[John Larkin]

[RST Engineering \(jw\)]

What are all the red, yellow, and purple horizontal striations?

Jim

--
"It is the mark of an educated mind to be able to entertain a thought
without accepting it."
--Aristotle


"John Larkin" wrote in message
...

[John Larkin]

On Sat, 17 May 2008 12:03:02 -0700, "RST Engineering \(jw\)"
wrote:

What are all the red, yellow, and purple horizontal striations?

Jim

That's a screen capture, and PADS does that to the display at certain
zoom levels. Maybe it's on purpose so that you can sort of see through
layers. They're solid copper pours, various power supply voltages,
mostly on layer 3, in real life.

I just finished moving all the output stage parts 0.7 inches to the
right - what a pain - to get closer to the output connector. That was
based on actually measuring the risetime of an inch of 50 ohm
microstrip on FR-4, which was ghastly.


John

[Joerg]

John Larkin wrote:
On Sat, 17 May 2008 12:03:02 -0700, "RST Engineering \(jw\)"
wrote:

What are all the red, yellow, and purple horizontal striations?

Jim

That's a screen capture, and PADS does that to the display at certain
zoom levels. Maybe it's on purpose so that you can sort of see through
layers. They're solid copper pours, various power supply voltages,
mostly on layer 3, in real life.

I just finished moving all the output stage parts 0.7 inches to the
right - what a pain - to get closer to the output connector. That was
based on actually measuring the risetime of an inch of 50 ohm
microstrip on FR-4, which was ghastly.


Hmm, back in the days when I still had hair we did 430MHz microstrip on
phenolic. In power amps there could be a slight discoloration after a
few years of usage ...

--
Regards, Joerg

http://www.analogconsultants.com/

"gmail" domain blocked because of excessive spam.
Use another domain or send PM.

[RST Engineering \(jw\)]

Good lord, the effective epsilon of FR-4 ought to be something on the order
of 2 or so (real er should be around 4.5 or so) and the propagation delay
ought to be femtoseconds or so. What is slowing down the risetime? 50 ohms
on FR4 ought to be a conductor about as wide as the thickness of the board
material as I vaguely remember.

Jim



That was
based on actually measuring the risetime of an inch of 50 ohm
microstrip on FR-4, which was ghastly.


John

[John Larkin]

On Sat, 17 May 2008 16:31:32 -0700, "RST Engineering \(jw\)"
wrote:

Good lord, the effective epsilon of FR-4 ought to be something on the order
of 2 or so (real er should be around 4.5 or so) and the propagation delay
ought to be femtoseconds or so. What is slowing down the risetime? 50 ohms
on FR4 ought to be a conductor about as wide as the thickness of the board
material as I vaguely remember.

Jim



The top dielectric layer is 12 mils, so a 50 ohm trace is only 20 mils
wide. That's cheap to fab from a quick proto house, but not optimum
for losses. I want this thing to have a risetime below 150 ps, and an
inch of 50 ohm line on this cheap crud is about 70 all by itself. A
real microwave laminate would help, and make the traces wider, but
that's expensive... this is really just a prototype. As Joerg says,
you can do fast stuff on junk boards, but you have to keep it short.

FR4 seems to have an Er around 4.5 at GHz speeds. Effective epsilon
for the microstrip is about 3.3.


John

[RST Engineering \(jw\)]

The top dielectric layer is 12 mils, so a 50 ohm trace is only 20 mils
wide.

My calculator says 22 mils at one Gig, and not knowing your frequency of
operation puts me at somewhat of a disadvantage. I wonder if the mismatch
between 20 and 22 is honking things up.

That's cheap to fab from a quick proto house, but not optimum
for losses. I want this thing to have a risetime below 150 ps, and an
inch of 50 ohm line on this cheap crud is about 70 all by itself.

Let me understand, please. I'm having a hard time relating line matching to
risetime. What does the loss tangent of the material or dielectric constant
have to do with risetime?


FR4 seems to have an Er around 4.5 at GHz speeds. Effective epsilon
for the microstrip is about 3.3.

I show 3.1, but that's picking at nits.

Jim

[For fun]

Hi,

I have a DMM that is suffering from dim LCD segments.
I beleive that the conductive rubber betwee the PCB and the LCD is causing
the issue.
What are the recommendations of cleaning the contacts on the PCB, LCD &
rubber?

Neil.

[Joerg]

RST Engineering (jw) wrote:
The top dielectric layer is 12 mils, so a 50 ohm trace is only 20 mils
wide.

