With the recent talk about oscillation I thought I would share a recent debug. Summary: have multiple ferrite bead types on hand.
?
I made a PCB for the Ikin Norton Amp:
?
?
It is a 2-layer PCB with ground fills. I built one using 2SC3357 transistors and it worked. I built a second using 2SC5551 and it oscillated at 660MHz. It also did not amplify. I checked and double checked the transformers but could find no difference between the two boards, but then I found the amplification problem: one end of the T1 primary was soldered but had not melted the enamel on the wire so it was not actually connected. But it still oscillated. The ferrite beads I used are Laird MI1206K310R, which are about 41 ohms at 1GHz. I tried replacing them with 68 ohm resistors but no change. I tried replacing the 10pF caps with 39pF, still no change. I then put the 10pF caps back and put Laird LI1206H121R-10 ferrite beads in and it tamed the oscillations. The new ferrite beads are around 135 ohms at 1GHz, so that is what it took.
?
So don't be afraid to use high fT transistors, but make sure you have a ferrite bead in the design as well as a variety of ferrites to test.
?
I just looked at the data sheets. The ZTX327 used in the original has an fT of 800MHz, the 2SC3357 is 6.5GHz and the 2SC5551 is 3.5GHz.
?
So even though the 2SC3357 has a higher fT than the 2SC5551, it did not oscillate.
?
--
===================================================================
Mike M