Having completed and further modified a kit amp, I next wanted to do something from scratch. And by scratch, I mean having a go at picking a topology, drawing up a schematic, choosing the components, and then hopefully cobbling together a working example. Obviously, nothing I've done here is original and is pretty much derived from picking bits and pieces from other circuits, but the final outcome is uniquely mine.
As far as picking a first project with reasonable chances for success, a transformer-coupled SE preamp seems a straightforward choice. And, conceptually, parafeed appears like a good idea so I went that direction. There are any number of examples out there: several appear in back issues of VALVE (Bill Petrowski's in vol.4 #5, the infamous 'Soul Sister' in vol.5 #4, Jim Dowdy's in vol.5 #5, and Rick Henthorn's 'Lefty' -- check out the Bottlehead forum on how to get back issues of VALVE) and also Jeff Lessard's Euridice. For a tube, I settled on the 27, an antique IDHT, which despite a reputation for microphonics (which I can now attest to) is purported to be a great tube for such a set-up. As far as an operating point for the 27, I went straight off the RCA datasheet: 135V Vpk, -9V Vgk, and 4.5mA Ip. One nice thing about this operating points versus either of the other 'book' values is that once you add the voltage drop from the plate choke, B+ is ~150V and thus ideal for adding a 0D3 gaseous regulator. For rectification, I went on Gordon Rankin's recommendation and slotted in the Bendix 6754 for those duties.
As far as iron goes, Magnequest makes a great product and is very local so I try and send as much of my business Mike's way as possible. I settled on the MQ R080 preamp power transformer very early on and ended up with a simple LC power supply using the special-offer MQ Abbie 100H 25mA choke. Because of the 27's 9k Rp, I went with the MQ 15k:500 B7 line stage transformer and the MQ EXO-010 276H 10mA choke for the plate load.
With the major building blocks of the circuit done, I went about filling in the details. I chose 100μF for the power supply filter cap because looking at the waveform on PSUD it didn't look to be either over or underdamped in the context of the overall PS design though rise time is a bit long (here is a discussion on the Q of the power supply). For the cathode bias resistor, 9V at 4.5mA works out to Rk = 2000Ω. For the cathode bypass cap, Paul Joppa of Bottlehead fame has a RofT: Ck = gm/30, which comes out to be ~33μF for the 27. I got some help from Paul again with figuring out the parafeed/ DC blocking cap, and plugged in 2.7μF. The VR dropping resistor has to drop ~85V at ~18mA, so 4k7Ω works. The 0D3 bypass cap is 0.1μF but it could easily be lower or replaced by a RC filter as suggested here. Lastly, because of some noise with the 27's, I added a voltage divider and biased the filaments up to around 22V.
Construction actually went quickly and with little if any drama. At the test bench the preamps fired right up sans smoke and voltages were almost exactly correct. Moved them into my system, and voilà music poured forth. I don't have much of a frame of reference, but I think they sound great. Whereas I spent a lot of time modding my 11BM8 amp, for these I don't have much planned. I might switch to a DHT like a 112A, but that's actually a relatively minor change. The VR dropping resistor will probably go, replaced by a C4S board from Bottlehead. And, the 100kΩ potentiometer will be swapped out for a MQ 10k:10k input transformer and 10k DACT attenuator. Outside of that, I'd like to build a nice enclosure for them and that would about wrap the project up. There's a couple threads about this preamp on the Magnequest forum here and here.
