7A11: Difference between revisions

From TekWiki
Jump to navigation Jump to search
No edit summary
No edit summary
 
(14 intermediate revisions by 4 users not shown)
Line 10: Line 10:
|designers=John Addis;Ron Peltola;Glenn Bateman
|designers=John Addis;Ron Peltola;Glenn Bateman
|manuals=
|manuals=
* [[Media:070-0984-00.pdf|Tektronix 7A11 Manual]]<br /><small>[[Media:070-0984-00 (2).pdf|Alternate copy]] (OCR)</small>
* [[Media:070-0984-00.pdf | Tektronix 7A11 Manual]]<br /><small>[[Media:070-0984-00 (2).pdf|Alternate copy]]</small>
}}
}}
The '''Tektronix 7A11''', [[introduced in 1969]], is a vertical plug-in for [[7000-series scopes]].
The '''Tektronix 7A11''', [[introduced in 1969]], is a vertical plug-in for [[7000-series scopes]].
Line 16: Line 16:
It includes a permanently attached FET probe.  The bandwidth of a 7A11 in a [[7904]] is 250 MHz.
It includes a permanently attached FET probe.  The bandwidth of a 7A11 in a [[7904]] is 250 MHz.
The front-end FET is a [[151-1034-00]].
The front-end FET is a [[151-1034-00]].
The probe can be used with a front panel BNC connector or the probe cable can be unwound from the plugin in six steps to a maximum of 7 feet in length.


The 7A11 contains a selectable three-pole 20 MHz bandwidth limit filter.
The 7A11 contains a selectable three-pole 20 MHz bandwidth limit filter.
The designers were [[John Addis]], [[Ron Peltola]], and [[Glenn Bateman]].


{{BeginSpecs}}
{{BeginSpecs}}
{{Spec | Bandwidth | 250 MHz (in [[7904]]) }}
{{Spec | Bandwidth | 250 MHz (in [[7903]], [[R7903]], [[7904]], [[7104]]) }}
{{Spec | Deflection | 5 mV/Div to 20 V/Div, 1–2–5 }}
{{Spec | Deflection | 5 mV/Div to 20 V/Div, 1–2–5 }}
{{Spec | Input resistance | 1 MΩ }}
{{Spec | Input resistance | 1 MΩ }}
Line 31: Line 35:
}}
}}
{{Spec | Features |  
{{Spec | Features |  
* Variable DC Offset (with output connector)
* Variable DC Offset (monitored with output connector)
* 20 MHz Bandwith Limit switch
* 20 MHz Bandwith Limit switch
* Invert switch
* Invert switch
}}
}}
{{EndSpecs}}
{{EndSpecs}}
==Links==
{{Documents|Link=7A11}}


==Internals==
==Internals==
Line 41: Line 48:
Regarding the 7A11, its designer [[John Addis]] says:
Regarding the 7A11, its designer [[John Addis]] says:
<blockquote>
<blockquote>
The 7A11 input capacitance is 5.8 pF from 5 mV/Div to 50 mV/Div, goes down to 3.4 pF from
The 7A11 input capacitance is 5.8 pF from 5 mV/Div to 50 mV/Div, goes down to 3.4 pF from 100 mV/Div to 1 V/Div, and down again to 2.0 pF from 2 V/Div to 20 V/Div.   
100 mV/Div to 1 V/Div, and down again to 2.0 pF from 2 V/Div to 20 V/Div.  Because the AC
Because the AC coupling is attained with a plug-on capacitor (that adds 1.2 pF to the input C), a DC offset is supplied to reduce the need for AC coupling.
coupling is attained with a plug-on capacitor (that adds 1.2 pF to the input C), a DC offset  
 
is supplied to reduce the need for AC coupling.
A nice thing about the 7A11 was that the probe cable length was adjustable in six steps to just over 7 feet in length.   
</blockquote>
You just uncoil however much cable length you want.
<blockquote>
 
