Date   
Re: 2465 Fan Collet Thingy

n4buq
 

Hi Chuck,

That's not what I meant. When the stud is turned CCW, provided the collet doesn't break first, the threads in the collet itself would push the flanged part of the stud against the inside surface of the impeller. Of course, 10 times out of 9, the collet breaks before the threads allow the stud to turn because the stud is locked somewhat tightly against the back face of the collet.

Thanks,
Barry - N4BUQ

----- Original Message -----
From: "Chuck Harris" <cfharris@...>
To: TekScopes@groups.io
Sent: Monday, November 11, 2019 2:16:22 PM
Subject: Re: [TekScopes] 2465 Fan Collet Thingy

Hi Barry,

The screw threads do NOT engage the impeller body. It is
a tight through hole. So, if you turned the screw slot,
all you would be doing is torquing the screw free of the
collet body. Which breaks the head of the collet, freeing
the collet's jaws.

When the collet is new, simply loosening the nut will allow
it to relax enough to be pulled off of the shaft... especially
given that pulling moves the collet in the direction of
release.

When the collet has been tightened for 2 or 3 decades, the
plastic of the impeller stretches out a little bit and
relaxes, leaving a pit for the collet to get stuck into.

That is why you need to loosen the nut, and give it a tap.

-Chuck Harris

n4buq wrote:
It's strange to me how the stud that holds the collet tight appears to have
been designed so that reversing the screw (with the screwdriver slot)
would naturally push the collar on the stud against the impeller and push
the collet open. The instructions don't say to do that and to simply grab
the impeller and pull it but that doesn't seem to work (at least it didn't
for me and, supposedly, that could easily break the impeller or strain the
motor bearing surfaces, etc).

Fun stuff...

Thanks,
Barry - N4BUQ

----- Original Message -----
From: "Chuck Harris" <cfharris@...>
To: TekScopes@groups.io
Sent: Monday, November 11, 2019 12:19:01 PM
Subject: Re: [TekScopes] 2465 Fan Collet Thingy

It should be pretty clear that even plastic works very
well.


Re: 2465 Fan Collet Thingy

Chuck Harris
 

Looks like a MIG welder without the argon.

The vacuum would certainly reduce the power requirements for
the melt.

I wonder how warpage is managed?

-Chuck Harris

Bruce Griffiths wrote:

You mean something like this?:
https://www.sciaky.com/additive-manufacturing/electron-beam-additive-manufacturing-technology

Bruce
On 12 November 2019 at 04:34 Chuck Harris <cfharris@...> wrote:


I have a CD stud welder that can weld a 3/8" threaded stud onto
a piece of heavy sheet metal without burning the paint on the
other side... takes about 5ms to do the weld. Clearly a much
smaller "weld" would take less time and less energy.

I have thought that using a CD method to fuse the stock metal onto
the part being made might be effective. The big problem is avoiding
fusing the feedstock metal to the electrode.

A pulsed laser to a bed of powdered metal granules is probably
an ideal way of avoiding that issue... and avoiding metal distortion
due to the heat of fusion.

-Chuck Harris

Early 2456 power supply ripple on 87V...

Chuck Harris
 

I remember a thread a few years back where there was excess ripple
on the 87V supply in a 2465.

I just fixed such a problem on a properly recapped 2465.

The issue was there was 80mVpp ripple on the 87V test pin, that was
discovered during calibration. 80mvpp happens to be the max value
of ripple allowed, but more typically the ripple at that point is
1/10th that amount.

The ripple looked like an oscillation that was all over the place in
frequency, but steady in amplitude. It went away when the supply was
hot, but came back when the opamp U1281 was given a squirt of
freeze spray.

The scope was a pre B020000 unit, and was built without any feedback
stabilization in the feedback loop. A stabilization cap was added
in all scopes B020000 and later.

Adding in the missing C1222, a 0.1uf cap, to slows down Q1222 fixing
the problem.

-Chuck Harris

Re: 2465 Fan Collet Thingy

Bruce Griffiths
 

On 12 November 2019 at 04:34 Chuck Harris <cfharris@...> wrote:


I have a CD stud welder that can weld a 3/8" threaded stud onto
a piece of heavy sheet metal without burning the paint on the
other side... takes about 5ms to do the weld. Clearly a much
smaller "weld" would take less time and less energy.

