DaveM said:
Dave,
You are an old timer and have post quite a bit; what I don’t like is seeing people who have barely post or newcomers which are only looking to grab what ever they can. For me diyAudio is sharing experiences and work done on the audio field of course plus hearing opinions and knowing people that share the same interest.
On the other hand the etching file is too big to post here which I don’t regret since I can now know who are really interested. I think it’s fair to the community to know who is working or has interest in this wonderful project shared by the Master itself, Nelson Pass, to whom we are all grateful.
Dave, PM me please.
Barnes and Noble carry AX here. I've been keeping an eye out for May's issue, but it hasn't hit the shelf yet. I did pick up the April copy though... very nice, I will subscribe from this point
I miss-cut my boards last night, so I'm back on it in a bit. I should be posting some pics tonight.
7/10
I miss-cut my boards last night, so I'm back on it in a bit. I should be posting some pics tonight.
7/10
Or do your own PCB artwork. It's not hard at all. I see posts from people who are scared, but I think of it as just another aspect of DIY. I do boards all the time (nine this week, ranging from the size of postage stamps [daughter boards for the Line Drive] to boards the size of my hand [Line Drive motherboard and the crossover variant]). Yes, I remember being scared the first time I did a board, but I also remember being scared the first time I rode a bicycle...and speaking from the perspective of having endured quite a few years of reconstructive efforts on my mouth, I will assure you that DIY circuit boards are a lot less hazardous to your health than bicycles. Cheaper, too.
We're not going to be intimidated by some dinky little piece of fiberglass! We're DIY! Show that thing who's boss.
Grey
We're not going to be intimidated by some dinky little piece of fiberglass! We're DIY! Show that thing who's boss.
Grey
steenoe said:
I dont think so......
As far as I recall, they are custom made extrucions. I am sure Jacco can remember the post, where NP said that....
the closest I've found for a pass labs clone heatsink is too small to do any good.
http://www.thermaflo.com/bin/cr_exdatasht.pl?Pnum=e1339&LengthUnits=in&ExLength=12
sad, really. it looks great till until you read the measurements of size
steenoe said:
Clap, clap, Grey
You will note that I said "DIY," which means Do It Yourself, not "DIG," being Do It for me, Grey. Jeez, only six transistors and a handful of resistors and people are already intimidated.
I'll give you a hint, though...there are far, far too many boards showing up on this site where people are trying to fit the circuit into a matchbox. When asked, they usually mumble something about a short signal path. While a short signal path is a laudable goal, I suggest that those who might wish to undertake PCB layout also keep in mind things like heat dissipation, serviceability, ability to modify, and the plain old-fashioned idea of pride of workmanship. Most circuit boards you see here look like they were designed by a committee comprised of drunken chimpanzees. Vow to yourself that every board you make will be better than its predecessors. When you look at a PCB in an amplifier, your proper frame of mind is not, "This was designed by audio gods!" it's "What did the designer do right and what did he do wrong?" Over time, you'll learn that certain things are that way for a reason. And you'll also learn that there are boneheads out there who really should have farmed out their PCB artwork.
A few quickie rules:
1) Lay out components in an orderly fashion. You do now arrive at a successful PCB layout by throwing parts at a piece of flypaper. Why should you do this? Because if something goes wrong, it's a lot easier to debug a board where you can find things without a road map.
2) Lay out components in a reasonably close resemblance to the schematic. (You do know how to make a decent schematic, don't you?) Why should you do this? Because if related parts are grouped near each other, you can isolate problems more quickly. This sounds the same as #1. It's not, though they're closely related. #1 means put the resistors in neat rows, not scattered in any old which-way. This one means populate those neat rows with closely related parts instead of one part from the current source next to another part from the power supply, next to the hookup to the feedback loop.
3) When you get around to populating the board, place all the parts so the nomenclature is on top where you can read it. Place all parts so they can be read from one or two orientations of the board--from six-o'clock and from 9 o'clock. Exceptions do occur. You can't get the part number on a transistor to change sides. It is the way it is. As such, this rule is subordinate to #1 and #2. However, try to keep all this stuff in mind as you do your layout. Why should you do this? So you can see the parts values without spraining your wrist or having to desolder the part from the board.
4) No connections to the middle of the board! All connections are to be made at the edges of the board. Why should you do this? Because it will make service and modification easier. Trust me on this one. Besides, you can literally make an unbuildable circuit in that you end up with a connection that you can't get to in order to solder it.
5) Put the mounting points in as rectangular and reasonable a pattern as possible. Better still, use even increments of inches/centimeters. Why should you do this? Because someday you might just get around to needing those holes when you get tired of looking at a PCB hanging in mid-air. And it's a bloody nuisance to have to unravel some oddball spacing so as to drill holes in a piece of metal in a manner where the two will work together.
There are other things that you'll learn and come to appreciate over time, but this will keep your boards from looking as though they were laid out by a five year-old.
6) Assuming that you intend to mount high dissipation parts (think output devices) directly on the board (as opposed to a sub-assembly somewhere else in the chassis), Do not lay out the board with the hot parts right next to each other! This is just downright silly. Pretend it's a hot day. Sweltering, in fact. You and your significant other are sitting side by side on a bench. You may love that person to death, but that doesn't mean you want to be skin to skin in the heat of the day. (Save that for the heat of the night...but I'll stop there and leave the rest to your overheated imaginations.) Why should you do this? I should think it's self-evident, but from the number of circuits I see with the output devices crowded together like refugees clinging to an overturned boat, apparently it's not. The laws of thermodynamics state that heat energy will flow from a place where there is more of it to a place where there is less of it...but they do not say that heat transfer takes place immediately. That's just a fantasy on the part of the DIYer. Spread the heat out. Please. For the sake of your outputs, I beg of you. Forsake this short signal path fetish at least long enough to allow your output devices to cool down.
7) If you can meet all of the above requirements, then you're (finally) allowed to obsess over signal path length. Why should you do this? Because the longer the trace on the board, the more likely the signal is to run afoul of EM or RF fields. No, you are not worried about the resistance of the copper trace under any but the most exceptional circumstances. No...sorry...your circuit is not the exception.
8) Use reasonable trace widths for high currents. Why should you do this? Because although PCB traces cool fairly efficiently, you don't want to lose power in a puny trace. Yes, this sometimes conflicts with #7. Sometimes it works with #7. Life ain't always fair. Deal with it.
9) You can get away with 1 oz board (or even lighter if necessary) for low level signals, but two ounce or greater is overkill for low level. It is mandatory (okay...almost) for high power. Why should you do this? Same as #8. Size your traces reasonably for the signal they will carry.
10) Whatever I forgot...I'm at work and I've got to go deal with a problem. This post is going in unedited and is likely to be ugly. Sorry 'bout that.
Grey
Useful stuff. A $2 increase in board price is negligible, so why make them so small just to increase yield? Also we are each tempted and limited by what we have and the latest component de jeur so often need more space for that fatter cap or longer resistor (John Curl now recommending half watt resistors, Nelson, the "silk" caps)
Speaking of Nelson, look how his First Watt boards are long and thin to spread out those output devices..
Speaking of Nelson, look how his First Watt boards are long and thin to spread out those output devices..
