30-11-2016, 04:44 PM
Yes they are, but I believe I have solved it. The problem was the valves, and not me! 
I have been well-aware for the past 55 years or so that G3 should be connected to the cathode. For that reason, I incorporated in my valve-holder test panel, a row of pin-linking terminals. When I tested the first few EF39s (three or four of them), I had the cathode/G3 link in. The valves (except one) all showed rather low in both gm and emission. I noticed that when I removed the link, the readings became more what I expected. Today, I tried with some hardly-ever-used EF36 and EF37 valves that have the same bases as EF39s and similar characteristics. These all gave excellent and accurate readings on both gm and emission. When I removed the link, the readings went haywire. I tried again with the EF39s and found that when I took the links off, the readings changed when the link was removed, so conluded that there was something wrong with the first few valves that I tested! I dug out some more EF39s that gave a good selection of results from a few poor, some average and several excellent conditions. When I removed the links on the excellent ones, the readings did go down a bit, but not much. I have quite store of boxed valves that have never even been unpacked yet, and it will be interesting to see how they show up on it. I suppose I shouldn't "experiment" by messing about with known facts, but experiments on things that shouldn't have worked have seen me out of a few tight spots at sea when no spares were available, and shore servicing had given up because the correct valve or component was not available!
Bob
I have been well-aware for the past 55 years or so that G3 should be connected to the cathode. For that reason, I incorporated in my valve-holder test panel, a row of pin-linking terminals. When I tested the first few EF39s (three or four of them), I had the cathode/G3 link in. The valves (except one) all showed rather low in both gm and emission. I noticed that when I removed the link, the readings became more what I expected. Today, I tried with some hardly-ever-used EF36 and EF37 valves that have the same bases as EF39s and similar characteristics. These all gave excellent and accurate readings on both gm and emission. When I removed the link, the readings went haywire. I tried again with the EF39s and found that when I took the links off, the readings changed when the link was removed, so conluded that there was something wrong with the first few valves that I tested! I dug out some more EF39s that gave a good selection of results from a few poor, some average and several excellent conditions. When I removed the links on the excellent ones, the readings did go down a bit, but not much. I have quite store of boxed valves that have never even been unpacked yet, and it will be interesting to see how they show up on it. I suppose I shouldn't "experiment" by messing about with known facts, but experiments on things that shouldn't have worked have seen me out of a few tight spots at sea when no spares were available, and shore servicing had given up because the correct valve or component was not available!Bob
