50 ohm termination

[Toaster]

Can I get away with terminating the far end of a transmission line or
do I need a series resistor at the source and a parallel resistor at
the destination?

Thank you,
Toaster

[john larkin]

Post by Toaster
Can I get away with terminating the far end of a transmission line or
do I need a series resistor at the source and a parallel resistor at
the destination?
Thank you,
Toaster

You can get clean wideband transmission with any impedance at the
source and a 50 ohm termination at the end. Voltages will be the same
all along the line. Assuming theoretical lossless trans lines of
course.

You can also source terminate, namely drive a 50 ohm line from a 50
ohm source. The signal at the end of the line, unterminated, will be
identical to the source internal voltage. But intermediate points will
be awful.

For extreme cases, terminate both ends at 50 ohms. That will minimize
reflections caused by imperfect connectors or terminators. It wastes a
lot of power.


You can Spice all this.

[Toaster]

Thank you for the advice. In my case I have a 10Mhz signal with very
sharp transitions (500ps, 5V) and wanted to make sure I did things
properly.

[Bill Sloman]

Post by Toaster
Thank you for the advice. In my case I have a 10Mhz signal with very
sharp transitions (500ps, 5V) and wanted to make sure I did things
properly.

Doing things properly may not be too difficult. Light travels at about
one foot (30cm) per nanosecond. Longer electromagnetic waves travels at
about two thirds of the pace in coax insulation, so 500psec is about 10cm.

You don't have have to make you terminating resistors all that compact
to have them behave predictably with those sorts of relatively slow edges.

Faster stuff can call for symetrical arrangements of compact surface
mount resistors and that can get messy.

Double termination - at both source and destination - will give you the
nicest looking waveforms, and since you would only see 2.5V at the
destination the 125mW you'd have to dissipate at the source and the
destination shouldn't be hard to cope with.

[john larkin]

Post by Toaster
Thank you for the advice. In my case I have a 10Mhz signal with very
sharp transitions (500ps, 5V) and wanted to make sure I did things
properly.

Interesting. What's generating the 5v signal? Lots of AC and Tiny
Logic chips are that fast, but might strain to drive 50 ohms. We use
several tiny triple buffers in parallel sometimes.

Regular thick-film surface-mount resistors are fine as terminators at
500 ps.

LVDS line receivers are great at the receive end.

[Toaster]

On Mon, 17 Mar 2025 19:17:13 -0700

Post by john larkin

Post by Toaster
Thank you for the advice. In my case I have a 10Mhz signal with very
sharp transitions (500ps, 5V) and wanted to make sure I did things
properly.

Interesting. What's generating the 5v signal? Lots of AC and Tiny
Logic chips are that fast, but might strain to drive 50 ohms. We use
several tiny triple buffers in parallel sometimes.
Regular thick-film surface-mount resistors are fine as terminators at
500 ps.
LVDS line receivers are great at the receive end.

I used a THS3111CD. Split up my project into a timing and driver board,
so i have some 50 ohm BNC cables between and wanted to be extra safe
about reflections at these higher frequencies.

[Phil Hobbs]

Post by Toaster
On Mon, 17 Mar 2025 19:17:13 -0700

Post by john larkin

Post by Toaster
Thank you for the advice. In my case I have a 10Mhz signal with very
sharp transitions (500ps, 5V) and wanted to make sure I did things
properly.

Interesting. What's generating the 5v signal? Lots of AC and Tiny
Logic chips are that fast, but might strain to drive 50 ohms. We use
several tiny triple buffers in parallel sometimes.
Regular thick-film surface-mount resistors are fine as terminators at
500 ps.
LVDS line receivers are great at the receive end.

I used a THS3111CD. Split up my project into a timing and driver board,
so i have some 50 ohm BNC cables between and wanted to be extra safe
about reflections at these higher frequencies.

Hmm. 5V in 500 ps is pretty good going for a part with 1300 V/us slew. ;)

Cheers

Phil Hobbs

[Toaster]

On Tue, 18 Mar 2025 22:46:55 -0000 (UTC)

Post by Phil Hobbs

Post by Toaster
On Mon, 17 Mar 2025 19:17:13 -0700

Post by john larkin

Post by Toaster
Thank you for the advice. In my case I have a 10Mhz signal with
very sharp transitions (500ps, 5V) and wanted to make sure I did
things properly.

Interesting. What's generating the 5v signal? Lots of AC and Tiny
Logic chips are that fast, but might strain to drive 50 ohms. We
use several tiny triple buffers in parallel sometimes.
Regular thick-film surface-mount resistors are fine as terminators
at 500 ps.
LVDS line receivers are great at the receive end.

I used a THS3111CD. Split up my project into a timing and driver
board, so i have some 50 ohm BNC cables between and wanted to be
extra safe about reflections at these higher frequencies.

