If I don't touch it , it does not go to rail voltage, however, should I touch the board or twist it....then she goes..
Lee,
You didn't mention which board you were touching so I will assume it is the Rev D board and not Don's DCP.
Follow the disassembly procedure in the Phase Linear 400 service manual.....
"Remove the (2) hold-down nuts from the Rev D board, flip the circuit board over and gently pull towards you. Before energizing the unit be sure that none of the PCB components are touching the rim of the chassis. "
The unit is now ready to be powered on.
Dean taught me many years ago to use the handle of a screwdriver and tap the side of the pc board to locate any intermittent's which may be caused by bad solder joints, traces, defective components, etc. We also heat and cool parts on the pc board to locate defective parts. Dean used the tip of a soldering iron placed in close proximity to parts to heat them as well as a circuit cooling spray to cool them.
The original wiring between the pc board and output stage when it is subject to flexing will start to break individual strands off and can also cause problems.
I was recently working on an 400 II amp while the owner was standing next to me and I broke one of the B+/B- wires leading to the pc board (due to flexing the board several times after performing the low THD mod and upgrading some parts on the pc board) and didn't see it right away. When I powered it back up one of the rails was not powering up and the sine wave was severely clipped on my oscilloscope. I remained calm and I told the owner that I had suspected one of the wires was bad and quickly located it. When the owner first showed up he mentioned some problem that was common to both channels and I took a look at the power supply filter caps. One of them was a FAO brand and the other was a replacement that occurred some time in its life and told him that the FAO brand had a higher than normal failure rate and was more than likely the culprit. Indeed when I measured both caps the replacement was measuring around 85 volts the FAO was at approx. 45 volts. That was a quick and easy fix as well.
As you probably already know, the 2N3403 bias transistors leads are also subject to breakage due to flexing of the leads as well.
Ed