It could be that Marce is really a top notch expert on PC design, but tends to get a bit long on trying to correct folks and often goes well over their level of understanding. As a result he tends to get knocked about a bit by those who don't get it. So he then over responds a bit.
I got him on this over reaction on another thread. He fell for it so I posted a "Gotcha."
So to give him a chance to even the score I posted what is an over simplified layout for an RF capable low noise amplifier.
So if he really does do a screen capture and transforms it via Paint into a JPEG we may all actually learn a bit.
My limit was attempting to keep some bit of a constant impedance line with terminations to distribute the signal.
Of course I seriously doubt Marce could do any better...
Otherwise he would show us.My PM
Sorry but I couldn’t resist….
An alternate version, aims being….
Decoupling caps next to the power pins.
Apart from a couple of supply tracks on bottom layer, most other tracks buried in between ground planes limiting susceptibility to noise pick up.
What I haven’t done:
Completely routed power, missed of the caps and connector, also used tracks, I would beef these up using copper pours but wanted to save time.
Other than that a fun little exercise.
Regards
Marc
I did say I would do each block squarer rather than elongated....
Sorry but I couldn’t resist….
An alternate version, aims being….
Decoupling caps next to the power pins.
Apart from a couple of supply tracks on bottom layer, most other tracks buried in between ground planes limiting susceptibility to noise pick up.
What I haven’t done:
Completely routed power, missed of the caps and connector, also used tracks, I would beef these up using copper pours but wanted to save time.
Other than that a fun little exercise.
Regards
Marc
I did say I would do each block squarer rather than elongated....
If you want Gerbers and it finishing of let us know, I'll have plenty of time over the next few days, working at the sssslowwest lab in the West
Thanks, but I'll only ask you do all the work if the cards coming in show any problems.
Constant impedance with tapoffs for distribution? How's that work?
Perhaps dropping the linewidth of the line at each tap, so that no reflections occur?
Unless you're talkin bog level dc type stuff..where line impedance really makes no difference on a pc board...like audio.
John
Directional couplers?Constant impedance with tapoffs for distribution? How's that work?
Perhaps dropping the linewidth of the line at each tap, so that no reflections occur?
Unless you're talkin bog level dc type stuff..where line impedance really makes no difference on a pc board...like audio.
John
google scholar found nothing..
I am searching with no space between the words, as scott typed.
Marce, bcarso,
It dawned on me that the fanout I describe, that of pulling traces off the main branch and reducing the main width will indeed maintain signal integrity by preventing reflections at each branch connection. And as I think of it, only in one direction, exactly as Bcarso stated. Is this a common PC trace thing for clock distribution? I've never seen it.
John
I am searching with no space between the words, as scott typed.
Marce, bcarso,
It dawned on me that the fanout I describe, that of pulling traces off the main branch and reducing the main width will indeed maintain signal integrity by preventing reflections at each branch connection. And as I think of it, only in one direction, exactly as Bcarso stated. Is this a common PC trace thing for clock distribution? I've never seen it.
John
Searching for something which is not a name, but a foreign word, may prove to be difficult either way.
Aka, eigenvalue is a rather stupid word.
('eigen' is a German and Dutch word, subtle difference between the two. In German, the meaning of 'eigen' is intrinsic. German parler example : Das ist mir eigen. Aka, it's something that belongs to me)
I have used it before for clock distribution using source series terminations. Motorola MECL design guide mentioned it. hydrid T IIRC? matching stub lengths for reflected waves, prop delay matching. i do not have howard johnson's book, should be covered in there i would think.
used HyperLynx & Interconnectix s/w to simulate this clock distribution/routing back in the day when i did pcb design
Last edited:
Distantly related, that reminds me of an assistant prof at UCLA who specialized in radio astronomy. He wanted me to build a 405MHz radio interferometer and suggested modifying a TV tuner. He was rather deficient in basic electronics and wanted something cheap and dirty and in a hurry, and expressed the belief that an array of dipoles could be interconnected in daisy chain fashion with twinlead. I focused on low noise front ends and double balanced mixers, and it bothered him that I built things out of copper clad board and soldered along the seams --- he felt that must explain why I was taking so long.google scholar found nothing..
I am searching with no space between the words, as scott typed.
Marce, bcarso,
It dawned on me that the fanout I describe, that of pulling traces off the main branch and reducing the main width will indeed maintain signal integrity by preventing reflections at each branch connection. And as I think of it, only in one direction, exactly as Bcarso stated. Is this a common PC trace thing for clock distribution? I've never seen it.
John
The idiot did not get tenure, appropriately in my opinion.
When a very bright guy came on board years later, he got a big dish and in short order we were able to detect bright radio sources in the hydrogen line at 1420MHz from the roof of the Math-Sciences building. One did have to worry about radar from San Pedro blowing out the front end when the thing was pointed at the galactic center.
google scholar found nothing..
I am searching with no space between the words, as scott typed.
John
The term appears in some German patents with no space as a MOSFET's intrinsic diode. It WAS a joke though.
- Status
- Not open for further replies.