Topics

067-0525-01 or equivalent needed.


Eric
 

I have some high quality T's on order so I will be able to get this done. Thanks for all the advice.

-----Original Message-----
From: TekScopes@groups.io <TekScopes@groups.io> On Behalf Of Dennis Tillman W7pF
Sent: Thursday, November 12, 2020 4:49 PM
To: TekScopes@groups.io
Subject: Re: [TekScopes] 067-0525-01 or equivalent needed.

Hi Eric,
As several people have pointed out already, the 1MHz bandwidth of the 7A22 Differential Amplifier is too low to require a T-cable matched to 0.1".

However, it occurred to me that you may have confused the 7A22 with the 7A13 Differential Amplifier. The 7A13 bandwidth is 100+MHz and it definitely does require a T-cable like the 067-0525 to test and adjust the plugin's high frequency common mode response.

Dennis Tillman W7pF

-----Original Message-----
From: TekScopes@groups.io [mailto:TekScopes@groups.io] On Behalf Of Eric
Sent: Wednesday, November 11, 2020 12:01 PM
To: TekScopes@groups.io
Subject: [TekScopes] 067-0525-01 or equivalent needed.

I need this calibration fixture or something like it to calibrate a 7A22 and some 400 series scopes. This fixture is a BNC Y cable the important part is the Y side of the cable is matched length to within 0.1 inch. This is to remove cable aberrations and phase difference to the inputs. I don’t necessarily need the 067-0525 if I can order a matched length Y cable from somewhere.




--
Dennis Tillman W7pF
TekScopes Moderator


Jean-Paul
 

We use a good quality BNC T and a pair of Tektronix precision 50 ohm 0.5 m cables, the type specific to SG503.

A 2-10X wideband passif attenuator like Mini Circuits HAT series can reduce abberations due to VSWR and reflection.

Jon


SCMenasian
 

4 GR874 to BNC adapters could be connected (in 2 pairs) to create two variable length lines. The 874 engagement can be adjusted slightly to create the desired delay.


 

On Fri, Nov 13, 2020 at 12:07 AM, Raymond Domp Frank wrote:


Experimenting with a line length difference of between 0.2" and 0.4" will
probably allow you to either double or halve the skew of 28 .. 40 ps.
Loosening the bayonet and slightly pulling the plug out will have a visible effect.

Raymond


 

On Thu, Nov 12, 2020 at 11:13 PM, gifsim wrote:


I zoomed in and with the cursors I read a difference of about 28 ps
tomorrow I will try to slightly vary the length of one of the lines
to see what happens
Experimenting with a line length difference of between 0.2" and 0.4" will probably allow you to either double or halve the skew of 28 .. 40 ps.
Have fun calibrating your 784A!

Raymond


Gif Sim
 

thanks Raymond

I'm no longer in the lab
from the latest tests
I zoomed in and with the cursors I read a difference of about 28 ps
tomorrow I will try to slightly vary the length of one of the lines
to see what happens
this scope is a tds754D upgraded (with resistors and capacitors) to tds784D .... now it is with original calibration data of the 754D at 500 Mhz ..... it must do the full calibration at 1024 mhz of the 784d ..... .
these days I'm having fun calibrating a 754A upgraded to 784A

a big greeting
Simon


 

Hi Eric,
As several people have pointed out already, the 1MHz bandwidth of the 7A22 Differential Amplifier is too low to require a T-cable matched to 0.1".

However, it occurred to me that you may have confused the 7A22 with the 7A13 Differential Amplifier. The 7A13 bandwidth is 100+MHz and it definitely does require a T-cable like the 067-0525 to test and adjust the plugin's high frequency common mode response.

Dennis Tillman W7pF

-----Original Message-----
From: TekScopes@groups.io [mailto:TekScopes@groups.io] On Behalf Of Eric
Sent: Wednesday, November 11, 2020 12:01 PM
To: TekScopes@groups.io
Subject: [TekScopes] 067-0525-01 or equivalent needed.

