MSO5000 Bode Plot Capability: Is it Good Enough? [Many tips about a 1st scope]

[balnazzar]

Hi Folks.
I'm still being kept awake at night by my nightmarish oscilloscope purchase.

As I mentioned a while ago, I have one grand to spend, And I'm leaning toward the (in)famous MSO5074, to be hacked, because it packs so many things into a single device. Ok, it does nothing really well, but:

- I can live with the dim display.
- I can also live with the sh*tty fan or snap a noctua into it.
- Maybe I can even live with the subpar front end and the noise (just maybe)..

But I need decent Bode plots. Now, of course I've read here and there, and the general opinion is somewhat twofold.. Some say that MSO5K's Bode plot are decent and people who say otherwise are just bashers. Others say that those BP are so bad that they are tantamount to being useless.

I'd like to hear the opinion of someone who has the '5K and actually use it to perform frequency response analysis quite often.

The alternative would be the sds1104x-e with a siglent wavegen, but spending almost the same amount of money for a 7" scope without separate channel control looks unexciting at best, AND I already have the gen (the UNI-T thingy).

Yeah I know that some people managed to make the sds1104x-e work with other brands, but I have no time right now to tinker, and with uncertain results...

And the new Rigol costs as much as the MSO5K without having anything apart from 12bit and a 10" screen (in particular, no bodes...)

Thanks!

[nctnico]

Have you considered getting a specific device for doing bode plots? What kind of frequency range are you after? Any amplitude & dynamic range requirements? The analog discovery is mentioned often as a tool for doing bode plots. From a comparison between various oscilloscopes made by one of the forum members, it seems Siglent's bode plot is rather slow so may not be ideal for doing a lot of bode plots in a short time.

[Fungus]

Some say that MSO5K's Bode plot are decent and people who say otherwise are just bashers. Others say that those BP are so bad that they are tantamount to being useless.

The truth is probably somewhere in between. 

[2N3055]

Some say that MSO5K's Bode plot are decent and people who say otherwise are just bashers. Others say that those BP are so bad that they are tantamount to being useless.

The truth is probably somewhere in between. 

Not really.
Both Siglent front end and Bode plot implementation is much better. And better, lower nose front end attributes to better dynamic range  in Bode plots.
If Bode plot is really important to you Siglent is better choice.

But MSO5000 has some good sides compared to SDS1104X-E.  It has MSO option and decoding more protocols. It might be better choice for microcontroller digital work.
But for pure analog work SDS1104X-E will outperform it.
Individual vertical knobs or single combined one are not important. They make no difference after you get used to it either way..

But let's get back to your incompletely formulated question:
Only thing clear is approximate budget.

What does "decent" Bode plot means? What are you plan to measure, dynamic range (lowest and highest signals), frequency range.... etc etc..  That will define what you need. Maybe you expect (or need) performance that no scope can do but you need to buy dedicated device that is out of your budget. Which is the point where you change expectations..

Best,

[mawyatt]

Have you considered getting a specific device for doing bode plots? What kind of frequency range are you after? Any amplitude & dynamic range requirements? The analog discovery is mentioned often as a tool for doing bode plots. From a comparison between various oscilloscopes made by one of the forum members, it seems Siglent's bode plot is rather slow so may not be ideal for doing a lot of bode plots in a short time.

Having considerable 1st hand experience with the Bode function in the Siglent 2000X+, can say it's quite capable indeed! A little slow, yes, but very versatile.

Siglent has put in considerable effort into this Function, as well as the excellent 2Mpt FFT, and it shows. These functions are for more serious work, not just checking the box for marketing, and not a gimmick function.

We've done various filters, closed loop analysis of complex analog functions, op-amps, and even injection locking oscillator analysis. Here's few threads.


https://www.eevblog.com/forum/testgear/bode-plots-useful/msg3580010/#msg3580010

https://www.eevblog.com/forum/testgear/diy-transformer-for-use-with-bode-plots/msg4182022/#msg4182022

https://www.eevblog.com/forum/testgear/capacitive-impedance-plots-with-sds2104x-plus-bode-function/msg4335745/#msg4335745

https://www.eevblog.com/forum/projects/injection-locked-peltz-oscillator-with-bode-analysis/msg4424434/#msg4424434

Best,

[nctnico]

I get that you are happy with the tool available to you, but have you tried other tools as a comparison? I have a DSO with bode plot as well but I always use my network analyser for doing bode plots. The network analyser is just more convenient to use.

