I was trying for generality in the PG article. If your rectifier tube socket will never have anything else plugged into it so the unused pins on the socket will always be unused, sure! Use the unused pins as tie points to make the additional diodes more mechanically robust. Series diodes just hanging in the air are prone to lead breakage, but it's guaranteed that you can make this work with any wire-leaded transformer. But supporting both ends of the diode(s) is going to be better.1992guitars wrote: ↑Thu Jun 11, 2026 9:54 pm 1. About the diode mounting.....In the Premier Guitar article it was written, "The simplest way to wire in the diodes is to unsolder the wire from the rectifier tube socket, then solder the cathode (bar) end of the diode to the rectifier tube. Bend a loop in the anode (opposite) end of the diode, then slip a length of heat shrink tubing over each transformer lead, soldering each transformer lead to the loop in one diode. When the solder cools, slip the heat shrink down over the diode and heat it. Don’t leave the wire leads uncovered – this is important for safety!" Looking over how the protection diodes were installed the other amp referred to above, I see they used the unused pins to make the connections. The cathode (bar end) went to pins 4 and 6 respectively, and then the anode went to the unused pins 3 & 7 (respectively) where the PT wires were moved to. Since the PG didn't mention that as a method, is there any reason to not use that mounting strategy, or is it fine to do?
"Fine" varies.2. If I recall correctly (and will definitely double check) the B+ in the amp is under 450vDC with the 5U4G last time I measured it (easily a couple years ago). I never measured the HV before the rectifier (and am aware the big bottle is dropping some voltage). Because you'd suggested two in-series (out of respect for max peak inverse voltage) when I asked about adding them to a Super Reverb given the 360vAC and higher than 450vDC B+ on the plates, is it advisable to measure the high voltage AC from the center tap *on the Louis Electric amp I want to add protection to* in order to make sure I get it right, or as long as the rectified B+ is still under 450vDC is it just fine to stick with a single (not in series) 1N5408 in series with each anode of the 5U4G?
The strict truth is that the conducting diode clamps one end of the transformer to the filter cap voltage, and the off-side transformer pulls the opposite diode's anode to an equal amount below ground. So the backup diodes will definitely see twice the B+ voltage as a reverse voltage. Whether a 1N4007 or 1N5408 will survive without dying depends on whether high line and/or transients cause the reverse voltage to exceed the > actual < breakover voltage of the diode. That's not an easy thing to guess. The diode datasheets amount to the manufacturer's guarantee that any diode they label as that type number will definitely withstand 1000.0V for these two parts. It is a truism of manufacturing that to make that claim, their diodes have to average higher than withstand voltage, or the weaker ones would die early. Any one 1N4007 might withstand maybe 1100, maybe 1200V. But you can't tell which ones will without testing every single one.
Given that 1N4007 are very cheap, and 1N5408s are not terribly expensive, it is cheap for the individual builder/modifier to just use two of them in series and not take chances. I would recommend just putting in two in series in each transformer lead and not trying to squint very fine voltage differences. I'd advise just using two in each lead and not bothering to measure the B+, as it is very cheap insurance.
