OT - CRT's

[Phil Hobbs]

On 5/23/19 9:10 PM, John Larkin wrote:

On Thu, 23 May 2019 20:47:28 -0400, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

On 5/23/19 11:09 AM, Joerg wrote:
On 2019-05-22 16:39, Chris Jones wrote:
On 23/05/2019 00:57, Joerg wrote:
On 2019-05-21 04:57, Bob Engelhardt wrote:
Is there still a use for CRT's?


Yes, in oscilloscopes. There is stuff that cannot be properly detected
on a digital oscilloscope, especially when it comes to random noise
from sources whose noise pattern isn't known a priori. The jobs where
you have to see "fuzz on fuzz" and sometimes figure out a pattern in
the unwanted fuzz to find out where that comes from. Many times I have
dragged an old analog scope scope from a client's clutter cabinet, did
a quick emergency "resurrection" on it and found their problem. Other
times we did an EBay rush order for a Tektronix 2465 or similar.

You can still buy analog scopes, in part for the above reason. Iwatsu
makes high-end ones and lower cost versions can be found on Amazon.

https://www.iti.iwatsu.co.jp/en/products/ss/ss_top_e.html
It says no longer available :-(


Indeed :-(

Maybe the number of people who are able to operate such scopes has
dropped into the noise. Which would not surprise me.

It could be like it is with Hammond tonewheel organs where production
stopped in 1975. Their sound is unsurpassed but luckily there are still
tons around. Restoration becomes tougher because they are old and most
of the more sensitive innards are mechanical parts. I needed a solid 20
hours to get one going again. Now the prices for used ones start to
rise. We just sold ours because my wife could not play it anymore due to
a wrist injury. In contrast to old pianos it found a buyer almost
instantly. Just like it is with Tektronix 2465 scopes.


https://www.amazon.com/dp/B00QXNY874
Gosh that is expensive.


Well, you can get cheaper ones:

https://www.amazon.com/Sinometer-Single-Channel-Oscilloscope-CQ5010C/dp/B00NB58FJE/



I have three Tek analogue scopes: the 475A and 2467 work, and the 466
storage scope doesn't. Fifteen or so years ago, I used to use a 2465A
quite a lot, but these days I mostly use DSOs. (*)

I still really like the 246x's triggering.

Cheers

Phil Hobbs

(*) a couple of TDS 784As (1 GHz 4GS/s), a TDS 694C (3 GHz 10 Gs/s), and
two 1180x sampling scope mainframes with about 20 plug-ins. Magic.

I have 20 SD-series plugins on my book shelf. Figure maybe $14K each
in the mid 1990s.

Mine are in a Rubbermaid tub under my bench. The ones that don't have
the original case are stuffed in a cut-out piece of soft plastic foam.

Do you have the 067-1331-00 Sampling Head Simulator?

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

 

[Phil Allison]

jurb...@gmail.com wrote:

Not many people make analog audio amplifiers these days.



** There is no other kind and millions are made each year.


OK fucker (no offense) let's pick this apart a bit.

Back in the 1970s you went to an audio store and bought a stereo. It was usually two channels but sometimes four. Now a 5.1 amp is technically six channels. A 7.1 is eight.

So let's say that the average number of channels in audio equipment has doubled. As such, if they only sold half as many units then they have sold just as many amps.

** 5 or more channel amps are only a fraction of ALL audio amplifiers sold.

Now we get to one of "those" questions. All amps are analog ? Well that is one way of looking at it but are you including totally digitally controlled class D amps ?

** Class D is an analogue technique - its a switching circuit not "digital".

Calling a class D amp " digital" is marketing puke.

A compact disk, regular CD, all digital.

** The signal storage is digital, ones and zeros representing binary numbers derived from sampling the audio - that is what the title refers to.



...... Phil

 

[Jan Panteltje]

On a sunny day (Thu, 23 May 2019 22:19:01 -0700 (PDT)) it happened
jurb6006@gmail.com wrote in
<138bb7da-5c88-4179-a0d7-a78e698aa957@googlegroups.com>:

.... snip...

Bottom line is my CROs made my life and lifestyle possible. I think I) was about
14 when i got my first one. It was a HUGE Hickok that I could barely carry
and had like a 3" screen. THEN my Uncle who worked for IBM gave me a Tek
310 that had been removed from service. It was the coolest thing I ever
had. then he gave me a 561A with a dual timebase plugin so I eventually figured
out how to work that.

No readouts, no digital, no nothing, if I have to make a choice between a meter
and a scope I will take the scope every time. If I have to I can make a
scope act as an ohmmeter. Also an ESR meter for lytics.

Me and scopes is brothers.

Scopes ANALOG scopes I have used to repair maybe a few thousand TV sets, VCRs, what have you,
studio equipment like quadruplex recorders, cameras, no end to it, system converters,
and I still have my Trio 10MHz dual channel TV repair scope as main scope and it still works
after 2019 - 1979 = 40 years!
ONE repair job when I accidently touched a booster output (few kV) with the probe,
and few years ago the soldering station dropped from the table on it
and cracked the graticule, so I made a new graticule.
Quality!

