Choosing between entry-level 12-bit DSOs

[awakephd]

Re-hacking after a firmware update is normal on many oscilloscope hacks. The popular ones get scripts to update them - just run the script after a firmware update. It's not as if there's firmware updates every day...

Things like the FFT hack are separate apps and won't be affected.

Also very helpful. Thanks!

[KungFuJosh]

Not on Siglent. Once a feature is unlocked, it stays that way.

"unlocked" features stay unlocked on Rigols, too.

We're talking about modifications

At least part of what you're talking about are modifications that bring some of the features that can simply be unlocked on the relevant Siglent being discussed in this thread.

The other part would be customizations to the UI.

[awakephd]

At least part of what you're talking about are modifications that bring some of the features that can simply be unlocked on the relevant Siglent being discussed in this thread.

The other part would be customizations to the UI.

Which features in particular? I ask out of ignorance, since I have yet to get my hands on even a basic DSO.

Are there features on the Rigol that are not on the Siglent, or which would have to be hacked to get on the Siglent? I have gained the impression that the Rigol might have more ability to create side-by-side windows showing different but related data ... ? If that is true - and please forgive my ignorance if it is not! - I'm not sure how useful that is in practice. Again, I plead ignorance - but I do feel a bit less ignorant as a result of the many helpful replies on this thread.

[Mechatrommer]

If you are highly technical person and dont develop your own tool like app in PC, siglent is more polished. And 2GSps is really something, its should be possible to look at near 1GHz RF, such as 433 or 800MHz china RF module digital ASK modulation, something 1GSps scope is incapable, even if frontend BW is hackable. But then which type of guy are you? 

[Fungus]

its should be possible to look at near 1GHz RF, such as 433 or 800MHz china RF module digital ASK modulation

Which I would never want to do...

[tggzzz]

If you are highly technical person and dont develop your own tool like app in PC, siglent is more polished. And 2GSps is really something, its should be possible to look at near 1GHz RF, such as 433 or 800MHz china RF module digital ASK modulation, something 1GSps scope is incapable, even if frontend BW is hackable. But then which type of guy are you? 

I would hope they are they type of guy that would use an appropriate tool.

• for many RF measurements that means a frequency-domain or modulation-domain instrument (a scope is time-domain)
• for amplitude shift keying (ASK), that means an "RF probe"

With one of those they could use any scope with a bandwidth that exceeds the signal bandwidth/baudrate (which has nothing to do with the RF carrier frequency).

Example RF probe: hp11096a/b. While that's only characterised to 700MHz, with modern components it would be trivial for an amateur to build a higher frequency equivalent.

[ebastler]

And 2GSps is really something, its should be possible to look at near 1GHz RF, such as 433 or 800MHz china RF module digital ASK modulation, something 1GSps scope is incapable, even if frontend BW is hackable. But then which type of guy are you? 

I would not stress potential hacked, extreme-bandwidth applications here. Very few users will try to go there. But even without considering hardware hacks, it is worth mentioning that 2 GSa/s (as in the SDS800X HD) are sufficient to support 200 MHz bandwidth in 4-channel mode, while 1.2 GSa/s (DHO800/900) are not.

Also, if one has any plans to use the logic analyser option in the future, the Siglent scopes will still have 2 GSa/s available for the analog channels, while the DHO900 will cut that sampling rate in half. (But the Siglent has other limitations, due to the fact that digital data are acquired by an external FPGA subsystem.)

[tggzzz]

I would hope they are they type of guy that would use an appropriate tool.

• for many RF measurements that means a frequency-domain or modulation-domain instrument (a scope is time-domain)

with what? 30K$ fieldfox? and 5K$ probe or cables? Sometime looking at presence of carrier will help. There's diy open loop antenna to detect rf etc.. and with cheap item we can do relative measurement of rf power, not necessarily absolute traceable measurement. if you cant do more with less like i can i wont force you, cheers.[/list]

That's a classic strawman argument, based on a cherry-picked quote and deliberate omission of the points which respond to your assertion - before you even made your assertion!

Here's the complete post to which you responded, with the relevant bits emphasised so you can't miss them.

I would hope they are they type of guy that would use an appropriate tool.

• for many RF measurements that means a frequency-domain or modulation-domain instrument (a scope is time-domain)
• for amplitude shift keying (ASK), that means an "RF probe"

With one of those they could use any scope with a bandwidth that exceeds the signal bandwidth/baudrate (which has nothing to do with the RF carrier frequency).

Example RF probe: hp11096a/b. While that's only characterised to 700MHz, with modern components it would be trivial for an amateur to build a higher frequency equivalent.

And apart from that, my 21GHz spectrum analysers (plural) only cost £350 each.

Finally, can I suggest you remember what you wrote here https://www.eevblog.com/forum/beginners/why-writing-style-and-grammar-matters-in-posts/msg5461151/#msg5461151 Other people shouldn't bother to follow the link

[tggzzz]

That's a classic strawman argument, based on a cherry-picked quote and deliberate omission of the points which respond to your assertion

cherry pick or not, thats what you typed. I dont have to or cant afford to respond to all of your arguments, i'm typing from phone... rather than run away from the point i picked from your post by accusing strawman etc.. why dont you respond with solid technical facts? So i ask again?  What appropriate frequency domain tool to use? It is you that strawman, pot talking to kettle. i talked about basic china module for hobby, not emc approved device.

Que?

Read what I wrote, then stop and think and understand. Yet again the answers to your questions are already there.

Nobody cares about you input device "...i'm typing from a phone...". If you do the equivalent of using a hammer to insert screws, is it surprising that the results are sub-optimal?

[Fungus]

I would not stress potential hacked, extreme-bandwidth applications here. Very few users will try to go there. But even without considering hardware hacks, it is worth mentioning that 2 GSa/s (as in the SDS800X HD) are sufficient to support 200 MHz bandwidth in 4-channel mode, while 1.2 GSa/s (DHO800/900) are not.

I don't see that as much of a problem in practice, especially if you're aware of it. Four simultaneous 200MHz signals aren't easy for hobbyists to find. It's really easy to drop down to two channels if you're in doubt. You'll probably see a difference on the Siglent, too, if it's anything more than a simple sine wave.

If you're always working at 200MHz on four channels then you need a better 'scope than either of these. Probably active probes, too (if you're using probes).

[Mechatrommer]

very few? i'm not sure in HAM community, but around here is just me.... this was the older model DS1054Z, 400MHz RF detection possible.. we aint talk about 4 channels...

[2N3055]

I would not stress potential hacked, extreme-bandwidth applications here. Very few users will try to go there. But even without considering hardware hacks, it is worth mentioning that 2 GSa/s (as in the SDS800X HD) are sufficient to support 200 MHz bandwidth in 4-channel mode, while 1.2 GSa/s (DHO800/900) are not.

I don't see that as much of a problem in practice, especially if you're aware of it. Four simultaneous 200MHz signals aren't easy for hobbyists to find. It's really easy to drop down to two channels if you're in doubt. You'll probably see a difference on the Siglent, too, if it's anything more than a simple sine wave.

If you're always working at 200MHz on four channels then you need a better 'scope than either of these. Probably active probes, too (if you're using probes).

On simple Arduino, I/O pins have sub 1ns transitions. Spectra of that goes over 1GHz...

If you are always working on 4ch and 200 Mhz you need better scope than Rigol that undersamples.
That better scope IS the SDS824xHD that can do that job competently.
Unlike DHO800/900 that cannot because of violating Nyquist.

[ebastler]

I don't see that as much of a problem in practice, especially if you're aware of it. Four simultaneous 200MHz signals aren't easy for hobbyists to find. It's really easy to drop down to two channels if you're in doubt. You'll probably see a difference on the Siglent, too, if it's anything more than a simple sine wave.

If you're always working at 200MHz on four channels then you need a better 'scope than either of these. Probably active probes, too (if you're using probes).

It's enough to have one 200 MHz signal, which you need to observe in context with two other (potentially lower-bandwidth) signals. Not that exotic a requirement for those who decided to get a 200 MHz, 4-channel scope. On the DHO800/900, with only 312 MSa/s per channel, you will be out of luck.

[KungFuJosh]

If you do the equivalent of using a hammer to insert screws, is it surprising that the results are sub-optimal?

OTOH "Hammer Drive Screws" do exist. 😉

[Fungus]

On the DHO800/900, with only 312 MSa/s per channel, you will be out of luck.

