Basic oscilloscope recommendations

[spasmous2]

I need to buy a 4-channel oscilloscope and am hoping someone can
suggestion what make/model might be suitable. I only require about
10MHz bandwidth for the transient signal, and it repeats every 0.1 to
10 seconds.

From a little experience using an old analog scope, the signal fades

out really quickly whereas I'd like it to stay on screen until the
next trigger. I was thinking an LCD screen might be the solution to
that problem? Does that necessitate a digitial scope?

From looking on google, the Tecktronix TDS2004B seems to fit. But

really I don't need anything fancy, like FFT or USB storage. I just
want 4-channels and a persistent trace on the screen. And 50 ohm input
impedence.

Can anyone point me in the right direction? Thanks.

 

[KC]

Take a look at the Hameg scopes. They sell excellent product without
the fancy gadgets for a fair price.

 

[mike]

I need to buy a 4-channel oscilloscope and am hoping someone can
suggestion what make/model might be suitable. I only require about
10MHz bandwidth for the transient signal, and it repeats every 0.1 to
10 seconds.

out really quickly whereas I'd like it to stay on screen until the
next trigger. I was thinking an LCD screen might be the solution to
that problem? Does that necessitate a digitial scope?

really I don't need anything fancy, like FFT or USB storage. I just
want 4-channels and a persistent trace on the screen. And 50 ohm input
impedence.

Can anyone point me in the right direction? Thanks.

Yes, you need digital storage.
Never, ever buy a scope with a bandwidth less than 100MHz.
Requirements change. You ALWAYS need more bandwidth than
you anticipated.
For slow signals, memory depth is an important parameter.
Draw the waveform you want to look at.
Put dots on it representing samples close enough together
so you can see what you want to see. Figure out the time
between samples. Then count how many of those samples
will be needed to cover the area of interest. If you expect
to see 10 MHz. bandwidth stuff over a whole 10 second interval,
you're gonna need a LOT of memory depth...times 4. If you can't afford the
memory, you're gonna need delayed triggering or some other
means to see what you want to see.

You haven't disclosed the nature of the waveforms you want to see,
but what you have said indicates that your situation may require
far more than "basic oscilloscope" features.

Don't skimp on a scope. It's your best tool for a lot of
design/debug situations. Requirements change, and waveforms
are NOT slowing down.
mike

 

[Hammy]

For slow signals, memory depth is an important parameter.
Draw the waveform you want to look at.
Put dots on it representing samples close enough together
so you can see what you want to see. Figure out the time
between samples. Then count how many of those samples
will be needed to cover the area of interest. If you expect
to see 10 MHz. bandwidth stuff over a whole 10 second interval,
you're gonna need a LOT of memory depth...times 4. If you can't afford the
memory, you're gonna need delayed triggering or some other
means to see what you want to see.

Do you really think memory depth is all that important? When you can
just hook up a DSO to your PC via USB or RS232 and record the waveform
for as long as you need, save it in a wmv file and review at your
leisure. Or am I missing something?

Don't skimp on a scope. It's your best tool for a lot of
design/debug situations. Requirements change, and waveforms
are NOT slowing down.
mike

I agree.

 

[mike]

Do you really think memory depth is all that important?

Only if you want high resolution for a long time.
Do the math. Acquisition time divided by the time between samples
equals the memory depth requirement.
Yes, there are different techniques, but not typically found on
entry-level scopes.

When you can

just hook up a DSO to your PC via USB or RS232 and record the waveform
for as long as you need, save it in a wmv file and review at your
leisure.

YOUbetcha! IFF your DSO has the capability to spit out real-time data.
AND your PC has the port and the capture software and the I/O bandwidth
to acquire the data. And if you have the software to transform it into
wmv, although I don't know why you'd want that.
Remember, we're talking 10MHz. analog scope bandwidth here.
Or am I missing something?
yes

 

[spasmous2]

Yes, you need digital storage.
Never, ever buy a scope with a bandwidth less than 100MHz.
Requirements change. You ALWAYS need more bandwidth than
you anticipated.
For slow signals, memory depth is an important parameter.
Draw the waveform you want to look at.
Put dots on it representing samples close enough together
so you can see what you want to see. Figure out the time
between samples. Then count how many of those samples
will be needed to cover the area of interest. If you expect
to see 10 MHz. bandwidth stuff over a whole 10 second interval,
you're gonna need a LOT of memory depth...times 4. If you can't afford the
memory, you're gonna need delayed triggering or some other
means to see what you want to see.

