Date   

Re: Bandwidth measurement.

ditter2
 

Bill,
The reason a leveled sine source is used to measure bandwidth rather than rise time of a fast step is to locate any "suck outs", which usually result from a resonance, that might appear somewhere in the passband. These would have an effect on the pulse shape of the step, but the effect may be too small to be easily seen in the waveform.

A leveled sine source is the easiest way to make this measurement, but if you lack one, any oscillator can be used if you can monitor the output level. This can be done using a RF power meter. To remove the effects of cable reflections, the amplitude should be measured as close to the scope input as possible. This can be done using a power divider at the end of the cable, with one end feeding the scope input, the other driving the power head of the power meter. You may have difficulty getting enough voltage to verify the higher V/div settings, but usually you won't find a problem that effects the frequency response with these, so they can be verified for attenuation accuracy only using a LF source, with the BW measurements at a lower V/Div setting. (This assumption is valid up through a few hundred MHz, but not with the 50 ohm inputs of multi GHz scopes. Verifying BW flatness of these is a challenge.)

- Steve

---In TekScopes@..., <wah_1003@...> wrote :

Hello Everyone:


I have been a member of the group for a few years now and always enjoy the posts, the combined experience here is awesome. I recently read a post that said the number of scopes to have was "one more", I agree. If it was not for the size of my work area I would have many more. Right now I have a Fluke Scopemeter, a Tek 2246A, two Tek 475s, a Tek 7904 and a small HP mainframe with a 8558B spectrum analyzer. I also have the normal potpourri of supporting equipment - probes, calibrators, etc. There always seems to be something missing.


The 7904 is a funny story. I had absolutely no room on my bench for something else, so I convinced myself I needed a Tek ScopeMobile cart to "put my hand tools in". Just a few weeks later I got a 7904 to put on top of it and then the unstoppable flow of plugins began. I still have a few extenders I got from John Griessen that I need to put together. Now I think I have really run out of space but you never know.


Now for my real question: One day I was using one of my 475s and there was a strange kind of popping noise and the trace went wild. After some diagnosis there was one large filter cap that had opened up. Since it was so hard to get to (you have to really take most everything out of these to get to the big caps) I replaced them all and also changed all of the incandescent lamps to LEDs with built-in internal dropping resistors. After this I was looking forward to using all of my equipment to re-calibrate the scope. Everything worked out fine but I had a question about the bandwidth test.


First of all one thing I do lack is an SG503/SG504 but if I did I was curious about what should be done if the test shows that the bandwidth is constrained? All of the other adjustments have variable pots and caps that are used to get it back into spec but if the bandwidth is not right what course of action would you take? Seems like the problem could literally be anywhere.


My second question is: Do you really need the leveled sine wave generators to determine bandwidth? Wouldn't just a really good measurement of the risetime be sufficient? I have a TD pulser.


Thank you for your time and keep up the good work,
Bill


Re: a sick 576 (and it's owner) needs help

Ed Breya
 

I may not get the whole picture here, but it seems from the messages so far that you put a resistor across the C-E terminals, but get unexpected results on the 576, then you went so far as to measure the current with a current probe. Now you believe the display appears to simply show the slope due to the current range sensing resistor. Right?

I did not see any mention of checking the voltage across the C-E terminals - only the actual current with a probe, and no mention of checking it without any device between the terminals. I suspect the problem may be a switch signal routing failure, and/or a possible short between the C-E terminals. This part would be simple to check - just run it without any DUT attached to confirm that no current flows as the voltage goes up. Also, check for proper right-left select switch operation, and the B-E mode switch too.

You should confirm that no current flows through the DUT when it shouldn't, and some definitely goes through when it should, and that there's voltage drop across it. If this is the case, then a C-E terminal short is unlikely, so there has to be a switching problem or maybe loss of a connection somewhere allowing the wrong signal to be displayed - maybe a lost ground connection, for example. Another possibility may be a broken switch, where one section or contact is stuck or out of sync with the rest of it.

