Tek 7934 mainframe...

[muvideo]

... Sorry if someone hoped for a teardown of such mainframe, not yet
I'm tempted to buy one, here a local seller has one 7934 for sale asking
little more than 100 euro, plus shipping.
It has no plugins and the seller says only that the unit powers on, no tests
performed on it.

I haven't had experience with analog storage scope,
how does it work?
I'm used to a philips 50MHz analog scope and only recently
I bought a DSO, both with two channels.
I'm attracted by the high bandwidth and the possibility
to use low voltage plugins like 7A22 for dc noise etc, and the possibility
to have more than 2 traces (if it's possible with it, i haven't understood yet).
Does anyone have experience with this scope?
In order to understand the total cost for such a system, what plugins should I
search for setting up a good working system?
Can I have more than 2 traces on it?
What do you think about the price?
What can be the possible weak points of this system
(something in particular to ask the seller before buying)?

Thank you.
Fabio.

[tekfan]

That's a very nice scope.

I haven't had experience with analog storage scope,
how does it work?

Analog storage works by first writing a beam on the screen the same way as a normal scope and then firing a second ''flood gun'' onto the screen so that the image is retained. This scope has quite a high writing rate so storing very fast single shot events should be easy. If you then want to zoom in and maipulate the image then youre out of luck. It can't do that.

The 7934 also has a variable persistence which does just that. It controls how long the waveform will appear on the screen. You can have a dot moving very slowly (or very fast) across the screen and it will draw a line behind it.

Can I have more than 2 traces on it?

You can have as little as 0 to as much as 4 analog channels. Depenting on plugins, you can even display 16 channels of digital data using the logic analyzer plugin.

You can select the left vertical plugin to be linked to the left time base and the right to the right one. You can select which channel you want to trigger off for each timebase. Both sweep speeds can be individually adjusted.
Or you can link all channels to a single time base (left or right).

With 2 dual trace plugins this mainframe can display 4 channels all at once. With a dual delayed time base you can then display 4 more.

What do you think about the price?

100EUR for this mainframe is a very very good price. Especially in Europe. I just realized that you said 7934. I was thinking of the 7834.

What can be the possible weak points of this system
(something in particular to ask the seller before buying)?

Not really.
The power supply is a bit harder to troubleshoot because it is switchmode and dangerous, but if it powers up then it is OK.
Tektronix also has a very good manual for troubleshooting this kind of power supply.

Ask the seller if he has some plugins.




Some good plugins would be:
7A26 - dual trace,general purpose 200MHz, 1Mohm input
7A24 - dual trace, 400MHz, 50ohm inputs
7A22 - low level differential, 1MHz, 10uV/div sensitivity
7A13 - differential comparator, 100MHz, see ripple on large DC voltage, switchable high impedance input
7A19 - single channel but 500MHz, 50ohm inputs

7B80 - general purpose 400MHz timebase
7B85 - delaying time (NOT dual), 400MHz, can also be used as a standard 7B80
7B92A - dual delayed, 500MHz, saves space for other plugins

7D14 - 500MHz frequency counter (or just use the signal output from the back of the scope and feed it into your Racal Dana 1992)

Of course you can buy only the ones that you need. You can always use your other scope for general work and use the 7934 for strange things (low level and HF measurments...)

and many many others depending on what you want: 14GHz sampling plugins? 60GHz spectrum analyzer? No problem.
http://amplifier.cd/Test_Equipment/Tektronix/Tektronix_Test_Equipment.htm

[muvideo]

Thank you very much Tekfan (hehe now I understand your nickname),
I'll contact the seller by phone for more details, what should the beam look like without
plugins, a point in the centre?
The seller has also a 7844 with 2x7A26 and 7B80 plugins, but the pic shows
7A18 and 7A19 instead of 7A26, maybe he has more plugins, the price he
is asking for this is 320euro.

That's a very nice scope.
...
and many many others depending on what you want: 14GHz sampling plugins? 60GHz spectrum analyzer? No problem.
http://amplifier.cd/Test_Equipment/Tektronix/Tektronix_Test_Equipment.htm

Now my wallet is seriously threatened 

Fabio.