Before that I first build a tube one with DG7-32 CRT early sixties, mostly from some circuit I found in some magazine.
the second one was my own design (1968 or so) and used RTL logic, the same CRT, had triggering,
and the third one was later, a 300 MHz bandwidth with an East German CRT and based on a Tek circuit they accidently published in some magazine.
Talked to that guy later at Tek, he had a lot of headwind because he publised that, he told me.
Then I moved to on other place an donated the whole stuff..

So, now, and yesterday after reading all this, I checked ebay for CRTs, found even my old DG7-32,
many different Russian CRTs, so there is noting in the world here why you cannot build a fast analog scope as to 2019!!!

It is : The player, NOT the instrument.
That is why I still have not bothered to buy boat anchors and Rigols and what have you, a new digital one appears every year from China.
Important is that you understand the circuit, understand how the scope works and what it really shows.
Maybe a wet finger can then replace many thousand of dollars of equipments, a clue will do,
I agree with Joerg, I use my SSB radio a lot... for precision things like locking to rubidium reference check etc,
Some people stuff their 'lab' full of multi million dollar impressive looking stuff.
But a clue, just ONE clue, is a lot cheaper.
And in broadcasting it was always the result that counted, the show must go on,
freaked out directors and artists is a heavy load if a camera or recorder breaks down,
transmitting 'black' is a nono, find a way.
I have done that for many years, and always the show went on.
Clues and in depth knowledge of the equipment you are facing is essential for fault finding.
So is improvisation if under pressure.
Just wanted to say that, most people work without a minutes timeline.
And 'tronix is simple!

 

[Phil Hobbs]

On 5/23/19 9:07 PM, John Larkin wrote:

On Thu, 23 May 2019 16:51:15 -0700, Joerg <news@analogconsultants.com
wrote:

On 2019-05-23 12:00, John Larkin wrote:
On Thu, 23 May 2019 09:38:50 -0700, Joerg <news@analogconsultants.com
wrote:

On 2019-05-23 09:11, John Larkin wrote:

[...]

That's another way to look at a signal, the spectrum of the jitter.
Sort of what HP used to call "modulation domain."


That is extremely hard with a DSO. Easier with analog scopes and a
darkened room.

This was assumed to be random, gaussian jitter. But with the scope
signal averaging some, and teasing the pulse rate, we can delicately
heterodyne what is actually a big periodic jitter contributor.

https://www.dropbox.com/s/gunkxsfa0ujbot3/Sync_Jitter.JPG?dl=0

(note the 50 ps/div sweep. Few analog scopes ever managed that.)

The board has a 40 MHz VCXO locked to a 10 MHz OCXO, and a Zynq SOC
with outboard DRAM and an army of PLLs inside. We see this heterodyne
jitter at trigger multiples of 1 MHz and 3.2 MHz. Yuk.


Is there RAM bank paging or read/write chunks happening every usec, or
maybe every 321nsec? Communication with other boards?

In the end you'll have to find those sources. That is where near field
probes and a good receiver or spectrum analyzer come in handy. You'll
have to go over the whole board plus all its supply rails. Like crime
scene investigators looking for that lone shell casing in the dirt.

No luck so far! I could cut the jitter at least in half if I can track
down that clock leakage.

And I have to confess that that scope HAS a CRT. It's a 50 GHz digital
sampling scope with a magnetic deflection (vertical scan!) cathode ray
tube.

Magnetic deflection produces a much sharper picture, because it can
control the aberrations of the electron beam. Electron optics (i.e.
electrostatic deflection and focusing) can't do that nearly as well,
because it would require a combination of negative and positive lenses,
as in optics-optics, and Laplace's equation says that you can't make a
negative electron lens.

(A negative lens would require a potential that reached its maximum
somewhere away from a boundary, and you can't do that with Laplace's
equation--it's known as the 'maximum principle'.)

Cheers

Phil Hobbs




--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

 

On Thu, 23 May 2019 22:56:39 -0700 (PDT), jurb6006@gmail.com wrote:

Not many people make analog audio amplifiers these days.



** There is no other kind and millions are made each year.

OK fucker (no offense) let's pick this apart a bit.

Back in the 1970s you went to an audio store and bought a stereo. It was usually two channels but sometimes four. Now a 5.1 amp is technically six channels. A 7.1 is eight.

So let's say that the average number of channels in audio equipment has doubled. As such, if they only sold half as many units then they have sold just as many amps.

Now we get to one of "those" questions. All amps are analog ? Well that is one way of looking at it but are you including totally digitally controlled class D amps ? Even though they put out an analog signal, does that prove they are analog ?

Class-D is simply Pulse Width Modulation (PWM) i.e. it has a constant
amplitude but continuously variable pulse width, often generated with
a sawtooth and a comparator.

In Pulse Amplitude Modulation (PAM) the pulse width is constant, but
the pulse amplitude is continuously variable. Now extend the pulse
width so that the duty cycle is 100 % and you have continuous analog.