Ed


Re: 465 oscilloscope problem

Siggi
 

On Sun, 10 Apr 2016 at 17:35 'Tewell, Kevin' tewell.ka@... [TekScopes] <
TekScopes@...> wrote:

The serial number is above 250000
Excellent I see you've noted this before and I missed it - such are the
perils of the centithread :).
I've gone ahead and bought the relevant manual from Artek so that we can be
on the same page - so to speak.


Re: 465 oscilloscope problem

Tewell, Kevin <tewell.ka@...>
 

The serial number is above 250000

Sent from my BlackBerry 10 smartphone.
From: Sigurður Ásgeirsson siggi@... [TekScopes]
Sent: Friday, April 8, 2016 10:17 AM
To: TekScopes@...
Reply To: TekScopes@...
Subject: Re: [TekScopes] 465 oscilloscope problem




On Thu, 7 Apr 2016 at 21:47 'Tewell, Kevin' tewell.ka@... [TekScopes] <
TekScopes@...> wrote:



I have a digital multimeter this one.


http://www.amazon.com/gp/aw/d/B000JKMTDM/ref=mp_s_a_1_sc_7?qid=1460078645&sr=8-7-spell&pi=SL75_QL70&keywords=digital+multimeter+madtech

That looks like a decent enough DMM.
I have a power supply up to 60 volts and a few amps.
Is this a proper lab supply - e.g. does it have a current limit, or just
settable voltage?

I bought the mantel and downloaded it.
Good, unfortunately I don't have a copy of the same manual, so we can't
talk in page numbers, nor can I be sure that you have the same version of
the document I'm going to be looking at, but I'm sure we can make tihs work.

Incidentally, what's the serial number of your scope? This matters because
Tek made changes and improvements over time, and the serial number can tell
us which version of a manual or of a schematic to look at.


I still have the two capacitors out of the scope and I put the meter on
like you said and I still get .9 ohms and I switched the probes and still
get .9 ohms. This is with one lead on test ground and one on the 55 volt
test lead. ‎
Where exactly are you measuring this?

In the power supply schematic (it's numbered schematic 12) I see a test
point labelled TP1536, is that where you're measuring?
If you're measuring this on the point in the schematic labeled "+55V
(UNREG)", then I'd say your resistance reading is likely normal.

So, if you haven't been measuring the resistance of TP1536 to and from
ground, then please measure that and report.

Where I get in trouble is I don't know the terminology dummy
resistor,consumers, bridge rectifier etc...I have to look them up. ‎
Yups, that's how you start. Starting with an oscilloscope repair is fairly
ambitious, but the documentation in the service manuals is so wonderful
that it's not a totally preposterous thing to do (IMHO).

The bridge rectifier is what turns AC to DC, it's the same rectifier for
the 55V and the 120V supplies in your scope - CR1611 on schematic diagram
12. Since your scope is blowing fuses, the bridge rectifier is one of the
natural suspects, as they can fail to a short.

"Consumer" here means everything that uses the 55V power supply, e.g. the
power supply is the producer, and the other boards in the scope are the
consumers.

The first order of business is to figure out whether problem is in the
power supply, or whether it's in one of the consumers, so let's get that
underways.

Good luck!


Re: a sick 576 (and it's owner) needs help

Bene's Mails
 

Yes, current is flowing definitively!

I had not much time the last days to make more measurement, but as I had
redone my measures, I found out, that the perfect describing of the failure
is: my 576 is measuring it’s own sensing shunts. I got perfectly the UI
trace of the series resistance of the collector current sensing circuit. I
proof that with DMM and a TCP202A directly clamped around the Black on white
wire which connects the current return connection from the test fixture with
the vertical switch assembly.



The big problem is, that I can not assume any other root cause of that
behavior as a wrong connection anywhere in the circuit between the shunts
and the vertical amplifier. But I can not imagine where. I checked all
connections between Vertical switch, Mode switch and Display Switch assy as
it was clear by the schematics in the manual without a result.



Tomorrow I will pack the 576 into my car and ask our last tracer guru Helmut
Hoevel here in Germany for help as he is now back from sickness. Maybe he
has a guess.

Von: TekScopes@... [mailto:TekScopes@...]
Gesendet: Donnerstag, 31. März 2016 19:46
An: TekScopes@...
Betreff: [TekScopes] Re: a sick 576 (and it's owner) needs help





Are you sure the current is flowing? I would first put a meter in series
with CE resistor in current mode and measure the current to make sure it was
what it should be then figure out why the voltage across current sense
resistor isn't making into the vert.


Re: 2465 channel two acts as I am going through low pass filter

almus_kenter
 

Yes I can hear the relays clicking away.

Thanks for the info--- I will try and
fix it in the next week or so.


Re: 2465 channel two acts as I am going through low pass filter

Siggi
 

On Sun, 10 Apr 2016 at 09:45 akenter@... [TekScopes] <
TekScopes@...> wrote:

Just checked R across BNC of channel two:

it is open regardless of DC/AC switch.

Channel one toggles between 1M and open (DC:AC)

Is this an easy fix?
It's most likely fixable - do you hear the relay click when you toggle
AC/GND/DC coupling? My 2430 had attenuators that were all over the place -
some of the relay pivots were stuck due to corrosion, and some of the
contacts weren't making contact. See <
https://groups.yahoo.com/neo/groups/TekScopes/photos/albums/501546574> for
the guts of the attenuators.

It's not as hard as it looks at first blush to get the CH1/2 attenuators
out.
From memory you need to get the front bezel off and get the front panel
switch board out.
You then need to pull the screws that hold the attenuators in place. Two
(IIRC) each to the front, a metal bar running across them, and one each
(IIRC) screw you access from the inside of the front-panel cavity pulling
them into the main board.
You then need to desolder two connections, one to the preamp, one to a
trimcap, and they'll lift straight off the board.

To get at the contacts into you just need to undo the screws holding the
shielding on, and then the screws holding the plastic bracket on.
I used IPA and DeOxit on the contact fingers (be careful with those) and on
the contact points on the ceramic substrate. I think perhaps using DeOxit
isn't the best of ideas if IPA will do the trick.


Advice on a sampling scope

mosaicmerc
 

Hi all:
I'm considering obtaining a sampling scope eg. 11801B chassis.
I'm looking for advice as to which heads are best. My interest is in wifi, mixers, assessing UHF VCOs etc.


Re: 2221 100MHz Scope repair project

Tom Gardner
 

On 10/04/16 14:59, satbeginner@... [TekScopes] wrote:

Still feels weird: to pay 30% of the price I got the scope for again for some missing knobs....
Yet another example of why people break scopes, illustrated books, companies, and cars up into their component pieces.


Re: 2221 100MHz Scope repair project

satbeginner
 

Just an update: I found most of the missing knobs on Qservice.eu .
They should arrive somewhere next week.

Still feels weird: to pay 30% of the price I got the scope for again for some missing knobs....

But hey, I want it to be complete and working!

Un saludo,

Leo


Re: 2465 channel two acts as I am going through low pass filter

almus_kenter
 

Just checked R across BNC of channel two:
it is open regardless of DC/AC switch.

Channel one toggles between 1M and open (DC:AC)

Is this an easy fix?

Thanks


Re: 2465 channel two acts as I am going through low pass filter

almus_kenter
 

Only channel two.

I have been playing around with it some more and
it is only when channel two is in dc coupling


Re: 2465 channel two acts as I am going through low pass filter

Siggi
 

On Sun, 10 Apr 2016 at 08:55 akenter@... [TekScopes] <
TekScopes@...> wrote:

I have a straight 2465. Everything seems to work except looking at
signals

(eg the onboard square wave) channel 2 acts as if I have a (very) low pass
filter engaged.