Hmm. 5V in 500 ps is pretty good going for a part with 1300 V/us slew. ;)
Cheers
Phil Hobbs

I'm an expert at misreading the divisions on my scope ill have you know
:)

[john larkin]

Post by Toaster
On Mon, 17 Mar 2025 19:17:13 -0700

Post by john larkin

Post by Toaster
Thank you for the advice. In my case I have a 10Mhz signal with very
sharp transitions (500ps, 5V) and wanted to make sure I did things
properly.

Interesting. What's generating the 5v signal? Lots of AC and Tiny
Logic chips are that fast, but might strain to drive 50 ohms. We use
several tiny triple buffers in parallel sometimes.
Regular thick-film surface-mount resistors are fine as terminators at
500 ps.
LVDS line receivers are great at the receive end.

I used a THS3111CD. Split up my project into a timing and driver board,
so i have some 50 ohm BNC cables between and wanted to be extra safe
about reflections at these higher frequencies.

Is the signal some analog thing, or a 10 MHz clock? The THS is an
opamp, but they can make good cable drivers too, even for clocks.

Lately I'm enamored of BUF602, a unity-gain 1 GHz beast.

[Toaster]

On Tue, 18 Mar 2025 16:02:45 -0700

Post by john larkin

Post by Toaster
On Mon, 17 Mar 2025 19:17:13 -0700

Post by john larkin

Post by Toaster
Thank you for the advice. In my case I have a 10Mhz signal with
very sharp transitions (500ps, 5V) and wanted to make sure I did
things properly.

Interesting. What's generating the 5v signal? Lots of AC and Tiny
Logic chips are that fast, but might strain to drive 50 ohms. We
use several tiny triple buffers in parallel sometimes.
Regular thick-film surface-mount resistors are fine as terminators
at 500 ps.
LVDS line receivers are great at the receive end.

I used a THS3111CD. Split up my project into a timing and driver
board, so i have some 50 ohm BNC cables between and wanted to be
extra safe about reflections at these higher frequencies.

Is the signal some analog thing, or a 10 MHz clock? The THS is an
opamp, but they can make good cable drivers too, even for clocks.
Lately I'm enamored of BUF602, a unity-gain 1 GHz beast.

I had a really hard time finding a good line driver. I might look into
this chip.

Thank you!

[john larkin]

Post by Toaster
On Tue, 18 Mar 2025 16:02:45 -0700

Post by john larkin

Post by Toaster
On Mon, 17 Mar 2025 19:17:13 -0700

Post by john larkin

Post by Toaster
Thank you for the advice. In my case I have a 10Mhz signal with
very sharp transitions (500ps, 5V) and wanted to make sure I did
things properly.

Interesting. What's generating the 5v signal? Lots of AC and Tiny
Logic chips are that fast, but might strain to drive 50 ohms. We
use several tiny triple buffers in parallel sometimes.
Regular thick-film surface-mount resistors are fine as terminators
at 500 ps.
LVDS line receivers are great at the receive end.

I used a THS3111CD. Split up my project into a timing and driver
board, so i have some 50 ohm BNC cables between and wanted to be
extra safe about reflections at these higher frequencies.

Is the signal some analog thing, or a 10 MHz clock? The THS is an
opamp, but they can make good cable drivers too, even for clocks.
Lately I'm enamored of BUF602, a unity-gain 1 GHz beast.

I had a really hard time finding a good line driver. I might look into
this chip.
Thank you!

Is your signal analog or digital?

[Toaster]

On Tue, 18 Mar 2025 19:01:51 -0700

Post by john larkin

Post by Toaster
On Tue, 18 Mar 2025 16:02:45 -0700

Post by john larkin

Post by Toaster
On Mon, 17 Mar 2025 19:17:13 -0700

Post by john larkin

Post by Toaster
Thank you for the advice. In my case I have a 10Mhz signal with
very sharp transitions (500ps, 5V) and wanted to make sure I
did things properly.

Interesting. What's generating the 5v signal? Lots of AC and
Tiny Logic chips are that fast, but might strain to drive 50
ohms. We use several tiny triple buffers in parallel sometimes.
Regular thick-film surface-mount resistors are fine as
terminators at 500 ps.
LVDS line receivers are great at the receive end.

I used a THS3111CD. Split up my project into a timing and driver
board, so i have some 50 ohm BNC cables between and wanted to be
extra safe about reflections at these higher frequencies.

Is the signal some analog thing, or a 10 MHz clock? The THS is an
opamp, but they can make good cable drivers too, even for clocks.
Lately I'm enamored of BUF602, a unity-gain 1 GHz beast.

I had a really hard time finding a good line driver. I might look
into this chip.
Thank you!

Is your signal analog or digital?

digital

[john larkin]

Post by Toaster
On Tue, 18 Mar 2025 19:01:51 -0700

Post by john larkin

Post by Toaster
On Tue, 18 Mar 2025 16:02:45 -0700

Post by john larkin

Post by Toaster
On Mon, 17 Mar 2025 19:17:13 -0700

Post by john larkin

Post by Toaster
Thank you for the advice. In my case I have a 10Mhz signal with
very sharp transitions (500ps, 5V) and wanted to make sure I
did things properly.