I have thought that using a CD method to fuse the stock metal onto
the part being made might be effective. The big problem is avoiding
fusing the feedstock metal to the electrode.

A pulsed laser to a bed of powdered metal granules is probably
an ideal way of avoiding that issue... and avoiding metal distortion
due to the heat of fusion.

-Chuck Harris

EricJ via Groups.Io wrote:
It's out there now, but it's not trickled down to the hobbyist level yet. May be a while.

https://gpiprototype.com/metal-3d-printing

--Eric

On Nov 11, 2019 8:09 AM, Chuck Harris <cfharris@...> wrote:




Very cool, but not useful as the strength of these composite
filaments is entirely in the plastic binder.

-Chuck Harris

Re: 7S14 7S11 SAMPLER bias QUESTION

Miguel Work
 

Thanks Raymond, for that I´m studying and asking what I don´t understand.

Yes, S3 head DC coupled, but the preamplifier is AC coupled, it samples the DC level.
There is no blowby circuit in S3 head?

The goal is to understand in detail why 7S14 needs floating bias, to be able to modify it.

Regards

Miguel

-----Mensaje original-----
De: TekScopes@groups.io [mailto:TekScopes@groups.io] En nombre de Raymond Domp Frank
Enviado el: lunes, 11 de noviembre de 2019 20:19
Para: TekScopes@groups.io
Asunto: Re: [TekScopes] 7S14 7S11 SAMPLER bias QUESTION

On Mon, Nov 11, 2019 at 07:37 PM, Miguel Work wrote:


I´m looking S1 diagrams, preamp board seems to be AC coupled, Q45, C49.
Even earlier in the circuit *the signal* is AC coupled at C40, directly after the "sample-and-hold FET"


I´m not an expert, but it seems that S1 is ac coupled, and 7s14 is DC coupled,
Yes, signal-wise. Apart from the fact that the S1 is just a sampler and the 7S14 is a complete "general purpose" 'scope plugin, including sampler.
The 7S14 has DC to 1 GHz BW, whereas the S1 is a specialized sampler for fast edges and high frequencies, AC only, not meant to view DC.


DC coupling could explain the need to use floating bias in 7S14?
No, not in and by itself: The S3(A) sampler plugin for the 7S11 goes from DC to 1 GHz and uses the same non-floating bias.


But, looking 7s14 diagrams and like S1 head, the difference is in R45,
8M2 resistor feedback?
No. There are more (significant) differences. R45 and C52/62 are part of the blowby (LF feed-through compensating) circuit, not general feedback.

You really need to read more about the workings of this kind of samplers...

Raymond

Re: Tek 576 Curve Tracer, 3 mechanical parts wanted; Does anyone have a parts unit ?

Dave Brown
 

Where are you located? We have one or possibly two 576 parts units at the museum. If local, stop by and we can see what we have. If not, send us an email at contactus@.... We can likely look this weekend when the museum is open.

Dave

Re: 2465 Fan Collet Thingy

Chuck Harris
 

Hi Barry,

The screw threads do NOT engage the impeller body. It is
a tight through hole. So, if you turned the screw slot,
all you would be doing is torquing the screw free of the
collet body. Which breaks the head of the collet, freeing
the collet's jaws.

When the collet is new, simply loosening the nut will allow
it to relax enough to be pulled off of the shaft... especially
given that pulling moves the collet in the direction of
release.

When the collet has been tightened for 2 or 3 decades, the
plastic of the impeller stretches out a little bit and
relaxes, leaving a pit for the collet to get stuck into.

That is why you need to loosen the nut, and give it a tap.

-Chuck Harris

n4buq wrote:

It's strange to me how the stud that holds the collet tight appears to have been designed so that reversing the screw (with the screwdriver slot) would naturally push the collar on the stud against the impeller and push the collet open. The instructions don't say to do that and to simply grab the impeller and pull it but that doesn't seem to work (at least it didn't for me and, supposedly, that could easily break the impeller or strain the motor bearing surfaces, etc).

Fun stuff...