[balnazzar]

Have you considered getting a specific device for doing bode plots? What kind of frequency range are you after? Any amplitude & dynamic range requirements? The analog discovery is mentioned often as a tool for doing bode plots. From a comparison between various oscilloscopes made by one of the forum members, it seems Siglent's bode plot is rather slow so may not be ideal for doing a lot of bode plots in a short time.

Thanks for the heads up about Siglent's sluggishness at bodeplotting, I wasn't aware of that.

Mh. TBH, I considered the AD2. It matches my requirements (barely, but it's workable). It has other perks, like differential probing (as long as you don't use the BNCs), and with the BNCs it can get to 30 MHz, which is kind of OK. Plus, it has other handy features.
Not to mention the good software, a thing that finds both Siglent and Rigol gravely wanting.

My problem with the AD2 is that I need a scope with at least 3 channels. So, considering some 400 eur for the AD2, I should also buy a 4-ch basic scope (5-600 bucks max). Any advice about this would be welcome, but I think it boils down to the usual couple.. The aforementioned Siglent, or the old (not good old, just old) 1054Z.

You seem to be an experienced contributor.. Tell me one thing: would it be possible to connect both the AD2 and a Keysight 1052A to a computer so that one can use the whole contraption as a 4 ch scope?

Thanks!

[balnazzar]

Not really.
Both Siglent front end and Bode plot implementation is much better. And better, lower nose front end attributes to better dynamic range  in Bode plots.
If Bode plot is really important to you Siglent is better choice.

But MSO5000 has some good sides compared to SDS1104X-E.  It has MSO option and decoding more protocols. It might be better choice for microcontroller digital work.
But for pure analog work SDS1104X-E will outperform it.
Individual vertical knobs or single combined one are not important. They make no difference after you get used to it either way..

[...]

What does "decent" Bode plot means? What are you plan to measure, dynamic range (lowest and highest signals), frequency range.... etc etc..  That will define what you need. Maybe you expect (or need) performance that no scope can do but you need to buy dedicated device that is out of your budget. Which is the point where you change expectations..

Right now I'm learning (as I said, that would be my first scope), so I'll start by tinkering with the frequency response of analog circuits. But an oscilloscope is not a thing that one buys each year. I'm hoping to grow up (electronically speaking) into it, and it has to last many years. But, possibly, without breaking the bank.

One solution would be buying the AD2 for the Bode plots along with the good software that comes with it (and logic & network analyzers) AND the siglent for having a decent 4chs scope with buttons, knobs and a screen...

[balnazzar]

Having considerable 1st hand experience with the Bode function in the Siglent 2000X+, can say it's quite capable indeed! A little slow, yes, but very versatile.

That scope, in its cheapest variant, is 3X more expensive than the SDS1104X-E: I don't think they are comparable.

[tautech]

Having considerable 1st hand experience with the Bode function in the Siglent 2000X+, can say it's quite capable indeed! A little slow, yes, but very versatile.

That scope, in its cheapest variant, is 3X more expensive than the SDS1104X-E: I don't think they are comparable.

Do you ?
Best you have a deeper look at datasheets and here where the 100dB+ of Bode plot dynamic range is displayed:
https://www.eevblog.com/forum/testgear/siglent-sds1x04x-e-bodeplot-ii-(sfra)-features-and-testing-(coming)/

You should also be aware SDS2kX Plus has an inbuilt 50 MHz AWG and therefore already has a stimulus source for Bode plot use whereas with X-E models a stimulus source for Bode plot use is an additional expenditure.

[nctnico]

Have you considered getting a specific device for doing bode plots? What kind of frequency range are you after? Any amplitude & dynamic range requirements? The analog discovery is mentioned often as a tool for doing bode plots. From a comparison between various oscilloscopes made by one of the forum members, it seems Siglent's bode plot is rather slow so may not be ideal for doing a lot of bode plots in a short time.

Thanks for the heads up about Siglent's sluggishness at bodeplotting, I wasn't aware of that.