Interesting. What's generating the 5v signal? Lots of AC and
Tiny Logic chips are that fast, but might strain to drive 50
ohms. We use several tiny triple buffers in parallel sometimes.
Regular thick-film surface-mount resistors are fine as
terminators at 500 ps.
LVDS line receivers are great at the receive end.

I used a THS3111CD. Split up my project into a timing and driver
board, so i have some 50 ohm BNC cables between and wanted to be
extra safe about reflections at these higher frequencies.

Is the signal some analog thing, or a 10 MHz clock? The THS is an
opamp, but they can make good cable drivers too, even for clocks.
Lately I'm enamored of BUF602, a unity-gain 1 GHz beast.

I had a really hard time finding a good line driver. I might look
into this chip.
Thank you!

Is your signal analog or digital?

digital

I use Tiny Logic triple buffers as line drivers, with all three
sections in parallel, and then sometimes two or three chips.

NL37WZ16US costs 10 cents.

Loading Image...

That US8 package is nasty to solder or probe.

[Phil Hobbs]

On 2025-03-19 10:41, john larkin wrote:> On Wed, 19 Mar 2025 06:23:25

Post by john larkin

Post by Toaster
On Tue, 18 Mar 2025 19:01:51 -0700

Post by john larkin

Post by Toaster
On Tue, 18 Mar 2025 16:02:45 -0700

Post by john larkin

Post by Toaster
On Mon, 17 Mar 2025 19:17:13 -0700

Post by john larkin

Post by Toaster
Thank you for the advice. In my case I have a 10Mhz signal with
very sharp transitions (500ps, 5V) and wanted to make sure I
did things properly.

Interesting. What's generating the 5v signal? Lots of AC and
Tiny Logic chips are that fast, but might strain to drive 50
ohms. We use several tiny triple buffers in parallel sometimes.
Regular thick-film surface-mount resistors are fine as
terminators at 500 ps.
LVDS line receivers are great at the receive end.

I used a THS3111CD. Split up my project into a timing and driver
board, so i have some 50 ohm BNC cables between and wanted to be
extra safe about reflections at these higher frequencies.

Is the signal some analog thing, or a 10 MHz clock? The THS is an
opamp, but they can make good cable drivers too, even for clocks.
Lately I'm enamored of BUF602, a unity-gain 1 GHz beast.

I had a really hard time finding a good line driver. I might look
into this chip.
Thank you!

Is your signal analog or digital?

digital

I use Tiny Logic triple buffers as line drivers, with all three
sections in parallel, and then sometimes two or three chips.
NL37WZ16US costs 10 cents.

https://www.dropbox.com/scl/fi/gw7wetgtovqc04as2gxol/NL37WZ16_5V_Pulse.JPG?rlkey=2eqbyhds8l1myrzfjsrqwn5b3&raw=1

Post by john larkin
That US8 package is nasty to solder or probe.

Just rereading this. John, the prop delay spreads in the datasheet are
all over the place--almost a factor of 2 from typical to max over
temperature.

I'd expect the three sections to match OK, but paralleling packages
seems quite a lot sportier. How well does that work in production?

Cheers

Phil Hobbs

[Gerhard Hoffmann]

Post by Phil Hobbs
On 2025-03-19 10:41, john larkin wrote:> On Wed, 19 Mar 2025 06:23:25

Post by john larkin

Post by Toaster
On Tue, 18 Mar 2025 19:01:51 -0700

Post by john larkin

Post by Toaster
On Tue, 18 Mar 2025 16:02:45 -0700

Post by john larkin

Post by Toaster
On Mon, 17 Mar 2025 19:17:13 -0700

Post by john larkin

Post by Toaster
Thank you for the advice. In my case I have a 10Mhz signal with
very sharp transitions (500ps, 5V) and wanted to make sure I
did things properly.

Interesting. What's generating the 5v signal? Lots of AC and
Tiny Logic chips are that fast, but might strain to drive 50
ohms. We use several tiny triple buffers in parallel sometimes.
Regular thick-film surface-mount resistors are fine as
terminators at 500 ps.
LVDS line receivers are great at the receive end.

I used a THS3111CD. Split up my project into a timing and driver
board, so i have some 50 ohm BNC cables between and wanted to be
extra safe about reflections at these higher frequencies.

Is the signal some analog thing, or a 10 MHz clock? The THS is an
opamp, but they can make good cable drivers too, even for clocks.
Lately I'm enamored of BUF602, a unity-gain 1 GHz beast.

I had a really hard time finding a good line driver. I might look
into this chip.
Thank you!