Thanks,
Barry - N4BUQ

----- Original Message -----
From: "Chuck Harris" <cfharris@...>
To: TekScopes@groups.io
Sent: Monday, November 11, 2019 12:19:01 PM
Subject: Re: [TekScopes] 2465 Fan Collet Thingy

It should be pretty clear that even plastic works very
well.

Re: Type 184 questions

bill K7WXW
 

On Sun, Nov 10, 2019 at 07:30 PM, Glenn Little wrote:


Are all of the power supply bypass capacitors in good condition, low
esr, correct value, low leakage?

Glenn
I am working through that now. Will check and fix any PSU problems (as I assume that 200mV of 10MHz oscillator is a problem) and then start looking for the other other potential sources of cross talk.

Re: 7S14 7S11 SAMPLER bias QUESTION

 

On Mon, Nov 11, 2019 at 07:37 PM, Miguel Work wrote:


I´m looking S1 diagrams, preamp board seems to be AC coupled, Q45, C49.
Even earlier in the circuit *the signal* is AC coupled at C40, directly after the "sample-and-hold FET"


I´m not an expert, but it seems that S1 is ac coupled, and 7s14 is DC coupled,
Yes, signal-wise. Apart from the fact that the S1 is just a sampler and the 7S14 is a complete "general purpose" 'scope plugin, including sampler.
The 7S14 has DC to 1 GHz BW, whereas the S1 is a specialized sampler for fast edges and high frequencies, AC only, not meant to view DC.


DC coupling could explain the need to use floating bias in 7S14?
No, not in and by itself: The S3(A) sampler plugin for the 7S11 goes from DC to 1 GHz and uses the same non-floating bias.


But, looking 7s14 diagrams and like S1 head, the difference is in R45,
8M2 resistor feedback?
No. There are more (significant) differences. R45 and C52/62 are part of the blowby (LF feed-through compensating) circuit, not general feedback.

You really need to read more about the workings of this kind of samplers...

Raymond

Re: 2465 Fan Collet Thingy

n4buq
 

It's strange to me how the stud that holds the collet tight appears to have been designed so that reversing the screw (with the screwdriver slot) would naturally push the collar on the stud against the impeller and push the collet open. The instructions don't say to do that and to simply grab the impeller and pull it but that doesn't seem to work (at least it didn't for me and, supposedly, that could easily break the impeller or strain the motor bearing surfaces, etc).

Fun stuff...

Thanks,
Barry - N4BUQ

----- Original Message -----
From: "Chuck Harris" <cfharris@...>
To: TekScopes@groups.io
Sent: Monday, November 11, 2019 12:19:01 PM
Subject: Re: [TekScopes] 2465 Fan Collet Thingy

It should be pretty clear that even plastic works very
well.

The only broken collets were broken by people that didn't
read the manual, and guessed wrongly about how to remove
the impeller... Usually their wrong guess involved what
the slot on the threaded shaft was for.

If you hold the shaft by the slot in the screw, loosen
the nut with a wrench, and then give the end of the shaft
a tap with the butt of your screwdriver's handle, you
will never break another collet.

-Chuck Harris

GerryR wrote:
I'm pretty sure that the slots can be done, too, but from the drawing, it
looks like
the part is a perfect candidate for a Swiss Screw Machine. If I was to
make this
part, I actually would try making it in two pieces, the main body and I
would insert
a threaded stud into the base, thereby giving threads all the way up to the
body and
not having to thread it on the lathe. Loctite on the approximately 4
threads in the
body would hold it. Any reason it can't be made from steel; brass would be
fine, as
well, but aluminum might be a little fragile for my liking. Just a
thought.


----- Original Message ----- From: "Renée" <@rjdeeter>
To: <TekScopes@groups.io>
Sent: Monday, November 11, 2019 11:29 AM
Subject: Re: [TekScopes] 2465 Fan Collet Thingy


I forgot....sorry
Renée

On 2019-11-11 7:55 a.m., Chuck Harris wrote:
Attachments are not allowed on this group.

The collet is in the files section. Log on to groups, and
search collet. It will come right up.

-Chuck Harris

GerryR wrote:
I can't see any attachment. If you have the drawing, can you email to
totalautomation1 at gee mail dot com. Thank you.

