While I don't smoke a pipe I do enjoy an occassional cigar and drink about 4-6 cups of coffee a day. I spent today listening to Mozart in my garage dividing my time between projects and the computer. I am putting together some misc power supplies. One for Aleph X amp testing and and one for mosfet matching. I am really getting stoked about the X V.1 boards. I have been going back through the threads and keep running into pictures of Peter's projects. 
Those are enough to motivate anyone even remotely interested in building an amp.
Those are enough to motivate anyone even remotely interested in building an amp.Mark, I wish I could have access to that kind of machinery. I have a pretty complete woodworking shop but I am a little limited in metalworking equipment. Unfortunately, the Navy frowns on using their equipment for personal projects. You are in a lucky situation. I have been thinking about buying a 3in1 Smithy and a combo 30" sliproll/brake/bender but some thing big would have to go. My wife would love me to sell my Harley to make room for more equipment. I'd stay home more and wouldn't make so much noise at 5:15 in the morning when I rode it in to work. 
Looking at Carr's designs it's easy to see who impresses you in construction practices Peter. Machining the frame out of a 5" thick billet of aluminum. That's really going the extra mile. Bet you wouldn't need to spend much time on vibration isolation with that preamp.Peter Daniel said:
Phil
Not at all. I did check into what it would cost to buy all the aluminum out right and once I did the math the total seemed a little high. Unfortunately, I would have to buy the tubing in 10' or 20' lengths which would give me a lot of left over if the sinks weren't efficient enough. I have been hitting all the local metal supply shops and buying their short pieces at $1 to $1.50/lb. Once I get a couple more sizes I might only have to invest in 1 size at full length. Probably 2.5" because it is an unpopular size. A 10' length of that costs about $32. When I went to the first place I picked up some other pieces to finish another project so I have been involved in that. Plus there is that holiday thing that my wife thinks I should come out of the garage for an hour or so... 
My goal is to come up with a few designs the average DIY'er can do, yours truely included. I have been checking into home anodizing for which I would have to build a heavy duty 35-50 amp DC supply. I might just bite the bullet and buy a used manual truck battery charger, though.
My goal is to come up with a few designs the average DIY'er can do, yours truely included. I have been checking into home anodizing for which I would have to build a heavy duty 35-50 amp DC supply. I might just bite the bullet and buy a used manual truck battery charger, though.
On the side note of DIY heatsinks, that's what I use for my Class-A amps. I'll start a thread describing them and with pictures (my dad got a brand new digital camera to play with for christmas) if anyone's interested. It's a very simple design that should be relatively cheap (I got the metal for free).
jh6you said:
I agree.
I have two p2p amps from AudioNote (kits) and they are P2P.
The only reason that people use PCB's is to make possible a more complex circuit.
I saw 15 days ago a 50(?) years old valve amp that worked fine with a layer of dust(and bugs too!!
) of 3cm!!.I don't think a pcb design could withstand such treatment.
PCB's components solder crack very often too, so I really don't think that a proffesional product is better. no way!
I think also that a self made amp is better, at least to the creator, like a child. And that's what matters.
In my amps (good or bad they may be) I used stuff like heatsinks, caps, trafos, wire, chassis that are inimaginable in any amp with less then 10x the cost.
DIYers don't think in comercial budjets. Nor timmings.
WE just think in OUR pleasure, in building and listening.
Merry Xmas to you all!
DIY heatsink thread
JoeBob,
I am interested in seeing your heatsinks. I wont be able to get my sinks together for a couple of weeks because of holiday plans but will post some pics of them on your thread then. I made some drawings up and put them on the Aleph X heatsink thread. Those will give you an idea of what I am doing.
JoeBob,
I am interested in seeing your heatsinks. I wont be able to get my sinks together for a couple of weeks because of holiday plans but will post some pics of them on your thread then. I made some drawings up and put them on the Aleph X heatsink thread. Those will give you an idea of what I am doing.
I completely agree Pedro!! A DIY'er can build anything as good as he or she wants it to be. Many DIY'er Alephs will probably sound a tad different than the run of the mill(you know what I mean Nelson) production Aleph. Some may sound better, and some worse. Some are using a power transformer that is much larger than is called for and some barely squeek by.