Is your signal analog or digital?

digital

I use Tiny Logic triple buffers as line drivers, with all three
sections in parallel, and then sometimes two or three chips.
NL37WZ16US costs 10 cents.

https://www.dropbox.com/scl/fi/gw7wetgtovqc04as2gxol/NL37WZ16_5V_Pulse.JPG?rlkey=2eqbyhds8l1myrzfjsrqwn5b3&raw=1

Post by john larkin
That US8 package is nasty to solder or probe.

Just rereading this.  John,  the prop delay spreads in the datasheet are
all over the place--almost a factor of 2 from typical to max over
temperature.
I'd expect the three sections to match OK, but paralleling packages
seems quite a lot sportier.  How well does that work in production?
Cheers
Phil Hobbs

This here are 2 single TI 74LVC inverters, 100 Ohms on each output,
RG174 / RG188-like Coax into the 50 Ohm of an Agilent 2.4GHz scope.

<

Post by Phil Hobbs
+ 2 pics to the right.

Sorry for the blue on black trace; it is default for input 3 that
happened to be unused. Looks better if downloaded.
Having a GND/VCC pair for each output is friendlier to GND bounce
and temp rise is also smaller in comparison to multichannel gates.

Gerhard

[john larkin]

Post by Gerhard Hoffmann

Post by Phil Hobbs
On 2025-03-19 10:41, john larkin wrote:> On Wed, 19 Mar 2025 06:23:25

Post by john larkin

Post by Toaster
On Tue, 18 Mar 2025 19:01:51 -0700

Post by john larkin

Post by Toaster
On Tue, 18 Mar 2025 16:02:45 -0700

Post by john larkin

Post by Toaster
On Mon, 17 Mar 2025 19:17:13 -0700

Post by john larkin

Post by Toaster
Thank you for the advice. In my case I have a 10Mhz signal with
very sharp transitions (500ps, 5V) and wanted to make sure I
did things properly.

Interesting. What's generating the 5v signal? Lots of AC and
Tiny Logic chips are that fast, but might strain to drive 50
ohms. We use several tiny triple buffers in parallel sometimes.
Regular thick-film surface-mount resistors are fine as
terminators at 500 ps.
LVDS line receivers are great at the receive end.

I used a THS3111CD. Split up my project into a timing and driver
board, so i have some 50 ohm BNC cables between and wanted to be
extra safe about reflections at these higher frequencies.

Is the signal some analog thing, or a 10 MHz clock? The THS is an
opamp, but they can make good cable drivers too, even for clocks.
Lately I'm enamored of BUF602, a unity-gain 1 GHz beast.

I had a really hard time finding a good line driver. I might look
into this chip.
Thank you!

Is your signal analog or digital?

digital

I use Tiny Logic triple buffers as line drivers, with all three
sections in parallel, and then sometimes two or three chips.
NL37WZ16US costs 10 cents.

https://www.dropbox.com/scl/fi/gw7wetgtovqc04as2gxol/NL37WZ16_5V_Pulse.JPG?rlkey=2eqbyhds8l1myrzfjsrqwn5b3&raw=1

Post by john larkin
That US8 package is nasty to solder or probe.

Just rereading this.  John,  the prop delay spreads in the datasheet are
all over the place--almost a factor of 2 from typical to max over
temperature.
I'd expect the three sections to match OK, but paralleling packages
seems quite a lot sportier.  How well does that work in production?
Cheers
Phil Hobbs

This here are 2 single TI 74LVC inverters, 100 Ohms on each output,
RG174 / RG188-like Coax into the 50 Ohm of an Agilent 2.4GHz scope.
<

Post by Phil Hobbs
+ 2 pics to the right.

Sorry for the blue on black trace; it is default for input 3 that
happened to be unused. Looks better if downloaded.
Having a GND/VCC pair for each output is friendlier to GND bounce
and temp rise is also smaller in comparison to multichannel gates.
Gerhard

What scope is that? The gate may be a tad faster.

Splitting the 50r source terminator into 100's has virtues.

[Gerhard Hoffmann]

Post by john larkin
What scope is that? The gate may be a tad faster.
Splitting the 50r source terminator into 100's has virtues.

Agilent 54846B 2.4 GHz, tr = 156 ps
<
https://www.flickr.com/photos/***@N07/50657566657/in/dateposted-public/lightbox/
I use the fast tr=9ps sampler only when really needed.
It would be an invitation to disaster :-)

[Phil Hobbs]