GerryR
KK4GER







Re: 2465 Fan Collet Thingy

Chuck Harris
 

Hi Tom,

If you didn't buy the scope new from tektronix,
you have no idea what it has been through since
it was made.

I think a probable scenario is the last person that
removed the fan turned the screw slot, and heard a
cracking sound, quickly stopped turning, and went off
to read the manual...

I have also heard of people breaking the collet, and
"repairing" the break with crazy glue. Once you get
the collet to pull up, it doesn't matter if all of the
collet's jaws are broken.

Except to the guy that got stuck holding the bag.

But we will never know for sure.

-Chuck Harris

Tom Gardner wrote:

On 11/11/19 18:19, Chuck Harris wrote:
It should be pretty clear that even plastic works very
well.

The only broken collets were broken by people that didn't
read the manual, and guessed wrongly about how to remove
the impeller... Usually their wrong guess involved what
the slot on the threaded shaft was for.

If you hold the shaft by the slot in the screw, loosen
the nut with a wrench, and then give the end of the shaft
a tap with the butt of your screwdriver's handle, you
will never break another collet.
Not entirely :(

I was well aware of that, and have removed a small number successfully using that
technique.

Nonetheless, the last one fell apart: all four parts of the tapered section fractured
at their base and, as they say, "it just came away in my hand".



Re: 7S14 7S11 SAMPLER bias QUESTION

Miguel Work
 

DC coupling could explain the need to use floating bias in 7S14?

-----Mensaje original-----
De: TekScopes@groups.io [mailto:TekScopes@groups.io] En nombre de Raymond Domp Frank
Enviado el: lunes, 11 de noviembre de 2019 19:20
Para: TekScopes@groups.io
Asunto: Re: [TekScopes] 7S14 7S11 SAMPLER bias QUESTION

On Mon, Nov 11, 2019 at 06:53 PM, Miguel Work wrote:


But, looking 7s14 diagrams and like S1 head, the difference is in R45,
8M2 resistor feedback? Is de S1 DC coupled?
Hi Miguel,
- AFAIK, the S1 contains no 8M2 resistor, certainly not in R45's location. R45 is 30K in my SM (070-0763-00). I don't know what "difference" you're referring to.
- All samplers for use in the 7S11 are DC coupled in the sense as we use it here.

Raymond

Re: 2465 Fan Collet Thingy

Tom Gardner
 

On 11/11/19 18:19, Chuck Harris wrote:
It should be pretty clear that even plastic works very
well.

The only broken collets were broken by people that didn't
read the manual, and guessed wrongly about how to remove
the impeller... Usually their wrong guess involved what
the slot on the threaded shaft was for.

If you hold the shaft by the slot in the screw, loosen
the nut with a wrench, and then give the end of the shaft
a tap with the butt of your screwdriver's handle, you
will never break another collet.
Not entirely :(

I was well aware of that, and have removed a small number successfully using that technique.

Nonetheless, the last one fell apart: all four parts of the tapered section fractured at their base and, as they say,  "it just came away in my hand".

Re: 7S14 7S11 SAMPLER bias QUESTION

Miguel Work
 

On Mon, Nov 11, 2019 at 10:19 AM, Raymond Domp Frank wrote:


AFAIK
Sorry is R45 in 7S14.

I´m looking S1 diagrams, preamp board seems to be AC coupled, Q45, C49. I´m not an expert, but it seems that S1 is ac coupled, and 7s14 is DC coupled, with dual fets. 8M2 seems DC feedback loop.

Re: 7S14 7S11 SAMPLER bias QUESTION

 

On Mon, Nov 11, 2019 at 06:53 PM, Miguel Work wrote:


But, looking 7s14 diagrams and like S1 head, the difference is in R45, 8M2
resistor feedback? Is de S1 DC coupled?
Hi Miguel,
- AFAIK, the S1 contains no 8M2 resistor, certainly not in R45's location. R45 is 30K in my SM (070-0763-00). I don't know what "difference" you're referring to.
- All samplers for use in the 7S11 are DC coupled in the sense as we use it here.