My calculator says 22 mils at one Gig, and not knowing your frequency of
operation puts me at somewhat of a disadvantage. I wonder if the mismatch
between 20 and 22 is honking things up.

That's cheap to fab from a quick proto house, but not optimum
for losses. I want this thing to have a risetime below 150 ps, and an
inch of 50 ohm line on this cheap crud is about 70 all by itself.

Let me understand, please. I'm having a hard time relating line matching to
risetime. What does the loss tangent of the material or dielectric constant
have to do with risetime?



Take John's example. 150psec means that you'll have spectral components
in excess of 3GHz in there. If they are dissipated a lot more than the
lower ones your risetime goes to pots.


FR4 seems to have an Er around 4.5 at GHz speeds. Effective epsilon
for the microstrip is about 3.3.

I show 3.1, but that's picking at nits.


It's also losses that matter.

--
Regards, Joerg

http://www.analogconsultants.com/

"gmail" domain blocked because of excessive spam.
Use another domain or send PM.

[John Larkin]

On Sat, 17 May 2008 22:39:43 -0700, "RST Engineering \(jw\)"
wrote:



The top dielectric layer is 12 mils, so a 50 ohm trace is only 20 mils
wide.

My calculator says 22 mils at one Gig, and not knowing your frequency of
operation puts me at somewhat of a disadvantage. I wonder if the mismatch
between 20 and 22 is honking things up.

It's wideband, DC to maybe 2.5 GHz, corresponding to roughly 150 ps
edges. I don't want something as simple as the pcb degrading what the
circuit can do, hence the desire to keep the traces very short. That
beats spending $1000 or so on better boards. Things are not yet honked
up; the board hasn't even been built.


That's cheap to fab from a quick proto house, but not optimum
for losses. I want this thing to have a risetime below 150 ps, and an
inch of 50 ohm line on this cheap crud is about 70 all by itself.

Let me understand, please. I'm having a hard time relating line matching to
risetime. What does the loss tangent of the material or dielectric constant
have to do with risetime?

Dielectric and copper losses gobble up high frequency components, so
naturally degrade the risetime of pulse leading/falling edges. I
suspect that, with 20 mil, 50 ohm lines, the copper skin loss is worse
than the dielectric loss. Added bummer is that they crud up the bottom
of the copper ("black oxide" process) to make it bond to the laminate
better, and that radically worsens skin losses. Most of the current is
headed down, so the shiny top side, working against air, doesn't help
that much.

Of course, impedance mismatches degrade risetimes too.




FR4 seems to have an Er around 4.5 at GHz speeds. Effective epsilon
for the microstrip is about 3.3.

I show 3.1, but that's picking at nits.

I'm using Appcad. If it has any systematic trend, it seems to design
traces that come out a bit lower impedance than intended. But then
most of the board houses take liberties with stackups too.

John

[Joerg]

John Larkin wrote:
On Sat, 17 May 2008 22:39:43 -0700, "RST Engineering \(jw\)"
wrote:

The top dielectric layer is 12 mils, so a 50 ohm trace is only 20 mils
wide.
My calculator says 22 mils at one Gig, and not knowing your frequency of
operation puts me at somewhat of a disadvantage. I wonder if the mismatch
between 20 and 22 is honking things up.

It's wideband, DC to maybe 2.5 GHz, corresponding to roughly 150 ps
edges. I don't want something as simple as the pcb degrading what the
circuit can do, hence the desire to keep the traces very short. That
beats spending $1000 or so on better boards. Things are not yet honked
up; the board hasn't even been built.


Check with your fab houses whether they can do partial-Rogers boards.
Might be the ticket here.

[...]

--
Regards, Joerg

http://www.analogconsultants.com/

"gmail" domain blocked because of excessive spam.
Use another domain or send PM.

[Phil Allison]

"For fun"


I have a DMM that is suffering from dim LCD segments.
I beleive that the conductive rubber betwee the PCB and the LCD is causing
the issue.
What are the recommendations of cleaning the contacts on the PCB, LCD &
rubber?


** I have used IPA = Iso Propyl Alcohol with success.

Or else try good old methylated spirits = denatured alcohol.

Wet a cloth and used that to gently wipe the surfaces clean.



........ Phil

[Oppie[_5_]]

I believe that the rubber connector is called a 'zebra connector' and is
made of alternating layers of silicone rubber and conductive elastomer. The
pitch is usually much smaller than the trace spacing on the pwb or display
so that placement is not critical.

I have used clean flux remover to clean the components. Try to Blot rather
than rub with a soft cloth or tissue. The elastomer will sometimes leach out
plasticizer which interferes with the connection.