In other words, at < 1kHz frequencies I have a square wave, but it rapidly
turns
into a very slow RC type rise and fall.
Hey there,

does this affect all channels, or only one?
f only one channel is afflicted, maybe the AC/DC coupling relay if off? The
easiest way to check is to measure the resistance of the input (from center
conductor of the BNC to ground).

Siggi


2465 channel two acts as I am going through low pass filter

almus_kenter
 

I have a straight 2465. Everything seems to work except looking at signals
(eg the onboard square wave) channel 2 acts as if I have a (very) low pass filter engaged.

In other words, at < 1kHz frequencies I have a square wave, but it rapidly turns
into a very slow RC type rise and fall.


New file uploaded to TekScopes

TekScopes@...
 

Hello,


This email message is a notification to let you know that
a file has been uploaded to the Files area of the TekScopes
group.


File : /TEKWARE_GPIB_ca_1988.zip
Uploaded by : ulf_r_k <ulf_r_k@...>
Description : Tekware: Tektronix GPIB software dated ca 1988 for DOS


You can access this file at the URL:
https://groups.yahoo.com/neo/groups/TekScopes/files/TEKWARE_GPIB_ca_1988.zip


To learn more about file sharing for your group, please visit:
https://help.yahoo.com/kb/index?page=content&y=PROD_GRPS&locale=en_US&id=SLN15398


Regards,


ulf_r_k <ulf_r_k@...>


Tek 1401A / 323 combo

Jeff Woolsey
 

I managed to snag the subject equipment at a flea market this morning
(almost didn't go; I've been looking for these ever since I saw youtube
videos about them). They're in less-than-perfect condition; the 323 has
no power supply, and the fine-tuning knob on the 1401A is missing its
outer shell, and the indicator doesn't turn with the knob. I have other
scopes, so the 323 will not hold me back. No handles for either. No
weird power cord, either. (job for 3d printer?)

I tried firing up the 1401A this evening. Why are the + and - banana
jacks just a hair too far apart to use a duplex banana plug? Eventually
got power to it, and it's stone dead (a pattern emerges). (Why isn't
there a pilot light?) I opened it up and found that the blown fuse was
holding a piece of tinfoil in the contacts.

I've ordered the service manual from Greece and will see what it says.
Meanwhile, sources for both the power supply (identical in both, I
wonder??) and the fine-tuning knob would be nice....

I checked out more videos, including one regarding removing the NiCd
batteries. In mine, they're wrapped with a more-modern material (foam
plastic sheets) than its age would suggest, so I think they've already
been replaced at least once.

==

On the other hand, I also took home a complete bitscope (it works, but I
had no idea what it was when I spotted it). It seems I accidentally
took home TWO spectrum analyzers, of roughly the same size, but widely
varied generations.

--
Jeff Woolsey {{woolsey,jlw}@jlw,first.last@{gmail,jlw}}.com
Nature abhors straight antennas, clean lenses, and empty storage.
"Delete! Delete! OK!" -Dr. Bronner on disk space management
Card-sorting, Joel. -Crow on solitaire


Re: Bandwidth measurement.

mosaicmerc
 

Craig,
Why-o-why don't u have an SG504?

Anyhow I exercised both the SG503 & 504 yesterday mapping out the bandwidth of a 'hacked' Rigol DS2072A.
Officially rated @ 70 Mhz, I am now getting 400Mhz @ 3dB down bandwidth.
300Mhz = 2 dB down and @ < 150Mhz it's spot on. Not bad for a $800 unit with I2C/SPI comm. decode protocols and 2G SA/s.

Ancel


Re: 2235 First Assessment

 

Two things I would do:

1. Set the sweep to 2ms/div with the baseline trace of channel 1 and
set the trigger source to line. Then if the vertical variation on the
trace becomes fixed, it is related to the AC line input and probably a
power supply issue.