Post by Gerhard Hoffmann

Post by Phil Hobbs
On 2025-03-19 10:41, john larkin wrote:> On Wed, 19 Mar 2025 06:23:25
 >
 >> On Tue, 18 Mar 2025 19:01:51 -0700
 >>
 >>>
 >>>> On Tue, 18 Mar 2025 16:02:45 -0700
 >>>>
 >>>>>
 >>>>>> On Mon, 17 Mar 2025 19:17:13 -0700
 >>>>>>
 >>>>>>>
 >>>>>>>> Thank you for the advice. In my case I have a 10Mhz signal with
 >>>>>>>> very sharp transitions (500ps, 5V) and wanted to make sure I
 >>>>>>>> did things properly.
 >>>>>>>
 >>>>>>> Interesting. What's generating the 5v signal? Lots of AC and
 >>>>>>> Tiny Logic chips are that fast, but might strain to drive 50
 >>>>>>> ohms. We use several tiny triple buffers in parallel sometimes.
 >>>>>>>
 >>>>>>> Regular thick-film surface-mount resistors are fine as
 >>>>>>> terminators at 500 ps.
 >>>>>>>
 >>>>>>> LVDS line receivers are great at the receive end.
 >>>>>>>
 >>>>>>>
 >>>>>>
 >>>>>> I used a THS3111CD. Split up my project into a timing and driver
 >>>>>> board, so i have some 50 ohm BNC cables between and wanted to be
 >>>>>> extra safe about reflections at these higher frequencies.
 >>>>>
 >>>>> Is the signal some analog thing, or a 10 MHz clock? The THS is an
 >>>>> opamp, but they can make good cable drivers too, even for clocks.
 >>>>>
 >>>>> Lately I'm enamored of BUF602, a unity-gain 1 GHz beast.
 >>>>>
 >>>>
 >>>> I had a really hard time finding a good line driver. I might look
 >>>> into this chip.
 >>>>
 >>>> Thank you!
 >>>
 >>> Is your signal analog or digital?
 >>>
 >>
 >> digital
 >
 > I use Tiny Logic triple buffers as line drivers, with all three
 > sections in parallel, and then sometimes two or three chips.
 >
 > NL37WZ16US costs 10 cents.
 >
 >
https://www.dropbox.com/scl/fi/gw7wetgtovqc04as2gxol/NL37WZ16_5V_Pulse.JPG?rlkey=2eqbyhds8l1myrzfjsrqwn5b3&raw=1
 >
 > That US8 package is nasty to solder or probe.
 >
Just rereading this.  John,  the prop delay spreads in the datasheet
are all over the place--almost a factor of 2 from typical to max over
temperature.
I'd expect the three sections to match OK, but paralleling packages
seems quite a lot sportier.  How well does that work in production?
Cheers
Phil Hobbs

This here are 2 single TI 74LVC inverters, 100 Ohms on each output,
RG174 / RG188-like Coax into the 50 Ohm of an Agilent 2.4GHz scope.
<
     >    + 2 pics to the right.
Sorry for the blue on black trace; it is default for input 3 that
happened to be unused. Looks better if downloaded.
Having a GND/VCC pair for each output is friendlier to GND bounce
and temp rise is also smaller in comparison to multichannel gates.
Gerhard

Thanks. We do similar things, but only with single packages--we often
drive fast (500 ps to 5 ns) LEDs with 74AHC04s or 74LVC04s in QFN.
(QFNs are dramatically faster for this than TSSOPs or SOICs, on account
of the much lower package inductance in the power and ground.)

For that job, the trick is to parallel four sections with one resistor
at the output, and drive them with the other two sections in cascade, to
get a bunch of voltage gain. That sharpens up slow input edges very
nicely. (In bad cases, one might want a 1G14 Schmitt in there as well.)

Cheers

Phil Hobbs

[john larkin]

On Sat, 29 Mar 2025 19:53:04 -0400, Phil Hobbs

Post by Phil Hobbs
On 2025-03-19 10:41, john larkin wrote:> On Wed, 19 Mar 2025 06:23:25

Post by john larkin

Post by Toaster
On Tue, 18 Mar 2025 19:01:51 -0700

Post by john larkin

Post by Toaster
On Tue, 18 Mar 2025 16:02:45 -0700

Post by john larkin

Post by Toaster
On Mon, 17 Mar 2025 19:17:13 -0700

Post by john larkin

Post by Toaster
Thank you for the advice. In my case I have a 10Mhz signal with
very sharp transitions (500ps, 5V) and wanted to make sure I
did things properly.

Interesting. What's generating the 5v signal? Lots of AC and
Tiny Logic chips are that fast, but might strain to drive 50
ohms. We use several tiny triple buffers in parallel sometimes.
Regular thick-film surface-mount resistors are fine as
terminators at 500 ps.
LVDS line receivers are great at the receive end.

I used a THS3111CD. Split up my project into a timing and driver
board, so i have some 50 ohm BNC cables between and wanted to be
extra safe about reflections at these higher frequencies.

Is the signal some analog thing, or a 10 MHz clock? The THS is an
opamp, but they can make good cable drivers too, even for clocks.
Lately I'm enamored of BUF602, a unity-gain 1 GHz beast.

I had a really hard time finding a good line driver. I might look
into this chip.
Thank you!

Is your signal analog or digital?

digital

I use Tiny Logic triple buffers as line drivers, with all three
sections in parallel, and then sometimes two or three chips.
NL37WZ16US costs 10 cents.

https://www.dropbox.com/scl/fi/gw7wetgtovqc04as2gxol/NL37WZ16_5V_Pulse.JPG?rlkey=2eqbyhds8l1myrzfjsrqwn5b3&raw=1

Post by john larkin
That US8 package is nasty to solder or probe.

