Yes they could. If they are going to be in the signal path anyway, why not use them for the buffer and save yet another circuit! They may not sound a sgood as some buffers but then buffers won't divide up the signal either.
You could try improving the X-overs with different opamps or whatever.
I'm not familiar with that one but from the NS site I found:
See Also:
LMC662 - improved bandwidth, slew rate and minimum supply voltage.
LMC6032 - improved supply current per channel and input bias current.
LMC6082 - improved bandwidth, slew rate, input bias current, offset voltage.
Perhaps somebody else can recommend an alternative.
See Also:
LMC662 - improved bandwidth, slew rate and minimum supply voltage.
LMC6032 - improved supply current per channel and input bias current.
LMC6082 - improved bandwidth, slew rate, input bias current, offset voltage.
Perhaps somebody else can recommend an alternative.
Generally it sounds like a mute circuit that normally can be found, say, at the output of CD players but with JFET instead of more common BJTs and controlled by the RC network. I think it can be made to work but such mute circuits are not sonically invisible. No idea if JFETs based ones could be. Switch is the cleaner solution. Though you must do the effort of switching…Sheldon said:
What does supply your buffers, i.e. what does drive their regulators? Are they powered from the same transformer used for the power chips?
Pedja
Pedja said:
Originally, and because all this transistor stuff is new to me, I was considering some sort of timed relay. That seems to require more parts, though if I did the effort to search, I guess that pre-made circuits are probably available. The transistor thing seems simple and would take little space. Sonics are the issue. I assumed (there's that word again) that since transistors are used in these circuits that they might possibly be used for this kind of function without hurting sonics. I thought that perhaps the jfet might give a steep enough cut-off to be the sonic equivalent of a mechanical switch. Might have to try, but I don't think I have very golden ears or maybe enough patience to do a useful study. But who knows? Does anyone know how it's implemented in the 3886?
Sheldon
As far as I can see the voltage reference of one of its current sources is made to be supplied externally.Sheldon said:
If I did not misunderstood you, your buffers have their own (additional) regulators hooked to the +/-24V lines(?). Connect the grounds of these regs and caps following them by some decently sized conductor to the main ground point. Keep the overall layout tight and do not make another ground connections, that is IMO about everything you can do about the buffer and grounding in this case.Sheldon said:
If you use an additional shunt resistor at the buffer’s input, it belongs to the signal ground.
Filter in fact rather belongs to the power chip and the ground side of the filter shunt caps goes the chip’s signal ground.
Pedja
Pedja said:
Sincere thanks Pedja,
I hope that great evening weather is keeping you up so late and that you find some time to sleep. You'll spoil us (me certainly) with your dedication.
Yes, the set up is as you describe. So I will separate the filter and signal grounds from the buffer reg grounds, and run the filter grounds to the signal ground on the power amp. The shunt resistor is already on the power amp board and does go to the signal ground on that board.
Almost there,
Sheldon
Uhmm, Huh?
Sheldon (& Co.):
The posts on this thread are so far over my head it isn't funny. It sounds like you're pretty far along in refining buffers for use in LM3875 Gainclones. I'm planning on buying a bunch of PCBs from Brian GT within the week. It seems that your PCBs are designed to work with Brian's (and not in substitution for his boards). When do you think your project will be available to the group?
Sorry to be so slow about this stuff.
Regards,
Scott
Sheldon (& Co.):
The posts on this thread are so far over my head it isn't funny. It sounds like you're pretty far along in refining buffers for use in LM3875 Gainclones. I'm planning on buying a bunch of PCBs from Brian GT within the week. It seems that your PCBs are designed to work with Brian's (and not in substitution for his boards). When do you think your project will be available to the group?
Sorry to be so slow about this stuff.
Regards,
Scott
Re: Uhmm, Huh?
My name doesn't belong in the front of your post message. I'm not as much over your head as you think (as in just barely). The real work here in design has been done by Pejda, Carlos, and many others. I'm just taking what the've done and building one example. I'm using Brian's board and plugging Pejda's power supply and jfet buffer into it. You can see from my questions, that I don't have anywhere near the level of expertise of many contributors to this board. I'm operating on what I've learned here, reading a basic text on electronics (("The Art of Electronics" by Horowitz and Hill), plus some common sense logic about how things work (which can get one in big trouble, by the way).
I don't have the capability to really do a group design and sourcing, such as Brian has done. Actually, I was sort of hoping we could sucker Brian into doing it if enough interest could be generated. Even though a group buy of boards hasn't come out of the thread, a lot of useful information has resulted, including some board designs which could be made by individual users. Brian started his project to motivate as many people as possible to take the plunge and build an amp. I checked out the DIY site originally intending to build speakers (one done, one to start on) and here I am building a one of a kind (at least for a short time) integrated amp. The last thing like that I did was a Dynakit back in the early '70's. So Brian's subversive intentions were realized.
