Questions on Tektronix S-1, 7S11, and 7T11

[Electro Fan]

Just checking to see if the item in the photo might be the right part to help extract a S-1 from a 7S11?  Somehow I managed to get a 7S11 with a S-1 that is missing the knob.  Assuming it's the right part, is it just a matter of screwing it in and pulling or is their some something else (special technique - pushing/pulling/twisting, etc) needed to remove the S-1 from the 7S11?  (Not sure why this part screws in and locks/unlocks vs simply screwing in/out).

On a related matter, any advice on making the 7S11 with the S-1 play nicely with a 7T11 next to it in a 7904 when measuring rise times?

[TerraHertz]

Looks like it is.
I have an S1, and took it apart for photos of the locking mechanism. See below.
The silvery metal thing with a bevel at the other end, is what moves downwards (up in photo) to lock against a tongue that slides into the S1 case.

With the scope upright, and that screw missing, the silvery metal part will fall down, into lock position.

Maybe you could try turning the scope upside down (or front face down?), pushing the S1 fully in, and giving it a few gentle taps to let that stop drop down. Then the S1 should pull out OK.

Incidentally, that screw on mine has a small bend in the middle where the thread starts. Visible in the photos. I think it's deliberate, to give the stop a bit of extra up and down movement. Oh, and it's a reverse thread.

Worth noting that there was a metal circlip, which may now be loose in the S1, or scope.

[Electro Fan]

TerraHertz, Thanks! 

Is the circlip the item that is shown in the red circle?  If so, maybe it was designed to keep someone (who apparently circumvented or broke it) from removing the screw/knob - in which case I think the screw/knob might work without the circlip.  Either way, thanks for the heads-up on a part possibly floating around.  A  loose part in the circuit could be a problem 

I tried turning the plug-in upside down to see if that let the bevel part drop loose but no luck (easier turning the plug-in upside down than the mainframe  ).

Given where the bevel part resides - any chance that could have fallen out of place - and the screw might not mate? In which case the S-1 could be major stuck?

Hopefully when I get the screw/knob installed it will allow removal of the S-1 from the 7S11 and then I can open the S-1 to see if the circlip is loose inside.


PS, to anyone reading this who is not yet fully experienced with EEVblog, there is probably no other place on the planet where a person could post such a question and get an accurate answer (or any answer) to something like this, much less fully documented with pictures, much less within minutes.  EEVblog Rocks!

[TerraHertz]

Is the circlip the item that is shown in the red circle?

Yes.
When turning the knob clockwise (lock) it's pushing that metal part further in. Which means the circlip is being pressed against the back of the front face. I guess if someone turns the screw too firmly, the circlip will pop off.

Here's another photo, of the inside of a 7S12 module, with a sampler partially removed. Shows the plastic clips that slide into the sampler. Never paid attention to this before. Now I realize the clip goes between that end bevel and the sampler center PCB. The bevel pushes the plastic clip against the pcb corner, locking it against that end web.

But do you see the other way to get it apart?
Those two screws on the other side of the bulkhead secure that white plastic clip to the edge connector on the other side. if you can get at them in your module, remove them and the sampler will come out, even if the plastic clip is still wedged inside.
Then you can take the sampler apart.

[Electro Fan]

Cool

First plan is to try the screw knob when it arrives - maybe yet this week.

Backup plan is to try getting at the two screws holding the white plastic clip.

I'm about 1 win, 1 loss, and 1 tie on fixing plug-ins so I kind of need a win here to get the average up.

Will keep you posted.  Thx

[Electro Fan]

Ok, Goofed up.

The screw knob arrived.  And so did my new (actually decades old) 7T11.

The screw head didn't reach far enough into the S1 to attach to anything - or maybe there is nothing inside the sampling head to attach to.  If that wasn't bad enough my next move probably was.

Rather than dealing with trying to remove the sampling head from the 7S11 I decided to install the 7S11 with the sampling head inside into the 7904 along with the 7T11. 

To explain the fuller mess of things, the 7S11 doesn't have it's pull tab installed.  So (silly me) I figured I just remove the plug-in next to it (where the 7T11 goes) and I'd carefully slide the 7S11 into it's slot (2nd from the left) with the hope that without the pull tab the 7S11 might pressure fit into place and then I could always pull it out with good grip via the empty 3rd slot.  Right up to the last mm or so this was working but on the last push I heard the 7S11 (with no pull tab) click into place.  Uh oh.  - I'm hoping there is a way to get that bad boy out of the mainframe....