Both PWM and PAM are analog system. This is just sampling, but not
digitizing.

A true digital amplifier would have several independent switches, one
delivering current I to the load, a second switch 2I, the third one
4I, the fourth 8I and so on into the load.

A proportional digital hydraulic valve has multiple simple valves with
1, 2, 4, 8 and so on times hydraulic flow rates based on digital
control.

People seem to forget that ADCs do two separate things, first
sampling, which produce still an analog value and then quantization to
a discrete number of quantization steps, discarding variations outside
these steps (quantization noise). The reverse process is then in the
DAC.

A compact disk, regular CD, all digital. They yelled it from every mountaintop. But the signal at the jacks in the back is definitely analog unless you are using the digital, like optical or SPDIF or what the fuck ever.

But say you got a CD player that only has analog outputs, left and right. Is it then an analog device ?

Enquiring minds want to know but don't really care, just have too much time on their hands.

 

[Phil Hobbs]

On 5/23/19 10:10 PM, AK wrote:

On Thursday, May 23, 2019 at 8:10:10 PM UTC-5, John Larkin wrote:
On Thu, 23 May 2019 20:47:28 -0400, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

On 5/23/19 11:09 AM, Joerg wrote:
On 2019-05-22 16:39, Chris Jones wrote:
On 23/05/2019 00:57, Joerg wrote:
On 2019-05-21 04:57, Bob Engelhardt wrote:
Is there still a use for CRT's?


Yes, in oscilloscopes. There is stuff that cannot be
properly detected on a digital oscilloscope, especially
when it comes to random noise from sources whose noise
pattern isn't known a priori. The jobs where you have to
see "fuzz on fuzz" and sometimes figure out a pattern in
the unwanted fuzz to find out where that comes from. Many
times I have dragged an old analog scope scope from a
client's clutter cabinet, did a quick emergency
"resurrection" on it and found their problem. Other times
we did an EBay rush order for a Tektronix 2465 or similar.

You can still buy analog scopes, in part for the above
reason. Iwatsu makes high-end ones and lower cost versions
can be found on Amazon.

https://www.iti.iwatsu.co.jp/en/products/ss/ss_top_e.html
It says no longer available :-(


Indeed :-(

Maybe the number of people who are able to operate such scopes
has dropped into the noise. Which would not surprise me.

It could be like it is with Hammond tonewheel organs where
production stopped in 1975. Their sound is unsurpassed but
luckily there are still tons around. Restoration becomes
tougher because they are old and most of the more sensitive
innards are mechanical parts. I needed a solid 20 hours to get
one going again. Now the prices for used ones start to rise. We
just sold ours because my wife could not play it anymore due
to a wrist injury. In contrast to old pianos it found a buyer
almost instantly. Just like it is with Tektronix 2465 scopes.


https://www.amazon.com/dp/B00QXNY874
Gosh that is expensive.


Well, you can get cheaper ones:

https://www.amazon.com/Sinometer-Single-Channel-Oscilloscope-CQ5010C/dp/B00NB58FJE/





I have three Tek analogue scopes: the 475A and 2467 work, and the 466
storage scope doesn't. Fifteen or so years ago, I used to use a
2465A quite a lot, but these days I mostly use DSOs. (*)

I still really like the 246x's triggering.

Cheers

Phil Hobbs

(*) a couple of TDS 784As (1 GHz 4GS/s), a TDS 694C (3 GHz 10
Gs/s), and two 1180x sampling scope mainframes with about 20
plug-ins. Magic.

I have 20 SD-series plugins on my book shelf. Figure maybe $14K
each in the mid 1990s.


Speaking of CRTs.

I just found out that my state does not let you put CRTS or TVs out
for trash pickup because of the lead, mercury, etc.

A average TV can have 5 lbs of lead in the glass.

I could not find any free recyclers, they charge a fee.

Which is completely stupid. Lead oxide doesn't leach out of big chunks
of glass, and heavy metals don't go anywhere in landfills. The Oklo
natural reactor in Gabon went critical about 1.5 billion years back, and
ran for a few hundred thousand years, moderated by flowing ground water.
The resulting plume of heavy metal fission products has travelled only
a few hundred yards in a billion years. The plutonium hasn't even left
the individual uranium oxide grains where it was formed.

When I was a kid, I used to build stuff out of parts salvaged from old
TVs. Getting rid of the CRTs was fun--I'd stuff them in a big old
plastic trash can and shoot them with my slingshot from a safe distance.
Hitting the face plate hard enough to break it reduced the tube to
fragments of manageable size.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

 

[Sylvia Else]

On 21/05/2019 9:57 pm, Bob Engelhardt wrote:
> Is there still a use for CRT's?

Didn't you know? Every last bit of any limited resource has to be
divvied out to the majority chattering hordes, especially when doing so
helps the government's budget. The idea that some should be kept back
for minority use... that's communism, or racism, or whichever ism is
most likely to have the effect of suppressing dissent. Will no one think
of the children?

Sylvia.

