Hi,

I was taking a look at the datasheets of the 2N4401 and 2N3904. All two TO-92 and 40V containers from Vceo though:

- the 3904 can dissipate up to 625mW and has an Ic of 200mA
- the 4401 can dissipate up to 250mW and has an Ic of 600mA

obviously they are maximum values but they are not few 250mW with an Ic of 600mA. Why would that mean that with only 250mW of dissipation it would only be possible with a maximum of 6.2mA? But then why so much difference on the dissipation equal to the container? The question arises from the need to understand which of the two is more suitable for a small class A in push pull with complementary PNPs in parallel.





The second question is related to the connection of the load which in this case is 8 Ohm but it is not important (I think) at the end of the question.
Normally the load is connected like screen 2 but can it be connected like screen 3?


Thank you!

 

I see the dissipation listed as 625mW on both. That is typical for a TO92 package.

 

In my book the two are interchangeable. The data sheet pages you posted seem to suggest the same. I have been using them interchangeably for over 30 years, into the tens of MHz for amplifiers and over 100 MHz for oscillators.

If you use configuration B (as in your circuit above) it looks like the signals going into one input needs to be inverted with respect to the signal going into the other input. But why do that -I don't see any benefit to it, but I do recognize and opportunity to have DC voltage across the load.

 

When I see an 8Ω load resistance I immediately assume Audio, and audio requires an AC output signal.
Circuit #2 can produce an AC output signal, and Circuit #3 can't.

 

I see the dissipation listed as 625mW on both. That is typical for a TO92 package.

Yes it's true but why but how come one can handle 200mA and the other 600mA?

 

In my book the two are interchangeable. The data sheet pages you posted seem to suggest the same. I have been using them interchangeably for over 30 years, into the tens of MHz for amplifiers and over 100 MHz for oscillators.

If you use configuration B (as in your circuit above) it looks like the signals going into one input needs to be inverted with respect to the signal going into the other input. But why do that -I don't see any benefit to it, but I do recognize and opportunity to have DC voltage across the load.

Do you know if there is a database with all the pairs of pnp and npn transistors that can be used as complementary?

 

Do you know if there is a database with all the pairs of pnp and npn transistors that can be used as complementary?

Normally, it shows the part number of the complement in the datasheet.

 

Yes it's true but why but how come one can handle 200mA and the other 600mA?

If you compare the two data sheets from the same manufacturer, it shows the 2N4403 with larger capacitances, suggesting it has a larger chip.
Once upon a time, manufacturers included a dimensioned photo of the chip in the datasheet.

 

When I see an 8Ω load resistance I immediately assume Audio, and audio requires an AC output signal.
Circuit #2 can produce an AC output signal, and Circuit #3 can't.

Sorry if it is not specified. I forgot to mention that the signal is AC. So correct .

 

If you compare the two data sheets from the same manufacturer, it shows the 2N4403 with larger capacitances, suggesting it has a larger chip.
Once upon a time, manufacturers included a dimensioned photo of the chip in the datasheet.

Here's why! . Then the 440x pair might be more suitable than the 390x for a current buffer stage as a power amp if I'm not mistaken. While the 300x as a vbe amplifier (driver stage?)

 

Yes it's true but why but how come one can handle 200mA and the other 600mA?

They are different transistors. One was designed for use at higher collector currents than the other. If you look at many of the performance parameters, you see that the 2n3904 is better optimized for operation at a lower current than the 2n4401. For lots of applications, the difference isn't enough to draw attention and so they are effectively interchangeable for those applications.

 

The 2N4403 is famous for being a really low noise amplifier for very low impedance sources such as moving coil cartridges, capable of sub 1nV/√Hz

 

Yes it's true but why but how come one can handle 200mA and the other 600mA?

The power dissipation limit has only to do with the package and how much heat it can dissipate without raising the temperature of the silicon above its max. This has nothing to do with the max current the transistor can handle.

 

The 2N4403 is famous for being a really low noise amplifier for very low impedance sources such as moving coil cartridges, capable of sub 1nV/√Hz

So is pretty suitable also for small signal like phono preamp (turnable).

 

Do you know if there is a database with all the pairs of pnp and npn transistors that can be used as complementary?

No, haven't seen such a thing.

 

Did you read the specs and see the curves in the datasheets?
1) The 2N3904/2N3906 have a max allowed current of 200mA but their specs and curves show poor performance above 50mA.
2) The 2N4401/2N4403 have a max allowed current of 600mA but their specs and curves show poor performance above 200mA.

Instead of many weak little transistors, why not use two power transistors with little heatsinks? Also their emitter resistors can be 0.22 ohms for much less voltage wasted.

The transistors have a max allowed voltage of 40V so make your circuit safe with 30V. Your paralleled transistors have 20 ohms/5= 4 ohms of emitter resistors in series with the 8 ohms speaker then the max peak current in each group of 5 transistors is 14V/(8ohms+4ohms)= 1.17A or 233mA for each transistor.

Your amplifier with a 30VDC supply might produce 8.2W RMS total in the 8 ohms speaker in series with the 4 ohms emitter resistors or 5.5W RMS in the speaker plus 2.7W in the emitter resistors. The transistors heat with about 800mW each. Then they will overheat and be destroyed.
Using TIP31/TIP32 power transistors with a 30V supply, the amplifier output power will be about 12W RMS and each output transistor needs a little 8W heatsink.

 

Did you read the specs and see the curves in the datasheets?
1) The 2N3904/2N3906 have a max allowed current of 200mA but their specs and curves show poor performance above 50mA.
2) The 2N4401/2N4403 have a max allowed current of 600mA but their specs and curves show poor performance above 200mA.

Unfortunately it all depends on whose datasheet you are looking at. Checking the Fairchild 2N3904 and 2N4401 data sheets, they both list 800 ma as the absolute maximum current rating.

 

My 2001 Rev- A Fairchild datasheet shows 200mA max for the 2N3904/2N3906 and the graphs stop at 100mA.
The Fairchild 2N4401 datasheet shows 600mA absolute max and the graphs stop at 500mA.

 

Do you know if there is a database with all the pairs of pnp and npn transistors that can be used as complementary?

I have run across a few charts that list complementary pairs, including one only a couple weeks ago. I'm sure it's far from all-inclusive.

I tried doing a search just now and didn't find anything quickly -- but they are out there.

 

And this difference between different manufacturers of the same transistor model that got me confused.