[tekfan]

Ah yes you're looking at radiosurplus.it. I drool over the pictures of test eqipment every time I see that site. Sadly (luckily?) I don't buy anything there.

They definitely have more plugins (go to their site > press accessori on the left > press accessori strumentazione on center of screen > scroll down > drool)

I don't think the mainframe will show anything without any plugins (atleast my 7904 doesn't). All you can do is to turn up the graticule intensity (often bulbs are burnt out so that won't work either). It should illuminate the entire screen green if the var persist button is in and the persistence knob is set to minimum. That will indicate if there is HV on the tube. If you put in a working timebase then there will be a trace and readouts for the sweep speed.

Readouts will indicate that the vertical amplifier is actually working correctly and so is the horizontal amp.

[muvideo]

Ah yes you're looking at radiosurplus.it. I drool over the pictures of test eqipment every time I see that site. Sadly (luckily?) I don't buy anything there.

Usually their prices are a little high, many years ago I bought the philips scope from them and it was ok.

They definitely have more plugins (go to their site > press accessori on the left > press accessori strumentazione on center of screen > scroll down > drool)

Oh I didn't noticed all that, but it's not good: I'm asking why they just didn't insert a timebase in that mainframe for a quick
photo showing it working?
After all it's better specs than 7844.

Fabio.

[rf-loop]

After all it's better specs than 7844.

Fabio.

maybe you mean 7834.

But if you really mean 7844 then:

Yes and no.

7844 is real dual beam oscilloscope! There is two independent beams. Others have one beam (but shared to separate channels)

And if look this feature, there is no competition, 7844 wins all others and sit just alone in its own class.

<smile>

[muvideo]

After all it's better specs than 7844.

Fabio.

maybe you mean 7834.

But if you really mean 7844 then:

Yes and no.

7844 is real dual beam oscilloscope! There is two independent beams. Others have one beam (but shared to separate channels)

And if look this feature, there is no competition, 7844 wins all others and sit just alone in its own class.

<smile>

You are right, I didnt think about dual beam feature.
How many channels can be assembled in a 7844 mainframe?
I imagine that one could use two totally independent timebases,
what could be useful scenarios with a dual beam scope, examples
of use where a dual beam can make the difference?

Fabio

[tekfan]

Yes you can have two totally independent time bases but you can also do this on a single beam oscilloscope. The number of channels is limited to the plugins and available plugin spaces and not the mainframe itself.

With a dual beam oscilloscope you can have one beam drawing a slow X-Y plot and the other beam displaying a 400MHz signal. That's basically it.

Maybe it would be advantageous in slightly better screen brightnes since you are not chopping or alternating the beam between channels.

When using only one beam for displaying a waveform the other beam will be used to write the readouts, so you don't get flicker in the other waveform. Even using a single beam unit the display flicker is very minimal and only shows in the audio range waveforms.

[jerry507]

Tekfan,

It was always my understanding that dual beams were useful because they could produce two very bright traces which made them work well with cameras. But the human eye can see even dim traces very well, so is it really important for "normal" use?

[tekfan]

Exactly my point. In the old days with film the trace had to be very bright to get caught onto the film especially when dealing with high sweep speeds with low repetition rates. Even more important when trying to capture a single shot event on film.

Cameras have improved so much thad you don't really need a very bright trace at all. I've actually had to turn the intensity down to barely visible level to the eyes so that the camera could capture all the small abberations of the trace. Too high intensity causes the trace to blind the camera and cause a halo around the trace.

[dfnr2]

It was always my understanding that dual beams were useful because they could produce two very bright traces which made them work well with cameras. But the human eye can see even dim traces very well, so is it really important for "normal" use?

Dual-beam scopes are (were) so much more expensive, I doubt that many were ever bought for brighter or crisper traces, or other aesthetic reasons.  They were bought because sometimes, they could make certain measurements better than single beam scopes.