You haven't disclosed the nature of the waveforms you want to see,
but what you have said indicates that your situation may require
far more than "basic oscilloscope" features.

Don't skimp on a scope. It's your best tool for a lot of
design/debug situations. Requirements change, and waveforms
are NOT slowing down.
mike

--

Thanks mike for your thoughts. Digital scopes seem to start around
$2000, which is steeper than I was hoping. On ebay there are many
analog scopes that are high quality for much less than this.

I would get an analog but I just don't know if the screen persistence
is adequate. The signal I am looking at is around 20ms duration and
repeats every second or so, so the update of the screen would be
very slow on an analog scope. I really want the trace to stay on the
screen until the next trigger. Can analog scopes accomodate this?

 

[Jean-Yves]

I would get an analog but I just don't know if the screen persistence
is adequate. The signal I am looking at is around 20ms duration and
repeats every second or so, so the update of the screen would be
very slow on an analog scope. I really want the trace to stay on the
screen until the next trigger. Can analog scopes accomodate this?

sure it will not...
you can get quite cheaps digital scopes on ebay too...
but for that job you sure need a digital scope.

 

[spasmous2]

sure it will not...
you can get quite cheaps digital scopes on ebay too...
but for that job you sure need a digital scope.

I am a little confused about digital storage oscilloscopes. I have
seen some DSO's that appear to have an analog screen (eg.
http://www.harlanlabs.com/2430a.jpg) whereas others are obviously
digital because of the LCD color display (eg.
http://www.testequity.com/Images/new/tektronix/tds2024.jpg).

I may be just mixed up, but can the first model (Tektronix 2430A)
acquire a 20ms transient, then hold it on the screen until the next
trigger, which may be a second later - or even hold it on the screen
indefinitely?

 

[DaveM]

sure it will not...
you can get quite cheaps digital scopes on ebay too...
but for that job you sure need a digital scope.

I haven't been following the entire thread, but hopefully I won't get flamed for
breaking in here.
I understand that you're looking for a 4-channel scope. Don't know if this is
an absolute requirement, but if you can get by with a 2-channel scope, an analog
storage scope is certainly usable at the time intervals you describe.
They have several storage modes; one is a timed storage, where the trace is
reset after a preset elapsed time, or triggered reset, where the trace stays on
the screen until the next sweep trigger. The brightness is quite good,
especially on slow sweep speeds.
If you want to try a cheapie to begin, I recommend that you look at Ebay for a
Tektronix storage scope. The model 265 scopes were a staple in their day, and
there are lots of them on Ebay, usually in the $200-$300 range. If you can find
one that the seller will verify that it is a working scope, they can be a
bargain. On the other hand, they are obsolete, no longer supported by
Tektronix, and some parts are hard to find.

There are other Tektronix models of analog storage scopes on the market, usually
for reasonable prices.

--
Dave M
MasonDG44 at comcast dot net (Just substitute the appropriate characters in the
address)

Life is like a roll of toilet paper; the closer to the end, the faster it goes.

 

[mike]

Thanks mike for your thoughts. Digital scopes seem to start around
$2000, which is steeper than I was hoping. On ebay there are many
analog scopes that are high quality for much less than this.

I would get an analog but I just don't know if the screen persistence
is adequate. The signal I am looking at is around 20ms duration and
repeats every second or so, so the update of the screen would be
very slow on an analog scope. I really want the trace to stay on the
screen until the next trigger. Can analog scopes accomodate this?

The absolutely positively most important thing is to define with
precision what you want your scope to do. Your current definition
I interpret to take about 3 orders of magnitude less memory depth
than my interpretation of your original requirement.

How much of the signal do you need to see on the screen at one time from
one acquisition? And how fine a detail do you need to see when you
magnify it? That determines what you need from a digital scope.

Second thing to worry about is how much your time is worth.
If you make money with your time, and your time gets $100 an hour,
your break-even point is 20 hours. That's a no-brainer.
If lack of measurement delays a project, it's mismanagement.
A system that's difficult to use on an occasional problem is
frustrating. A system that won't easily do your primary requirement
is EXPENSIVE. Stated another way, look at the life-cycle costs
of the project, not just the $$$ spent on a scope.

If it's a hobby, that's another story. Hobbies are DESIGNED to waste
your time. It's their job...
With a hobby, you might have more fun trying to squint at a
flashing display. Heck, I once built a spot welder. Was trying
to look at the current waveform on an analog scope. I went as far
as hooking a usb webcam up to the trigger-out on my scope and
captured the trace afterglow on the computer. I had great fun
for several months messing around with it. Then a digital scope
fell into my lap and I had the welder problem solved in a matter
of a few minutes. Bummer...I had to find something else to waste
my time.