A nice thing about the 7A11 was that the probe cable length was adjustable in six steps  
The 7A11 was admittedly clumsy to use because of the probe size.  It had some trouble — DC drift, RF pickup at 5 mV, 100 mV and 2 V/Div.  
to just over 7 feet in length.  You just uncoil however much cable length you want.
It was not very popular partially because it was expensive, $850 for single channel vs. dual channel [[7A12]] (105 MHz) at $700.   
</blockquote>
But there was never another 1 MΩ 7000-series plugin as fast as the 7A11 ... and it was one of the originals!  
<blockquote>
 
The 7A11 was admittedly clumsy to use because of the probe size.  It had some trouble —
DC drift, RF pickup at 5 mV, 100 mV and 2 V/Div. It was not very popular partially  
because it was expensive, $850 for single channel vs. dual channel [[7A12]] (105 MHz) at $700.   
But there was never another 1 MΩ 7000-series plugin as fast as the 7A11 ... and it was  
one of the originals!  
</blockquote>
<blockquote>
The [[7A16]] (single channel plugin) was also 150 MHz, but that was a year later.   
The [[7A16]] (single channel plugin) was also 150 MHz, but that was a year later.   
Then the [[7904]] came out in late 1971. That made the 7A11 a 250 MHz plugin (for $950).   
Then the [[7904]] came out in late 1971. That made the 7A11 a 250 MHz plugin (for $950).   
The [[7A16]] was then 225 MHz (for $625). The 7A16 disappeared rapidly and became the [[7A16A]],  
The [[7A16]] was then 225 MHz (for $625). The 7A16 disappeared rapidly and became the [[7A16A]], still 225 MHz in 1973.  
still 225 MHz in 1973.  
 
</blockquote>
<blockquote>
In 1972, the [[485]] came out with a 250 MHz, 1 MΩ input.  
In 1972, the [[485]] came out with a 250 MHz, 1 MΩ input.  
The 485 used a faster IC process  
The 485 used a faster IC process ("[[SH2]]", 3.5 GHz) than was available for first generation 7K instruments but the 485's 1 MΩ to 50 Ω converter used only discrete devices. (See Electronics June, 1972.)
("[[SH2]]", 3.5 GHz) but the 1 MΩ to 50 Ω converter used only discrete devices.  
 
(See Electronics June, 1972.)
By 1974, the 7A11 cost $950 and the new, popular [[7A26]] (dual trace) was $1,050, not counting probes.   
</blockquote>
But that would still only get you to 50 mV/div with a (9.5 pF, 3 foot, × 10) probe, five years after the 7A11 introduction.
<blockquote>
[[Tom Rousseau]] designed the 7A26 which used the faster SH-2 IC process and a [[155-0078-00 | vertical IC I designed for the 485]].  
By 1974, the 7A11 cost $950 and the new, popular [[7A26]] (dual trace) was $1,050, not  
The 7A26 sold so well that Tektronix presented Tom with an entirely gold plated 7A26!   
counting probes.  But that would still only get you to 50 mV/div with a (9.5 pF, 3 foot, × 10)  
Obviously, he still has it.
probe, five years after the 7A11 introduction. [[Tom Rousseau]] designed the 7A26 which used the  
 
faster IC process and a [[155-0078-00|vertical IC I designed for the 485]]. The 7A26 sold so well that  
Remember also that the 7A11 was introduced at the same show as the HP183A/1830A/1840A, faster (250 MHz), smaller, lighter, and less expensive with better triggering than 7000 series had.   
Tektronix presented Tom with an entirely gold plated 7A26!  Obviously, he still has it.
</blockquote>
<blockquote>
Remember also that the 7A11 was introduced at the same show as the HP183A/1830A/1840A,  
faster (250 MHz), smaller, lighter, and less expensive with better triggering than 7000 series had.   
Their secrets were Al DeVilbiss and a faster IC process.  We had neither.
Their secrets were Al DeVilbiss and a faster IC process.  We had neither.
</blockquote>
 