If all is clean and you still have dim/missing segments, try shimming to
press the display against the zebra slightly tighter.

"For fun" wrote in message
node...
Hi,

I have a DMM that is suffering from dim LCD segments.
I beleive that the conductive rubber betwee the PCB and the LCD is causing
the issue.
What are the recommendations of cleaning the contacts on the PCB, LCD &
rubber?

Neil.

[Joel Koltner[_2_]]

"Joerg" wrote in message
...
Check with your fab houses whether they can do partial-Rogers boards. Might
be the ticket here.

I had a quote a couple weeks back on a 4 layer board done all-Roger vs.
half-and-half... there wasn't a huge difference; half-and-half was only ~15%
cheaper. This was prototype quantities, though, so I imagine you're just
seeing how little the board material itself plays in the overall cost.

[Joerg]

Joel Koltner wrote:
"Joerg" wrote in message
...
Check with your fab houses whether they can do partial-Rogers boards. Might
be the ticket here.

I had a quote a couple weeks back on a 4 layer board done all-Roger vs.
half-and-half... there wasn't a huge difference; half-and-half was only ~15%
cheaper. This was prototype quantities, though, so I imagine you're just
seeing how little the board material itself plays in the overall cost.


Even if it was only 15%, in my world that's HUGE.

--
Regards, Joerg

http://www.analogconsultants.com/

"gmail" domain blocked because of excessive spam.
Use another domain or send PM.

[John Larkin]

On Sun, 18 May 2008 14:19:34 -0700, Joerg
wrote:

John Larkin wrote:
On Sat, 17 May 2008 22:39:43 -0700, "RST Engineering \(jw\)"
wrote:

The top dielectric layer is 12 mils, so a 50 ohm trace is only 20 mils
wide.
My calculator says 22 mils at one Gig, and not knowing your frequency of
operation puts me at somewhat of a disadvantage. I wonder if the mismatch
between 20 and 22 is honking things up.

It's wideband, DC to maybe 2.5 GHz, corresponding to roughly 150 ps
edges. I don't want something as simple as the pcb degrading what the
circuit can do, hence the desire to keep the traces very short. That
beats spending $1000 or so on better boards. Things are not yet honked
up; the board hasn't even been built.


Check with your fab houses whether they can do partial-Rogers boards.
Might be the ticket here.

We did that once. The board was curled up like a potato chip. I could
lay it flat on my desk, give it a twirl, and it would spin for about a
minute. Cool.

I still want to do this as a cheap proto-house board on FR-4. It's
really a test circuit, although I might call it a product if it works
well.

John

[Joerg]

John Larkin wrote:
On Sun, 18 May 2008 14:19:34 -0700, Joerg
wrote:

John Larkin wrote:
On Sat, 17 May 2008 22:39:43 -0700, "RST Engineering \(jw\)"
wrote:

The top dielectric layer is 12 mils, so a 50 ohm trace is only 20 mils
wide.
My calculator says 22 mils at one Gig, and not knowing your frequency of
operation puts me at somewhat of a disadvantage. I wonder if the mismatch
between 20 and 22 is honking things up.
It's wideband, DC to maybe 2.5 GHz, corresponding to roughly 150 ps
edges. I don't want something as simple as the pcb degrading what the
circuit can do, hence the desire to keep the traces very short. That
beats spending $1000 or so on better boards. Things are not yet honked
up; the board hasn't even been built.

Check with your fab houses whether they can do partial-Rogers boards.
Might be the ticket here.

We did that once. The board was curled up like a potato chip. I could
lay it flat on my desk, give it a twirl, and it would spin for about a
minute. Cool.

I still want to do this as a cheap proto-house board on FR-4. It's
really a test circuit, although I might call it a product if it works
well.


No idea what the old Duroid material would cost nowadays but if it were
my project I'd probably pick a really good material first, see what the
circuit can do and then ratchet down material qualities to reach a
compromise.

Or if you only have one or two longer stretches place rigid coax. That
looks really industrial and gives copycats a good scare ;-)

--
Regards, Joerg

http://www.analogconsultants.com/

"gmail" domain blocked because of excessive spam.
Use another domain or send PM.

