Replacement for 2N5160 transistor?

[autophile]

Hi all,

In trying to repair an HP 8165A signal generator, I found that one of the output transistors is bad. HP's test equipment division, which is now Agilent, had the wonderful corporate mentality of using custom components, even going to the extreme of manufacturing their own transistors.

The transistors listed for their push-pull x2 50 MHz 1W output amplifier (which has to amplify 50 MHz square waves as well as sine and triangle) are:

NPN: 1854-0332
PNP: 1853-0315

The NPN is listed as having PD = 1W, fT = 800 MHz.
The PNP is listed as having PD = 1W, fT = 1 GHz.

Both transistors have a TO-39 case.

No HP cross-reference lists these transistors.

After examining the circuit diagrams, I've come to the conclusion that any complementary pair of transistors will do, so long as their beta is above 30 or so, PD of at least 1 watt, and fT of around 500 MHz or higher.

After looking around, it seems the best PNP to fit is the 2N5160, whose complement is the 2N5109.

The 2N5160 is extremely difficult to find, since it is not being manufacured by anyone except a few die-hards (who probably want to sell them in lots of 500).

Does anyone know of a reasonable replacement for the 2N5160? The replacement MUST have a complementary NPN part.

Again, the specs are:

PD around 1W
fT > 500 MHz
beta > 30
TO-39 case preferred

Remember, I'm not interested in actually procuring a 2N5160. I'm interested in a *replacement* or an *equivalent*.

Thanks,

--Rob

 

[felixc]

You did not mention the working voltage. You may want to take a look at the Zetex Super E-line of transistors. They have a plastic package but can dissipate more power than a TO-39. (they can work at 200C) As an example look at the ZTX451 and 551 pair. The Ft is not as high as 500MHz, but you can give it a try. Available at Digikey, in qty 1 if you wish.

 

[autophile]

Hi felixc,

The working voltage is 24 volts.

The ZTX451 is okay, but the fT is a bit low for the application, as you've pointed out. Since we're only dealing with a x2 amplification at max 50 MHz, an fT of 150 MHz should probably work for sine wave output, but it would distort the triangle and square wave output.

It's better than nothing, but I'd like to see if there are any other suggestions before going with a "reduced performance" part.

Thanks for the suggestion!

--Rob

 

[autophile]

It seems I'm about to answer my own question, sort of.

Texas instruments makes a THS4041/2 opamp (see

http://focus.ti.com/lit/ds/symlink/ths4042.pdf

for datasheet) which has a gain-bandwidth product of 165 MHz, power supply max +/- 15V, power dissipation of at least 1W (with proper PCB cooling).

I should be able to just bypass the whole discrete amplifier in the 8165A and use this opamp.

Six bucks from digikey plus a handful of components and I should be in business!

--Rob