I need this calibration fixture or something like it to calibrate a 7A22 and some 400 series scopes. This fixture is a BNC Y cable the important part is the Y side of the cable is matched length to within 0.1 inch. This is to remove cable aberrations and phase difference to the inputs. I don’t necessarily need the 067-0525 if I can order a matched length Y cable from somewhere.




--
Dennis Tillman W7pF
TekScopes Moderator


 

On Thu, Nov 12, 2020 at 09:19 PM, gifsim wrote:


I wish I had a TDR !!
I put some photos with a test (@ 800 mhz +0 dBm) just for fun
the strange thing is that even if I invert the two bnc of the coupler the
tracks do not move (always first the white track and then the green)
I didn't understand this thing .......
You're triggering on CH1, the white trace. If swapping the split-cable plugs between CH1 and CH2 results in the same image, i.e. CH1 the same amount before CH2, it means:
1. The cable's signal lengths for both plugs are about equal
2. What you're seeing is the delay between both channels in the 'scope and that's exactly what the cable is for, when adjusting the 'scope! The difference is about 40 ps.

Raymond


Gif Sim
 

hello ed
I wish I had a TDR !!
I put some photos with a test (@ 800 mhz +0 dBm) just for fun
the strange thing is that even if I invert the two bnc of the coupler the tracks do not move (always first the white track and then the green)
I didn't understand this thing .......

https://groups.io/g/TekScopes/album?id=256551


Ed Breya
 

I have to put my one-cent-worth in here. I'd vote for the simplest suggestion of using a BNC Tee with "identical" cables connected, out to the DUT. If you really want to build something "permanent" or dedicated, it should be easy to do, with a small enclosure like a Pomona box, or just soldering the leads all together in a neat splice that can be tape-wrapped or heat-shrinked up, and even foil-wrapped if shielding is desired. To make cables with BNC at one end and wires to solder at the other, get a single high grade BNC-BNC cable assembly a little over twice the length needed, and cut it at the center. The final matching can be trimmed as close as you want and can measure (even TDR if capable), and you'll know that both pieces are the exact same kind of cable.

Ed


Gif Sim
 

yes colin ...... building it as tek is almost impossible ... very very precise ...
however, you can have fun getting close to their accuracy

https://groups.io/g/TekScopes/photo/256562/0?p=Created,,,20,2,0,0


Colin Herbert
 

Sorry, I made an error in deleting some text. It should read:

"a BNC tee-piece with three female connectors, plus two equal lengths of BNC male to BNC male coax could be used."

Colin.

-----Original Message-----
From: TekScopes@groups.io [mailto:TekScopes@groups.io] On Behalf Of Colin Herbert via groups.io
Sent: 12 November 2020 17:24
To: TekScopes@groups.io
Subject: Re: [TekScopes] 067-0525-01 or equivalent needed.

I have two of these (none for sale!). The tricky bit in "building" one would be in sourcing the "joint" which has a female BNC and the two identical coax wires going to the two male BNCs. Of course, if you are talking about something similar which will do the job, then a BNC tee-piece with one three female connectors, plus two equal lengths of BNC male to BNC male coax could be used.
Colin.

-----Original Message-----
From: TekScopes@groups.io [mailto:TekScopes@groups.io] On Behalf Of Glenn Little
Sent: 11 November 2020 21:35
To: TekScopes@groups.io
Subject: Re: [TekScopes] 067-0525-01 or equivalent needed.

Why not build one?

Glenn

On 11/11/2020 3:01 PM, Eric wrote:
I need this calibration fixture or something like it to calibrate a 7A22 and some 400 series scopes. This fixture is a BNC Y cable the important part is the Y side of the cable is matched length to within 0.1 inch. This is to remove cable aberrations and phase difference to the inputs. I don’t necessarily need the 067-0525 if I can order a matched length Y cable from somewhere.