<blockquote>
The star-crossed [[7A12]] designed by [[Roy Hayes]], which was supposed to be the dual trace flagship of the original plugins, was only 105 MHz, not the hoped-for 150 MHz.   
The star-crossed [[7A12]], which was supposed to be the dual trace flagship of the original plugins,  
That was partially due to the fact that it used the existing Tek IC process (about 1 GHz).  That was a bad decision on day one. The HP IC process was about 3 GHz.   
was only 105 MHz, not the hoped-for 150 MHz.  That was partially due to the fact that it used the  
The 7A11 used discrete NPN and PNP transistors with 4 GHz f<sub>t</sub>.
existing Tek IC process (about 1 GHz).  The HP IC process was about 3 GHz.  The 7A11 used discrete  
 
NPN and PNP transistors with 4 GHz f<sub>t</sub>.
There is one noteworthy point I would like to make about the 7A11.  In those days, the different V/div settings were generally attained using fixed high impedance attenuators, usually stacked one after another and few with more than ×10 attenuation.   
</blockquote>
I wanted the 7A11 to be able to handle the full useful range of sensitivites that other plugins attained when adding a ×10 probe ... that meant going to 20 V/Div.   
<blockquote>
But you could certainly get to 5 mV/Div, which you could not attain with a 5 mV/Div plugin and a ×10 passive probe.   
There is one noteworthy point I would like to make about the 7A11.  In those days, the different  
 
V/div settings were generally attained using fixed high impedance attenuators, usually stacked one  
But using that scheme, if the 7A11 went to 20 V/Div, it would have to stack three ×10 attenuators inside the probe, making it a behemoth, and even if you did, that would still leave the ×2 and ×5 attenuations up to gain switching in the amplifier.   
after another and few with more than ×10 attenuation.  I wanted the 7A11 to be able to handle the  
 
full useful range of sensitivites that other plugins attained when adding a ×10 probe ... that  
Problem with gain switching is, the bandwidth and transient response tend to change when you change amplifier gain.  If gain switching was accomplished by changing emitter resistors, dc balance changed with every gain change.  That meant that the VAR gain control and fixed gain selection required two dc balance controls.  That would be unpleasant (see 7A13).
meant going to 20 V/Div.  But you could certainly get to 5 mV/Div, which you could not attain with  
The greater the gain change, the greater the bandwidth/transient response change in the amplifier.   
a 5 mV/Div plugin and a ×10 passive probe.   
The more gain settings you had, the worse it got due to longer leads and more parasitics.
</blockquote>
 
<blockquote>
To get a 1-2-5 sequence from 5 mV/Div to 20 V/Div, it had the worst of all possible worlds.   
But using that scheme, if the 7A11 went to 20 V/Div, it would have to stack three ×10 attenuators  
The 7A11 needed two ×20 attenuators in the probe and that meant that it needed not just ×1, ×2 and ×5 gains in the amplifier, but ×1, ×2, ×2.5, ×4, ×5 and ×10!   
inside the probe, making it a behemoth, and even if you did, that would still leave the ×2 and ×5  
This had not been done before, but the 7A11 does it without any change in transient response!
attenuations up to gain switching in the amplifier.   
 
</blockquote>
The secret is three entirely passive, [[miniature relays | relay-switched]], O-pad attenuators:  ×2, ×2, and ×2.5 in a balanced transmission line environment.   
<blockquote>
Stack them up (as you can do with matched pads) and you get all the combinations you need: ×1, ×2, ×2.5, ×4, ×5, and ×10.  From 5 mV/Div to 20 V/Div, twelve different V/div settings, more than any other high speed plug in!   
Problem with gain switching is, the bandwidth and transient response tend to change when you change  
 
amplifier gain.  The greater the gain change, the greater the bandwidth/transient response change in  
I got a patent on the variable attenuator, which was just a JFET shunting the O-pads.   
the amplifier.  The more gain settings you had, the worse it got due to longer leads and more parasitics.
The patent had to do with making the gain vs. control rotation linear, which does not simply happen with linear gate-source voltage control.   
</blockquote>
The JFET causes a small change in transient response, but not a bad one.
<blockquote>
 
