Biggest mistakes reading BJT data sheets, What counts the most?

Thread Starter

MikeJacobs

Joined Dec 7, 2019
226
Hey all,
So in regards to BJT's and data sheets, there is so much information on them sometimes its hard to figure out what is even what.

If I were to ask you what are the most important piece of BJT data sheets what would you say?

Let contain this strictly to DC only. So totally disregarding the important stuff for amplifiers. This should make the list much shorter.
 

bertus

Joined Apr 5, 2008
22,266
Hello,

There are also databooks with reduced content.
Those will show tables of the transistors with basic information.
This is a scan of a part of a page as example:

EN_types.png

Bertus
 

Thread Starter

MikeJacobs

Joined Dec 7, 2019
226
It all depends on the application. Any single piece of the data could be vital for one application but irrelevant for another.
True

but what I am asking are, what are the most common things misses that will cause a part failure.

e.g VCE max etc etc ICE max etc etc
Power ?
DC applications only
 

Papabravo

Joined Feb 24, 2006
21,094
I would say the Absolute Maximum Ratings. It would be foolish in the extreme to place a device in a situation where one of those ratings were violated. As to the remainder I would say β and Vce(sat) would be next. You need to realize that those numbers are not tightly controlled, but rather are like normally distributed random variables with a mean and a variance. So most design are done in such a way that a particular value of a parameter is not required.
 

ronsimpson

Joined Oct 7, 2019
2,954
most important
PNP /NPN

C-E max voltage.

Current. Not max current but useable current. (I like to use the transistor where the collector current is near the max current gain) Also watch Vset verses Collector current.

Current Gain is a mystery to many people. (gain min./gain typical/gain max) Just because the top line said "gain=100" does not indicate all the parts are like that. Also that gain=100 is at the best current. Not at the current you will be using the part at. Also that is for DC not at 1mhz where the gain is much less. Gain is measured at some voltage C-E not at a very low Vce.

Power is often measured with a "worlds biggest heatsink" not a reasonable heat sink and certainly not standing in free air.
 

crutschow

Joined Mar 14, 2008
34,201
VCE max etc etc ICE max etc etc
Power ?
Those are among the most important, as exceeding any of those will likely zap the part.

Along with power, is thermal resistance to case/air for power devices, since the power rating is often given with the case at 25°C, which is not typically obtainable in practice, even with a good heatsink.

Also those maximum ratings/limits are typically derated by 25-50% for good part reliability.
 

Thread Starter

MikeJacobs

Joined Dec 7, 2019
226
Many simulator programs violate maximum allowed Vce, Ice and power dissipation.
That is a very good point. In a lot of ways I think the simulator model specific parts to be useless. If the sim is going to allow you to do whatever you want without adhering to the mode, what's the point of the model. Just use an ideal part
 

Thread Starter

MikeJacobs

Joined Dec 7, 2019
226
So , if someone said, lets make a checklist of all the things needed to properly implement a BJT for a simple DC application without zapping a part.
What would be on it?
 

MrChips

Joined Oct 2, 2009
30,618
I think that you are attempting to simplify something whereas there is no such simple answer.

If you do not want to kill a device then you would operate the device within its recommended limits which would be:

voltage
current
power dissipation
temperature range
 

Thread Starter

MikeJacobs

Joined Dec 7, 2019
226
I think that you are attempting to simplify something whereas there is no such simple answer.

If you do not want to kill a device then you would operate the device within its recommended limits which would be:

voltage
current
power dissipation
temperature range
I don't want to simplify it as much as I want to make a verification list. IF you do, X Y and Z your part will live
If you do X and Y and not Z your part dies
 

MrChips

Joined Oct 2, 2009
30,618
I don't want to simplify it as much as I want to make a verification list. IF you do, X Y and Z your part will live
If you do X and Y and not Z your part dies
You are making assumptions that exceeding manufacturer's specifications will destroy the device.

For example, manufacturer's specifications may say maximum operating voltage is 200V. That does not mean that the device will be destroyed if the applied voltage is 250V,

Another example, manufacture's specifications state that the device operating temperature range is 0-70°C. There is nothing that says the device will be destroyed if the temperature were 100°C.
 

MrChips

Joined Oct 2, 2009
30,618
Before you get me wrong, it is important to read manufacturer's datasheets and to stay well within recommended operating conditions.

For example, if a resistor is rated as a 1W resistor, it would be wise to keep power dissipation below ½W.
In other words, if your resistor power dissipation is calculated to be less than ¼W, use a ½W resistor.
 

OBW0549

Joined Mar 2, 2015
3,566
So , if someone said, lets make a checklist of all the things needed to properly implement a BJT for a simple DC application without zapping a part.
What would be on it?
Whatever's listed under "Absolute Maximum Ratings" on the data sheet for that part. Nothing more, nothing less.

ALWAYS CONSULT THE DATA SHEET.
 

SamR

Joined Mar 19, 2019
5,024
Not only Max Voltage and Current, but their product also has to be below Max power. Remember the Data Sheet for a transistor is a guide as they vary greatly in ratings due to manufacturing variations. So the Data Sheet is a guide and not absolute and it all changes with temperature.
 
Last edited:

OBW0549

Joined Mar 2, 2015
3,566
Remember the Data Sheet for a transistor is a guide as they vary greatly in ratings due to manufacturing variations. So the Data Sheet is a guide and not absolute and it all changes with temperature.
That's true for device characteristics (gain, leakage current, frequency response, switching times, etc.) but Absolute Maximum Ratings are just that: absolute. They are not just a guide, and to suggest they are is really, REALLY bad advice.
 

SamR

Joined Mar 19, 2019
5,024
While the Max may not be technically absolute, I would not suggest attempting to run anything close to the DataSheets Max. Yes, they posted a Max value for a very good reason and it would not be wise to exceed that value (or come close to it). And even that Max rating becomes derated by temperature increase which is going to happen when current flows through the device.
 
Top