--
-----------------------------------------------------------------------
Glenn Little ARRL Technical Specialist QCWA LM 28417
Amateur Callsign: WB4UIV wb4uiv@arrl.net AMSAT LM 2178
QTH: Goose Creek, SC USA (EM92xx) USSVI LM NRA LM SBE ARRL TAPR
"It is not the class of license that the Amateur holds but the class
of the Amateur that holds the license"


Colin Herbert
 

I have two of these (none for sale!). The tricky bit in "building" one would be in sourcing the "joint" which has a female BNC and the two identical coax wires going to the two male BNCs. Of course, if you are talking about something similar which will do the job, then a BNC tee-piece with one three female connectors, plus two equal lengths of BNC male to BNC male coax could be used.
Colin.

-----Original Message-----
From: TekScopes@groups.io [mailto:TekScopes@groups.io] On Behalf Of Glenn Little
Sent: 11 November 2020 21:35
To: TekScopes@groups.io
Subject: Re: [TekScopes] 067-0525-01 or equivalent needed.

Why not build one?

Glenn

On 11/11/2020 3:01 PM, Eric wrote:
I need this calibration fixture or something like it to calibrate a 7A22 and some 400 series scopes. This fixture is a BNC Y cable the important part is the Y side of the cable is matched length to within 0.1 inch. This is to remove cable aberrations and phase difference to the inputs. I don’t necessarily need the 067-0525 if I can order a matched length Y cable from somewhere.




--
-----------------------------------------------------------------------
Glenn Little ARRL Technical Specialist QCWA LM 28417
Amateur Callsign: WB4UIV wb4uiv@arrl.net AMSAT LM 2178
QTH: Goose Creek, SC USA (EM92xx) USSVI LM NRA LM SBE ARRL TAPR
"It is not the class of license that the Amateur holds but the class
of the Amateur that holds the license"


Gif Sim
 

thanks eric of the info

I figured it had to be like this
but when in doubt I asked the fantastic group
I have attached link to a photo
can it be built like this?

https://groups.io/g/TekScopes/photo/256551/0?p=Created,,,20,2,0,0


Eric
 

Simon,
Yea this one really is that simple. A BNC T + a matched set of cables should do the trick. I expect the way this was built is the cables are soldered in the junction box after their length is confirmed by TDR. Using a connecter at the time would have caused a slight but measurable change in the length of the cable that could impact results. Should be doable the only peace of information I am missing at this point is variance in the T its self. I expect something from a good brand will not be an issue bot for this trying to stay away from the cheap stuff.

Eric

-----Original Message-----
From: TekScopes@groups.io <TekScopes@groups.io> On Behalf Of gifsim via groups.io
Sent: Thursday, November 12, 2020 8:53 AM
To: TekScopes@groups.io
Subject: Re: [TekScopes] 067-0525-01 or equivalent needed.

thank you very much Zenith
so is it possible to build it yourself with a T bnc and two pieces of cable respecting the lengths?
what's inside T tektronix ? (only 2 cables connected in parallel?)
is it really that simple?
Simon


Gif Sim
 

thank you very much Zenith
so is it possible to build it yourself with a T bnc and two pieces of cable respecting the lengths?
what's inside T tektronix ? (only 2 cables connected in parallel?)
is it really that simple?
Simon


 

As Glenn said, it is a BNC-T with two fixed coaxes of equal lengths.
The 067-0525-01 had a total length of 11.5" and the -02 was 15.5".
I have no info on the -00 but I think it was similar to the -01.
The -02 replaced both the other.
A picture here: http://hakanh.com/dl/temp/0525.jpg
/Håkan


Gif Sim
 

067-0525-02 is required in the tds784 calibration procedure
I did a search on the net but did not find anything
in the service manual it is called "Coupler, Dual-Input"
Does anyone in the group have some coupler photos or do they know how it is made (wiring diagram)?
thanks
simon


Glenn Little
 

It appears a BNC T adapter, two equal lengths is 50 Ohm coax and another 50 Ohm coax would create a dual input cable to serve this need.