Mh. TBH, I considered the AD2. It matches my requirements (barely, but it's workable). It has other perks, like differential probing (as long as you don't use the BNCs), and with the BNCs it can get to 30 MHz, which is kind of OK. Plus, it has other handy features.
Not to mention the good software, a thing that finds both Siglent and Rigol gravely wanting.

My problem with the AD2 is that I need a scope with at least 3 channels. So, considering some 400 eur for the AD2, I should also buy a 4-ch basic scope (5-600 bucks max). Any advice about this would be welcome, but I think it boils down to the usual couple.. The aforementioned Siglent, or the old (not good old, just old) 1054Z.

You seem to be an experienced contributor.. Tell me one thing: would it be possible to connect both the AD2 and a Keysight 1052A to a computer so that one can use the whole contraption as a 4 ch scope?

No. But depending on your location, a GW Instek GDS1054B could be a good alternative to buy a decent 4 channel oscilloscope (options can be enabled for free using a key generator available on this forum). For example: https://eleshop.eu/gw-instek-gds-1054b.html

[Warpspeed]

What you require, really depends on the application and test envioronment.
What are you planning to Bode plot ?

Things such as frequency range, noise floor and dynamic range are pretty obvious, and many measuring instruments these days have some excellent features. If you are studying just analog type circuits, amplifiers, filters, servo systems etc... that is all pretty straightforward.

These days switching power supplies are a significant area of interest, and Bode plots are just about a must have, if you wish to maintain your sanity.
Switching supplies are are very Bode plotter unfriendly by being significant producers of wideband noise by the power components.

Trying to resolve a sine wave of a couple of millivolts submerged in hundreds of millivolts of spikes and ripple is something very simple equipment is just not capable of doing.  Its all very well the software being able to plot a nice smooth gain curve over 100db dynamic range, but unless millivolt or microvolt level signals can be resolved beneath huge levels of noise, the pretty plot on the screen is just not going to happen.

Many if not all low end Bode plot systems just use direct analog to digital conversion and some clever number crunching. That works fine for demo purposes by the plotter salesman, but not when trying to make real world measurements on switching power supplies or in really noisy applications.

The only way to do it is with a couple of very narrow band swept filters, similar in concept to a spectrum analyser.  That can be done either with hardware or very sophisticated software to reject everything but the exact test frequency being run through the equipment under test.
All very clever, but it does not come cheap unfortunately.

So a simple low end Bode plotter may be all you need, or it may not work for you at all, depending on the test environment.

[mawyatt]

I get that you are happy with the tool available to you, but have you tried other tools as a comparison? I have a DSO with bode plot as well but I always use my network analyser for doing bode plots. The network analyser is just more convenient to use.

We have a VNA, but often prefer the Bode Function because of the higher impedances and/or voltages and lower frequencies involved. Haven't used any of the other brands Bode functions tho, so can't speak for them. Sure a dedicated High Input Z Lower Frequency Network Analyzer is a better instrument, but that's another instrument and expense, for us the SDS2000X+ has filled that need nicely!!

If you actually use the Bode function in the SDS2000X+ and spend some time with it you will appreciate the performance they've achieved. Like the comment just above about the Frequency Selective measurement capability, don't think they use the Synchronous Sampling, but whatever technique they employed does work well rejecting signals not of interest!!

Best,

[Fungus]

One solution would be buying the AD2 for the Bode plots along with the good software that comes with it (and logic & network analyzers) AND the siglent for having a decent 4chs scope with buttons, knobs and a screen...

Obvious first question: What frequencies and sensitivities do you need for your bode plots? What are you measuring?

Having considerable 1st hand experience with the Bode function in the Siglent 2000X+

When did the 2000X series come into it? That's way over budget.

[2N3055]

One solution would be buying the AD2 for the Bode plots along with the good software that comes with it (and logic & network analyzers) AND the siglent for having a decent 4chs scope with buttons, knobs and a screen...

Obvious first question: What frequencies and sensitivities do you need for your bode plots? What are you measuring?

Having considerable 1st hand experience with the Bode function in the Siglent 2000X+

When did the 2000X series come into it? That's way over budget.