Putting off that problem for the time being I figured I'd plug in the 7T11 next to the 7S11 and see if by any chance this stuff would work as is.  Well, the 7904 seems happy enough - the 7A26 to the far left and the 7B92A to the far right work together as always.

So, I tried to get a signal up on the 7S11/S1 and the 7T11 but no luck.  As far as I can tell it could easily be operator error, or it could be the 7S11 or it could be the S1 or it could be the 7T11 - or a combination of all of the above.

I tried to get 7S11/S1 to play with the 7B92A just in case that is even a workable combo but no luck there.  Likewise I tried to get the 7A26 to play with the 7T11 but nada with this combination too.

So, the overall question is:  is there some way to trouble shoot my way out of this mess?

Any way to use the known good 7A26 and 7B92A to help trouble shoot the 7S11/S1 and the 7T11 or are the sampling plug-in completely independent of the 7A26and 7B92A?

Something I noticed on the 7T11 is that the Time Position knob is multi-turn; likewise on the 7S11 the DC Offset knob seems to go about 10 turns before it stops on either end.  So finding a trace might require those two to be properly dialed-in?

Any particular type of input signal to the S1 that might be easiest to display?

Thanks for helping me dig out of the good size hole I have dug.

EF

- the photo shows a pulse signal displaying properly using the 7A26 and the 7B92A.... just need to get the same type of pulse to show up using the 7S11/S1 and the 7T11...

[Electro Fan]

Update:  as suspected, the 7S11 DC Offset and the 7T11 Time Position knobs are delicate little buggers.  With the these two plugins selected (right horizontal and A vertical) I can center the dot in the middle of the display when selecting 200mV and 1us.  So, something is working.

Update:  at 50mV and 2 us I and my pulse signal plugged into the S1 I can get a straight up and down some "dottish" line indicating 300mV pk-pk.  The 7S11 Dot Response knob seems to spread or condense dots up and down the vertical line.  Any thoughts on how to use the Dot Response knob?  (The 7S11 is currently in Normal rather than Smooth mode.) 

The 7T11 has a very wide range from 5ms/Div to 10ps/Div.  Pretty exciting seeing 10ps register as the timdiv label on the display.  Now if something other than the labels and vertical line would display....

Update:  Kind of like rock, scissors, paper something has to win... while trying to disconnect the GR874 to BNC adapter from the GR874 on the S-1 I found that the S-1 slides out of the 7S11 - so maybe one small step in the mechanical realm of part removal capability (I'm still a little concerned about how I'm going to get the 7S11 out of the 7904 but we'll deal with that later, I guess).  I have a hunch that of all the parts the 7T11 might be the most reliable, but that's just a guess.  Still looking for a way to get a signal to display across the S1/7S11 and 7T11....

Update on the S-1:  The metal bevel inside the S1 is completely Missing In Action; the S1 is strictly resident in the 7S11 on a pressure fit basis.

Update:  The image shows a 200mV 1MHz sine wave fed into the S1/7S11 with the 7T11.  The sampling plugins seem to be talking to the 7B92A - when I use the DC Offset to move the signal up or down the green trigger light on the 7B92A will go off, and back on when the signal is centered.  Either there is an issue - or more likely I don't yet understand how to operate the sampling system.....  I hope.  Suggestions?  Thx

[Electro Fan]

Some progress.... setting the 7T11 to EXT HF SYNC seemed to make things happier.  Still can get a square wave to look like a square wave but it seems happy to show AM Modulation.

[Electro Fan]

Ok, at very low speeds (10kHz) using the Trigger Input (from another scope also connected to the sig gen) on the 7T11 the system will display a square wave.  I guess it's progress but the whole point of this configuration was to be able to measure sub nanosecond rise times - theoretically the limiting factor in the system should be the S1 at 350ps.  (I'd really like to find a reason to turn the 7T11 to 10ps per div. )  Not sure how much more user error I have to overcome but I'm wondering if the S1 and plugins are working properly.  Suggestions for how to get things dialed-in very welcome.  Thx

[Electro Fan]

Here is 20kHz; looks like dot city.  I get that the sampling system is pretty dot oriented but it is behaving like it's running at a fraction of it's rated performance.  Might still be user error or maybe something in one of the components is goofed.