 

[Sylvia Else]

On 24/05/2019 6:12 pm, Sylvia Else wrote:

On 21/05/2019 9:57 pm, Bob Engelhardt wrote:
Is there still a use for CRT's?

Didn't you know? Every last bit of any limited resource has to be
divvied out to the majority chattering hordes, especially when doing so
helps the government's budget. The idea that some should be kept back
for minority use... that's communism, or racism, or whichever ism is
most likely to have the effect of suppressing dissent. Will no one think
of the children?

Sylvia.

Er, that ended up in the wrong place, somehow - probably my ageing
fingers. It was meant to be about 24GHz.

Sylvia.

 

jurb6006@gmail.com wrote in
news:f685f301-b9d4-45a0-9acc-b592b0bcedf5@googlegroups.com:

But say you got a CD player that only has analog outputs, left and
right. Is it then an analog device ?

Enquiring minds want to know but don't really care, just have too
much time on their hands.

Remember before DVDs that a CD had a rating of AAD ADD or DDD.
This refered to the path the signal took to get put onto the disk.
That declaration of how the CD was made was important say in classical
renderings

A DDD disc is rare as the musical instrument itself is analog.

And still, after all the quaintisization, the finished product is an
analog signal 100% representative of the original instrument's analog
product.

 

AK <scientist77017@gmail.com> wrote in news:5ff11d39-9745-43cd-9652-
26014a3066b7@googlegroups.com:

A average TV can have 5 lbs of lead in the glass.

I could not find any free recyclers, they charge a fee.

Leaded glass is not an environmental hazard. It has NEVER been one.

So my question is How can "science" be part of your email address
when "science" is no part of your understanding of the world?

 

John Larkin <jjlarkin@highland_snip_technology.com> wrote in
news:gmgeeepord0ncnlkv9isbi95vi58e8gc87@4ax.com:

On Thu, 23 May 2019 16:51:15 -0700, Joerg
news@analogconsultants.com> wrote:

On 2019-05-23 12:00, John Larkin wrote:
On Thu, 23 May 2019 09:38:50 -0700, Joerg
news@analogconsultants.com> wrote:

On 2019-05-23 09:11, John Larkin wrote:

[...]

That's another way to look at a signal, the spectrum of the
jitter. Sort of what HP used to call "modulation domain."


That is extremely hard with a DSO. Easier with analog scopes
and a darkened room.

This was assumed to be random, gaussian jitter. But with the
scope signal averaging some, and teasing the pulse rate, we can
delicately heterodyne what is actually a big periodic jitter
contributor.

https://www.dropbox.com/s/gunkxsfa0ujbot3/Sync_Jitter.JPG?dl=0

(note the 50 ps/div sweep. Few analog scopes ever managed that.)

The board has a 40 MHz VCXO locked to a 10 MHz OCXO, and a Zynq
SOC with outboard DRAM and an army of PLLs inside. We see this
heterodyne jitter at trigger multiples of 1 MHz and 3.2 MHz.
Yuk.


Is there RAM bank paging or read/write chunks happening every
usec, or maybe every 321nsec? Communication with other boards?

In the end you'll have to find those sources. That is where near
field probes and a good receiver or spectrum analyzer come in
handy. You'll have to go over the whole board plus all its supply
rails. Like crime scene investigators looking for that lone shell
casing in the dirt.

No luck so far! I could cut the jitter at least in half if I can
track down that clock leakage.

And I have to confess that that scope HAS a CRT. It's a 50 GHz
digital sampling scope with a magnetic deflection (vertical scan!)
cathode ray tube.

It was my understanding that CRTs ALL have magnetic field beam
deflection... in both vectors.

 

Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote in
news:QNOdnRgW4eK8onrBnZ2dnUU7-anNnZ2d@supernews.com:

On 5/23/19 11:09 AM, Joerg wrote:
On 2019-05-22 16:39, Chris Jones wrote:
On 23/05/2019 00:57, Joerg wrote:
On 2019-05-21 04:57, Bob Engelhardt wrote:
Is there still a use for CRT's?


Yes, in oscilloscopes. There is stuff that cannot be properly
detected on a digital oscilloscope, especially when it comes to
random noise from sources whose noise pattern isn't known a
priori. The jobs where you have to see "fuzz on fuzz" and
sometimes figure out a pattern in the unwanted fuzz to find out
where that comes from. Many times I have dragged an old analog
scope scope from a client's clutter cabinet, did a quick
emergency "resurrection" on it and found their problem. Other
times we did an EBay rush order for a Tektronix 2465 or
similar.

You can still buy analog scopes, in part for the above reason.
Iwatsu makes high-end ones and lower cost versions can be found
on Amazon.

https://www.iti.iwatsu.co.jp/en/products/ss/ss_top_e.html
It says no longer available :-(


Indeed :-(

Maybe the number of people who are able to operate such scopes
has dropped into the noise. Which would not surprise me.