I believe Mike simply speaks of his experience with the product he owns.

Truth is  BodePlot II on SDS2000X + is same algorithms as the one on SDS1000X-E. So Mike's good experience with it is very relevant for both scopes, one within the budget and one that is not.

[mwb1100]

But I need decent Bode plots. Now, of course I've read here and there, and the general opinion is somewhat twofold.. Some say that MSO5K's Bode plot are decent and people who say otherwise are just bashers. Others say that those BP are so bad that they are tantamount to being useless.

I'd like to hear the opinion of someone who has the '5K and actually use it to perform frequency response analysis quite often.

I don't know a damn thing about Bode plots, but there is a youtube with someone generating a Bode plot on an MSO5000 and showing some Bode plot results for various opamps.  The "results" part of the video just shows the final graphs - the time spent plotting the graphs isn't in the video.



Hopefully there will be some information there that is useful to you.

- I can live with the dim display.
- I can also live with the sh*tty fan or snap a noctua into it.

I got an MSO5074 a couple weeks ago (haven't hacked it yet); the display doesn't seem dim to  me and the fan is barely noticeable (it is silent compared to the noise the fan on my cheap "Dr. Meter" power supply makes).  I don't know if Rigol addressed those issues or if I'm just not sensitive to them.

[Fungus]

- I can live with the dim display.
- I can also live with the sh*tty fan or snap a noctua into it.

I got an MSO5074 a couple weeks ago (haven't hacked it yet); the display doesn't seem dim to  me and the fan is barely noticeable (it is silent compared to the noise the fan on my cheap "Dr. Meter" power supply makes).  I don't know if Rigol addressed those issues or if I'm just not sensitive to them.

- Maybe I can even live with the subpar front end and the noise (just maybe).

And they've added a hires mode that will reduce noise to within spitting distance of a Siglent in a pinch thanks to the beast's massive sample rate.

(but not on all channels simultaneously)

[balnazzar]

I don't know a damn thing about Bode plots, but there is a youtube with someone generating a Bode plot on an MSO5000 and showing some Bode plot results for various opamps.  The "results" part of the video just shows the final graphs - the time spent plotting the graphs isn't in the video.

I got an MSO5074 a couple weeks ago (haven't hacked it yet); the display doesn't seem dim to  me and the fan is barely noticeable (it is silent compared to the noise the fan on my cheap "Dr. Meter" power supply makes).  I don't know if Rigol addressed those issues or if I'm just not sensitive to them.

Thanks for reporting about the display and the noise.

As for the Bodes, I'd like to see the 5K's Bode plots compared to a scope (or analyzer) that's known for producing high quality Bode plots. Seeing it standalone is a bit pointless without nothing to compare with.
But thanks for the video, it's interesting by its own.

[balnazzar]

You should also be aware SDS2kX Plus has an inbuilt 50 MHz AWG and therefore already has a stimulus source for Bode plot use whereas with X-E models a stimulus source for Bode plot use is an additional expenditure.

Is it inbuilt? I thought one had to buy the little black boxy thing in order to enable the WG...

[balnazzar]

No. But depending on your location, a GW Instek GDS1054B could be a good alternative to buy a decent 4 channel oscilloscope (options can be enabled for free using a key generator available on this forum). For example: https://eleshop.eu/gw-instek-gds-1054b.html

Interesting. Mh. How does it compare to the Siglent? Snappier UI?

The screen looks smaller than Siglent's, but it's still 7". Probably it's the scope body that's larger.. I'm seeing that people on this forum adapted a 8" screen, but it's far from being straightforward.

Tell me, is that Eleshop a reliable vendor? Thanks!

[balnazzar]

Things such as frequency range, noise floor and dynamic range are pretty obvious, and many measuring instruments these days have some excellent features. If you are studying just analog type circuits, amplifiers, filters, servo systems etc... that is all pretty straightforward.

These days switching power supplies are a significant area of interest, and Bode plots are just about a must have, if you wish to maintain your sanity.
Switching supplies are are very Bode plotter unfriendly by being significant producers of wideband noise by the power components.