[Electro Fan]

PS, still looking for a way to remove the 7S11 from the mainframe without the pull tab in case anyone can rescue me from that goof.    Thx

[cncjerry]

I've been looking at a 7904 to add to my collection of equipment I wonder why I bought.

Pull the 7a26 on the left.  You should be able to pop the latch with a thin blade screwdriver, credit card thick plastic strip or other thin material.  Make sure the power is off.

Also, if at a later date you remove the latch because of the missing pull tab, which is a bad idea because you can pull the module out by accident with power on, make sure to be careful connecting and removing probes. I have a few modules that have broken tabs and if you pull the bnc connected device off,  you can slide the module out with the next move of pushing it back in automatically, popping something in the module. Lots of repairs to one volt meter module from an out/in with power on.

[Electro Fan]

I've been looking at a 7904 to add to my collection of equipment I wonder why I bought.

LOL - I'm sure you aren't the only one here who wonders such things.

Pull the 7a26 on the left.  You should be able to pop the latch with a thin blade screwdriver, credit card thick plastic strip or other thin material.  Make sure the power is off.

The pull tab on the outside of the plug-in is of course is attached to a nylon strip that attaches with a spring to the underside of the plug-in.  As simple as this mechanism appears I don't understand the theory of operation.  Where is the engagement occurring that latches the plugin into the mainframe?  Since the latching mechanism takes place without the pull tab with the nylon strip and the spring (as evidenced by the fact that my plug-in without those parts is firmly engaged in the mainframe), where is the "latch" point occurring that can be popped with a thin blade screw driver, credit card plastic, etc?  Thanks for the help.

Also, if at a later date you remove the latch because of the missing pull tab, which is a bad idea because you can pull the module out by accident with power on, make sure to be careful connecting and removing probes. I have a few modules that have broken tabs and if you pull the bnc connected device off,  you can slide the module out with the next move of pushing it back in automatically, popping something in the module. Lots of repairs to one volt meter module from an out/in with power on.

Thanks for the reminder on powering down before plugging/unplugging.

[Electro Fan]

Upon closer examination the pull tab nylon's bevel shape appears to run over a small black "speedbump latch" that forces the front of the plugin to rise maybe a mm or so which lifts the speedbump out of the speedbump latch "catcher" in the black guide rail that the plugin sits upon.  So I guess that would be the area to try to lift from underneath with a very thin flat screw driver blade, credit card, etc.  Just checking in advance, is this a technique that has been successfully executed without munging up a mainframe rail?   

[Electro Fan]

Sometimes it's better to be lucky than good.

Since the Sampling Head can be removed in this particular plugin it's possible to observe the Speedbump Latch from inside the plugin.  It appears that the speedbump is held down by a fairly stiff leafspring.  While it might be possible to push the speedbump latch up from underneath (it seems there is barely enough tolerance between the mainframe rail and the bottom of the plugin to get a business card in the slack space) it might be easier to use a needle nose pliers or some other tool to lift the leafspring up to take the tension off the speedbump latch.

Still hoping to hear from someone who has done plugin extraction surgery and who can report that the mainframe rail lived through the procedure.   

PS, maybe Microsoft can provide something a tad better for annotations than the freehand drawing tool in the next version of their Snipping Tool 

[Electro Fan]

Well, all is well that ends (mostly) well.  (Just need to figure out how to add a latch to the plugin and the Sampling Head, or find a couple new units.)

Attached you can see the black Speedbump Latch that is pressed down into the Speedbump Catcher in the mainframe rail.  It was extracted with the help of the tools shown. 

I wouldn't say it's impossible to remove a plugin without the good fortune of being able to extract the Speedbump Catcher from the opening made possible via the removal of the Sampling Head, but I can't see any good way to have done it from the bottom (inside the chassis from an open adjacent slot) with a conventional (non Sampling Head) plugin (not to mention one open slot is barely enough space to fit a hand).  More importantly (and I could be wrong) - given how the mainframe rails fit into the recessed rail slot on the bottom of the plugins I don't see how the speedbump could have been reached and pushed up with any kind of tool I can imagine.  I guess if Houdini could pick locks all tied-up underwater this might be an easier trick.

