I mentioned servo only because you had posted it here (as SCP-1.pdf) :
http://www.diyaudio.com/forums/loun...ch-preamplifier-part-ii-1350.html#post4317902
So I thought servo was part of it.
And if so then you might have other arguments than CCS drift.
Patrick
http://www.diyaudio.com/forums/loun...ch-preamplifier-part-ii-1350.html#post4317902
So I thought servo was part of it.
And if so then you might have other arguments than CCS drift.
Patrick
Distortion and DC drift are secondary to changing the supply voltage. However the supply resistor would have to increase to 600 ohms or more to keep the Idle current of both the first stage and the cascode stage about the same current.
Let's see if any of you can recommend a QUIET 25ma current source, that will replace the supply resistor without adding noise. Any suggestions?
Let's see if any of you can recommend a QUIET 25ma current source, that will replace the supply resistor without adding noise. Any suggestions?
This current source would work IF you don't care about input noise. Unfortunately, current sources have noise gain, and it would probably overwhelm the noise contribution of the input stage since they sum together. Think of the current source and the input fet as part of a mixer where both contribute their noise.
@Esperado
I took a different approach to Alex. He contacted me and offered to make the board as a favor. In principle, I don't like accepting favors from people I can't return the favors to, and in the end, that one board morphed into three boards, main, piggyback board and the main PSU board. In my judgement, too much work to do as a favor. So Alex did it, I asked him for his address and sent him €200. That sum because that's what it would have cost me locally. Please remember that Romania has a lot of economic problems, so labor rates are way below what you would probably expect them to be as in France.
He came back to me saying that was way too much, but he and his wife could now contemnplate a proper summer vacation ?!? That really touched me, because in my book, his work was well worth it. Besides, I like to keep a clean tab, no leftovers, makes me feel better. As I see it, we both gained fair and square. And we'll do it again soon enough.
@Elektroj
In my experience, the key problem with PCB artists is that most of them do digital only and don't know much about analog audio. Alex is the different one, he is obviously well acquainted with analog audio, so his approach to designing the board is a clear advantage. You have very little need to explain anything, you just need to make your part sizes well known, unless written in the schematic he cannot know whether you want some capacitor rated at 100V or 1,000V, which is a whopper compared to the 100V version. Ditto for electrolytic caps. And you have to make it very clear how you want your power section connected, as through hole connection, as longer one side connection, or what. Do that and you're home and dry.
I took a different approach to Alex. He contacted me and offered to make the board as a favor. In principle, I don't like accepting favors from people I can't return the favors to, and in the end, that one board morphed into three boards, main, piggyback board and the main PSU board. In my judgement, too much work to do as a favor. So Alex did it, I asked him for his address and sent him €200. That sum because that's what it would have cost me locally. Please remember that Romania has a lot of economic problems, so labor rates are way below what you would probably expect them to be as in France.
He came back to me saying that was way too much, but he and his wife could now contemnplate a proper summer vacation ?!? That really touched me, because in my book, his work was well worth it. Besides, I like to keep a clean tab, no leftovers, makes me feel better. As I see it, we both gained fair and square. And we'll do it again soon enough.
@Elektroj
In my experience, the key problem with PCB artists is that most of them do digital only and don't know much about analog audio. Alex is the different one, he is obviously well acquainted with analog audio, so his approach to designing the board is a clear advantage. You have very little need to explain anything, you just need to make your part sizes well known, unless written in the schematic he cannot know whether you want some capacitor rated at 100V or 1,000V, which is a whopper compared to the 100V version. Ditto for electrolytic caps. And you have to make it very clear how you want your power section connected, as through hole connection, as longer one side connection, or what. Do that and you're home and dry.
Last edited:
It is wonderful to see how some people with such a talent, taste for beauty and knowledge can be so often so generous people. That is what we wanted the world of freeware to be: the best stuff that can be done for pleasure with passion.
This said, I can't believe there is no commercial company who want to hire people like him to design their products and that such talented guys can stay half a second unemployed !
Alas, one of my best friends is a fantastic guitarist. Some kind of Larry Carlton for the technique, taste and inventiveness, plus the lyricism of some Eric Clapton at his best period. How to understand he lives in misery ?
Well, i tried to help him during my life, recommending-him for studio sessions each time it was possible. Very often, his great guitar solos or chorus were not used. Imagine, this poor singer hooting just after him ;-)
On stage with his groups, after all his solos, the public was standing...
Since Brad has answered and not mentioned the SY in the room, magnetic field is proportional to number of turns times the applied current. Resistance (not impedance) is proportional to turns. So when you double the number of turns, you double the resistance and halve the current so the field stays the same. (Roughly a 30% or less error.)
So why 8 ohms? In the days of glowing transistors the output transformer was required. The size and number of windings on the primary was pretty much a function of current and frequency response. The windings on the secondary were desired to be a minimum.
18 gauge wire, the standard of the day, had a resistance of .004 ohms per running foot. So for a public auditorium a wire loss of about 10% on a 50' run would be acceptable. Other folks of course used 4, 6 or 16 ohms.
The big exception was intercom loudspeakers that were 45 ohms. This allowed 350' runs in school buildings and gave a greater voltage back when the transducer was run backwards.
The upper limit is of course the rising inductance will limit high frequency performance.
I prefer 16 ohm loudspeakers when available for large installations and for background music 32 ohm custom made loudspeakers worked well. Although it confused just about everyone else who didn't understand how you could put 4-8 loudspeakers in a joint without 70 volt transformers or burning out the amplifier. (One BGM supplier did provide current output amplifiers just so they could do the same with 8 ohm loudspeakers.)
ES
Last edited:
Interesting Ed! I could see a diy-er building OTL tube amps and rewinding driver voice coils to say 100 - 200 ohms to match that amp! My OTL tube amp used the then-available 400 ohms drivers from Philips.
jan
PCB artists... never liked that term, its more science than art and always has been, SIV EMC etc a schematic is only a diagram, the layout will determine how the design performs. Do most of them only do digital, not really, there is much more digital these days but there is usually analogue in there somewhere and quite often RF with everyone wanting near field communications.
more later when I can get some internet!!!! Got a reply for Kindhornman, but cant upload it with docs.....
Also analogue layout is analogue layout, there is no difference between analogue layout for instrumentation, audio or anything else that requires analoge.
I never have nor never will put my name on a PCB design, It goes in the border text only (Drawn, checked, approved)...
Last edited:
Never seen or heard of distortion due to solder in any other field, including audio!!!! strange
As to the PTH holes that sounds like barrel cracking, I would suspect this is a one off problem and not a recurring one these days, if the correct grade of PCB material is used.... I have seen this on cheepo cheepo designs with lead free processes where they have used laminates with a Tg rating only suitable for lead processes.
Last edited:
Having trouble replying due to bad internet, working away from home as usual in a hotel.....
Trace width, distance from return path, copper thickness (weight) and dielectric are all important for calculating the various trace parameters, as the two pictures show for half and one ounce copper. The PDF is from my SIV software help, where you can go one step further and model traces, depending on how serious you want to get.
will post attachments later.....
- Status
- Not open for further replies.