Raymond

Re: 2465 Fan Collet Thingy

Chuck Harris
 

It should be pretty clear that even plastic works very
well.

The only broken collets were broken by people that didn't
read the manual, and guessed wrongly about how to remove
the impeller... Usually their wrong guess involved what
the slot on the threaded shaft was for.

If you hold the shaft by the slot in the screw, loosen
the nut with a wrench, and then give the end of the shaft
a tap with the butt of your screwdriver's handle, you
will never break another collet.

-Chuck Harris

GerryR wrote:

I'm pretty sure that the slots can be done, too, but from the drawing, it looks like
the part is a perfect candidate for a Swiss Screw Machine. If I was to make this
part, I actually would try making it in two pieces, the main body and I would insert
a threaded stud into the base, thereby giving threads all the way up to the body and
not having to thread it on the lathe. Loctite on the approximately 4 threads in the
body would hold it. Any reason it can't be made from steel; brass would be fine, as
well, but aluminum might be a little fragile for my liking. Just a thought.


----- Original Message ----- From: "Renée" <@rjdeeter>
To: <TekScopes@groups.io>
Sent: Monday, November 11, 2019 11:29 AM
Subject: Re: [TekScopes] 2465 Fan Collet Thingy


I forgot....sorry
Renée

On 2019-11-11 7:55 a.m., Chuck Harris wrote:
Attachments are not allowed on this group.

The collet is in the files section. Log on to groups, and
search collet. It will come right up.

-Chuck Harris

GerryR wrote:
I can't see any attachment. If you have the drawing, can you email to
totalautomation1 at gee mail dot com. Thank you.

GerryR
KK4GER





Re: 7S14 7S11 SAMPLER bias QUESTION

 

On Mon, Nov 11, 2019 at 06:23 PM, Miguel Work wrote:

(Actually, Steve Ditter is quoted here:)

The circuit needs extremely low capacitance to ground.
That supports my idea that batteries in the sampler housing is a problem for significantly higher frequencies.


In Ed's suggested circuit, LEDs are used for a shunt regulator. They work well for this, but they will have a temp coefficient
for the forward voltage, which will be much larger than the Hg cell itself.
I don't know which topic contains Steve's quoted text. It seems that he refers to an application that needs no floating voltage but output voltage stability, whereas in the 7S14, exact voltage and output voltage vs. temperature aren't critical. Electrically floating, small size, long service life at low load current is, and the led solution fulfills that requirement (as did the Hg cells).

In my first 7S14, I had used two head-to-head leds as photovoltaic voltage sources/sensors for each channel, unlike Ed's "five couplers in series" design (Yahoo message March 15, 2012). As I had experienced and Ed reported later (ISTR), temperature dependence of light output only resulted in an easily corrected offset drift.
I used visible light leds because of their much higher voltage output than infrared leds: one correctly chosen led will do as a receiver.


Anytime one considers modifying a circuit in an instrument to "improve it" you really need to understand all the artifacts effecting its operation. It is never an easy task.
Sometimes it's not so much a desire to improve as a desire to continued use, as in this case. In this case, it turns out it was an easy task.

Raymond

Re: 2465 Fan Collet Thingy

GerryR
 

I'm pretty sure that the slots can be done, too, but from the drawing, it looks like the part is a perfect candidate for a Swiss Screw Machine. If I was to make this part, I actually would try making it in two pieces, the main body and I would insert a threaded stud into the base, thereby giving threads all the way up to the body and not having to thread it on the lathe. Loctite on the approximately 4 threads in the body would hold it. Any reason it can't be made from steel; brass would be fine, as well, but aluminum might be a little fragile for my liking. Just a thought.

----- Original Message -----
From: "Renée" <@rjdeeter>
To: <TekScopes@groups.io>
Sent: Monday, November 11, 2019 11:29 AM
Subject: Re: [TekScopes] 2465 Fan Collet Thingy


I forgot....sorry
Renée

On 2019-11-11 7:55 a.m., Chuck Harris wrote:
Attachments are not allowed on this group.

The collet is in the files section. Log on to groups, and
search collet. It will come right up.

-Chuck Harris

GerryR wrote:
I can't see any attachment. If you have the drawing, can you email to
totalautomation1 at gee mail dot com. Thank you.