It could be like it is with Hammond tonewheel organs where
production stopped in 1975. Their sound is unsurpassed but
luckily there are still tons around. Restoration becomes tougher
because they are old and most of the more sensitive innards are
mechanical parts. I needed a solid 20 hours to get one going
again. Now the prices for used ones start to rise. We just sold
ours because my wife could not play it anymore due to a wrist
injury. In contrast to old pianos it found a buyer almost
instantly. Just like it is with Tektronix 2465 scopes.


https://www.amazon.com/dp/B00QXNY874
Gosh that is expensive.


Well, you can get cheaper ones:

https://www.amazon.com/Sinometer-Single-Channel-Oscilloscope-
CQ501
0C/dp/B00NB58FJE/



I have three Tek analogue scopes: the 475A and 2467 work, and the
466 storage scope doesn't. Fifteen or so years ago, I used to use
a 2465A quite a lot, but these days I mostly use DSOs. (*)

I still really like the 246x's triggering.

Cheers

Phil Hobbs

(*) a couple of TDS 784As (1 GHz 4GS/s), a TDS 694C (3 GHz 10
Gs/s), and two 1180x sampling scope mainframes with about 20
plug-ins. Magic.

Tek makes extremely good gear!

 

[Phil Hobbs]

On 5/24/19 7:51 AM, DecadentLinuxUserNumeroUno@decadence.org wrote:

John Larkin <jjlarkin@highland_snip_technology.com> wrote in
news:gmgeeepord0ncnlkv9isbi95vi58e8gc87@4ax.com:

On Thu, 23 May 2019 16:51:15 -0700, Joerg
news@analogconsultants.com> wrote:

On 2019-05-23 12:00, John Larkin wrote:
On Thu, 23 May 2019 09:38:50 -0700, Joerg
news@analogconsultants.com> wrote:

On 2019-05-23 09:11, John Larkin wrote:

[...]

That's another way to look at a signal, the spectrum of the
jitter. Sort of what HP used to call "modulation domain."


That is extremely hard with a DSO. Easier with analog scopes
and a darkened room.

This was assumed to be random, gaussian jitter. But with the
scope signal averaging some, and teasing the pulse rate, we can
delicately heterodyne what is actually a big periodic jitter
contributor.

https://www.dropbox.com/s/gunkxsfa0ujbot3/Sync_Jitter.JPG?dl=0

(note the 50 ps/div sweep. Few analog scopes ever managed that.)

The board has a 40 MHz VCXO locked to a 10 MHz OCXO, and a Zynq
SOC with outboard DRAM and an army of PLLs inside. We see this
heterodyne jitter at trigger multiples of 1 MHz and 3.2 MHz.
Yuk.


Is there RAM bank paging or read/write chunks happening every
usec, or maybe every 321nsec? Communication with other boards?

In the end you'll have to find those sources. That is where near
field probes and a good receiver or spectrum analyzer come in
handy. You'll have to go over the whole board plus all its supply
rails. Like crime scene investigators looking for that lone shell
casing in the dirt.

No luck so far! I could cut the jitter at least in half if I can
track down that clock leakage.

And I have to confess that that scope HAS a CRT. It's a 50 GHz
digital sampling scope with a magnetic deflection (vertical scan!)
cathode ray tube.


It was my understanding that CRTs ALL have magnetic field beam
deflection... in both vectors.

Nope. Scopes are all-electrostatic. That's why they don't focus as
well as CRT TVs.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

 

[John Larkin]

On Thu, 23 May 2019 22:41:09 -0700 (PDT), whit3rd <whit3rd@gmail.com>
wrote:

On Thursday, May 23, 2019 at 12:00:14 PM UTC-7, John Larkin wrote:
On Thu, 23 May 2019 09:38:50 -0700, Joerg <news@analogconsultants.com
wrote:

On 2019-05-23 09:11, John Larkin wrote:

That's another way to look at a signal, the spectrum of the jitter.
Sort of what HP used to call "modulation domain."


That is extremely hard with a DSO. Easier with analog scopes and a
darkened room.

This was assumed to be random, gaussian jitter. But with the scope
signal averaging some, and teasing the pulse rate, we can delicately
heterodyne what is actually a big periodic jitter contributor.

https://www.dropbox.com/s/gunkxsfa0ujbot3/Sync_Jitter.JPG?dl=0

(note the 50 ps/div sweep. Few analog scopes ever managed that.)

The board has a 40 MHz VCXO locked to a 10 MHz OCXO, and a Zynq SOC
with outboard DRAM and an army of PLLs inside. We see this heterodyne
jitter at trigger multiples of 1 MHz and 3.2 MHz. Yuk.

So, feed the Z axis of an analog scope with some of your suspect jitter
inducers, and look at X/Y plots where X = signal, Y = signal-through-a-delay-line.
and you have a fine analog sensor for that effect.

That sweep is 50 ps/div!


