Class A/B amplifier transistor voltage rating

Thread Starter

eric stevens

Joined Dec 10, 2008
My question is about a Mono amplfier design with rail voltages of +/- 60 volts for a total potential of 120 volts.

Am I correct that each transistor will only see 60 volts in a class A/B bipolar push pull amplifier design because both transistors will never be on at the same time?

The transistors in question woudl be the TIP35C/36C which are rated for 100 volts.

I am being told that we need to use transistors that are rated at 150 Volts and think that this isnt necessary even to stay well within the SOA of the devices.

Thanks for you help,




Joined Jul 3, 2008
You seems correct to me. Did they say why 150 V was needed? Is it for reliability saftey margin, or do they really believe 120V will be across the device?

Is the other end of the load connected to ground? If so, it would seem that 120V on the load, or on either transistor would not be possible.

Perhaps they are worried about a possible open load condition?

EDIT: Actually, scratch that. If you drive the load with one transitor to maximum voltage, you will put 120 V on the other transistor.

Additional EDIT: Yeah it's easy to get fooled by this one. There is a tendency to think only of the driving transistor. While the transistor is driving the load, the voltage will be 60 V or less, but the other "off" transistor takes up the 60 V plus the load voltage. Even though that transistor is off with no power dissipation, the 120V will exceed the breakdown voltage of the transistor and kill it instantly.
Last edited:


Joined Nov 9, 2007
Yes 150 volt transistors would be very good.

You can see from the bare bones class AB amp that when TR1 is on, point D in the circuit is nearly at +60 volts. This is connected to the emitter of TR2, whose collector is at -60 volts.



Joined Aug 24, 2008
Those responding to this question figured it out very well, I think. But when dealing with transistor specs as given by vendors beware. A rule of thumb is to use a device with twice the pertinent parameter, be it voltage, current, power, or whatever.

Thread Starter

eric stevens

Joined Dec 10, 2008
Thank you to everyone for your responses. I now understand well the forces at work. I am a mechanical engineer and working on amplifier product development.



Joined Aug 8, 2005
If I stretch my memory and recall correctly, most of the TIP35 type are in TO-220 style packages, and there have been many amp manufacturers using those types for saving space. They will work fine until some rocker turns up the volume to 11, then they blow for reasons described above. The owner takes it into the repair shop, where the repairman dutifully replaces the blown parts with the same type. The amp blows again, and the repair person gets blamed for a piece of crap (or at the least for being incompetent).

In high-voltage amps like you describe, you are much better off using the TO-3 style can, because higher voltage translates into higher current into the load/speaker. For applications such as yours, the 2N5631 from On-semi should work well.


Joined Nov 9, 2007
It should perhaps be pointed out that the A,B,C suffixes for the TIP series refer to voltage ratings not gain groups as is common with BC series.

So for the TIP35 we have CE voltage ratings of


You should be able to find a suitable version in this list.


Joined Dec 20, 2007
The TIP35 is in a TO-218 package that is much bigger than a little TO-220.
Its max allowed dissipation is 125W which is more than a 2N3055 that has a TO-3 metal case.
Its max continuous current rating is 25A and its max peak current rating is 40A.

An amplifier with a 120V total supply will produce about 55V peak into its load. If the load is 8 ohms then the peak current is only (!) 6.9A. The amplifier's output will be about 190W RMS at clipping into 8 ohms. The amplifier will dissipate a max of about 150W so the heatsink must be enormous for continuous full power output.


Joined Nov 9, 2007
The amplifier will dissipate a max of about 150W so the heatsink must be enormous for continuous full power output.
Whilst the TIP series of transistors were often used in 1970s/1980s designs, they were rarely individually at this power level.

More often they were used with two or more parallel complementary pairs. It was common to see a whole row of power transistors on the massive heatsinks. Fan cooling was not popular then either.

Because of gain requirements substantial driver transistors were also required.

I reliased this, but since the discussion was about voltage ratings, I kept to a simplified circuit, adequate for that purpose.