![]() ![]() When the cathode resistors are 1K8 then 268K is closer to perfect value.Įdit 2: Initially the floating paraphase inverter was adjusted to equal output voltages. + /- 7% variation is acceptable according to the Radio Designers Handbook 4th edition. Later the capacitors were removed (I found some pictures of the original populated AN PCB's where they space for the capacitors was filled and later another picture showed up where the space was empty.) However the resistor values were not adjusted for the lower amplification (about 14* instead of approx 40* when bypassed).If the 12AX7 has an amplification factor of 100 and the cathode resistor is 2K2 then 273K total is spot on. That value was based on the cathode being bypassed. If really worried then the 7025 can be used but I am happy enough with the Sovtek after I changed the cathode resistors to 1K8.Įdit: Originally 270K + 22K was specified in the floating paraphase inverter. the Sovtek 12AX7 LPS) or some of the more exotic $$ (e.g. (Omron has good Octal and Belton good Noval - I put in NOS Cinch).īecause the 12AX7 cathodes are not bypassed with a capacitor hum can be a problem, best to use a valve with a spiral filament (e.g. If I had realised that it was not that easy to find high quality PCB tube sockets then I would have build it point to point instead. I am still in two minds if I will implement an electronic filter (choke) for the drivers / 6V6 screens (aka capacitor multiplier). The inverter can be adjusted and if the feedback is temporary disconnected then the adjustment can include the 6SN7 and the finals (rather than just adjusting a single stage). Many people are critical of the design but I prefer moderate amplification for each stage (so each stage is working at the optimal operating point) and the floating paraphase inverter. Cheerio!Ĭlick to expand.You already discovered that I posted this earlier. That's it from me folks, I'm back to enoying my music and spending less time on the internet. Around the 6V6 finals the temperature gets up to around 50C between chassis and PCB.Īlso a picture of how to re-route the screen connections (100R 3/4W SMD close to tube socket, cut track). Temperature in the 12AX7 region is in low 40C and the trimpots are rated for use up to 70C so they are fine and within operating tolerance (room temperature approx 25C). I managed to make the floating paraphase inverter adjustable by mounting two 50K trimpots in series with 250K metal film resistors. Feedback increased slightly by approx 0.8 dB to 9.9dB 1kHz and hum disappeared into the noise level and all is well, I am now very happy. I changed the cathode resistors to 1K8 and the feedback resistor to 120R. The 2K2 is correct for 250V and 1K5 for 300V. Checking the 12AX7 bias and the B+ voltages that are applied to the 12AX7 (approx 280-285V at idle, around 270-275V at 9.5 W RMS) suggested that the bias was not optimum according to the MAZDA datasheet for the 12AX7. Studying the circuit diagram the cathode is not bypassed with a capacitor which increases the risk of hum introduced. I went round and round in circles trying this, that and the other. Getting back to my Audio Note clone and the PCB that I bought on ePray from Jim's audio: I did not use DC to the filaments of the 12AX7 but was suffering from a slight hum in one channel. Electric plate is safer, you can even use one of those immersible heaters that can be used to heat water. You are lucky it did not catch fire, scary prospect. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |