Topics

REVISED, UPDATED, EXPANDED, and IMPROVED Evaluation of the Tektronix 012-0482-00 cable for the SG503

 

I uploaded a completely revised and improved version of my paper on the 012-
0482-00 cable for the SG503 signal generator.

It is located at

https://groups.io/g/TekScopes/files/Evaluation%20of%20the%20Tektronix%20012-
0482-00%20cable%20for%20the%20SG503%20REVISED%20and%20UPDATED.pdf



It now contains the EXACT SPECIFICATIONS Tek defined for the cable. In
addition I evaluated several 50Ω terminators and I show how to build a
simple 50ohm terminator in 15 minutes that outperforms most of the ones you
can buy.



Dennis Tillman W7PF

Jim Ford
 

Uh, Dennis, I don't see the 50 ohm termination evaluations, nor the 15-minute high-performance termination in the .pdf.  Do d I miss something?Thanks.Jim Ford Sent from my Verizon, Samsung Galaxy smartphone

-------- Original message --------From: Dennis Tillman W7PF <@Dennis_Tillman_W7PF> Date: 5/25/19 8:35 PM (GMT-08:00) To: TekScopes@groups.io Subject: [TekScopes] REVISED, UPDATED, EXPANDED, and IMPROVED Evaluation of the Tektronix 012-0482-00 cable for the SG503 I uploaded a completely revised and improved version of my paper on the 012-0482-00 cable for the SG503 signal generator.It is located at https://groups.io/g/TekScopes/files/Evaluation%20of%20the%20Tektronix%20012-0482-00%20cable%20for%20the%20SG503%20REVISED%20and%20UPDATED.pdf It now contains the EXACT SPECIFICATIONS Tek defined for the cable. Inaddition I evaluated several 50Ω terminators and I show how to build asimple 50ohm terminator in 15 minutes that outperforms most of the ones youcan buy. Dennis Tillman W7PF

Craig Sawyers
 

I uploaded a completely revised and improved version of my paper on the 012-
0482-00 cable for the SG503 signal generator.

It is located at

https://groups.io/g/TekScopes/files/Evaluation%20of%20the%20Tektronix%20012-
0482-00%20cable%20for%20the%20SG503%20REVISED%20and%20UPDATED.pdf



It now contains the EXACT SPECIFICATIONS Tek defined for the cable. In addition I evaluated
several
50 terminators and I show how to build a simple 50ohm terminator in 15 minutes that outperforms
most of the ones you can buy.



Dennis Tillman W7PF
That is really, really interesting Dennis. Your homebrew termination reminded me of a homemade 1GHz
Z0 probe described here http://emcesd.com/1ghzprob.htm

Smith makes the point that his probe could be made even better by using microwave resistors, and it
would be interesting to discover if they would make an even better 50-ohm termination.

Craig

 

Hi Jim,
It starts on page 10 and runs to the end of the document (page 17)
Dennis

-----Original Message-----
From: Jim Ford, Sent: Saturday, May 25, 2019 10:31 PM

Uh, Dennis, I don't see the 50 ohm termination evaluations, nor the 15-minute high-performance termination in the .pdf. Do d I miss something?

Thanks.Jim Ford
--
Dennis Tillman W7PF
TekScopes Moderator

 

Hi Craig,
Yes, that homebrew 1GHz probe article at http://emcesd.com/1ghzprob.htm was
one of the most useful things I ever came across.
Everyone should read it. It is so practical. It overcame my fear of high
frequency design. It influenced a lot of things I have made since.

That article, and another one I read on how to make the perfect terminator,
have made it possible for me to improve my TDR skills to the point where I
am at the most sensitive scale on the TDR and able to see some very subtle
stuff. The 7S12 TDR is a wonderful plugin. It is definitely one of my very
favorite things to use. Extremely fast pulses have always fascinated me, and
when I discovered they can tell you so much about a cable I was hooked. I
check every new cable I get on a TDR before I will use it.

The article on terminator design contained a very simple, practical
explanation of how to arrive at the perfect shape.
I forgot all of the details but it goes something like this:
Start by attaching a 50 ohm resistor across the two contacts of a BNC
connector.
A TDR will show you the resistor has lead inductance and capacitance to the
BNC body. The result is big reflections and poor HF response.
That can be improved if you use two 100 ohm resistors in parallel across the
BNC. That improves the inductance but adds capacitance from the bulk of each
resistor to the BNC. So the inductance went down a little but the
capacitance went up a little.
Next try four 200 ohm resistors in parallel. Spread them around the center
pin to the outer ground to lower the capacitance between them. But there is
now a little more capacitance and a little less inductance.
On and on. Eventually you end up with a circular cylinder of resistive
material. The cylinder diameter reaches the outer ground connector of the
BNC and it has a tiny hole in the middle that connects it to the center pin
of the BNC.
The cylinder has no lead inductance but it has a lot of bulk capacitance. If
you turn the cylinder to a thin disk you will reduce the bulk capacitance to
an absolute minimum.
Now there is no more lead inductance and the only capacitance between the
center pin and the outer ground lead is swamped by the disk's intrinsic 50
ohm resistance.
A thin resistive disk turns out to be the ideal shape for the resistance of
a terminator. It applies to any kind of coax connector.

Notice that all of the so called inexpensive terminators use a single 50 ohm
resistor inside them. Many times you can spot those easily from their shape.

They are never going to have a good response. They will have the worst
possible response a terminator can have.

Microwave resistors might make a small improvement to the terminator but
since a BNC has very limited frequency response, 2GHz typical or up to 4GHz
best case, that will prevent microwave resistors from making a big
difference. The benefit to BNC connectors is they are big enough to work on
and there is enough room to put a few surface mount resistors around their
center pin.

You have to switch to SMA connectors to get higher performance. Their
diminutive size makes it impractical to build your own terminators without a
really good surface mount setup.

Dennis Tillman W7PF

-----Original Message-----
From: Craig Sawyers, Sent: Saturday, May 25, 2019 11:57 PM
Subject: Re: [TekScopes] REVISED, UPDATED, EXPANDED, and IMPROVED Evaluation
of the Tektronix 012-0482-00 cable for the SG503

<snip>

That is really, really interesting Dennis. Your homebrew termination
reminded me of a homemade 1GHz
Z0 probe described here http://emcesd.com/1ghzprob.htm

Smith makes the point that his probe could be made even better by using
microwave resistors, and it would be interesting to discover if they would
make an even better 50-ohm termination.

Craig
--
Dennis Tillman W7PF
TekScopes Moderator

Craig Sawyers
 

Next try four 200 ohm resistors in parallel. Spread them around the center pin to the outer ground
to
lower the capacitance between them. But there is now a little more capacitance and a little less
inductance.
On and on. Eventually you end up with a circular cylinder of resistive material.
Dennis Tillman W7PF
Quite a long time ago I bought some disc 50-ohm resistors that have a hole that takes a centre pin
of a BNC. Bought them from https://www.surplussales.com/ ages ago.

It would be interesting to see what the HF performance was as compared with four discrete surface
mount.

Craig

Bob Koller
 

I'd be curious what the ubiquitous 50 ohm feed through termination looks like under similar testing. Many scopes don't have a 50 ohm input setting requiring the use of a feed through.

hpxref
 

Craig said
<<Quite a long time ago I bought some disc 50-ohm resistors that have a hole that takes a centre pin
of a BNC. Bought them from https://www.surplussales.com/ ages ago.>>

As a teenager (15) I bought a very nice piece of WW11 surplus, contained within an olive green metal case with hinged lid, a variable coaxial
wave guide to 750MHz with a point contact microwave crystal detector , two short lengths of N connector coaxial cables clipped into the hinged cover,
the usual round micro ammeter calibrated in mW and a circular plastic calculator attached the lid, and a spare crystal detector diode
Probably a piece of radar test gear
The unit was lost many years ago on a move so have forgotten its military (APN?) number. It was beautifully made , I think by Westinghouse,
and had a battery compartment.
Whats relevant to Craig's post was that one of the short coax cables measured 50 Ohms . On dissembling as I at first thought, child like, it was faulty,
(the cable sheath was extra secured to its N connector by three tiny screws to an internal silvered ring through the barrel ) I found a circular
carbon button press fitted to the bottom of the N connector with a silvered pin pushed through the hole in its centre.
This was the 50 Ohm terminator resistance, so this form of termination was known during WW11.
I first used it to measure the output of an RF UHF signal generator I had made at the time and it worked well enough to tell me that I needed a piston attenuator
to replace the hopeless shielded rotary switch one I had innocently implemented.
So there nothing new under the sun, even N connectors had them then and the microwave engineers of the time really knew there stuff.
Seems to me this form of termination would be superior to a radially spoked terminator.