To get a 1-2-5 sequence from 5 mV/Div to 20 V/Div, it had the worst of all possible worlds.  The 7A11  
[[Ron Peltola]] (of [[Peltola connector]] fame) designed the probe attenuator, Glenn Bateman and Ron designed the probe amplifier.  I helped a little on the probe amplifier.
needed two ×20 attenuators in the probe and that meant that it needed not just ×1, ×2 and ×5 gains in  
the amplifier, but ×1, ×2, ×2.5, ×4, ×5 and ×10!  This had not been done before, but the 7A11 does it  
without any change in transient response!
</blockquote>
<blockquote>
The secret is three entirely passive, [[miniature relays | relay-switched]], O-pad attenuators:  ×2, ×2, and ×2.5 in a  
balanced transmission line environment.  Stack them up (as you can do with matched pads) and you get  
all the combinations you need: ×1, ×2, ×2.5, ×4, ×5, and ×10.  From 5 mV/Div to 20 V/Div, twelve  
different V/div settings, more than any other high speed plug in!   
</blockquote>
<blockquote>
I got a patent on the variable attenuator, which was just a JFET shunting the O-pads.  The patent had  
to do with making the gain vs. control rotation linear, which does not simply happen with linear  
gate-source voltage control.  The JFET causes a small change in transient response, but not a bad one.
</blockquote>
<blockquote>
[[Ron Peltola]] (of [[Peltola connector]] fame) designed the probe.  I helped on its amplifier.
</blockquote>
</blockquote>


Line 146: Line 122:
|align=right| $2,700
|align=right| $2,700
|-
|-
! 2019 value
! 2023 value
|align=right| $5,340
|align=right| $6,400
|align=right| $5,320
|align=right| $6,400
|align=right| $6,600
|align=right| $7,900
|-
|-
|}
|}
Line 156: Line 132:
<gallery>
<gallery>
Tek-7a11-front.jpg
Tek-7a11-front.jpg
7A11_2.JPG| right side, probe unplugged
7A11_2.JPG                                     | right side, probe unplugged
7A11_3.JPG| probe accessoires: tip cover, AC coupling capacitor, [[GR-874]] adapter ([[017-0088-00]])
7A11_3.JPG                                     | probe accessoires: tip cover, AC coupling capacitor, [[GR-874]] adapter ([[017-0088-00]])
7a11-right.jpg| right side detail, cover removed
7a11-right.jpg                                 | right side detail, cover removed
7a11-left.jpg| left side detail, cover removed
7a11-left.jpg                                   | left side detail, cover removed
7a11-input.jpg| input jack with probe cable  
7a11-input.jpg                                   | input jack with probe cable  
7a11-probe-interior.jpg| interior of probe
7a11-probe-interior.jpg                         | interior of probe
7a11-probe-interior-back.jpg| interior of probe, back side with range relays
7a11-probe-interior-back.jpg                   | interior of probe, back side with range relays
Tek 7a11 on.JPG
Tek 7a11 on.JPG
Tek 7a11 probe.jpg|Probe
Tek 7a11 probe.jpg                             | Probe
Tek 7a11 input amplifier.jpg|Input Amplifier
Tek 7a11 input amplifier.jpg                   | Input Amplifier
Tek 7a11 output amplifier.jpg|Output Amplifier
Tek 7a11 output amplifier.jpg                   | Output Amplifier
Tek 7a11 offset generator.jpg|Offset Generator
Tek 7a11 offset generator.jpg                   | Offset Generator
Tek 7a11 volts-per-div switch and readout.jpg|Volts/Div Switch and Readout
Tek 7a11 volts-per-div switch and readout.jpg   | Volts/Div Switch and Readout
Tek 7a11 power supply and output connectors.jpg|Power Supply and Output Connectors
Tek 7a11 power supply and output connectors.jpg | Power Supply and Output Connectors
</gallery>
</gallery>
==Components==
{{Parts|7A11}}