Single-beam analog scopes display multiple traces by multiplexing the signals in one of two ways: "chop mode" and "alterating mode"

Chop mode: This mode rapidly switches from one channel to the next hundreds of times, or more, during each sweep, "chopping" each trace into hundreds of discreet pieces very quickly so the pieces are on the order of the spot size, and look contiguous.  This is effectively a sampling mode, and can produce artifacts, essentially multiplying the signal by a pulse train, and may cause sampling artifacts on signals, especially periodic signals on the order of the chop frequency (for example, serial signals, PWM, switch-mode supplies).  You can usually eliminate the artifacts by moving to a faster sweep (turning the timebase knob), but occasionally you may be looking at some phenomenon that requires a timebase setting susceptible to artifacts.  Also, if you are unaware of this sampling effect, you may not know an artifact is present. 

Alternating mode: This mode traces an entire sweep on one channel, then the next sweep on the next channel, etc.  There is no sampling artifact.  However, since each trace is triggered separately, synchronization between the traces may not be perfect.  If you need to time two edges on different channels very accurately, your measurements may be unreliable on this setting.

Usually your application will dictate the most appropriate mode, and you can avoid artifacts.  However, you find yourself measuring a response time, or other time interval on a periodic signal, and the timebase setting required to measure the response time puts the chop frequency near the period of some component of the signal.  In that case, you will be at risk of artifact in either chop or alternating mode.  That is where there is no substitute for a dual beam scope. 

That is why it's really important to understand your scope, whether analog or digital, and make sure that you control and minimize the artifacts, and understand the resulting waveforms.

By the way, I would not spend eu. 100 for any unproven tek mainframe, even if they are excellent scopes.  If it doesn't work perfectly, it may be hard to fix due to custom IC's.  Power supply problems are not so bad, but you're essentially gambling.  In fact, nowadays, a fully functioning, configured  mainframe should be in the USD$100-$300 range, and a working bare mainframe, except for the more unusual varieties such as the 7844 dual beam, or the 7104 GHz mainframe, should be less than USD$100.  By the way, the best bet is a 7704A, the workhorse of the series, although 7904's are nice with BW up to 500 MHz, but less reilable, and the 7834's are nice as well, with storage and 400 MHz BW.

Dave

[muvideo]

By the way, I would not spend eu. 100 for any unproven tek mainframe, even if they are excellent scopes.  If it doesn't work perfectly, it may be hard to fix due to custom IC's.  Power supply problems are not so bad, but you're essentially gambling.  In fact, nowadays, a fully functioning, configured  mainframe should be in the USD$100-$300 range, and a working bare mainframe, except for the more unusual varieties such as the 7844 dual beam, or the 7104 GHz mainframe, should be less than USD$100.  By the way, the best bet is a 7704A, the workhorse of the series, although 7904's are nice with BW up to 500 MHz, but less reilable, and the 7834's are nice as well, with storage and 400 MHz BW.

Dave

I asked the seller to insert a timebase and vertical and check scope's behaviour.
Can you advice some rapid test to do check the scope?
As for the prices you quoted, I imagine you are talking about USA's prices.

Fabio.

[vk6zgo]

True Dual Beam  Oscilloscopes with two separate electron gun assemblies,or in some cases, a beam splitter of some kind,are as rare as hen's teeth, because in most circumstances a normal alternate or chopped display gives the same facilities without the rather horrific scan geometry problems,& extra cost.

The last real Dual Beam I had anything to do with was an old 1950s Cossor!

Although there are circumstances where a true Dual Beam may be necessary,they are quite rare,& in many cases there are work-arounds.
If you are worried about whether the  method of display will give you a problem,you can  often look at a single signal on both channels & produce a correction factor------OK,this won't work in all cases,which seems to be why Tek produced the 7844.

As far as photographic film is concerned,the old black & white Polaroid film was able to reproduce quite dull traces.
It was very much faster than the standard colour film most of us used for everyday photography.
It was messy & disgusting,but there was little room for complaint about its performance.