measure its voltages and see the voltage that is incorrect due to the leakage current. It is that simple.
But the incorrect voltage might be caused by a resistor that has a value that drifted or something else that is feeding an incorrect voltage to the diode or transistor.

If the voltages in a diode or transistor circuit are wrong and you suspect it is caused by a leaky diode or transistor then would you build a circuit to test them?
When a diode costs 5 cents and a transistor costs only 10 cents?? Wouldn't you simply replace them without testing them??

 

[danny davis]

Re: Testing Diode leakage & transistor leakage

You simply measure its voltages and see the voltage that is incorrect due to the leakage current. It is that simple.
But the incorrect voltage might be caused by a resistor that has a value that drifted or something else that is feeding an incorrect voltage to the diode or transistor.

Yes this is what I have been doing, is there another way to test the leakage current?

But not all transistors leakage are just steady current leakage it's flaky , intermittent leakages I mean also

How do you test intermittent leakages? or Flaky leakages?

When a diode costs 5 cents and a transistor costs only 10 cents?? Wouldn't you simply replace them without testing them??

Cause u have to present to a manager and the company why these components went bad , everything is ISO9001 at every company now a days , you can't just replace parts like that for no reason , they will get made

look up ISO9001 and the FORMS you have to fill out for it

 

[Audioguru]

Re: Testing Diode leakage & transistor leakage

ISO9001 deals with the quality of products and services.
Now we are talking about the same thing as on the other website where we asked, "OMG! Why is there a technician who knows NOTHING about electronics (YOU) working on electronic circuit boards for airliners?"

Aren't you supposed to be a certified electronic technician who passed manufacturer's courses about the product you are servicing?

 

[danny davis]

Re: Testing Diode leakage & transistor leakage

Not for the jobs I get , iso9000 makes u fill out forms, u have no idea because u been away from electronics for 50yyears

 

[betwixt]

Re: Testing Diode leakage & transistor leakage

Some of us have been IN electronics for 50 years and still can't work out what you are on about 8-O

Back to technicalities:
1. in almost all cases you can not measure leakage while a semiconductor is in circuit. It has to be removed or what you measure is conduction around it through surrounding components.
2. leakage IS temperature dependant but it would be extremely unusual for it to be intermittent or vary randomly
3. to test leakage you apply a test voltage with the device in a non-conducting state and measure how much current flows. In theory nothing flows, that's why anything you meausre is called 'leakage'.
4. for diodes, you apply the voltage in the non-conducting direction and measure the current.
5. for bipolar transistors, connect the base and emitter together so it cannot have any bias current then measure any current passing through it.
6. some tests for bipolar transistors are done with the base left disconnected but still measure collector current.
7. there is absolutely no point in trying to do any kind of leakage test on an IC because they intentionally have current paths through them to allow them to work.
8. for all devices, the leakage is specified at a temperature and applied voltage. The data sheet tells you the limit that is allowed. (IR for diodes, ICbo for transistors)

The junction and leakage properties are essentially the same regardless of package type. It is rare for leakage to cause problems, it doesn't normally change with age. It is also normal for an overheated component to become leaky but if it is kept within it's absolute maximum temperature rating, the leakage should return to normal as it cools down.

Brian.

 

[dick_freebird]

Re: Testing Diode leakage & transistor leakage

Leakage can be a consequence of EOS damage.

In circuit, you'd simply "think like a transistor" and figure
out what the signature of a leaking one would be. If it
does not leak enough to move the node off normal, it is
not leaky enough to matter.

Datasheets always have the conditions somewhere in
them, along with the limits - limits otherwise are not
meaningful.

However some independent thought may be required.