Sheldon
SRMcGee said:
My name doesn't belong in the front of your post message. I'm not as much over your head as you think (as in just barely). The real work here in design has been done by Pejda, Carlos, and many others. I'm just taking what the've done and building one example. I'm using Brian's board and plugging Pejda's power supply and jfet buffer into it. You can see from my questions, that I don't have anywhere near the level of expertise of many contributors to this board. I'm operating on what I've learned here, reading a basic text on electronics (("The Art of Electronics" by Horowitz and Hill), plus some common sense logic about how things work (which can get one in big trouble, by the way).
I don't have the capability to really do a group design and sourcing, such as Brian has done. Actually, I was sort of hoping we could sucker Brian into doing it if enough interest could be generated. Even though a group buy of boards hasn't come out of the thread, a lot of useful information has resulted, including some board designs which could be made by individual users. Brian started his project to motivate as many people as possible to take the plunge and build an amp. I checked out the DIY site originally intending to build speakers (one done, one to start on) and here I am building a one of a kind (at least for a short time) integrated amp. The last thing like that I did was a Dynakit back in the early '70's. So Brian's subversive intentions were realized.
Sheldon
just a note, since I know it's been a bit since I have posted any updated layouts... I'm still working on these. I also finally got a job for a bit again, so I don't have all day every day to tweak and post. But I will soon have funds to start trying them out, so it's not at all a bad thing.
One question:
If I'm able to do a double sided board myself (probably would be toner transfer method no less), and documented it, would there be a comfort level with that? Because there are a few little things that I think double sided would help clean up that little bit extra. No vias would be required except perhaps at a through component (if even then), just solder top or bottom.
C
I also finally got a job for a bit again, so I don't have all day every day to tweak and post. But I will soon have funds to start trying them out, so it's not at all a bad thing.One question:
If I'm able to do a double sided board myself (probably would be toner transfer method no less), and documented it, would there be a comfort level with that? Because there are a few little things that I think double sided would help clean up that little bit extra. No vias would be required except perhaps at a through component (if even then), just solder top or bottom.
C
It works!
Well, after a few minor setbacks it's working and sounds pretty nice at first go. Haven't had time yet to listen critically, but no abvious nasty's. To recap, I'm using Brian Gt's board and resistors (premium pack), but with 47uf Cerafine caps on the board, Pedja's ultimate discreet regulated ps, and Pedja's jfet buffers. This is all stuffed (and I mean that - I'll try to post some pics later today) into one of Peter Daniels fancy chassis.
One might ask why I chose this combo. Just in case you do, here's the story. I bought Brians kit mostly on a lark. I'm in this primarily to learn, not to save money. So I figured I might as well complicate my life, and I liked the idea of building a couple of discreet circuits (have never done that before) and getting the whole thing to work. Plus, a fair number of builders have fournd the regulated and/or buffered "clones" preferable. I'm not taking sides in that debate, and I probably won't test one against the other. I'm getting out of it what I want.
As for the start-up problems. The first I had was a loose solder joint on the neg. ps j309, so no output from that side. Can you say DC offset boys and girls? I had used smd j309's to which I attached fine wire, then mounted. Those are tiny. Fiddling around, I accidently detached one of the wires. Once fixed, I discovered that I had wired the source selector switch wrong. Fixed that, and it works. By the way, I wired the selector switch so that one position grounds the buffer output, in order to eliminate the turn-on thump. That works.
I still have a little tuning to do. I'm getting about 90mv offset in one channel and some hum, about 75mv in the other but very low hum. Maybe a bad joint or grounding arrangement to chase down. Time for a break before I tackle those.
Thanks to all the DIY members who inspired me to give this a shot. It's been slow going, because I'm a complete novice, but it's been fun.
Sheldon
Well, after a few minor setbacks it's working and sounds pretty nice at first go. Haven't had time yet to listen critically, but no abvious nasty's. To recap, I'm using Brian Gt's board and resistors (premium pack), but with 47uf Cerafine caps on the board, Pedja's ultimate discreet regulated ps, and Pedja's jfet buffers. This is all stuffed (and I mean that - I'll try to post some pics later today) into one of Peter Daniels fancy chassis.
One might ask why I chose this combo. Just in case you do, here's the story. I bought Brians kit mostly on a lark. I'm in this primarily to learn, not to save money. So I figured I might as well complicate my life, and I liked the idea of building a couple of discreet circuits (have never done that before) and getting the whole thing to work. Plus, a fair number of builders have fournd the regulated and/or buffered "clones" preferable. I'm not taking sides in that debate, and I probably won't test one against the other. I'm getting out of it what I want.
As for the start-up problems. The first I had was a loose solder joint on the neg. ps j309, so no output from that side. Can you say DC offset boys and girls? I had used smd j309's to which I attached fine wire, then mounted. Those are tiny. Fiddling around, I accidently detached one of the wires. Once fixed, I discovered that I had wired the source selector switch wrong. Fixed that, and it works. By the way, I wired the selector switch so that one position grounds the buffer output, in order to eliminate the turn-on thump. That works.
I still have a little tuning to do. I'm getting about 90mv offset in one channel and some hum, about 75mv in the other but very low hum. Maybe a bad joint or grounding arrangement to chase down. Time for a break before I tackle those.
Thanks to all the DIY members who inspired me to give this a shot. It's been slow going, because I'm a complete novice, but it's been fun.
Sheldon
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.