--

John Larkin Highland Technology, Inc

lunatic fringe electronics

 

[John Larkin]

On Fri, 24 May 2019 08:55:39 -0400, Phil Hobbs
<pcdhSpamMeSenseless@electrooptical.net> wrote:

On 5/24/19 7:51 AM, DecadentLinuxUserNumeroUno@decadence.org wrote:
John Larkin <jjlarkin@highland_snip_technology.com> wrote in
news:gmgeeepord0ncnlkv9isbi95vi58e8gc87@4ax.com:

On Thu, 23 May 2019 16:51:15 -0700, Joerg
news@analogconsultants.com> wrote:

On 2019-05-23 12:00, John Larkin wrote:
On Thu, 23 May 2019 09:38:50 -0700, Joerg
news@analogconsultants.com> wrote:

On 2019-05-23 09:11, John Larkin wrote:

[...]

That's another way to look at a signal, the spectrum of the
jitter. Sort of what HP used to call "modulation domain."


That is extremely hard with a DSO. Easier with analog scopes
and a darkened room.

This was assumed to be random, gaussian jitter. But with the
scope signal averaging some, and teasing the pulse rate, we can
delicately heterodyne what is actually a big periodic jitter
contributor.

https://www.dropbox.com/s/gunkxsfa0ujbot3/Sync_Jitter.JPG?dl=0

(note the 50 ps/div sweep. Few analog scopes ever managed that.)

The board has a 40 MHz VCXO locked to a 10 MHz OCXO, and a Zynq
SOC with outboard DRAM and an army of PLLs inside. We see this
heterodyne jitter at trigger multiples of 1 MHz and 3.2 MHz.
Yuk.


Is there RAM bank paging or read/write chunks happening every
usec, or maybe every 321nsec? Communication with other boards?

In the end you'll have to find those sources. That is where near
field probes and a good receiver or spectrum analyzer come in
handy. You'll have to go over the whole board plus all its supply
rails. Like crime scene investigators looking for that lone shell
casing in the dirt.

No luck so far! I could cut the jitter at least in half if I can
track down that clock leakage.

And I have to confess that that scope HAS a CRT. It's a 50 GHz
digital sampling scope with a magnetic deflection (vertical scan!)
cathode ray tube.


It was my understanding that CRTs ALL have magnetic field beam
deflection... in both vectors.

Nope. Scopes are all-electrostatic. That's why they don't focus as
well as CRT TVs.

Cheers

Phil Hobbs

Here's the CRT from a Tek 547. The vertical deflection plates are
actually a delay line so the deflection voltage tracks the electron
velocity. Beautiful piece of glass. The electron optics was probably
the best of any scope.

https://www.dropbox.com/s/evoq6p2nvzyl6wo/547_crt.JPG?dl=0

547 was a beautiful scope. I still have a few.

The 519 tube is more extreme, 1 GHz bandwidth, but the vertical
deflection goodies are hidden inside a box.

https://www.dropbox.com/s/r6c3zkwlqrayt53/519_CRT.JPG?dl=0

It looks like they tuned the location of each deflection segment.
Operating? In a vacuum? Did the tech wear a space suit?

Later tube scopes used post-deflection-mesh CRTs to increase vertical
sensitivity and make the tube shorter. That really wrecked spot size.

Hamamatsu made a tube-based sampling scope with optical input,
basically a streak tube with a slit and a PMT as output. I saw a demo
at the factory, but I don't think they ever commercialized it.


--

John Larkin Highland Technology, Inc

lunatic fringe electronics

 

[John Larkin]

On Fri, 24 May 2019 11:45:41 +0000 (UTC),
DecadentLinuxUserNumeroUno@decadence.org wrote:

Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote in
news:QNOdnRgW4eK8onrBnZ2dnUU7-anNnZ2d@supernews.com:

On 5/23/19 11:09 AM, Joerg wrote:
On 2019-05-22 16:39, Chris Jones wrote:
On 23/05/2019 00:57, Joerg wrote:
On 2019-05-21 04:57, Bob Engelhardt wrote:
Is there still a use for CRT's?


Yes, in oscilloscopes. There is stuff that cannot be properly
detected on a digital oscilloscope, especially when it comes to
random noise from sources whose noise pattern isn't known a
priori. The jobs where you have to see "fuzz on fuzz" and
sometimes figure out a pattern in the unwanted fuzz to find out
where that comes from. Many times I have dragged an old analog
scope scope from a client's clutter cabinet, did a quick
emergency "resurrection" on it and found their problem. Other
times we did an EBay rush order for a Tektronix 2465 or
similar.

You can still buy analog scopes, in part for the above reason.
Iwatsu makes high-end ones and lower cost versions can be found
on Amazon.

https://www.iti.iwatsu.co.jp/en/products/ss/ss_top_e.html
It says no longer available :-(


Indeed :-(

Maybe the number of people who are able to operate such scopes
has dropped into the noise. Which would not surprise me.

It could be like it is with Hammond tonewheel organs where
production stopped in 1975. Their sound is unsurpassed but
luckily there are still tons around. Restoration becomes tougher
because they are old and most of the more sensitive innards are
mechanical parts. I needed a solid 20 hours to get one going
again. Now the prices for used ones start to rise. We just sold
ours because my wife could not play it anymore due to a wrist
injury. In contrast to old pianos it found a buyer almost
instantly. Just like it is with Tektronix 2465 scopes.


https://www.amazon.com/dp/B00QXNY874
Gosh that is expensive.


