Date   

TLS216 at Singer

stefan_trethan
 

Yours for only 1422eur....
<http://www.helmut-singer.de/stock/-128652307.html>

I didn't even know this thing existed.
It looks to me like 16 real analog channels, not digital like in current MSOs?
16 input channels at 8 bit and 2Gs, that's quite something, must be a
challenge to keep that cool.

I thought some of you guys might enjoy looking at it...

There is also a video on the tube, haven't watched it yet so can't say
if it is any good:
<https://www.youtube.com/watch?v=1VyXamEvi_U>

ST


Re: problem with parametric measurements on 2445B

John Ferguson
 

good idea. I'll try it. thanks much, john

On 04/05/2016 09:04 PM, Lee Vogt @Greenie [TekScopes] wrote:

You may have try without the scope terminated with the 50 ohm internal
termination switch on.

Sent from my Verizon Wireless 4G LTE DROID

"john jferg977@... [TekScopes]" <TekScopes@...> wrote:


Let me try this question again. I have a 2445B. If I feed a 400 Hz Sine
wave into it at 10 volts from my FG501, I can measure both voltage and
frequency with the Cursors. If i try the MEASURE button and pick Volts,
I get a reasonable response. If I try for frequency, I get a note "CH1:
SMALL OR LOW REP RATE SIGNAL"
and,of course no frequency.

I thought I might have some other setting wrong so I read the manual
again, set offset to Minimum and tried everything else I could think of,
but no joy. Rise time and Fall time show same "small or low rep rate"
error message.

I get no errors on the POST, and as nearly as I can tell everything else
works.

Am I doing something wrong, or does my scope need some work?

John




------------------------------------

------------------------------------


------------------------------------

Yahoo Groups Links


[Non-text portions of this message have been removed]


Tek 2465 DMM Option - Part Assistance Requested

MJATS@...
 

Tek 2465 with DMM option. I have been slowly getting the DMM fully functional and calibrated and everything was going well until recently. I fixed a drift issue by re-capping the inverter and power supply boards. I managed to get the DCV calibrated up thru 190 VDC and the Ohms was fully calibrated. ACV was pending. I wanted to calibrate the 500 VDC range with a unit that I had cobbled together but the results were disastrous. I think the unit went unstable and threw a high voltage spike into the DMM and I now have an offset error on both DCV and Ohms. AC is OK. I checked every resistor in the attenuator and input network and they are OK. So now I'm going after the active components. Troubleshooting in this area is very difficult due to the high impedances. Trying to check functions with another DMM or scope loads down the circuits. So I'd like to source the following list of parts and replace them as a matter of course.

I have these parts:

Q4972 and Q4973 IRFD113
Q4980 IRF530

Looking for these parts:

Q4970 and Q4971 Tek 151-1103-00 N Channel FET in T0-72 package, MFG is Siliconix with a number of
DM1001 but I cannot find any info. Appears to be an obsolete part. Any recommend a substitute?


These parts were made by Tek:

Q5070 Tek 151-1077-01 2 N Channel FET's in one package
U4970 Tek 151-1838-01 Op amp
U5060 Tek 151-1191-01 Dual B-FET op amp

Any help would be greatly appreciated.....Mike


Re: problem with parametric measurements on 2445B

 

You may have try without the scope terminated with the 50 ohm internal termination switch on.

Sent from my Verizon Wireless 4G LTE DROID

"john jferg977@... [TekScopes]" <TekScopes@...> wrote:


Let me try this question again. I have a 2445B. If I feed a 400 Hz Sine
wave into it at 10 volts from my FG501, I can measure both voltage and
frequency with the Cursors. If i try the MEASURE button and pick Volts,
I get a reasonable response. If I try for frequency, I get a note "CH1:
SMALL OR LOW REP RATE SIGNAL"
and,of course no frequency.

I thought I might have some other setting wrong so I read the manual
again, set offset to Minimum and tried everything else I could think of,
but no joy. Rise time and Fall time show same "small or low rep rate"
error message.

I get no errors on the POST, and as nearly as I can tell everything else
works.

Am I doing something wrong, or does my scope need some work?

John




------------------------------------

------------------------------------


------------------------------------

Yahoo Groups Links



Re: For sale in UK: 549 with 1A4 plugin

Brian <brianclarke01@...>
 

And if you want a spare manual for the TEK 1A4, I have one available for AU$20 plus postage.
73 de Brian, VK2GCE.