[[Category:7000 series vertical plugins]]
[[Category:7000 series vertical plugins]]
[[Category:Active oscilloscope probes]]
[[Category:Active oscilloscope probes]]

Latest revision as of 19:15, 5 September 2024

Tektronix 7A11
250 MHz FET-probe amplifier
7A11 front view

Compatible with 7000-series scopes

Produced from 1969 to 1984

Manuals
(All manuals in PDF format unless noted otherwise)
Manuals – Specifications – Links – Pictures

The Tektronix 7A11, introduced in 1969, is a vertical plug-in for 7000-series scopes.

It includes a permanently attached FET probe. The bandwidth of a 7A11 in a 7904 is 250 MHz. The front-end FET is a 151-1034-00.

The probe can be used with a front panel BNC connector or the probe cable can be unwound from the plugin in six steps to a maximum of 7 feet in length.

The 7A11 contains a selectable three-pole 20 MHz bandwidth limit filter.

The designers were John Addis, Ron Peltola, and Glenn Bateman.

Key Specifications

Bandwidth 250 MHz (in 7903, R7903, 7904, 7104)
Deflection 5 mV/Div to 20 V/Div, 1–2–5
Input resistance 1 MΩ
Input capacitance
  • 5.8 pF @ 5 mV/Div to 50 mV/Div
  • 3.4 pF @ 100 mV/Div to 1 V/Div
  • 2.0 pF @ 2 V/Div to 20 V/Div
  • Add 1.2 pF for BNC input adapter
  • Add 1.2 pF for AC coupling plug-on capacitor
Features
  • Variable DC Offset (monitored with output connector)
  • 20 MHz Bandwith Limit switch
  • Invert switch

Links

Documents Referencing 7A11

Document Class Title Authors Year Links
Tekscope 1969 V1 N5 Oct 1969.pdf Article Introducing the New Generation 1969
Tekscope 1969 V1 N6 Dec 1969.pdf Article A New Logic for Oscilloscope Displays 1969
7000 series brochure March 1973.pdf Brochure 7000 series brochure, March 1973 1973

Internals

Regarding the 7A11, its designer John Addis says:

The 7A11 input capacitance is 5.8 pF from 5 mV/Div to 50 mV/Div, goes down to 3.4 pF from 100 mV/Div to 1 V/Div, and down again to 2.0 pF from 2 V/Div to 20 V/Div. Because the AC coupling is attained with a plug-on capacitor (that adds 1.2 pF to the input C), a DC offset is supplied to reduce the need for AC coupling.

A nice thing about the 7A11 was that the probe cable length was adjustable in six steps to just over 7 feet in length. You just uncoil however much cable length you want.

The 7A11 was admittedly clumsy to use because of the probe size. It had some trouble — DC drift, RF pickup at 5 mV, 100 mV and 2 V/Div. It was not very popular partially because it was expensive, $850 for single channel vs. dual channel 7A12 (105 MHz) at $700. But there was never another 1 MΩ 7000-series plugin as fast as the 7A11 ... and it was one of the originals!

The 7A16 (single channel plugin) was also 150 MHz, but that was a year later. Then the 7904 came out in late 1971. That made the 7A11 a 250 MHz plugin (for $950). The 7A16 was then 225 MHz (for $625). The 7A16 disappeared rapidly and became the 7A16A, still 225 MHz in 1973.

In 1972, the 485 came out with a 250 MHz, 1 MΩ input. The 485 used a faster IC process ("SH2", 3.5 GHz) than was available for first generation 7K instruments but the 485's 1 MΩ to 50 Ω converter used only discrete devices. (See Electronics June, 1972.)