Maybe if you turned the mainframe upside down and removed the bottom it could be done - but even that doesn't look super doable given how the rails are installed in the chassis.  I think I was VERY lucky to have plugged in a plugin that didn't have the pull tab but happened to be the 7S11 with the opening for the Sampling Head.

Moral of this story:  DO NOT PLUG IN A TEKRONIX 7000 PLUGIN IF IT DOESN'T HAVE THE PULL TAB.  AND LIKEWISE, DON'T PLUG A SAMPLING HEAD INTO A PLUGIN IF THE SAMPLING HEAD DOESN'T HAVE IT'S LOCK/UNLOCK KNOB.  It might be possible to recover from a snafu with the Sampling Head getting stuck inside the plugin but it would be a major hassle to deal with a stuck plugin and it would be a bummer to mung up a rail inside the mainframe.   

[tautech]

Maybe if you turned the mainframe upside down and removed the bottom it could be done - but even that doesn't look super doable given how the rails are installed in the chassis. 

That's what I would have tried FIRST.
Not that I have one of these but an old CRO that I had with plugins, with the bottom panel removed most of what you needed to see for the latching mechanism was visible.

[Electro Fan]

Maybe if you turned the mainframe upside down and removed the bottom it could be done - but even that doesn't look super doable given how the rails are installed in the chassis. 

That's what I would have tried FIRST.
Not that I have one of these but an old CRO that I had with plugins, with the bottom panel removed most of what you needed to see for the latching mechanism was visible.

Seeing the parts is one thing, getting them to release is another.  I'm sure other folks have recovered from such a snafu but my strong recommendation would be - if a plugin doesn't have a pull tab DO NOT plug it in - or you will for sure be trying to fix it by taking it apart (or giving up a slot to a permanently installed plugin).

[Electro Fan]

Ok, so back to Square 1:  I can get a signal but it's very dot oriented - no normal line-like traces, and it's hard to display a signal at much over a 100kHz without either dots that are had to imagine as a square wave or no signal at all.  Anyone have any ideas on how to isolate the problem between the S1, 7S11, and 7T11?  Thanks, EF

[TerraHertz]

Upon closer examination the pull tab nylon's bevel shape appears to run over a small black "speedbump latch" that forces the front of the plugin to rise maybe a mm or so which lifts the speedbump out of the speedbump latch "catcher" in the black guide rail that the plugin sits upon.  So I guess that would be the area to try to lift from underneath with a very thin flat screw driver blade, credit card, etc.  Just checking in advance, is this a technique that has been successfully executed without munging up a mainframe rail?   

Ha ha... You mean the sampler was never actually locked in? You just didn't pull hard enough? And now you really locked in a plugin with no pull tab? Ha ha... Well live and learn.

Anyway, no, the black block is spring loaded (down), and the nylon pull-tab bevel pushes the black block up, so it clears the catcher in the mainframe rail. It doesn't move the whole plugin.

Extracting a plugin that's missing the pull-tab is easy. All you need is a 1/16" drill bit, or any other thin strong rod of that diameter.
See pics. Proceedure:
Push the plugin all the way back in, so the black block is not binding on that rail latch edge.
Insert the 1/16" metal rod in through the facia hole where the pull tab was, far enough that the end is definitely at the black block.
Lever the extended part of the rod DOWN. You can feel the springy movement of the block retracting. Lever the block up till you feel it bottom (not far) while being careful not to snap your thin metal rod.
While still applying that force on the rod, grab something solid on the front of the plugin, and pull firmly. It will come out with the same force as usually required on the pull tab.

As for operation of the sampler, I can't help. A while ago I was in a phase of buying Tek 7000 gear (while I could afford to) but the second stage, of getting it all tested & debugged, hasn't happened yet. Due to some unexpected life dramas. Some of which are now resolving, so hopefully soon. I collected original manuals for almost all the gear, which will make things a lot easier.
I strongly recommend buying original paper op-service manuals. For simultaneous learning curve and repair work, the greater ease and clarity of a paper book over crappy pdf on screen, makes a vast difference.