Just rereading this. John, the prop delay spreads in the datasheet are
all over the place--almost a factor of 2 from typical to max over
temperature.
I'd expect the three sections to match OK, but paralleling packages
seems quite a lot sportier. How well does that work in production?
Cheers
Phil Hobbs

I have one product with three chips, nine gates, in parallel. But each
package has its own output resistor, so they sorta share nicely.

Seems to work. Maybe parts on the same reel match pretty well.

I'd love to have a chip that was specifically a brutal fast driver.

[Bill Sloman]

Post by Toaster
On Tue, 18 Mar 2025 19:01:51 -0700

Post by john larkin

Post by Toaster
On Tue, 18 Mar 2025 16:02:45 -0700

Post by john larkin

Post by Toaster
On Mon, 17 Mar 2025 19:17:13 -0700

Post by john larkin

Post by Toaster
Thank you for the advice. In my case I have a 10Mhz signal with
very sharp transitions (500ps, 5V) and wanted to make sure I
did things properly.

Interesting. What's generating the 5v signal? Lots of AC and
Tiny Logic chips are that fast, but might strain to drive 50
ohms. We use several tiny triple buffers in parallel sometimes.
Regular thick-film surface-mount resistors are fine as
terminators at 500 ps.
LVDS line receivers are great at the receive end.

I used a THS3111CD. Split up my project into a timing and driver
board, so i have some 50 ohm BNC cables between and wanted to be
extra safe about reflections at these higher frequencies.

Is the signal some analog thing, or a 10 MHz clock? The THS is an
opamp, but they can make good cable drivers too, even for clocks.
Lately I'm enamored of BUF602, a unity-gain 1 GHz beast.

I had a really hard time finding a good line driver. I might look
into this chip.
Thank you!

Is your signal analog or digital?

digital

5V is a big swing for a modern digital system, but there are lots of
fast switching transistors out there that can cope with a 5V swing.

Discrete surface mount devices can be pretty compact, and there are some
fast integrated circuit devices designed to drive them.

Some of the ECL-to-TTL level shifters did generate a very fast full 0V
to 5V swing. I got stuck with up-dating a very fast TTL-based timing
circuit in the early 1990's, and used a bit of ECLinPS ECL to get rid of
the usual TTL faults, and used 100k ECL-to-TTL converters to push out
the TTL house-keeping signals.

They were a lot better than the original TTL signals

It involved adding -4.5V rail to drive the ECL, but with surface mount
parts we could squeeze the additional stuff onto same sized printed
circuit board that the original system had used.

[Toaster]

On Thu, 20 Mar 2025 01:41:29 +1100

Post by Bill Sloman

Post by Toaster
On Tue, 18 Mar 2025 19:01:51 -0700

Post by john larkin

Post by Toaster
On Tue, 18 Mar 2025 16:02:45 -0700

Post by john larkin

Post by Toaster
On Mon, 17 Mar 2025 19:17:13 -0700

Post by john larkin

Post by Toaster
Thank you for the advice. In my case I have a 10Mhz signal
with very sharp transitions (500ps, 5V) and wanted to make
sure I did things properly.

Interesting. What's generating the 5v signal? Lots of AC and
Tiny Logic chips are that fast, but might strain to drive 50
ohms. We use several tiny triple buffers in parallel sometimes.
Regular thick-film surface-mount resistors are fine as
terminators at 500 ps.
LVDS line receivers are great at the receive end.

I used a THS3111CD. Split up my project into a timing and driver
board, so i have some 50 ohm BNC cables between and wanted to be
extra safe about reflections at these higher frequencies.

Is the signal some analog thing, or a 10 MHz clock? The THS is an
opamp, but they can make good cable drivers too, even for clocks.
Lately I'm enamored of BUF602, a unity-gain 1 GHz beast.

I had a really hard time finding a good line driver. I might look
into this chip.
Thank you!

Is your signal analog or digital?

digital

5V is a big swing for a modern digital system, but there are lots of
fast switching transistors out there that can cope with a 5V swing.
Discrete surface mount devices can be pretty compact, and there are
some fast integrated circuit devices designed to drive them.
Some of the ECL-to-TTL level shifters did generate a very fast full
0V to 5V swing. I got stuck with up-dating a very fast TTL-based
timing circuit in the early 1990's, and used a bit of ECLinPS ECL to
get rid of the usual TTL faults, and used 100k ECL-to-TTL converters
to push out the TTL house-keeping signals.
They were a lot better than the original TTL signals
It involved adding -4.5V rail to drive the ECL, but with surface
mount parts we could squeeze the additional stuff onto same sized
printed circuit board that the original system had used.

fyi, im an amateur and may have made a mistake in my design, reason for
5V is so the logic chip in the other end gets around 2.5V after going
through the two 50 ohm resistors (voltage divider) and can trigger.

