Originally Posted by scootsit
With the newest board, the switch is necessary? I didn't realize that. I assume any SPST switch will work. Which set of contacts are used when the switch is opened/closed? Sorry for being such a drain, I'm just a little confused.
Thanks as always for all of your help.
If you populate the input relay-switching, the default (relays de-energized) is only one of the signal input punchdown blocks. You can never change the input to the other punchdown block unless you throw the switch. Something is needed to energize the relays to switch to the other input - that "something" is Switch S1. There are only two terminals, so it can be any old SPST switch - as long as it isn't momentary - that would be quite irritating.
I believe the left input punchdown block is the one that is in the circuit with the relays de-energized, but you can't really tell once it's wired in and if you've used a toggle like I did (no way to physically tell ON or OFF), so don't take that as gospel. Maybe jdkJake knows which is which. Here's my 2nd prototype (production PCB's are RED):
EDIT: the two white wires curving around behind the heat sink are the wires to the two-position terminal block for Switch S1. I used long wires on mine so I could mount the switch on the front plate. It's easy to tell the output wires from the BantamDAC going to the left-hand input punchdown block. The orange, purple and black wires come from a "test" pair of RCA jacks. So, two input sources - one the onboard BantamDAC (could now be a GrubDAC or SkeletonDAC, too) and the other to a standard pair of RCA jacks, whichh could be connected to anything.
The two mini-relays are energized in an either-or fashion by throwing the switch. However, one of the relays is energized by default when the power is on to the amp. The resistors are there to sink an un-used relay's output to ground, which cuts down on the cross-talk between the contacts. Both are fed from a TO-92 linear regulator at 24VDC. The jumper is there because cetoole (the designer) didn't want to break the groundplane above the signal traces. There is also a suppressor diode that works better than the old 1N4148 to cut down on the potential arcing between relay contacts and will promote long-life in the relays.
The relay circuit changes were also incorporated into the output relay-delay circuit for headphone protection. The old Zener-diode and high-power transistor arrangement with its legacy in the e22 design is gone. Instead, a single TO-92 linear regulator is used and the headphone relay is 24V. This change significantly cut down on the transient voltage that many people see at cut-off. (The time delay prevents headphones from seeing any transients on power on, but power off has always been a bit of an issue with headphone delay-relays.)
Edited by tomb - 2/10/12 at 8:30am