GerryR
KK4GER

Re: Tek 576 noise and looping

peter bunge
 

I suspected that I could not attach photos.
Any one interested please see the file I uploaded with examples of looping
on pages 5 and 6.
File is: Tektronix 576 Curve Tracer Repair Log 1.pdf
It also covers several repairs and my current attempt to find out about the
noise.
PeterB

On Mon, Nov 11, 2019 at 12:19 PM Mlynch001 <@mlynch001> wrote:

"I'm not sure if pictures show up, let me know.
[image: 2N3568 low current.jpg]"

Peter,

Attachments are not allowed within our posts. However, you can post your
picture(s) in the "PHOTOS" section. Make a File and share the filename in
a post. Then anyone can see the photos. Please edit and reduce the size
of the files to save our limited storage space on the forum.

--
Michael Lynch
Dardanelle, AR



Re: 7S14 7S11 SAMPLER bias QUESTION

Miguel Work
 

But, looking 7s14 diagrams and like S1 head, the difference is in R45, 8M2 resistor feedback? Is de S1 DC coupled?

-----Mensaje original-----
De: TekScopes@groups.io [mailto:TekScopes@groups.io] En nombre de Miguel Work
Enviado el: lunes, 11 de noviembre de 2019 18:24
Para: TekScopes@groups.io
Asunto: Re: [TekScopes] 7S14 7S11 SAMPLER bias QUESTION

Thanks, the answer seems to be in preamplifier stage, I´m going to study it. Anyway my 7S14 is working well now with infrared leds.

Searching discussions, ditter2 said:

"
This may not be such a good idea for the Keithley. Check the circuit and see what they are doing with the Hg cells. In the case of the 7S14, they are used for a floating power supply to bias the sample gates. The circuit needs extremely low capacitance to ground.

I suspect the Keithley is using them as a voltage reference. My guess is this application does not require a floating supply with low C to ground. Hg cells are extremely stable and have very low temperature coefficients. In Ed's suggested circuit, LEDs are used for a shunt regulator. They work well for this, but they will have a temp coefficient for the forward voltage, which will be much larger than the Hg cell itself.

Anytime one considers modifying a circuit in an instrument to "improve it" you really need to understand all the artifacts effecting its operation. It is never an easy task.

"

-----Mensaje original-----
De: TekScopes@groups.io [mailto:TekScopes@groups.io] En nombre de Raymond Domp Frank Enviado el: lunes, 11 de noviembre de 2019 17:48
Para: TekScopes@groups.io
Asunto: Re: [TekScopes] 7S14 7S11 SAMPLER bias QUESTION

On Mon, Nov 11, 2019 at 05:51 AM, Ed Breya wrote:


Sampling gates and other front-end stuff have traditionally been
biased with constant current sources (high resistances from fairly
high supply voltages) and associated circuitry.
Hi Ed,
Thanks for your detailed explanation.

In my posts, I have tried to answer Miguel's original question, which he repeated a few times:

***Quote***

Hi, why 7S14 need cells to bias bridge diodes, and 7S11 no?
I possible to feed bias in 7S14 with +-50v and resistors, just like the 7S11?
***Unquote***

I haven't found an answer to that question in your post, please correct me if I'm wrong. I found the question interesting because it implies that the battery-less solution in the (older) 7S11 samplers is simpler and therefore would have been preferable to the battery-solution in the later 7S14.

My answer(s) concentrated on the following:
- Different designs are conceivable for one solution
- The 7S11 samplers solution, although not "needing" (floating) batteries but using the traditional current-source biasing, was in fact more complex than the 7S14 and probably had to be because of its older, more general and, especially, faster samplers.
- The 7S14 "got away" with the simpler solution because of its smaller target bandwidth; a counter-argument in a way.

One key consideration I had for the last point is, that the batteries in the 7S14, being physically rather large and at the "strobed, switching end" of the circuit would present too much of a load/distortion in sampling, at much higher frequencies than in the 7S14. Again, a counter-argument to the OP's premise in fact but a relevant point to me, if correct.

Can you confirm or reject my assumption?

Thanks in advance for your response.

Raymond