From what everyone is saying I dont even need to do that and can get a
reliable termination just by using one 50 ohm resistor and avoid
dealing with the voltage divider side effect.

the hard part is coming up...soldering these tiny smd components...i
bought a microscope and a little platform to hold the boards. going to
try hot air soldering as i havent shelled out for a reflow oven yet.

the logic chips im using in my project are 74VHC series.

[Bill Sloman]

Post by Toaster
On Thu, 20 Mar 2025 01:41:29 +1100

Post by Bill Sloman

Post by Toaster
On Tue, 18 Mar 2025 19:01:51 -0700

Post by john larkin

Post by Toaster
On Tue, 18 Mar 2025 16:02:45 -0700

Post by john larkin

Post by Toaster
On Mon, 17 Mar 2025 19:17:13 -0700

Post by john larkin

Post by Toaster
Thank you for the advice. In my case I have a 10Mhz signal
with very sharp transitions (500ps, 5V) and wanted to make
sure I did things properly.

Interesting. What's generating the 5v signal? Lots of AC and
Tiny Logic chips are that fast, but might strain to drive 50
ohms. We use several tiny triple buffers in parallel sometimes.
Regular thick-film surface-mount resistors are fine as
terminators at 500 ps.
LVDS line receivers are great at the receive end.

I used a THS3111CD. Split up my project into a timing and driver
board, so i have some 50 ohm BNC cables between and wanted to be
extra safe about reflections at these higher frequencies.

Is the signal some analog thing, or a 10 MHz clock? The THS is an
opamp, but they can make good cable drivers too, even for clocks.
Lately I'm enamored of BUF602, a unity-gain 1 GHz beast.

I had a really hard time finding a good line driver. I might look
into this chip.
Thank you!

Is your signal analog or digital?

digital

5V is a big swing for a modern digital system, but there are lots of
fast switching transistors out there that can cope with a 5V swing.
Discrete surface mount devices can be pretty compact, and there are
some fast integrated circuit devices designed to drive them.
Some of the ECL-to-TTL level shifters did generate a very fast full
0V to 5V swing. I got stuck with up-dating a very fast TTL-based
timing circuit in the early 1990's, and used a bit of ECLinPS ECL to
get rid of the usual TTL faults, and used 100k ECL-to-TTL converters
to push out the TTL house-keeping signals.
They were a lot better than the original TTL signals
It involved adding -4.5V rail to drive the ECL, but with surface
mount parts we could squeeze the additional stuff onto same sized
printed circuit board that the original system had used.

fyi, im an amateur and may have made a mistake in my design, reason for
5V is so the logic chip in the other end gets around 2.5V after going
through the two 50 ohm resistors (voltage divider) and can trigger.
From what everyone is saying I dont even need to do that and can get a
reliable termination just by using one 50 ohm resistor and avoid
dealing with the voltage divider side effect.
the hard part is coming up...soldering these tiny smd components...i
bought a microscope and a little platform to hold the boards. going to
try hot air soldering as i haven't shelled out for a reflow oven yet.
the logic chips im using in my project are 74VHC series.

I can sympathise with the difficulty of coming to terms with soldering
smd chips. Around 1989 I got stuck with introducing smd parts to
Cambridge Instruments in the UK, because the GaAs chips that I needed to
use only came in surface mount packages. We bought a fairly expensive
Groatmore hot-air reflow machine that would reflow individual packages.

When I wanted to use similar parts - Motorola ECLinPS devices - at
Nijmegen University in the Netherlands, nearly ten years later, they
just bought a much cheaper and smaller work station (but didn't let me
use it). At Haffmans BV in the Netherlands around 2002 I just used a
fine tipped soldering iron under a cheap binocular microscope.

It was fiddly work, but perfectly practical.

[Toaster]

On Thu, 20 Mar 2025 16:28:38 +1100

Post by Bill Sloman

Post by Toaster
On Thu, 20 Mar 2025 01:41:29 +1100

Post by Bill Sloman

Post by Toaster
On Tue, 18 Mar 2025 19:01:51 -0700

Post by john larkin

Post by Toaster
On Tue, 18 Mar 2025 16:02:45 -0700

Post by john larkin

Post by Toaster
On Mon, 17 Mar 2025 19:17:13 -0700

On Mon, 17 Mar 2025 19:50:17 -0400, Toaster

Post by Toaster
Thank you for the advice. In my case I have a 10Mhz signal
with very sharp transitions (500ps, 5V) and wanted to make
sure I did things properly.

Interesting. What's generating the 5v signal? Lots of AC and
Tiny Logic chips are that fast, but might strain to drive 50
ohms. We use several tiny triple buffers in parallel
sometimes.
Regular thick-film surface-mount resistors are fine as
terminators at 500 ps.
LVDS line receivers are great at the receive end.

