High-end analog or Low-end Digital scope ?

[Pawelr98]

Hello

I'm going to purchase a new oscilloscope.
When browsing local deals I found a high-end analog scope:
SCHLUMBERGER 5228

It seems to have 3 channels,each with 250MHz bandwidth,some basic math functions and other decent functions.
No storage.

With similar money I can purchase a new digital scope.Most beeing 2 channel 100MHz @1GSps.

I'm wondering which one to choose.
For now I'm dealing with low frequency analog circuitry(audio) but I'm planning to jump into high frequency circuits.
That is radio equipment(for now I have built some AM radio) as well as some digital circuitry.
As far as I know even digital scopes can go higher than nominal.
Saw Rigol 1054Z syncing 300MHz sine wave.
I'm aware of "hacking" possible on low-end scopes.

Analog has very high bandwidth but doesn't have any storage.
No goodies like stopping the waveform or single-shot.
With digital scope the sample rate goes down as more inputs are connected.
And no Z-modulation input which I want to use with certain projects(controlling CRT with a computer).
But I get the single shot and ability to analyse a stopped waveform.

When it comes to repair abilities(as saw in a similar thread).
I have one low-end scope. 5MHz single channel analog.
As for tools I have weller PU-81 and one cheap chinese hotair station.

Budget wise it's ~1000-ish PLN or ~280USD (3.54PLN/USD).
I highly prefer local deals over buying from other countries (shipping adds a ton of costs).
VAT is 23% over here.

SCHLUMBERGER 5228 costs 1100PLN=310USD
The seller gives schematics for the scope just in case.

I have few soviet probes(two 1X and one 10X) and one "100MHz" chinese probe.

[rstofer]

That 5228 has some nice features.  Just be careful to limit the input voltage if you use the 50 Ohm setting.

If you're looking at a 100 MHz digital, the DS1054Z just about owns that market.  And that 'hacking' thing isn't a hack at all.  That's just a buzzword somebody stuck on the process to make it sound cool.  All it involves is getting another secret squirrel code (Google for riglol, no that isn't misspelled) and entering it into the scope.  This process is very well understood and the option code is DSER.  I know it isn't in the list.  There is no downside in unlocking the features.  If you have a warranty claim, you can reset the scope (if it works at all).

Just because the 1054Z can display a 300 MHz signal doesn't tell you anything about signal amplitude.  Yes, you will have wiggly lines but the measurements will be way off.

I'm not a fan of used <anything> although I did buy a used Tek 485.  If you can get the DS1054Z for anywhere near the price of the 5228, go for it.  I'm not sure what other 100 MHz DSOs you are looking at.

It really gets down to features and used versus new.  I would seriously consider the 5228 as long as it was working.  I wouldn't be looking for a repair project..

[Fungus]

Saw Rigol 1054Z syncing 300MHz sine wave.
I'm aware of "hacking" possible on low-end scopes.

A hacked DS1054Z will display a 300MHz sine wave but realistic bandwidth is only about 130Mhz.

[nctnico]

That SCHLUMBERGER 5228 only has 250MHz bandwidth at high V/div settings. At maximum sensitivity the bandwidth drops down to 100MHz so IMHO it is not that special.
All in all I'd get a low end digital scope and there is more out there than just the 1054Z.

[tggzzz]

With similar money I can purchase a new digital scope.Most beeing 2 channel 100MHz @1GSps.

That's only one dimension to an ADC's specification; the other is the number of bits. After that, see my .sig!

I believe that many low-end scopes have 8-bit ADC; you may decide that 256 levels is insufficient for audio work.

Note that the Digilent Analog Discovery is a 14-bit ADC (specced to 10MHz), plus an AWG and digital i/o. Within its explicitly stated limitations, it is very good.

[Pawelr98]

I'm aware that amplitude will decrease when going over the range.
My friend had an 35MHz analog which was doing fine above 100MHz but he had to measure the amplitude in a different way.

As for alternatives:
Fluke PM3082-used for 1150PLN
Siglent SDS1052DL-new for 1250PLN
Uni-T UTD2052CL-new for 1100PLN

Rigol 1054Z is around 1650PLN, that's bit high.

There are also some CRT digital scopes (90's era) but I don't like these as they often are old and complicated.
Analog can be repaired with some effort but these old digital scopes are neither easy nor cheap to repair if they break down.
New scopes at least are not worn out.

[nctnico]

Analog scopes can be filled with custom chips and lots of digital circuits as well. Finding a fault in such a device isn't as straightforward as you might think.

[Pawelr98]

Analog scopes can be filled with custom chips and lots of digital circuits as well. Finding a fault in such a device isn't as straightforward as you might think.

Depends on construction and maker.

My current scope has entire multimeter done on TTL gates and discrete parts.
https://www.eevblog.com/forum/repair/c1-107-multimeter-repair/
The oscilloscope part is fully analog.
If I had a second scope right now then I could attempt repair (using documentation which has all critical waveforms listed).
Most parts are clones of western devices.Apart from some custom logic chips (like said 2xNAND + 2xNOR chip).