John

 

Hi John,
The ideal shape for a 50 ohm terminator is exactly like what you described at the bottom of the N connector that you found as a child: a disk with the same diameter as the connector's outer shell with a metallized hole in the middle for the center conductor to be soldered to and metallization around the edge so that can be soldered all around to the shell of the connector. The disk should only be thick enough to be able to dissipate the power that will be applied.
Dennis Tillman W7PF

-----Original Message-----
From: hpxref, Sent: Sunday, May 26, 2019 9:46 PM
Subject: Re: [TekScopes] REVISED, UPDATED, EXPANDED, and IMPROVED Evaluation of the Tektronix 012-0482-00 cable for the SG503

Craig said
<<Quite a long time ago I bought some disc 50-ohm resistors that have a hole that takes a centre pin of a BNC. Bought them from https://www.surplussales.com/ ages ago.>>

As a teenager (15) I bought a very nice piece of WW11 surplus, contained within an olive green metal case with hinged lid, a variable coaxial wave guide to 750MHz with a point contact microwave crystal detector , two short lengths of N connector coaxial cables clipped into the hinged cover, the usual round micro ammeter calibrated in mW and a circular plastic calculator attached the lid, and a spare crystal detector diode Probably a piece of radar test gear The unit was lost many years ago on a move so have forgotten its military (APN?) number. It was beautifully made , I think by Westinghouse, and had a battery compartment.
Whats relevant to Craig's post was that one of the short coax cables measured 50 Ohms . On dissembling as I at first thought, child like, it was faulty,
(the cable sheath was extra secured to its N connector by three tiny screws to an internal silvered ring through the barrel ) I found a circular
carbon button press fitted to the bottom of the N connector with a silvered pin pushed through the hole in its centre.
This was the 50 Ohm terminator resistance, so this form of termination was known during WW11.
I first used it to measure the output of an RF UHF signal generator I had made at the time and it worked well enough to tell me that I needed a piston attenuator to replace the hopeless shielded rotary switch one I had innocently implemented.
So there nothing new under the sun, even N connectors had them then and the microwave engineers of the time really knew there stuff.
Seems to me this form of termination would be superior to a radially spoked terminator.

John
--
Dennis Tillman W7PF
TekScopes Moderator

Adrian
 

We had an old box full of those in the Cavendish Radio Astronomy stores back in the 70's, they had "Erie Great Yarmouth, England" on the box IIRC. Had them in both 50 and 75 ohms, color coded red for 75 and green for 50 if memory serves. I tried them as 1.4GHz cold loads dunked in LN2 - they didn't like it one bit!

On 5/28/2019 9:29 PM, Dennis Tillman W7PF wrote:
Hi John,
The ideal shape for a 50 ohm terminator is exactly like what you described at the bottom of the N connector that you found as a child: a disk with the same diameter as the connector's outer shell with a metallized hole in the middle for the center conductor to be soldered to and metallization around the edge so that can be soldered all around to the shell of the connector.

 

Hi Craig,
I looked at the Surplus Sales link you included. Right now all he has is
non-inductive thick disc shaped resistors with contacts on either side.
There is a very big hole in the middle so these are not something that could
be used. They are meant to be wired up from each side.

The disk we need must have a connection in the middle and all along the
perimeter. What we would be looking for was a disc about the diameter of the
BNC shell with silver or metal deposited all around the outer edge that
could be soldered to. There would need to be a small hole in the middle with
silver or metal deposited to form a contact that could be soldered to. The
resistive material would preferably be ceramic rather than carbon composite.
Ceramic would be stronger.

If the disks you bought fit that description send me two and I will mount
them up and test them.
Contact me off list for my address.

Dennis

-----Original Message-----
From: Craig Sawyers, Sent: Sunday, May 26, 2019 11:08 AM

Next try four 200 ohm resistors in parallel. Spread them around the
center pin to the outer ground
to
lower the capacitance between them. But there is now a little more
capacitance and a little less inductance.
On and on. Eventually you end up with a circular cylinder of resistive
material.
Dennis Tillman W7PF
Quite a long time ago I bought some disc 50-ohm resistors that have a hole
that takes a centre pin of a BNC. Bought them from
https://www.surplussales.com/ ages ago.

It would be interesting to see what the HF performance was as compared with
four discrete surface mount.

Craig
--
Dennis Tillman W7PF
TekScopes Moderator