I used a THS3111CD. Split up my project into a timing and
driver board, so i have some 50 ohm BNC cables between and
wanted to be extra safe about reflections at these higher
frequencies.

Is the signal some analog thing, or a 10 MHz clock? The THS is
an opamp, but they can make good cable drivers too, even for
clocks.
Lately I'm enamored of BUF602, a unity-gain 1 GHz beast.

I had a really hard time finding a good line driver. I might
look into this chip.
Thank you!

Is your signal analog or digital?

digital

5V is a big swing for a modern digital system, but there are lots
of fast switching transistors out there that can cope with a 5V
swing.
Discrete surface mount devices can be pretty compact, and there are
some fast integrated circuit devices designed to drive them.
Some of the ECL-to-TTL level shifters did generate a very fast full
0V to 5V swing. I got stuck with up-dating a very fast TTL-based
timing circuit in the early 1990's, and used a bit of ECLinPS ECL
to get rid of the usual TTL faults, and used 100k ECL-to-TTL
converters to push out the TTL house-keeping signals.
They were a lot better than the original TTL signals
It involved adding -4.5V rail to drive the ECL, but with surface
mount parts we could squeeze the additional stuff onto same sized
printed circuit board that the original system had used.

fyi, im an amateur and may have made a mistake in my design, reason
for 5V is so the logic chip in the other end gets around 2.5V after
going through the two 50 ohm resistors (voltage divider) and can
trigger.
From what everyone is saying I dont even need to do that and can
get a reliable termination just by using one 50 ohm resistor and
avoid dealing with the voltage divider side effect.
the hard part is coming up...soldering these tiny smd components...i
bought a microscope and a little platform to hold the boards. going
to try hot air soldering as i haven't shelled out for a reflow oven
yet.
the logic chips im using in my project are 74VHC series.

I can sympathise with the difficulty of coming to terms with
soldering smd chips. Around 1989 I got stuck with introducing smd
parts to Cambridge Instruments in the UK, because the GaAs chips that
I needed to use only came in surface mount packages. We bought a
fairly expensive Groatmore hot-air reflow machine that would reflow
individual packages.
When I wanted to use similar parts - Motorola ECLinPS devices - at
Nijmegen University in the Netherlands, nearly ten years later, they
just bought a much cheaper and smaller work station (but didn't let
me use it). At Haffmans BV in the Netherlands around 2002 I just used
a fine tipped soldering iron under a cheap binocular microscope.
It was fiddly work, but perfectly practical.

Very fiddly, my hands shake too much for stuff this fine. I have a
driver and a mosfet:

BD2311NVX-LBE2 - driver
GAN190-650FBE - mosfet

That driver is so tiny I'm wondering if the 10Mhz switching speeds are
worth it! Researching some strange field interactions so I needed
something that could switch a decent voltage (300V) at repetition rates
close to 10Mhz. Odd requirements but until I can narrow down parameters
I need to sweep up the frequency range as far as I can.

Thanks for all of your help.

[Bill Sloman]

Post by Toaster
On Thu, 20 Mar 2025 16:28:38 +1100

Post by Toaster
On Thu, 20 Mar 2025 01:41:29 +1100

Post by Toaster
On Tue, 18 Mar 2025 19:01:51 -0700

Post by Toaster
On Tue, 18 Mar 2025 16:02:45 -0700

Post by Toaster
On Mon, 17 Mar 2025 19:17:13 -0700

On Mon, 17 Mar 2025 19:50:17 -0400, Toaster

<snip>

Post by Toaster
That driver is so tiny I'm wondering if the 10Mhz switching speeds are
worth it! Researching some strange field interactions so I needed
something that could switch a decent voltage (300V) at repetition rates
close to 10Mhz. Odd requirements but until I can narrow down parameters
I need to sweep up the frequency range as far as I can.

300V is a lot.

When I was blanking and unblanking an an electron beam in a electron
microscope, the fast version - which turned the beam on for just 0.5nsec
with 100psec transition times - only generated +/-7.5V.

The slow version - in an electron beam microfabricator - had to swing
from 60V to OV at up to 10 MHz, because we couldn't get the blanking
plates all that close to the beam in the electron beam microfabrictor
(which was a million dollar machine, and we only sold a handful per year
- ten per year in the good year).

300V shouldn't be all that difficult, but getting it to happen fast
means charging up the capacitances in the MOSFET switch calls for quite
a lot of current, and turning it on and off even faster.

[Phil Hobbs]

Post by Toaster
Can I get away with terminating the far end of a transmission line or
do I need a series resistor at the source and a parallel resistor at
the destination?
Thank you,
Toaster

Depends.

In general, when you just have one source and one load, with no taps along
the way, the best approach is to series-terminate the source.

That is, you put a 50Ω resistor in series with the output connector, and
let the load be essentially an open circuit.

The load sees the full signal amplitude, and there’s no huge power
dissipation.

Anyplace in between, the waveform is more complicated.

Cheers

Phil Hobbs