Well, you can get cheaper ones:

https://www.amazon.com/Sinometer-Single-Channel-Oscilloscope-
CQ501
0C/dp/B00NB58FJE/



I have three Tek analogue scopes: the 475A and 2467 work, and the
466 storage scope doesn't. Fifteen or so years ago, I used to use
a 2465A quite a lot, but these days I mostly use DSOs. (*)

I still really like the 246x's triggering.

Cheers

Phil Hobbs

(*) a couple of TDS 784As (1 GHz 4GS/s), a TDS 694C (3 GHz 10
Gs/s), and two 1180x sampling scope mainframes with about 20
plug-ins. Magic.


Tek makes extremely good gear!

Not always. I've had bugs in newer Tek scopes, and support has been
terrible. We buy Rigol now.


--

John Larkin Highland Technology, Inc

lunatic fringe electronics

 

[John Larkin]

On Fri, 24 May 2019 02:40:17 -0400, Phil Hobbs
<pcdhSpamMeSenseless@electrooptical.net> wrote:

On 5/23/19 9:10 PM, John Larkin wrote:
On Thu, 23 May 2019 20:47:28 -0400, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

On 5/23/19 11:09 AM, Joerg wrote:
On 2019-05-22 16:39, Chris Jones wrote:
On 23/05/2019 00:57, Joerg wrote:
On 2019-05-21 04:57, Bob Engelhardt wrote:
Is there still a use for CRT's?


Yes, in oscilloscopes. There is stuff that cannot be properly detected
on a digital oscilloscope, especially when it comes to random noise
from sources whose noise pattern isn't known a priori. The jobs where
you have to see "fuzz on fuzz" and sometimes figure out a pattern in
the unwanted fuzz to find out where that comes from. Many times I have
dragged an old analog scope scope from a client's clutter cabinet, did
a quick emergency "resurrection" on it and found their problem. Other
times we did an EBay rush order for a Tektronix 2465 or similar.

You can still buy analog scopes, in part for the above reason. Iwatsu
makes high-end ones and lower cost versions can be found on Amazon.

https://www.iti.iwatsu.co.jp/en/products/ss/ss_top_e.html
It says no longer available :-(


Indeed :-(

Maybe the number of people who are able to operate such scopes has
dropped into the noise. Which would not surprise me.

It could be like it is with Hammond tonewheel organs where production
stopped in 1975. Their sound is unsurpassed but luckily there are still
tons around. Restoration becomes tougher because they are old and most
of the more sensitive innards are mechanical parts. I needed a solid 20
hours to get one going again. Now the prices for used ones start to
rise. We just sold ours because my wife could not play it anymore due to
a wrist injury. In contrast to old pianos it found a buyer almost
instantly. Just like it is with Tektronix 2465 scopes.


https://www.amazon.com/dp/B00QXNY874
Gosh that is expensive.


Well, you can get cheaper ones:

https://www.amazon.com/Sinometer-Single-Channel-Oscilloscope-CQ5010C/dp/B00NB58FJE/



I have three Tek analogue scopes: the 475A and 2467 work, and the 466
storage scope doesn't. Fifteen or so years ago, I used to use a 2465A
quite a lot, but these days I mostly use DSOs. (*)

I still really like the 246x's triggering.

Cheers

Phil Hobbs

(*) a couple of TDS 784As (1 GHz 4GS/s), a TDS 694C (3 GHz 10 Gs/s), and
two 1180x sampling scope mainframes with about 20 plug-ins. Magic.

I have 20 SD-series plugins on my book shelf. Figure maybe $14K each
in the mid 1990s.



Mine are in a Rubbermaid tub under my bench. The ones that don't have
the original case are stuffed in a cut-out piece of soft plastic foam.

Do you have the 067-1331-00 Sampling Head Simulator?

Cheers

Phil Hobbs

No, but I have a couple of the extension cables, which are still rare
and expensive. It's great to get the sampling head a few inches from
the signal source. Really fast signals don't travel well over feet of
coax. Which is why I wonder what is the utility of that million dollar
100 GHz LeCroy scope someone posted a link to... how go you get a
signal to it?


https://www.dropbox.com/s/wkiehn6iowq3emf/TDR_3Vplane.JPG?dl=0


I have posted pics of the guts of the SD24 TDR head. I can do that
again if anyone is interested.

Hey, the pics are online:

http://w140.com/tekwiki/wiki/SD-24


--

John Larkin Highland Technology, Inc

lunatic fringe electronics

 

[Joerg]

On 2019-05-24 07:18, John Larkin wrote:

On Fri, 24 May 2019 08:55:39 -0400, Phil Hobbs
pcdhSpamMeSenseless@electrooptical.net> wrote:

On 5/24/19 7:51 AM, DecadentLinuxUserNumeroUno@decadence.org wrote:
John Larkin <jjlarkin@highland_snip_technology.com> wrote in
news:gmgeeepord0ncnlkv9isbi95vi58e8gc87@4ax.com:

On Thu, 23 May 2019 16:51:15 -0700, Joerg
news@analogconsultants.com> wrote:

On 2019-05-23 12:00, John Larkin wrote:
On Thu, 23 May 2019 09:38:50 -0700, Joerg
news@analogconsultants.com> wrote:

On 2019-05-23 09:11, John Larkin wrote:

[...]

