------------------------------ Date: Fri, 12 Feb 93 09:45:53 EST From: Johannes Chiu Subject: Reflections on Joe Curcio's ST-70 Mod (Glass Audio) PART I PART 1 0. Prelude ========== I finished modifying my Dynaco ST-70 using Joe Curcio's Mod from Glass Audio about two months ago. Since then, I have been tweaking and fine tuning it, and I think that I have come to the point where there is something worth writing about. I believe these articles will be helpful to any person already owning or thinking about doing the mod. Hopefully it will also contain something new for the more sophisticated tube-philes. I bought only the board from Old Colony and tried to get as high quality parts I could get for the money I had. When I finished, everything was done according to the article except that I used higher power supply caps for the plate regulators. The input tubes are Yugoslavian EI 6DJ8s and the output tubes matched GE 6CA7s all bought from New Sensor Corp. While I was doing the mod, I was listening to another "almost" stock ST-70 I had (except for rewiring the input socket to accept 6AN8s and rewiring the EL-34s for triode output operation) driving my Rogers LS3/5s. 1. Initial Comparison ===================== Well, was it worth it? Yes, the bass was way better, the dynamics were better, and it was a lot more quiet. BUT, the soundstage width and depth, the imaging, the beautiful midrange, and the air was gone. I this respect, I was very disappointed. The highs had some ring, but it wasn't all that bad. I guess it's because the EI 6DJ8s don't ring that much. In other words,the sound became somewhat unreal for me. Especially when one listens to a live recording with hand clappings, the hand clapping just seems to sound muffled in. I also seriously doubted that my new amp had better highs, because compared to my old amp, it didn't sound more extended at all. So I sat down for a long while to figure out what to do next. 2. Triode Ectasy ================ Since I have lived with my old amp connected in triode mode for a long time, the first most obvious thing was to connect my new amp this way, too. This is easily done by removing the screen taps fron the screen regulator (S-Reg) and reconnect them to the plate. Just to be shure, at that time I also added a 100pF cap across the feedback resistor, to make it more stable, as someone here already mentioned before. Needless to say, this is recommended to anyone who can live with the reduced power output. Of course, the improvement was signifficant as many other people have documented and described, so I won't go into boring you. I reckoned that if the output tubes are more linear now, I could reduce the global feedback a bit, which I did, by about 3dBs. The sound was a tiny bit more relaxed, but nothing spectacular. At this point, I couldn't figure what else I could do. I felt that the amp still had a lot more potential, but there was nothing I could do without changing things signifficantly. This brought me to a reader response in Glass Audio 3/91. He varied the cathode resistor in his EL-34 output stage and listened to it. He came to the conclusion that the tubes sounded best with a cathode resistor of 14.1 ohms. The El-34s were pentode connected, so I was wondering if it also applied to triode connected EL-34s. Which brings me to my first question to the experienced tube-phile out there: Would the same optimum cathode resistor apply? My guess is that it wouldn't, since the transconductance and plate resistance are both different now. But then, in theory, should the cathode resistor be higher or lower now for the triode connected amp? Well, things stayed like this for a couple of weeks, and was getting more and more annoyed with the new amp. However, I would not give up. So I went to the university library to see if I could find anything helpful in understanding the ciruitry better. What puzzled me the most is that a lot of people were using the cascoded differential amp in their design, and it seemed to be the optimum circuit, but yet the sound I got really left a lot to be desired. Two days ago, I finally tried something that made sense to me, and it worked! It really solved a lot of the pitfalls, but you will have to wait for it next time, because it takes a lot longer to explain and it is not obvious at all why one should do so. I also need to listen to it a bit longer, to confirm my observations. So, in part two, I will tell you about my battle with the cascoded differential front end, and it's really going to be a lot more interesting if not flaming. References: =========== I think since a lot of people don't know the magazines like Audio Amateur and Glass Audio, I believe anyone mentioning the magazine should include an address so that people don't have to ask for them again. So here it it: Glass Audio, published by Ed Dell Circulation Dept. P.O. Box 176 Peterborough, NH 03458 Tel: (603) 924-9464 Fax: (603) 924-9467 ------------------------------------------------------------------------- Johannes S. Chiu ta-jc155@cunixa.cc.columbia.edu "Once a tube lover, always a tube lover. Once a triode lover, always a triode lover." --------------------------------------------------------------------------