Oh, sorry. Just a pc board if you'll be making them available.
It looks like a great project.
Neither of us will know that until I get the project finished (or at least breadboarded) and actually listen to it.
At any rate, I don't offer boards or kits for anything that I show here at DIYA, as I barely have enough time to realize the ideas in question. I then move on to the next one. There's a pile of projects in the basement (lots of them vacuum state) begging to be finished. If you are interested enough, a lot of the things I present can be readily thrown together on Vector/Vero/perf board. It may not be as pretty, but it works just as well.
At any rate, I don't offer boards or kits for anything that I show here at DIYA, as I barely have enough time to realize the ideas in question. I then move on to the next one. There's a pile of projects in the basement (lots of them vacuum state) begging to be finished. If you are interested enough, a lot of the things I present can be readily thrown together on Vector/Vero/perf board. It may not be as pretty, but it works just as well.
I'm going to dump this project on a relatively small heat sink and use an auxiliary fan for cooling. I picked up some 12V fans from Electronic Goldmine that draw less than 100ma full tilt, and are dead quiet. Just a whisper of air is enough to get things done if the heat sink isn't too small. Next I have to find the Transcendar 3k output transformers I had in mind for this project.
The project is marching along - the boards are mounted to heat sinks with all I/O and power leads/jacks installed. Output transformers will be spliced in using wire nuts so that I can yank them out if the sound displeases me. Final feature will be a shroud around the heat sink to allow cooling via a teensy 48V DC fan powered from the 30V supply. I may give in to expediency and rig something out of cardboard rather than bending up sheet metal.
Here's something to think about it in the mean time. I'm doing the HV/transformer coupled routine to get the best bang out of a limited number of jfets and a relatively low total bias current (100ma). I pooh-poohed doing a lower voltage version due to the huge number of jfets involved. However, I do have some J107s that are lounging around looking for gainful employment. The have a minimum 100mA IDSS, so one jfet could easily support 30ma bias current if you clamped the VDS to 4V with a cascode. So, instead of 100 fets, you would only need 33 for a reasonable bias current target of 1A. The fets also have a hefty gate capacitance of ~140-150 pF apiece, so the assemblage would really need a follower for drive. I don't know if this approach would be good or "SLS". I'm not even thinking of doing something like this until I get a bunch of other stuff off my plate - maybe not even then. However, the more ambitious among you might want to give it a try.
Note -the other reason I mentioed the J107 in my previous post is that that are still around cheaply as NOS with a little searching. There are other chopper/switch fets with similar ratings.
Tonight I took the hot-doggers approach, carting the amp down to the basement and powering it up a side at a time, using an electrophoresis supply for the HV and a Topward bench supply for the LV and a small DC fan. I used a Discman for a signal source.
No smoke was emitted, bias was stable (~100ma/side), and sounds were made. Next up is carting the mess in to work for characterization and engaging the global feedback. Right now, the gain with just the partial feedback is pretty high, so that adjusting the volume is touchy. I'll probably need to adjust the ratio between partial and global feedback.
No smoke was emitted, bias was stable (~100ma/side), and sounds were made. Next up is carting the mess in to work for characterization and engaging the global feedback. Right now, the gain with just the partial feedback is pretty high, so that adjusting the volume is touchy. I'll probably need to adjust the ratio between partial and global feedback.
Well I tried hooking up the global feedback tonight, and the Beast didn't like it either way - can you say motorboat? Also, though the thing does play music open loop with a CD player, the sound is not as clean as I'd like, and when I switched in another player, it started picking up German-style oompah music from some radio station. The amp seriously needs some bench time with a scope, and the gate resistors on the paralleled fets probably need to be a lot bigger.
I had this problem before on the Shrine amp with the cascoded fet input stage wanting to take off - the situation is no doubt worsened by all those fets in parallel. The fets are probably yakking away amongst themselves at a few MHz - maybe more than a few....
I had this problem before on the Shrine amp with the cascoded fet input stage wanting to take off - the situation is no doubt worsened by all those fets in parallel. The fets are probably yakking away amongst themselves at a few MHz - maybe more than a few....
- Status
- Not open for further replies.