In my amp for instance I had the option to use 16ga solid silver wire but instead opted for 16ga stranded ofc silver plated teflon. Which sounds better? Who knows......The teflon insulated stuff will NEVER dry out in the heat of these amplifiers though. I started with 54000 mfd per rail but ended up going to 112000 because there was a tad too much residual noise in the first finished amp. Going to 112000 at least has made it inaudible for the most part....even with your ears right up to the drivers in my speakers. Anyway, there are many routes to go building your own gear, and thats the reason we enjoy doing it ourselves.
Merry Christmas and Happy Holidays to all!!!!!!
Mark
In my amp for instance I had the option to use 16ga solid silver wire but instead opted for 16ga stranded ofc silver plated teflon. Which sounds better? Who knows......The teflon insulated stuff will NEVER dry out in the heat of these amplifiers though. I started with 54000 mfd per rail but ended up going to 112000 because there was a tad too much residual noise in the first finished amp. Going to 112000 at least has made it inaudible for the most part....even with your ears right up to the drivers in my speakers. Anyway, there are many routes to go building your own gear, and thats the reason we enjoy doing it ourselves.
Merry Christmas and Happy Holidays to all!!!!!!
Mark
Per Anders:
I use both machine (pick&place, wave solder etc.) and hand assembly techniques. Both are good at different things. The machine is fast and consistent. But it is neither versatile nor flexible, and places severe limitations on how you can design the physical construction of your circuits.
Hand work is slower, and sometimes the wrong part may be installed, or with reversed polarity. With good solder, good soldering equipment and a skilled assembler, the quality of solder joints is not an issue. The major advantage of hand work, however, is that you can do things that a machine simply can't.
For surface mount, I use pick&place and cream solder in an infrared oven. For simple, regular stuff, I use wave soldering. But for all difficult work (custom footprints, special lead bending patterns, point-to-point assembly, layering components on top of each other...), I use hand work. These are all things that I would have to give up in my products if I were to limit myself to machine assembly.
For example, here is an amplifier module from one of my 4.0 preamps. This is the product that Michael Fremer reviewed in Stereophile, and Roy Gregory reviewed in HiFi+. (I won't post any photos of the 3.0 because you can see it on our website). The top board holds the amplifier circuits and the bottom board houses the voltage regulators.
The two boards were designed as a single set and plug into each other. The "pins" are the component leads on both boards, and in some cases the components themselves (vertically oriented resistors, diodes, etc). The requisite leads are kept long and uncut during component installation and soldering, and are plugged into matching receiver pads on the opposing board during assembly.
You cannot do this with machine assembly techniques.
I use both machine (pick&place, wave solder etc.) and hand assembly techniques. Both are good at different things. The machine is fast and consistent. But it is neither versatile nor flexible, and places severe limitations on how you can design the physical construction of your circuits.
Hand work is slower, and sometimes the wrong part may be installed, or with reversed polarity. With good solder, good soldering equipment and a skilled assembler, the quality of solder joints is not an issue. The major advantage of hand work, however, is that you can do things that a machine simply can't.
For surface mount, I use pick&place and cream solder in an infrared oven. For simple, regular stuff, I use wave soldering. But for all difficult work (custom footprints, special lead bending patterns, point-to-point assembly, layering components on top of each other...), I use hand work. These are all things that I would have to give up in my products if I were to limit myself to machine assembly.
For example, here is an amplifier module from one of my 4.0 preamps. This is the product that Michael Fremer reviewed in Stereophile, and Roy Gregory reviewed in HiFi+. (I won't post any photos of the 3.0 because you can see it on our website). The top board holds the amplifier circuits and the bottom board houses the voltage regulators.
The two boards were designed as a single set and plug into each other. The "pins" are the component leads on both boards, and in some cases the components themselves (vertically oriented resistors, diodes, etc). The requisite leads are kept long and uncut during component installation and soldering, and are plugged into matching receiver pads on the opposing board during assembly.
You cannot do this with machine assembly techniques.
Attachments
Peter Daniel:
I like your amplifier - both the casing and the way that you have built the circuit. Congratulations on a design that looks nice and undoubtedly sounds super!
Regarding p2p, as with many other things, it is good for some things, and not so good for others. If you want to do impedance control, it is much better to go with a multilayer pcb and ground planes. P2p is best done with leaded components rather than surface-mount, but unfortunately, many high-performance components are available only as SMD, so you may have to choose between p2p or SMD. And given the limited ability of the human being to maintain full alertness and concentration for long periods of time, p2p places a realistic upper limit on the complexity of the circuits that can be built with this technique. Simple circuits are no problem, but when you are building complex circuits with lots of components and connections, having a ready-laid-out pcb to make sure that you don't forget any components or don't wire up the wrong connections is very reassuring.