Yes, we know sampling theory, thanks...

[Fungus]

It's enough to have one 200 MHz signal, which you need to observe in context with two other (potentially lower-bandwidth) signals. Not that exotic a requirement

Why stop at exactly 200MHz? Where did that magic number come from?

Seems to me like you're inventing use cases that just happen to match the Siglent.

eg. What if somebody feeds in a 250MHz signal by mistake?

[shapirus]

Multi-ADC scopes with higher sampling rate are better for looking at higher frequency signals than single-ADC interleaving scopes with lower sampling rate. Not exactly a surprise.

[2N3055]

If you are always working on 4ch and 200 Mhz you need better scope than Rigol that undersamples.

Oh, FFS... I knew this was coming, again...

How many people meet that criteria of looking at exactly 200MHz on 4 channels, AND who wouldn't be far better served with (eg.) a 350MHz 'scope?

That's such a narrow user profile that it seems silly to me to even argue about it.

[KungFuJosh]

On the DHO800/900, with only 312 MSa/s per channel, you will be out of luck.

Yes, we know sampling theory, thanks...

Not everybody does. If you know the info in the reply, maybe it wasn't for you? 😉

Why stop at exactly 200MHz? Where did that magic number come from?

Seems to me like you're inventing use cases that just happen to match the Siglent.

Of the 2 scopes in this thread, the max is 200MHz, it seems appropriate to discuss the limits of the scopes considered in the thread, doesn't it? DHO800 maxes at 100MHz, and SDS800X HD maxes at 200MHz. Seems relevant.

Also, didn't you hack your 800 to a 900? How fast can it max out in that case?

[ebastler]

Why stop at exactly 200MHz? Where did that magic number come from?

From the scope's datasheet (SDS824X HD). The point is that it can actually do what it's specified to do, in contrast to the DHO924 (or an "upgraded" DHO8x4).

Seems to me like you're inventing use cases that just happen to match the Siglent.

Seems to me like you are getting defensive about the Rigol again, trying to rationalize its awkward design compromises. 

eg. What if somebody feeds in a 250MHz signal by mistake?

Wouldn't it be cool to have a scope which automatically switches in a proper 200 MHz input lowpass when three or four channels are activated, to avoid aliasing? Want to venture a guess which scope might do that?

[Fungus]

What if somebody feeds in a 250MHz signal into a Siglent by mistake?

[ebastler]

What if somebody feeds in a 250MHz signal into a Siglent by mistake?

That's what I just tried to answer with the last paragraph in my previous post. If you have one or two channels active, the Siglent will see the 250 MHz signal just fine and will reconstruct it properly because it's sampling with 1 or 2 GSa/s. If you have three or four channels active, the Siglent will switch in a steeper 200 MHz low-pass filter, and will hence suppress the higher frequency components which would otherwise cause aliasing at 500 MSa/s.

Edit: And yes, this has been measured. Please see the first post of Performa01's review & demonstration thread. (Scroll down to the end of the "table of contents" in that post, to the "Bandwidth" headline.) https://www.eevblog.com/forum/testgear/sds800x-hd-review-demonstration-thread/

[Fungus]

That's what I just tried to answer with the last paragraph in my previous post. If you have one or two channels active, the Siglent will see the 250 MHz signal just fine and will reconstruct it properly because it's sampling with 1 or 2 GSa/s.

No it won't Sin(x)/x reconstruction doesn't work above the "2.5x" rule of thumb.

You might be sampling that signal on the peaks, you might be sampling it on the zero-crossing points. It's just luck.

[ebastler]

No it won't Sin(x)/x reconstruction doesn't work above the "2.5x" rule of thumb.

 

I'm afraid I don't follow. What exactly would go wrong if you sample a 250 MHz signal at 1 GSa/s?

[2N3055]

That's what I just tried to answer with the last paragraph in my previous post. If you have one or two channels active, the Siglent will see the 250 MHz signal just fine and will reconstruct it properly because it's sampling with 1 or 2 GSa/s.

No it won't Sin(x)/x reconstruction doesn't work above the "2.5x" rule of thumb.

You might be sampling that signal on the peaks, you might be sampling it on the zero-crossing points. It's just luck.