By 1974, the 7A11 cost $950 and the new, popular 7A26 (dual trace) was $1,050, not counting probes. But that would still only get you to 50 mV/div with a (9.5 pF, 3 foot, × 10) probe, five years after the 7A11 introduction. Tom Rousseau designed the 7A26 which used the faster SH-2 IC process and a vertical IC I designed for the 485. The 7A26 sold so well that Tektronix presented Tom with an entirely gold plated 7A26! Obviously, he still has it.

Remember also that the 7A11 was introduced at the same show as the HP183A/1830A/1840A, faster (250 MHz), smaller, lighter, and less expensive with better triggering than 7000 series had. Their secrets were Al DeVilbiss and a faster IC process. We had neither.

The star-crossed 7A12 designed by Roy Hayes, which was supposed to be the dual trace flagship of the original plugins, was only 105 MHz, not the hoped-for 150 MHz. That was partially due to the fact that it used the existing Tek IC process (about 1 GHz). That was a bad decision on day one. The HP IC process was about 3 GHz. The 7A11 used discrete NPN and PNP transistors with 4 GHz ft.

There is one noteworthy point I would like to make about the 7A11. In those days, the different V/div settings were generally attained using fixed high impedance attenuators, usually stacked one after another and few with more than ×10 attenuation. I wanted the 7A11 to be able to handle the full useful range of sensitivites that other plugins attained when adding a ×10 probe ... that meant going to 20 V/Div. But you could certainly get to 5 mV/Div, which you could not attain with a 5 mV/Div plugin and a ×10 passive probe.

But using that scheme, if the 7A11 went to 20 V/Div, it would have to stack three ×10 attenuators inside the probe, making it a behemoth, and even if you did, that would still leave the ×2 and ×5 attenuations up to gain switching in the amplifier.

Problem with gain switching is, the bandwidth and transient response tend to change when you change amplifier gain. If gain switching was accomplished by changing emitter resistors, dc balance changed with every gain change. That meant that the VAR gain control and fixed gain selection required two dc balance controls. That would be unpleasant (see 7A13). The greater the gain change, the greater the bandwidth/transient response change in the amplifier. The more gain settings you had, the worse it got due to longer leads and more parasitics.

To get a 1-2-5 sequence from 5 mV/Div to 20 V/Div, it had the worst of all possible worlds. The 7A11 needed two ×20 attenuators in the probe and that meant that it needed not just ×1, ×2 and ×5 gains in the amplifier, but ×1, ×2, ×2.5, ×4, ×5 and ×10! This had not been done before, but the 7A11 does it without any change in transient response!

The secret is three entirely passive, relay-switched, O-pad attenuators: ×2, ×2, and ×2.5 in a balanced transmission line environment. Stack them up (as you can do with matched pads) and you get all the combinations you need: ×1, ×2, ×2.5, ×4, ×5, and ×10. From 5 mV/Div to 20 V/Div, twelve different V/div settings, more than any other high speed plug in!

I got a patent on the variable attenuator, which was just a JFET shunting the O-pads. The patent had to do with making the gain vs. control rotation linear, which does not simply happen with linear gate-source voltage control. The JFET causes a small change in transient response, but not a bad one.

Ron Peltola (of Peltola connector fame) designed the probe attenuator, Glenn Bateman and Ron designed the probe amplifier. I helped a little on the probe amplifier.

The group that did the probe design was Ken Holland's High Frequency Component Design Group. Ron Peltola was the engineer.

Links

Prices

Year 1971 1980 1984
Catalog price $850 $1,725 $2,700
2023 value $6,400 $6,400 $7,900

Pictures

Components

Some Parts Used in the 7A11

Part Part Number(s) Class Description Used in
148-0034-00 148-0034-00 Discrete component miniature DPDT relay 5A48 7A11 7A13 7A14 7B70 7B71 7503 7904 7904A
148-0034-03 148-0034-03 Discrete component miniature DPDT relay 7A11
148-0058-00 148-0058-00 Discrete component miniature SPDT relay 7A11
151-1034-00 151-1034-00 Discrete component dual N-channel JFET 7A11 P6051