Trying to resolve a sine wave of a couple of millivolts submerged in hundreds of millivolts of spikes and ripple is something very simple equipment is just not capable of doing.  Its all very well the software being able to plot a nice smooth gain curve over 100db dynamic range, but unless millivolt or microvolt level signals can be resolved beneath huge levels of noise, the pretty plot on the screen is just not going to happen.

That's the point. I'll be studying the frequency response of amplifiers, filters, servo systems, etc, for a time. The next step forward would be power supplies. I'd like to have an instrument that scales up well, or at least acceptably..

If that's not possible with my budget, ok, but at least it has to be really good with the basic analog stuff.

[balnazzar]

No. But depending on your location, a GW Instek GDS1054B could be a good alternative to buy a decent 4 channel oscilloscope (options can be enabled for free using a key generator available on this forum). For example: https://eleshop.eu/gw-instek-gds-1054b.html

Interesting. Mh. How does it compare to the Siglent? Snappier UI?

The screen looks smaller than Siglent's, but it's still 7". Probably it's the scope body that's larger.. I'm seeing that people on this forum adapted a 8" screen, but it's far from being straightforward.

Tell me, is that Eleshop a reliable vendor? Thanks!

Addendum: How is the software?

[nctnico]

No. But depending on your location, a GW Instek GDS1054B could be a good alternative to buy a decent 4 channel oscilloscope (options can be enabled for free using a key generator available on this forum). For example: https://eleshop.eu/gw-instek-gds-1054b.html

Interesting. Mh. How does it compare to the Siglent? Snappier UI?

The screen looks smaller than Siglent's, but it's still 7". Probably it's the scope body that's larger.. I'm seeing that people on this forum adapted a 8" screen, but it's far from being straightforward.

Tell me, is that Eleshop a reliable vendor? Thanks!

IMHO Eleshop is a reliable vendor. I have bought some test equipment from them (and returned one as well because it didn't meet my expectations). The firmware for the GDS1054B is mature. If you do run into a problem, the service department from GW Instek is responsive and they will fix bugs (if you happen to find one). Personally I find GW Instek's UI easier to use compared to Keysight. The key difference is that GW Instek has a horizontal and vertical row of buttons which makes navigating through the menus much easier. The seperate 'select' button prevents you from turning the rotary encoder while pressing it and selecting a different menu item than the one you wanted to select (which typically happens to me when the rotary encoder and select button are integrated in one).

[mawyatt]

Having considerable 1st hand experience with the Bode function in the Siglent 2000X+

When did the 2000X series come into it? That's way over budget.

Maybe you should read the entire post and include what this was responding too!!

Here's the prior comment by nctnico:

"Have you considered getting a specific device for doing bode plots? What kind of frequency range are you after? Any amplitude & dynamic range requirements? The analog discovery is mentioned often as a tool for doing bode plots. From a comparison between various oscilloscopes made by one of the forum members, it seems Siglent's bode plot is rather slow so may not be ideal for doing a lot of bode plots in a short time."

So nctnico brought Siglent into the discussion, and we responded to such.

Regarding relevance, think 2N3055 covered that pretty well!!

Best,

[2N3055]

No. But depending on your location, a GW Instek GDS1054B could be a good alternative to buy a decent 4 channel oscilloscope (options can be enabled for free using a key generator available on this forum). For example: https://eleshop.eu/gw-instek-gds-1054b.html

Interesting. Mh. How does it compare to the Siglent? Snappier UI?

The screen looks smaller than Siglent's, but it's still 7". Probably it's the scope body that's larger.. I'm seeing that people on this forum adapted a 8" screen, but it's far from being straightforward.

Tell me, is that Eleshop a reliable vendor? Thanks!

IMHO Eleshop is a reliable vendor. I have bought some test equipment from them (and returned one as well because it didn't meet my expectations). The firmware for the GDS1054B is mature. If you do run into a problem, the service department from GW Instek is responsive and they will fix bugs (if you happen to find one). Personally I find GW Instek's UI easier to use compared to Keysight. The key difference is that GW Instek has a horizontal and vertical row of buttons which makes navigating through the menus much easier. The seperate 'select' button prevents you from turning the rotary encoder while pressing it and selecting a different menu item than the one you wanted to select (which typically happens to me when the rotary encoder and select button are integrated in one).

Little GW Instek 1000 has bode plotting?