That's another way to look at a signal, the spectrum of the
jitter. Sort of what HP used to call "modulation domain."


That is extremely hard with a DSO. Easier with analog scopes
and a darkened room.

This was assumed to be random, gaussian jitter. But with the
scope signal averaging some, and teasing the pulse rate, we can
delicately heterodyne what is actually a big periodic jitter
contributor.

https://www.dropbox.com/s/gunkxsfa0ujbot3/Sync_Jitter.JPG?dl=0

(note the 50 ps/div sweep. Few analog scopes ever managed that.)

The board has a 40 MHz VCXO locked to a 10 MHz OCXO, and a Zynq
SOC with outboard DRAM and an army of PLLs inside. We see this
heterodyne jitter at trigger multiples of 1 MHz and 3.2 MHz.
Yuk.


Is there RAM bank paging or read/write chunks happening every
usec, or maybe every 321nsec? Communication with other boards?

In the end you'll have to find those sources. That is where near
field probes and a good receiver or spectrum analyzer come in
handy. You'll have to go over the whole board plus all its supply
rails. Like crime scene investigators looking for that lone shell
casing in the dirt.

No luck so far! I could cut the jitter at least in half if I can
track down that clock leakage.

And I have to confess that that scope HAS a CRT. It's a 50 GHz
digital sampling scope with a magnetic deflection (vertical scan!)
cathode ray tube.


It was my understanding that CRTs ALL have magnetic field beam
deflection... in both vectors.

Nope. Scopes are all-electrostatic. That's why they don't focus as
well as CRT TVs.

Cheers

Phil Hobbs

Here's the CRT from a Tek 547. The vertical deflection plates are
actually a delay line so the deflection voltage tracks the electron
velocity. Beautiful piece of glass. The electron optics was probably
the best of any scope.

https://www.dropbox.com/s/evoq6p2nvzyl6wo/547_crt.JPG?dl=0

547 was a beautiful scope. I still have a few.

The 519 tube is more extreme, 1 GHz bandwidth, but the vertical
deflection goodies are hidden inside a box.

https://www.dropbox.com/s/r6c3zkwlqrayt53/519_CRT.JPG?dl=0

It looks like they tuned the location of each deflection segment.
Operating? In a vacuum? Did the tech wear a space suit?

They probably did that in TDR-fashion in open air.

Later tube scopes used post-deflection-mesh CRTs to increase vertical
sensitivity and make the tube shorter. That really wrecked spot size.

Hamamatsu made a tube-based sampling scope with optical input,
basically a streak tube with a slit and a PMT as output. I saw a demo
at the factory, but I don't think they ever commercialized it.

It's all going digital which, for nasty noise hunts, isn't so great.

--
Regards, Joerg

http://www.analogconsultants.com/

 

[John Larkin]

On Wed, 22 May 2019 14:59:44 -0700 (PDT), Phil Allison
<pallison49@gmail.com> wrote:

Steve Wilson wrote:




Not many people make analog audio amplifiers these days.



** There is no other kind and millions are made each year.

Playing fast and loose with the facts is Steve's middle name.




..... Phil

I have a customer who wants us to make a 3-phase 400 Hz power source
to simulate a PM alternator. I was thinking we could buy some
already-built class-D amp boards to do the power bits, say 150 watts
per channel or so. I'd rather buy than build that part.

We could use three of this eval board (we have a couple and they work
great) but they are big and have goofy connectors...

https://www.digikey.com/product-detail/en/texas-instruments/TPA3255EVM/296-46782-ND/7219409

Like most eval boards, it has a zillion jumper options.

Can anyone suggest a good-quality small equivalent class-D amp board?
We wouldn't want to use the really cheap ebay stuff which might not be
available next month.





--

John Larkin Highland Technology, Inc

lunatic fringe electronics

 

[John Larkin]

On Fri, 24 May 2019 18:13:49 +1000, Sylvia Else <sylvia@email.invalid>
wrote:

On 24/05/2019 6:12 pm, Sylvia Else wrote:
On 21/05/2019 9:57 pm, Bob Engelhardt wrote:
Is there still a use for CRT's?

Didn't you know? Every last bit of any limited resource has to be
divvied out to the majority chattering hordes, especially when doing so
helps the government's budget. The idea that some should be kept back
for minority use... that's communism, or racism, or whichever ism is
most likely to have the effect of suppressing dissent. Will no one think
of the children?

Sylvia.

Er, that ended up in the wrong place, somehow - probably my ageing
fingers. It was meant to be about 24GHz.

Sylvia.

OK, we'll forget the children.


--

John Larkin Highland Technology, Inc

lunatic fringe electronics