I was just thinking about a grid supply , batteries say a 9 volt would require maintenance,wouldnt last very long and involve some fairly ugly switching hash to generate the 50 or more volts required .
What if you picked up 6.3volts ac from one of the heater sockets on the amp ,then back in your test jig box install a small low volt transformer in reverse to step up the volts before the usual recification,smoothing and dc voltage adjust to grid neg voltages
Certainly would make it a breeze install , no external mains or battery power source needed ,totally powered from the amp your testing , and you know your bias voltage is always present as soon as the Lt on your amp is energised .
A small 6 volt transformer using a single 110v tap primary might be in the right ball park for the job allowing for some losses ,only requires small current anyway , a balance and current adjust control like in the reference below looks like a nice addition ,only requires an extra pot ,now we can effectively zero out imbalance between push pull pairs , at one point on the scale at least , a null test by driving both grids in phase could also be interesting for test purposes .
https://www.vtadiy.com/book/chapter-5-power-supply-unit/5-3-power-supply-for-the-fixed-grid-bias/
I have a bunch of NOS russian ceramic multi wafer rotaries lying around for decades , should be just about right for the job.
I think 1 or 10 ohm resistor are usual as cathode current shunt ,maybe increaseing the wattage and heatsinking of these resistors could be a good plan ,then there should be no need to worry if the amp is driven while on test or not , I see the Weber bias rite uses copper tubing , I might try a small metal clad resistor heat sinked to a lenght copper pipe to make the adapters .
Makes sense to provide a sheilded cable for the driver stage output and grid inputs from and to the amp socket adapters
When output stage test signals arent required i could have it so driver output is routed back to power tube grid via shorting jacks, only bias control remains in the test jig .
Wow that all seems to good to be true , everything I need in a simple passive switch box, external metering and signal source means it fits in just fine with any bench setup ,no micro processors to go wrong , only electronics involved is the transformer bridge and bias circuit which can be passively smoothed .
Below find added a general idea of how it might look ,
Im a bit unsure about all the interactions in the cables ,or how best to minimise them , cathode current is of course measured as a voltage, 1mv corresponding to 1ma with one ohm resistor , would stopper resistors do as a current shunts to measure grid and screen current also ?
If everything could be read in voltage mode on an auto multimeter it would make it quick and easy to opperate ,the meter would autorange to the most appropriate range for the voltage drop/currents your measuring and give a meaningfull reading in mV for mA on screen or cathode current ranges. What would be the most appropriate resistor to set the scale for grid current measurement on the multimeter?