My favorite scope is a TEK TDS500 series. Mostly because people
give them away dead. All you gotta do for most of them is
wash the boards and replace 140 electrolytic caps.

Analog storage scopes can be interesting if you get them VERY
cheap. I'd avoid 'em.
mike

 

[spasmous2]

The absolutely positively most important thing is to define with
precision what you want your scope to do. Your current definition
I interpret to take about 3 orders of magnitude less memory depth
than my interpretation of your original requirement.

How much of the signal do you need to see on the screen at one time from
one acquisition? And how fine a detail do you need to see when you
magnify it? That determines what you need from a digital scope.

Second thing to worry about is how much your time is worth.
If you make money with your time, and your time gets $100 an hour,
your break-even point is 20 hours. That's a no-brainer.
If lack of measurement delays a project, it's mismanagement.
A system that's difficult to use on an occasional problem is
frustrating. A system that won't easily do your primary requirement
is EXPENSIVE. Stated another way, look at the life-cycle costs
of the project, not just the $$$ spent on a scope.

I apologize for being unclear, I am trying to be as precise as I can!
The signal I expect is around 10-20 ms duration. It repeats every 1
second (approximately) but I only care about the first 10-20 ms. I
want around 2us resolution but probably 10us is more than adequate.
The scope will sit in a corner and display these types of waveforms
for the rest of its life and nothing else.

I will look at the waveforms - but hopefully not for the rest of my
life My viewing pleasure is paramount in this purchase - I would
like the 10-20ms of interest to be displayed brightly for 1 second, or
longer if necessary, until the next 10-20 ms of interest comes along.

My current (analog) scope triggers and shows the 10-20 ms but then
goes dark for the remaining 980-990 ms.

 

[Hammy]

"
I want around 2us resolution but probably 10us is more than adequate.
The scope will sit in a corner and display these types of waveforms
for the rest of its life and nothing else."

Then maybe you should be looking at transient recorders.
Heres just one combo dso Transient recorder.

http://www.maplin.co.uk/module.aspx?ITAG=SPEC&ModuleNo=30367&doy=15m6#spec

 

[Jean-Yves]

I am a little confused about digital storage oscilloscopes. I have
seen some DSO's that appear to have an analog screen (eg.
http://www.harlanlabs.com/2430a.jpg) whereas others are obviously
digital because of the LCD color display (eg.
http://www.testequity.com/Images/new/tektronix/tds2024.jpg).

I may be just mixed up, but can the first model (Tektronix 2430A)
acquire a 20ms transient, then hold it on the screen until the next
trigger, which may be a second later - or even hold it on the screen
indefinitely?

the 2430 will do that.

 

[peter]

really I don't need anything fancy, like FFT or USB storage. I just
want 4-channels and a persistent trace on the screen. And 50 ohm input
impedence.

50 ohm impedence?! Is this a typo? Oscilloscopes are designed to have high
input impedence on the order of mega ohms so as not to disturb the circuit
being monitored.

 

[Jean-Yves]

50 ohm impedence?! Is this a typo? Oscilloscopes are designed to have high
input impedence on the order of mega ohms so as not to disturb the circuit
being monitored.

my high end fluke combiscope has a 1Mohm input impedance
... and a special button for a 50 ohm impedance...
I dont know what this is for but as it is a special feature of a high
end scope it must be usefull for something...

 

[mike]

my high end fluke combiscope has a 1Mohm input impedance
.. and a special button for a 50 ohm impedance...
I dont know what this is for but as it is a special feature of a high
end scope it must be usefull for something...

It's done for signal quality. Active probes are designed to work into
50 ohms.

 

[Jim Yanik]

my high end fluke combiscope has a 1Mohm input impedance
.. and a special button for a 50 ohm impedance...
I dont know what this is for but as it is a special feature of a high
end scope it must be usefull for something...

above 250Mhz,scopes often have 50 ohm terminations.
TEK 485 and 2445/65 series have switchable 1M/50ohm input Z.
TEK 7A24,7A19,and 7A29 7K plug-ins all have only 50 ohm input Z.

Some active FET probes require a 50 ohm termination,and the TEK AM503/A6302
current probe system needs one,too.

 

[Jamie]

50 ohm impedence?! Is this a typo? Oscilloscopes are designed to have high
input impedence on the order of mega ohms so as not to disturb the circuit
being monitored.

my scope has a 50 ohm input option on it.. other than that, it's a 1 meg
input.