But p2p also has significant performance advantages. Many circuits need high-impedance nodes for gain, and the same node will be much higher in impedance as an air-dielectric p2p node than it will be on an FR4 pcb dielectric. Stray parasitic capacitances are also much lower with p2p. In my experience, the greatest advantages of p2p arise when it is used not by itself but in combination with a multilayer pcb with full ground-planing. The nodes that need to be low-impedance can be placed on the pcb, while the high-impedance or low-capacitance nodes can be p2p.
As much as possible, additional wiring should be avoided/minimized in p2p construction, because increasing the number of solder joints and the length of various paths are both sonically disadvantageous. I prefer to clamp parts together, bond them into solid blocks, then bend the leads onto each other and solder them at the parts. The solidity of the blocks and the firm mounting of the pcb also helps when you want to layer components, and layering can really help to cut down path lengths in a hurry. Making the circuit structure as compact/dense as possible is highly advantageous for both measured performance and sound, IME. Circuits built per above will often require less compensation for stability, and can be more resistant to RF interference than they otherwise would.
regards, jonathan carr
I like your amplifier - both the casing and the way that you have built the circuit. Congratulations on a design that looks nice and undoubtedly sounds super!
Regarding p2p, as with many other things, it is good for some things, and not so good for others. If you want to do impedance control, it is much better to go with a multilayer pcb and ground planes. P2p is best done with leaded components rather than surface-mount, but unfortunately, many high-performance components are available only as SMD, so you may have to choose between p2p or SMD. And given the limited ability of the human being to maintain full alertness and concentration for long periods of time, p2p places a realistic upper limit on the complexity of the circuits that can be built with this technique. Simple circuits are no problem, but when you are building complex circuits with lots of components and connections, having a ready-laid-out pcb to make sure that you don't forget any components or don't wire up the wrong connections is very reassuring.
But p2p also has significant performance advantages. Many circuits need high-impedance nodes for gain, and the same node will be much higher in impedance as an air-dielectric p2p node than it will be on an FR4 pcb dielectric. Stray parasitic capacitances are also much lower with p2p. In my experience, the greatest advantages of p2p arise when it is used not by itself but in combination with a multilayer pcb with full ground-planing. The nodes that need to be low-impedance can be placed on the pcb, while the high-impedance or low-capacitance nodes can be p2p.
As much as possible, additional wiring should be avoided/minimized in p2p construction, because increasing the number of solder joints and the length of various paths are both sonically disadvantageous. I prefer to clamp parts together, bond them into solid blocks, then bend the leads onto each other and solder them at the parts. The solidity of the blocks and the firm mounting of the pcb also helps when you want to layer components, and layering can really help to cut down path lengths in a hurry. Making the circuit structure as compact/dense as possible is highly advantageous for both measured performance and sound, IME. Circuits built per above will often require less compensation for stability, and can be more resistant to RF interference than they otherwise would.
regards, jonathan carr
Jonathan makes a good point, but the argument has two
edges. On the one hand, the X amp doesn't have any
exotic requirements about impedances and capacitance, and
on the other, it is simple enough to enable p2p.
It's interesting, after some experimenting we were able to
eliminate the compensation capacitances normally expected
of such a circuit through layout considerations. You would not
get this as a reproduceable thing in production from p2p.
Ah well, what's a few pf among friends?
edges. On the one hand, the X amp doesn't have any
exotic requirements about impedances and capacitance, and
on the other, it is simple enough to enable p2p.
It's interesting, after some experimenting we were able to
eliminate the compensation capacitances normally expected
of such a circuit through layout considerations. You would not
get this as a reproduceable thing in production from p2p.
Ah well, what's a few pf among friends?
As I see it, high end audio manufacturers have several advantages. They have research and development budgets, the ability to stock pile parts, state of the art part matching, cherry picked wiz bang people working for them, and a price point that takes them out of competition with Sony and Samsung. Their products have to be first class sonically and aesthetically, to stand up to products of equally high standards, judged by the industry pundits and the buying public. Rare it is that any DIY attempt would stand up to such scrutiny, let alone making every product produced meet the same very lofty standards, which just get more and more outrageously well done as time marches on. I love DIY, and it's idiosyncrasies, but making an actual copy of a Pass Labs amp in performance and fit and finish is very rarely seen.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.