[John Larkin]

On Mon, 19 May 2008 13:40:24 -0700, Joerg
wrote:

John Larkin wrote:
On Sun, 18 May 2008 14:19:34 -0700, Joerg
wrote:

John Larkin wrote:
On Sat, 17 May 2008 22:39:43 -0700, "RST Engineering \(jw\)"
wrote:

The top dielectric layer is 12 mils, so a 50 ohm trace is only 20 mils
wide.
My calculator says 22 mils at one Gig, and not knowing your frequency of
operation puts me at somewhat of a disadvantage. I wonder if the mismatch
between 20 and 22 is honking things up.
It's wideband, DC to maybe 2.5 GHz, corresponding to roughly 150 ps
edges. I don't want something as simple as the pcb degrading what the
circuit can do, hence the desire to keep the traces very short. That
beats spending $1000 or so on better boards. Things are not yet honked
up; the board hasn't even been built.

Check with your fab houses whether they can do partial-Rogers boards.
Might be the ticket here.

We did that once. The board was curled up like a potato chip. I could
lay it flat on my desk, give it a twirl, and it would spin for about a
minute. Cool.

I still want to do this as a cheap proto-house board on FR-4. It's
really a test circuit, although I might call it a product if it works
well.


No idea what the old Duroid material would cost nowadays but if it were
my project I'd probably pick a really good material first, see what the
circuit can do and then ratchet down material qualities to reach a
compromise.

Or if you only have one or two longer stretches place rigid coax. That
looks really industrial and gives copycats a good scare ;-)

I've got the critical output trace, the one feeding J5 on the right,
down to 130 mils long. That should be fine at a mere 2.5 GHz.

John

[Peter McMullin]

On Sun, 18 May 2008 19:55:58 +1000, "For fun"
wrote:

Hi,

I have a DMM that is suffering from dim LCD segments.
I beleive that the conductive rubber betwee the PCB and the LCD is causing
the issue.
What are the recommendations of cleaning the contacts on the PCB, LCD &
rubber?

Neil.

Pure Isopropyl alcohol. Don't swab it or wipe it (the zebra strip),
but bath it quickly (swish around for 5- 10 sec. with tweezers or
chopsticks) in a small dish of IPA, then set aside to dry on an
absorbent tissue.

Swab the mating surfaces on both glass and PCB with IPA, using only
swabs which will not leach adhesive, or use a lint-free tissue
(Kim-Wipes, or Kay-Dry).

Don't use the 70% rubbing alcohol! make sure it's 97% IPA.

If you have some of the dilute Stabilant 22 contact enhancer, swab a
hint of that on the mating surfaces (glass and PCB), but NOT ON THE
ZEBRA STRIP. Then, wipe off any visible residue. Ideally, we want
nearly a mono-molecular film. Then, carefully re-assemble.

I've done a few miles of these on ARRIFlex movie cameras and
accessories (535, 435, SR3, RCU1); it's factory approved service on
supposedly IP63 displays on movie cameras which work on sets from
Antarctica to the Sahara to East LA.

HTH,
Peter
** Posted from http://www.teranews.com **

[Joerg]

John Larkin wrote:
On Mon, 19 May 2008 13:40:24 -0700, Joerg
wrote:

John Larkin wrote:
On Sun, 18 May 2008 14:19:34 -0700, Joerg
wrote:

John Larkin wrote:
On Sat, 17 May 2008 22:39:43 -0700, "RST Engineering \(jw\)"
wrote:

The top dielectric layer is 12 mils, so a 50 ohm trace is only 20 mils
wide.
My calculator says 22 mils at one Gig, and not knowing your frequency of
operation puts me at somewhat of a disadvantage. I wonder if the mismatch
between 20 and 22 is honking things up.
It's wideband, DC to maybe 2.5 GHz, corresponding to roughly 150 ps
edges. I don't want something as simple as the pcb degrading what the
circuit can do, hence the desire to keep the traces very short. That
beats spending $1000 or so on better boards. Things are not yet honked
up; the board hasn't even been built.

Check with your fab houses whether they can do partial-Rogers boards.
Might be the ticket here.
We did that once. The board was curled up like a potato chip. I could
lay it flat on my desk, give it a twirl, and it would spin for about a
minute. Cool.

I still want to do this as a cheap proto-house board on FR-4. It's
really a test circuit, although I might call it a product if it works
well.

No idea what the old Duroid material would cost nowadays but if it were
my project I'd probably pick a really good material first, see what the
circuit can do and then ratchet down material qualities to reach a
compromise.

Or if you only have one or two longer stretches place rigid coax. That
looks really industrial and gives copycats a good scare ;-)

I've got the critical output trace, the one feeding J5 on the right,
down to 130 mils long. That should be fine at a mere 2.5 GHz.


130mils shouldn't even show on the Radar screen.

--
Regards, Joerg

http://www.analogconsultants.com/

"gmail" domain blocked because of excessive spam.
Use another domain or send PM.