Glenn

On 11/11/2020 7:31 PM, Raymond Domp Frank wrote:
On Wed, Nov 11, 2020 at 10:53 PM, Tom Lee wrote:

My guess (and it’s only that) is that they designed it to be enough better
than any DUT that you wouldn’t ever worry about the fixturing. A good stress
test, e.g, would be CMRR measurements at the upper end of the 485’s freq
range. Such a measurement is sensitive to phase match.
You guessed right as regards the 485, Tom.

A quick scan of the 485 Service Manual brings up the following:

Where test equipment for checking and adjusting is specified, a "Dual Input Cable" is listed, with a 067-0525-00 as an "Example of Applicable Test Equipment".

In the "Performance Check" chapter, CMR is checked, both at 50 Ohm and at 1 MOhm.

First, a 50 kHz sine wave is applied to both channels (using the 067-0525-00). Vertical sensitivity is set to 20 mV/div on both channels and amplitude is set to 8 divisions vertical deflection. With CH2 set to INVERTED and vertical mode display set to ADD, CH2 VARIABLE GAIN is set to minimize the displayed resulting amplitude.

Next, 50 MHz is applied and the resulting amplitude is checked against a 0.8 div. maximum spec.

My rough (very rough/wrong?) calculation suggests that apart from the 485's behavior, a 300 ps phase difference between both inputs could cause this. That'd be the result of a length difference of 2.5" between both cables.

The test is repeated with input impedance set to 1 MOhm with 50 Ohm terminations on the inputs.

The 067-0525-00 is also used for checking Added Mode Gain by applying equal amplitude signals to CH1 and CH1. Any good quality signal splitting cable would do the job, as it would for the 7A22

Raymond



--
-----------------------------------------------------------------------
Glenn Little ARRL Technical Specialist QCWA LM 28417
Amateur Callsign: WB4UIV wb4uiv@arrl.net AMSAT LM 2178
QTH: Goose Creek, SC USA (EM92xx) USSVI LM NRA LM SBE ARRL TAPR
"It is not the class of license that the Amateur holds but the class
of the Amateur that holds the license"


 

On Wed, Nov 11, 2020 at 10:53 PM, Tom Lee wrote:


My guess (and it’s only that) is that they designed it to be enough better
than any DUT that you wouldn’t ever worry about the fixturing. A good stress
test, e.g, would be CMRR measurements at the upper end of the 485’s freq
range. Such a measurement is sensitive to phase match.
You guessed right as regards the 485, Tom.

A quick scan of the 485 Service Manual brings up the following:

Where test equipment for checking and adjusting is specified, a "Dual Input Cable" is listed, with a 067-0525-00 as an "Example of Applicable Test Equipment".

In the "Performance Check" chapter, CMR is checked, both at 50 Ohm and at 1 MOhm.

First, a 50 kHz sine wave is applied to both channels (using the 067-0525-00). Vertical sensitivity is set to 20 mV/div on both channels and amplitude is set to 8 divisions vertical deflection. With CH2 set to INVERTED and vertical mode display set to ADD, CH2 VARIABLE GAIN is set to minimize the displayed resulting amplitude.

Next, 50 MHz is applied and the resulting amplitude is checked against a 0.8 div. maximum spec.

My rough (very rough/wrong?) calculation suggests that apart from the 485's behavior, a 300 ps phase difference between both inputs could cause this. That'd be the result of a length difference of 2.5" between both cables.

The test is repeated with input impedance set to 1 MOhm with 50 Ohm terminations on the inputs.

The 067-0525-00 is also used for checking Added Mode Gain by applying equal amplitude signals to CH1 and CH1. Any good quality signal splitting cable would do the job, as it would for the 7A22

Raymond