I found some other scopes.
Tektronix 2230(costs 1000PLN) seems to be analog scope with storage functionality.
How exactly does this work ?
As far as I read it's analog scope with digital circuit which probes the waveform and then "replays" it.
Am I right about it ?

This or similar scope would be about right for my needs.
Analog part which has Z modulation along with storage which helps on fast/changing waveforms.
Pricing is also fine.

I guess that with this particular scope the digital probing isn't too fast.
I have worked on similar-ish scope on lab activities in university but never had a chance to use the storage function.

[nctnico]

Analog scopes can be filled with custom chips and lots of digital circuits as well. Finding a fault in such a device isn't as straightforward as you might think.

Depends on construction and maker.

My current scope has entire multimeter done on TTL gates and discrete parts.
https://www.eevblog.com/forum/repair/c1-107-multimeter-repair/
The oscilloscope part is fully analog.
If I had a second scope right now then I could attempt repair (using documentation which has all critical waveforms listed).
Most parts are clones of western devices.Apart from some custom logic chips (like said 2xNAND + 2xNOR chip).

I found some other scopes.
Tektronix 2230(costs 1000PLN) seems to be analog scope with storage functionality.
How exactly does this work ?
As far as I read it's analog scope with digital circuit which probes the waveform and then "replays" it.
Am I right about it ?

No. I used to own a 2230 myself. It is a DSO and analog scope in one machine and it definitely is a complicated piece of equipment you won't be able to fix if you don't have a second scope. I always used mine in DSO mode.

[Pawelr98]

So the DSO and analog part are independent from each other ?

I managed to find some information. It seems to have 20MSps sampling rate and 4K samples memory.
Does this device have single-shot mode ?

I would mostly need the storage for analysing switching power supplies.Most often at around 100KHz or lower.

[nctnico]

Yes, the analog and DSO are seperate parts in the 2230 which share the CRT and analog inputs. However I wouldn't spend too much on a 2230 because they are 30 years old by now and thus very prone to breaking down. Expect a 2230 (or any very old oscilloscope) to break and become completely worthless. It quickly makes sense to save more money and buy a new entry level DSO which will work well for many years to come.

[Fungus]

...because they are 30 years old by now and thus very prone to breaking down.

At the very least you should be thinking of replacing all the capacitors, etc. in beasts of this age.

Plus the plastics go brittle, all sorts of things. Old gear needs a lot of TLC no matter how awesome they are in theory.

(plus there's the space/heat requirements of anything with a CRT in it)

[BravoV]

Analog can be repaired with some effort but these old digital scopes are neither easy nor cheap to repair if they break down.

Nope, effort, skill and experienced are not enough. What happened if the un-obtainium chip is broken say two days after you just brought it home ? 

Assuming that analog scope's components are easily replaceable or at least easy to find is a really weak argument.

Few well known un-obtainium chips from popular analog scope can cost almost like a new low end DSO.

Btw, to fix a broken scope, most cases you will need ANOTHER WORKING scope side by side. 

[Pawelr98]

I do have a working scope sitting on my bench.

Only the multimeter is not working correctly.
They are fully independent from each other. Cannot be used at the same time.
5MHz,single channel,XY mode,external trigger.

There's a few months warranty available(depends on seller) for most of these used scopes.

Managed to find more information on SCHLUMBERGER 5228.
I found a blog of some RF guy who was able to go above 500MHz on it.
5 V/div - 10 mV/div  -  250 MHz  - 1,4 ns
5 mV/div  -  200 MHz  - 1,75 ns
2 mV/div  -  100 MHz  - 3,5 ns

10mV will be fine for the bandwidth.

I have looked into the schematic.
All parts are TTL/CMOS parts(Worst is 82S129,TTL PROM) or analogs(worst beeing high frequency opamps).
No "special" parts.
Even PCB layouts are given.

For now I'm just planning FM band operation as I need to adjust few vintage superheterodyne radio receivers.
One was originally running on OIRT band and there are some problems in lower CCIR band.

The new scope will also help with the repair of my current scope multimeter.
Will be used as backup tool.

I guess I will go for this one.
I was somewhat leaning for DSO but then I saw the bugs which happen on low-end DSO.
Mostly waveforms "locking up" on some models.
There are bugs which happened on DSO's 10 years ago and somehow got copied to new models which are selling today.

[nctnico]

A TTL PROM is worse than a special chip! They are prone to failure, there is no way to replace it and if it goes bad it is likely bad on all units you can get your hands on.

[Pawelr98]

Here it is.

Fired it up for the first time and already noticed gain loss.
20% at each channel.
Used official repair manual and fixed the gain step-by-step.
Pretty much checking voltages and adjusting one potentiometer per channel.

20min of work and it runs fine now.

The only thing which may need replacement in the future is one of potentiometers.
B2 trigger level adjust.
It just runs bit hard compared to other potentiometers.

Most likely I will stick to B1 time base most of the time.