2. Measure level and ripple of the low voltage power supply outputs.


Re: Bandwidth measurement.

 

On 09 Apr 2016 17:21:55 -0700, you wrote:

Hello Everyone:

I have been a member of the group for a few years now and always enjoy the posts, the combined experience here is awesome. I recently read a post that said the number of scopes to have was "one more", I agree. If it was not for the size of my work area I would have many more. Right now I have a Fluke Scopemeter, a Tek 2246A, two Tek 475s, a Tek 7904 and a small HP mainframe with a 8558B spectrum analyzer. I also have the normal potpourri of supporting equipment - probes, calibrators, etc. There always seems to be something missing.

The 7904 is a funny story. I had absolutely no room on my bench for something else, so I convinced myself I needed a Tek ScopeMobile cart to "put my hand tools in". Just a few weeks later I got a 7904 to put on top of it and then the unstoppable flow of plugins began. I still have a few extenders I got from John Griessen that I need to put together. Now I think I have really run out of space but you never know.
I really like the Tektronix instrument carts. I do not have more
carts than oscilloscopes but I do have carts and no suitable
oscilloscope to place on them.

Now for my real question: One day I was using one of my 475s and there was a strange kind of popping noise and the trace went wild. After some diagnosis there was one large filter cap that had opened up. Since it was so hard to get to (you have to really take most everything out of these to get to the big caps) I replaced them all and also changed all of the incandescent lamps to LEDs with built-in internal dropping resistors. After this I was looking forward to using all of my equipment to re-calibrate the scope. Everything worked out fine but I had a question about the bandwidth test.

First of all one thing I do lack is an SG503/SG504 but if I did I was curious about what should be done if the test shows that the bandwidth is constrained? All of the other adjustments have variable pots and caps that are used to get it back into spec but if the bandwidth is not right what course of action would you take? Seems like the problem could literally be anywhere.
Low bandwidth without any other symptoms would be unusual. Low bias
current in an amplifier stage can cause this and some of the Tektronix
vertical output amplifier designs have a bias current trimmer which is
adjusted for maximum bandwidth. If I had an oscilloscope with low
bandwidth, I would start checking all of the resistors which set the
bias current for each stage.

Another thing which I have seen cause this is if the fixed
collector-emitter voltage of a transconductance amplifier is low
because the following cascode stage voltage is also low.

My second question is: Do you really need the leveled sine wave generators to determine bandwidth? Wouldn't just a really good measurement of the risetime be sufficient? I have a TD pulser.
At least at low frequencies including 300 MHz and below, my experience
is that a transition time test works just fine for determining
bandwidth. I have yet to find an oscilloscope with a good clean
transition time measurement that did not correlate exactly with a
bandwidth measurement. At higher frequencies I think this usually
applies but in some cases where the oscilloscope lacks a gaussian
response, it may not.

Note that you can also measure a high pass cutoff using exactly the
same transition time test. I did this to verify proper operation of
my 7A22 and my AC current probe.


Re: Bandwidth measurement.

David DiGiacomo
 

On Sat, Apr 9, 2016 at 6:21 PM, <wah_1003@...> wrote:
First of all one thing I do lack is an SG503/SG504 but if I did I was curious about what should be done if the test shows that the bandwidth is constrained? All of the other adjustments have variable pots and caps that are used to get it back into spec but if the bandwidth is not right what course of action would you take? Seems like the problem could literally be anywhere.
It's normally just a verification step to make sure you haven't messed
up the HF response while making the other vertical adjustments.

If the bandwidth is seriously below spec, and going back through the
vertical adjustments doesn't help, you would have to use signal
tracing or signal injection to locate the problem stage.

My second question is: Do you really need the leveled sine wave generators to determine bandwidth? Wouldn't just a really good measurement of the risetime be sufficient? I have a TD pulser.
That's fine unless the scope is misadjusted to a non-Gaussian
response... then the measured risetime can be misleading.