From: mailto:TekScopes@...
Sent: Wednesday, April 06, 2016 12:16 AM
To: TekScopes@...
Subject: [TekScopes] For sale in UK: 549 with 1A4 plugin


Hi Tek fans,

Due to lack of space, I reluctantly have to part with my 549 and its 1A4
plugin. I restored it to clean working order a couple of years ago and it's
still in excellent shape.

You can read the full restoration story starting here:

http://martin-jones.com/2013/06/08/t...oscope-part-1/

And there are lots of nice photos here:
http://martin-jones.com/2016/04/05/tektronix-549-storage-oscilloscope-restored/
and even a movie here:
https://www.youtube.com/watch?v=1SRtm7eva0A

The Tektronix 549 is the only storage scope in the Tektronix 500 series of
plug-in scopes. It dates from the late 1960s, and uses an analogue storage
tube. It's bistable rather than variable-persistence, and has the
enormously cool feature that storage mode can be used separately in the
upper and lower halves of the screen.

http://w140.com/tekwiki/wiki/549

It's a big, heavy, power-hungry monster. It weighs 65lbs (30kg) or so and
draws more than 500W from the mains. I wouldn't say it's a practical
everyday scope, and it's definitely not for beginners. The brightness from
the storage tube is low (it's best used in a darkened room) and getting the
best results from it needs patience. Using it is like playing a musical
instrument. It has all the usual Tek trimmings: dual timebases, multi-level
calibrator, and connections for every imaginable accessory. Its bandwidth
is about 30MHz but sadly it doesn't have tunnel diode triggering so really
fast stuff is a bit of a chore.

This example of the 549 comes with the elusive 1A4 plugin. This plugin has
four fully-featured channels, and allows dozens of combinations of
chopping/alternating between them. You can freely choose the trigger source
from any of the four channels. I know this plugin is particularly sought
after, and this one is in good working condition on all four channels,
albeit a bit drifty until it's warmed up.

http://w140.com/tekwiki/wiki/1A4

There are not many of these scopes around in this condition. It even has
the correct rectangular CRT bezel. This particular one, being built in
Guernsey, doesn't have the 'Bumble Bee' paper capacitor problem that
plagues US-built examples. It's also had a replacement hand-made EHT
transformer which doesn't suffer from the lossy epoxy problem that the
original transformer had.

Calibrated? No, not formally, but I've adjusted everything to within spec
as best I can. I'd be surprised if anything was more than a few percent
wrong. And the icing on the cake: I've got the printed paper manuals for it
and the plugin. They themselves are works of art.

The price for this slice of technology history? I'm asking £120, including
the plugin and manuals. Come and get it from Cambridge, UK. I will be
driving across Europe in a couple of weeks time, so may even be able to
deliver to somewhere between Cambridge and Warsaw.

Chris

[Non-text portions of this message have been removed]


--------------------------------------------------------------------------------
Posted by: Chris Jones <chris@...>
--------------------------------------------------------------------------------


--------------------------------------------------------------------------------

.



[Non-text portions of this message have been removed]


problem with parametric measurements on 2445B

John Ferguson
 

Let me try this question again. I have a 2445B. If I feed a 400 Hz Sine wave into it at 10 volts from my FG501, I can measure both voltage and frequency with the Cursors. If i try the MEASURE button and pick Volts, I get a reasonable response. If I try for frequency, I get a note "CH1: SMALL OR LOW REP RATE SIGNAL"
and,of course no frequency.

I thought I might have some other setting wrong so I read the manual again, set offset to Minimum and tried everything else I could think of, but no joy. Rise time and Fall time show same "small or low rep rate" error message.

I get no errors on the POST, and as nearly as I can tell everything else works.

Am I doing something wrong, or does my scope need some work?

John


Re: SG504 Frq. Display & Digital amplitude flattening.

mosaicmerc
 

Sure, but it takes a disciplined operator with a reasonably steady hand and a smooth SG504 dial knob. Optical feedback from the system may help. That's why I thought about a calibration phase during which the operator would be guided by arrows and a (rough) frequency delta to dial in and a visible indication when to hold steady (discrete calibration points).. Operator skill would develop quickly, even within one calibration cycle. Not even necessary to use a monotonous set of point settings. It would also simplify some algorithms. Open loop calibration could be V2. Anyway, experimentation during development will show what's achievable and practical

Optical feedback is a good idea. The head can light an LED at the sample points so the operator can pause and continue when the led goes off.


Re: AFG5101 Repair - photograph needed

eboytoronto
 

Egge,


That is great information !! exactly what I needed.


It seems that the back panel has the minimum number of holes. Two slots for the pcb connectors and one hole for the heatsink bracket.


Also included in the link, was the instructions for setting the contrast on the LCD. I have reproduced the instructions here:


Hint, when you turn it on the display will be barely legible. Press SPCL, then the up down buttons
until LCD is in the window, press enter, arrow up until you are at 99 (its easy, don t give up!), now
press enter again to turn on the back light. Still looks crappy, yes lcd displays have came a long ways
since the 80 s (see photo). Time to complete step, 10 seconds if you ve done it before; 5 or 10 minutes
if I didn't leave this hint.




Thank you !!


Regards,


John


Re: SG504 Frq. Display & Digital amplitude flattening.

 

If we assume a 1 minute sweep we're moving at about 13+Mhz per second. With a 4mS sampling window
we have a .054 Mhz movement, in the window. I don't see this as presenting a significant issue since
the amplitude mapping takes place at 3.2Mhz steps. Thus it's straightforward to do a moving
amplitude average over 16 samples (0.9 Mhz) and apply this moving avg as it crosses the 3.2Mhz
markers to create the correction map.
Sure, but it takes a disciplined operator with a reasonably steady hand and a smooth SG504 dial knob. Optical feedback from the system may help. That's why I thought about a calibration phase during which the operator would be guided by arrows and a (rough) frequency delta to dial in and a visible indication when to hold steady (discrete calibration points).. Operator skill would develop quickly, even within one calibration cycle. Not even necessary to use a monotonous set of point settings. It would also simplify some algorithms. Open loop calibration could be V2. Anyway, experimentation during development will show what's achievable and practical.

Raymond


Re: SG504 Frq. Display & Digital amplitude flattening.

mosaicmerc
 

I was considering the implications and complications of sampling a signal that is changing continously and unpredictably while calibration is going on.
Raymond

A valid point....

To the details then:

If we assume a 1 minute sweep we're moving at about 13+Mhz per second. With a 4mS sampling window we have a .054 Mhz movement, in the window. I don't see this as presenting a significant issue since the amplitude mapping takes place at 3.2Mhz steps. Thus it's straightforward to do a moving amplitude average over 16 samples (0.9 Mhz) and apply this moving avg as it crosses the 3.2Mhz markers to create the correction map.


Re: SG504 Frq. Display & Digital amplitude flattening.

 

Yes, the MC12080 certainly deserves consideration. A few years ago, I bought a few but never came to play with them.

I think for calibration, taking discrete points with uP (-display) feedback to the user on frequency- and time-intervals to employ may be workable and ease development. A second version may improve on it.

Raymond


Re: SG504 Frq. Display & Digital amplitude flattening.

mosaicmerc
 

@1: Is HiSpeed CMOS fast enough these days to allow 1 GHz frequency measurement and if so, at what power dissipation?

Maybe the AHC, but perhaps the MC12080 prescaler with a high enough Resitor load value to get the Vpp swing to trigger the uC counter input is better. It's min op. voltage is 4.5V though. Li ion is dependable to 3.6V or so. Battery power has to be looked at.

@ 3: I think sampling and timestamping the samples during calibration will need careful thought to make it work. Sampling takes the full gate time of the frequency measurement.

With 20Khz quantization=> a 4mS counter window (plus divide by 80 prescaler) with a 16 bit counter will do.
ADC oversampling to 12bit will take 160uSec . 10 bit is much better at 10uSec. Using a 4.096Vref built in to the uC and half wave Schottky detection I get a 8.192Vpp CW range to quantize with 10 bits at the fastest speeds.
That's about 8mVpp . Better than -30dB precision at the lowest SG504 signal level (worst case 10 bit quantization).

So a 4000uS freq sampling window with an interleaved 10 or 12bit ADC amplitude sample is required.

5 x the RC TC of the head CW amplitude schottky sampling with a 50 ohm into a 10nF S&Hold is 2.5 uSec. So that's not really much of a constraint.

I imagine one a fixed frq is dialed in the micro can over sample the frequency counting with a larger window and get better precision. To get to 1Khz will require an 80mS window, which is reasonable as updating the LCD is limited at around a 333mS period for legible data. That means a 4x averaging can take place on the freq. sample and be displayed at a 1/3 sec rate with 1Khz quantization.


--


Re: SG504 Frq. Display & Digital amplitude flattening.

 

DC sampling is @ 1uSec per bit. 10uSec per 10bit sample...Nyquist oversample 160 uSec
for a 12bit amplitude sample.
I was considering the implications and complications of sampling a signal that is changing continously and unpredictably while calibration is going on.

Raymond


Re: SG504 Frq. Display & Digital amplitude flattening.

 

On 05 Apr 2016 06:30:01 -0700, you wrote:

I think the idea is very nice in terms of usefulness, however, IMHO it may be taking things a bit too far. I see a few issues that would have to be dealt with, especially if you'd want to power the thing from a Li battery: The continuous display of the frequency would be the least of your problems. I mean the *displaying* itself only.

1. A digital frequency meter operating at up to 1+ GHz running for long hours on a lithium battery is no easy thing to accomplish, to say the least. Tricks may be possible because of the limited band.
2. Creating D/A (level control) and A/D (level measuring) circuits with resolution and accuracy matching the (desired) performance of the head and consuming very little energy (again, Li battery), may be a challenge.
The frequency counter signal conditioning is going to draw so much
power that everything else can be ignored if basic power saving is
implemented.

3. Digital calibration/normalisation is a nice thing. Your uP would have to recognise sweeping (too fast) during calibration, take care of all intermediate (non-calibrated) frequencies during normal use, creating a fluent interpolation. The simplest uP probably wouldn't do. I wonder whether sweeping during sampling frequencies would work at all.
Throughput will be limited by the frequency counter gate time so the
uP only has to deal with sample rates in the 0.1 to 1 Hz range. That
leaves plenty of time between samples for a low end 8 bit processor
lacking floating point and/or DSP capability. This would be a good
application for fixed point math anyway.

4. Noise from the digital side would have to be very well controlled. That takes up physical space.
I do not think noise would be an issue. It is a leveled signal
generator and not a local oscillator source.

5. I'm not convinced that putting the display in the head would be a good thing. The limited viewing angle comes to mind
I am not either. The leveling head has just one job and it should do
that while being as small as possible. Also the SG504 already has a
frequency monitor output; install a DC508 or DC508A frequency counter
beside it if you want to know the exact frequency. It is an
interesting thought exercise though.


Re: SG504 Frq. Display & Digital amplitude flattening.

mosaicmerc
 

@1: Is HiSpeed CMOS fast enough these days to allow 1 GHz frequency measurement and if so, at what
power dissipation?

Maybe AHC but I'm looking @ a MC12080 prescaler. Divide by 80 with a 4mS counter window should be good to handle the 1.050 Ghz freq. input. 20Khz quantization though.



@ 3: I think sampling and timestamping the samples during calibration will need careful thought to make it work. Sampling takes the full gate time of the frequency measurement.

yeah, 4mS.
ADC sampling is @ 1uSec per bit. 10uSec per 10bit sample...Nyquist oversample 160 uSec for a 12bit amplitude sample.


Re: 7S12 TDR Cable

 

Hi Raymond,
Thanks for the correction. You are correct, I meant the S-4 head.
So many heads, so little time to use them all! Sigh!

I even have an S42 Optical to Electrical head which is quite rare. They
couldn't make it fit in the sampling head form factor so they had to make it
an inch longer. As a result it sticks out about an inch from the front of
the 7S11 which makes it look funny / ugly. I imaging this gave Deane Kidd
fits when he had to approve it for final shipment.

I'm finally starting to play with electro-optics so I may have a chance to
use the S-42 head.

Dennis Tillman W7PF

-----Original Message-----
From: TekScopes@... [mailto:TekScopes@...]
Sent: Tuesday, April 05, 2016 1:59 PM
To: TekScopes@...
Subject: [TekScopes] Re: 7S12 TDR Cable

That 7S11 will require a sampling head such as an S-5 25pSec head
I guess that's an S-4 sampling head ;-(

Raymond





------------------------------------
Posted by: @Raymond
------------------------------------


------------------------------------

Yahoo Groups Links


Re: SG504 Frq. Display & Digital amplitude flattening.

 

On 05 Apr 2016 13:09:58 -0700, you wrote:

@1: Is HiSpeed CMOS fast enough these days to allow 1 GHz frequency measurement and if so, at what power dissipation?
No, a power hungry prescaler has to be used. Since the prescaler and
signal conditioning will draw a lot of power, they will have to be
powered down most of the time. Note that they have to be powered
during the frequency counter gate time so the gate time needs to be
minimized further limiting resolution.

Luckily high resolution is not needed in this application. A x100
prescaler and 0.1 second gate time will yield a resolution of 1 MHz.
If a 10 MHz resolution is sufficient, then the power can further be
reduced by another factor of 10. Variable and longer gate times could
be used to detect 6 MHz and 50 kHz.


Re: SG504 Frq. Display & Digital amplitude flattening.

 

Hi Ancel,

I find your energy level to be very refreshing. I wish there were more
'makers' on the forum like you. I do however, tend to agree with Raymond
that improving the SG504 with a more accurate readout or a digital level
indicator is impractical. It is a 1970s instrument with excellent
performance for its day. Your contribution to improving the flatness of the
leveling head to date is almost more than most people can measure. Doing
more than that would be appreciated by only a very few of us.

I think your talents would be better put to use designing something that no
one else has done yet, or turning you talents towards improving the
performance of other RF instruments. I would be very happy to help you do
that wherever I can.

Dennis Tillman W7PF

-----Original Message-----
Sent: Tuesday, April 05, 2016 2:20 AM
Subject: [TekScopes] SG504 Frq. Display & Digital amplitude flattening.

Hi all:
My recent efforts in optimizing the SG504 head has led to this concept
which addresses a lack of accurate freq. display and amplitude compensation
using digital controls.
These functions can be 'built in' to an upgraded leveling head (Li Ion
battery powered, rechargeable).

Concept statement subject to refinement.

In using the SG504, one of the areas I think could make the unit more
useful is a reasonably accurate frequency counter display. Perhaps to 1Khz
via a uC comparator triggered pulse counting.

I've come up with a concept 'head' which adds a battery operated low power
microcontroller to do the job via a small 8 x 2 LCD display (58mm x 32mm)
mounted in the head. LCD display triggers by detection of a FRQ sweep or
flashes the static FRQ one sec in every 5 secs to conserve power.

Now this approach will allow anyone to also 'flatten' the head output
relative to the ref. 6Mhz level to potentially within 0.01dB steps via a
0.31dB range of adjustment ((5 bit) DAC output). This requires an additional
schottky detector coupled @ the head output.

The process works by activating the cal. mode (pushbutton) while in the
6Mhz range. The controller then locks the amplitude ref. . Then the user
switches to the Low range and starts to sweep, slowly, then the high range,
repeat sweep. The controller monitors the frequencies and determines
amplitude offset points to 'map' the head amplitude flat. Once the frequency
stops changing for a few secs the head comes out of cal mode and applies
frequency mapped corrections (every 3.2 Mhz step) via it's (5 bit) DAC
output. No PWM.
It 'corrects' the head amplitude by modifying the DC level amplitude
signal to the SG504.

In terms of operation, I see it working like this.
The user sets the amplitude and selects the 6Mhz SG504 ref. The controller
detects the 6Mhz ref and grabs the ref. amplitude. Then it can properly
apply all the stored corrections to the current amplitude level.
Thus when changing amplitude the user would need to activate the 6Mhz ref,
set the amplitude and switch back to the frequency of interest for precision
leveling.(Perhaps using 12 or 14 bit oversampling)

The system can permit an RS232 to USB serial link to a PC to output live
frequency and ref. amplitude data.
I might look at a 10Mhz ref input as a feature for syncing to a common
source, which can then provide direct means to cal the SG504 6Mhz frq ref.
with just the head alone.

Idk whether this seems worthwhile to the forum, but it's an idea that is
interesting to me and seems economically viable and adds usefulness to the
instrument without having to 'hack' the instrument itself.

------------------------------------
Posted by: mosaicmerc@...
------------------------------------


Re: 7S12 TDR Cable

 

That 7S11 will require a sampling head such as an S-5 25pSec head
I guess that's an S-4 sampling head ;-(

Raymond


Yahoo was rejecting emails Monday night

 

I noticed Yahoo did not accept any of the emails I sent Monday night.

It appeared that whatever caused this problem occurred from 6:30PM until
about 11:30PM Monday night since there are no entries for that period of
time.

Dennis Tillman W7PF