Edit to add: Incidentally, who knew that the original pull-tab design was made of metal? See 4th pic. I looked through my boxes of 7000 plugins for one with a missing pulltab to take the other photos, found none. Only this one, missing the tab spring. Had to push out the little pin to get the tab bar out.
The point of the pic of the metal pull-tab bar: It's quite easy to make replacements like that. This one is hardened steel, so it's been bent to shape then tempered then plated. The bar is 1.4mm thick. If anyone wants I can do a dimensioned drawing of it.

This 7A13 module is missing some knobs too. See pic 5. So... does anyone have a spare pull-tab spring and those knobs, they'd sell me cheaply?

Edit AGAIN: On closer examination, never mind about the knobs. See pic 6. This module is the early type, with mechanical gears for the comparison voltage setting. Two of the gears are broken, one missing entirely. So, not worth attempting to repair, I think. I have a working later version (no gears) 7A13 anyway.

[Electro Fan]

TerraHertz, very clever approach to using the drill bit to release the plugin.  I'm betting that someone will surf the web and find your solution and be very happy they did.  Hope you get your 7000 put together just the way you want - keep us posted.

If anyone to have some insight on how to get the sampling system playing well please post.  Thx, EF

[TerraHertz]

TerraHertz, very clever approach to using the drill bit to release the plugin.  I'm betting that someone will surf the web and find your solution and be very happy they did.

Yeah, this is why I wrote it in detail with pics. It's surely not just you with a stuck-in plugin.

Hope you get your 7000 put together just the way you want - keep us posted.

"My 7000" heh. See https://www.eevblog.com/forum/testgear/why-does-everyone-have-twenty-oscilliscopes/msg791312/#msg791312
When I get to the 'fix all the scopes' item on my todo list, it will be a lot of work. Yes, I'll write it up.

If anyone to have some insight on how to get the sampling system playing well please post.  Thx, EF

1. Make sure the mainframe works with non-sampling plugins first.
2. Get the manuals for the samplers you have, read the operator manual (twice) then work through the test and diagnostics checklists.

You seem surprised to see dots, not linear traces? Granted they should be regularly spaced. Something wrong with the sampling timebase?

[Electro Fan]

TerraHertz, very clever approach to using the drill bit to release the plugin.  I'm betting that someone will surf the web and find your solution and be very happy they did.

Yeah, this is why I wrote it in detail with pics. It's surely not just you with a stuck-in plugin.

Hope you get your 7000 put together just the way you want - keep us posted.

"My 7000" heh. See https://www.eevblog.com/forum/testgear/why-does-everyone-have-twenty-oscilliscopes/msg791312/#msg791312
When I get to the 'fix all the scopes' item on my todo list, it will be a lot of work. Yes, I'll write it up.

If anyone to have some insight on how to get the sampling system playing well please post.  Thx, EF

1. Make sure the mainframe works with non-sampling plugins first.
2. Get the manuals for the samplers you have, read the operator manual (twice) then work through the test and diagnostics checklists.

You seem surprised to see dots, not linear traces? Granted they should be regularly spaced. Something wrong with the sampling timebase?

Well - I guess at 23 you are ahead of par if par is 20

The mainframe works A-OK with non-sampling plugins.  The pulse in the attached image is generated by a Tek board that supposedly has a rise time of <200ps.  With a 7B92A and the 7A26 200MHz plugin used in the image it's probably showing closer to 1-2ns; what I'd like to figure out is how to get the sampling system to do it's magic - supposedly the S1 will go to 350ps but I haven't seen anything visually intelligible anywhere near that speed from the sampling system yet. 

I'm not sure how to determine if the suspected problem is with the 7S11, S1, or 7T11....

[dom0]

Edit AGAIN: On closer examination, never mind about the knobs. See pic 6. This module is the early type, with mechanical gears for the comparison voltage setting. Two of the gears are broken, one missing entirely. So, not worth attempting to repair, I think. I have a working later version (no gears) 7A13 anyway.

Could always attempt to re-engineer the small DMM of the newer versions. These ultra-small 7-segment LEDs are still available NOS for calculators and stuff like that.

[Electro Fan]

Here is some images showing a pulse through a 7A19 - spec'd to 1 GHz but the 7904 is limited to 500 MHz.  It seems to show a rise time of about 700ps - which is pretty much what is expected at 500 MHz. 

So... I think the mainframe is working properly - still trying to sort through what might not be working so well among the S-1, 7S11, and 7T11.  Anyone have a way to methodically test these 3 components?