Difference between revisions of "S-20"

From TekWiki
Jump to: navigation, search
 
(One intermediate revision by the same user not shown)
Line 4: Line 4:
 
I designed that sampler while I worked in TekLabs.
 
I designed that sampler while I worked in TekLabs.
  
It was shown at the Research Forum interfaced to a SDXX dual channel sampler interface head plugged into the [[11801]] or [[CSA803]] mainframe.
+
It was shown at the Research Forum interfaced to a SDXX  
 +
dual channel sampler interface head plugged into the [[11801]] or [[CSA803]] mainframe.
  
 
Feedthrough design BW 30GHz 2mV sensitivity.
 
Feedthrough design BW 30GHz 2mV sensitivity.
  
4 diode beam lead Schottky diodes on a 5 mil duroid replaceable carrier, coupled to a beam lead [[Step_recovery_diode|SRD]] based driver also on 5 mil duroid substrate with pins to plug into sockets in the housing. This is all on the side of the lid. The other side of the housing has a polyimide flexible circuit connecting to the [[LEMO_S-series_connector|4 pin LEMO connector]].  
+
4 diode beam lead Schottky diodes on a 5 mil duroid replaceable carrier,  
 +
coupled to a beam lead [[Step_recovery_diode|SRD]] based driver also on 5 mil duroid substrate  
 +
with pins to plug into sockets in the housing.  
 +
This is all on the side of the lid.  
 +
The other side of the housing has a polyimide flexible circuit connecting to the [[LEMO_S-series_connector|4 pin LEMO connector]].  
  
A special flexible differential cable was designed to keep the delay change to 5ps while torqued/twisted 180 degree This cable drives the SRD differentially. The two extra wire provide the sampler bias and send back the average value of the sampled signal.
+
A special flexible differential cable was designed to keep the delay change to 5ps while torqued/twisted 180 degree.
 +
This cable drives the SRD differentially.  
 +
The two extra wire provide the sampler bias and send back the average value of the sampled signal.
 
</blockquote>
 
</blockquote>
  
 
[[John Rettig]] wrote this general description:
 
[[John Rettig]] wrote this general description:
 
<blockquote>
 
<blockquote>
As best I can recall – this was never a product, but a proposal that came out of Tek Labs after the introduction of the [[11800-series_scopes|11800 family of sampling products]]. It was likely just the [[sampling diodes]] and maybe the front end part of the charge amplifier, judging by the head size I see in your image link, and it would have suffered from the modularity of the entire system not having calibrated subcomponents. Remember that the 11800 was not supposed to need the user to tweak the loop gain, although that “feature” did sneak its way in to the “Enhanced Accuracy” menus.
+
As best I can recall – this was never a product,  
 +
but a proposal that came out of Tek Labs after the introduction of the [[11800-series_scopes|11800 family of sampling products]].  
 +
It was likely just the [[sampling diodes]] and maybe the front end part of the charge amplifier,  
 +
judging by the head size I see in your image link,  
 +
and it would have suffered from the modularity of the entire system not having calibrated subcomponents.  
 +
Remember that the 11800 was not supposed to need the user to tweak the loop gain,  
 +
although that “feature” did sneak its way in to the “Enhanced Accuracy” menus.
  
We instead productized the [[SD-20]] module in the Lab Scopes HFCD product area, which essentially was our 20 GHz sampler from the [[SD-23]] / [[SD-24]] designs but with the 50 Ohm coax fed back out to the front panel instead of having the built in internal 50 Ohm termination on the channel. And it was fully calibrated, since it didn’t have the modularity of this other proposal.
+
We instead productized the [[SD-20]] module in the Lab Scopes HFCD product area,  
 +
which essentially was our 20 GHz sampler from the [[SD-23]] / [[SD-24]] designs  
 +
but with the 50 Ohm coax fed back out to the front panel instead of having the built in internal 50 Ohm termination on the channel.  
 +
And it was fully calibrated, since it didn’t have the modularity of this other proposal.
  
This device, the SD-20, and for that matter the [[S-6]] of the 7K generation all suffered from customer inattention to ESD issues - there were a lot of repairs needed. As a result, our 8000-series never included a loop through feature, although we do support high BW [[TekConnect_connector|TekConnect]] probes through the [[80A03]]. And furthermore, all of our samplers now run open loop, with on-the-fly point correction for the front end nonlinearity, so there is no loop gain correction needed.
+
This device, the SD-20, and for that matter the [[S-6]] of the 7K generation  
 +
all suffered from customer inattention to ESD issues - there were a lot of repairs needed.  
 +
As a result, our 8000-series never included a loop through feature,  
 +
although we do support high BW [[TekConnect_connector|TekConnect]] probes through the [[80A03]].  
 +
And furthermore, all of our samplers now run open loop,  
 +
with on-the-fly point correction for the front end nonlinearity,  
 +
so there is no loop gain correction needed.
 
</blockquote>
 
</blockquote>
  
Line 32: Line 54:
  
 
[[Category:Sampling_plugins]]
 
[[Category:Sampling_plugins]]
 +
[[Category:Prototypes]]

Latest revision as of 12:05, 12 January 2020

Agoston Agoston wrote this detailed description of what is inside the S-20 that he designed:

I designed that sampler while I worked in TekLabs.

It was shown at the Research Forum interfaced to a SDXX dual channel sampler interface head plugged into the 11801 or CSA803 mainframe.

Feedthrough design BW 30GHz 2mV sensitivity.

4 diode beam lead Schottky diodes on a 5 mil duroid replaceable carrier, coupled to a beam lead SRD based driver also on 5 mil duroid substrate with pins to plug into sockets in the housing. This is all on the side of the lid. The other side of the housing has a polyimide flexible circuit connecting to the 4 pin LEMO connector.

A special flexible differential cable was designed to keep the delay change to 5ps while torqued/twisted 180 degree. This cable drives the SRD differentially. The two extra wire provide the sampler bias and send back the average value of the sampled signal.

John Rettig wrote this general description:

As best I can recall – this was never a product, but a proposal that came out of Tek Labs after the introduction of the 11800 family of sampling products. It was likely just the sampling diodes and maybe the front end part of the charge amplifier, judging by the head size I see in your image link, and it would have suffered from the modularity of the entire system not having calibrated subcomponents. Remember that the 11800 was not supposed to need the user to tweak the loop gain, although that “feature” did sneak its way in to the “Enhanced Accuracy” menus.

We instead productized the SD-20 module in the Lab Scopes HFCD product area, which essentially was our 20 GHz sampler from the SD-23 / SD-24 designs but with the 50 Ohm coax fed back out to the front panel instead of having the built in internal 50 Ohm termination on the channel. And it was fully calibrated, since it didn’t have the modularity of this other proposal.

This device, the SD-20, and for that matter the S-6 of the 7K generation all suffered from customer inattention to ESD issues - there were a lot of repairs needed. As a result, our 8000-series never included a loop through feature, although we do support high BW TekConnect probes through the 80A03. And furthermore, all of our samplers now run open loop, with on-the-fly point correction for the front end nonlinearity, so there is no loop gain correction needed.


Abbreviations: