Author Topic: Oscilloscope Bandwidth Question  (Read 3317 times)

0 Members and 1 Guest are viewing this topic.

Offline jlamTopic starter

  • Contributor
  • Posts: 11
  • Country: au
Oscilloscope Bandwidth Question
« on: August 02, 2019, 02:07:58 am »
Hello Everybody first post.

I am a looking to buy my first Oscilloscope for Audio Repair and Mod Work.

My goal is to work with 55Mhz to 100Mhz maybe even 300Mhz Opamps and check them for Oscillation and do some Synth Repair work too.

Because the the Opamps will be using feedback capacitors set to roll off just above in 200KHz to I need 100Mhz or 300Mhz bandwidth?

Also is a Dual Trace scope enough for this type of work? Or should opt for a 4 channel scope?

Thanks in Advance.

 
« Last Edit: August 02, 2019, 02:34:05 am by jlam »
 

Offline TK

  • Super Contributor
  • ***
  • Posts: 1722
  • Country: us
  • I am a Systems Analyst who plays with Electronics
Re: Oscilloscope Bandwidth Question
« Reply #1 on: August 02, 2019, 02:23:54 am »
The Rigol MSO5074 might be the best option for you... 4 channels, can be hacked to 350MHz with tons of features.
 
The following users thanked this post: jlam

Offline 0culus

  • Super Contributor
  • ***
  • Posts: 3032
  • Country: us
  • Electronics, RF, and TEA Hobbyist
Re: Oscilloscope Bandwidth Question
« Reply #2 on: August 02, 2019, 02:24:34 am »
I'm confused...even 55 MHz is way above audio ranges....  :-//


Anyway, one way to think about it is to consider the possible harmonic content of your signals of interest. If you care about looking at non sinusoidal waveforms, they often either are made up of odd or even harmonics. If those harmonics aren't within the analog bandwidth of your 'scope, the scope simply won't see them and the waveform you get will likely not resemble what you think it should.
 

Offline jlamTopic starter

  • Contributor
  • Posts: 11
  • Country: au
Re: Oscilloscope Bandwidth Question
« Reply #3 on: August 02, 2019, 02:44:09 am »
Lol true.

I think that refers to unity gain bandwidth. Even the lowly TL072 has a 10Mhz bandwidth.
 

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 8177
  • Country: us
Re: Oscilloscope Bandwidth Question
« Reply #4 on: August 02, 2019, 02:44:59 am »
You might just want to start with a basic older analog scope since even the lowest bandwidths will be more than adequate.  I don't know your market there, but here I've picked a few up under $50 that would be more than suitable.  I don't see why you would need 4 channels or advanced features at this point.
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline jlamTopic starter

  • Contributor
  • Posts: 11
  • Country: au
Re: Oscilloscope Bandwidth Question
« Reply #5 on: August 02, 2019, 03:17:47 am »
I would prefer a Digital Scope at this Point since I will be working with Power Supplies and would like to check their transient response and would like the ability to do screen captures.

Thanks
 

Offline 0culus

  • Super Contributor
  • ***
  • Posts: 3032
  • Country: us
  • Electronics, RF, and TEA Hobbyist
Re: Oscilloscope Bandwidth Question
« Reply #6 on: August 02, 2019, 03:25:07 am »
Just my 2 cents, but IMO a decent analog cathode-ray scope makes a better first 'scope for learning the basics of running one. No fancy features to distract you; learn to use those tools and those skills will easily transfer over to a DSO.
 

Online bdunham7

  • Super Contributor
  • ***
  • Posts: 8177
  • Country: us
Re: Oscilloscope Bandwidth Question
« Reply #7 on: August 02, 2019, 03:26:39 am »
I would prefer a Digital Scope at this Point since I will be working with Power Supplies and would like to check their transient response and would like the ability to do screen captures.

Thanks

That's what we call "additional information".  :)

I can't think of any modern DSO (I mean an actual instrument, not some $20 hobby kit) that wouldn't work for what you want.  Are the power supplies SMPS or linear?
A 3.5 digit 4.5 digit 5 digit 5.5 digit 6.5 digit 7.5 digit DMM is good enough for most people.
 

Offline jlamTopic starter

  • Contributor
  • Posts: 11
  • Country: au
Re: Oscilloscope Bandwidth Question
« Reply #8 on: August 02, 2019, 03:57:45 am »
Sorry I have been tinkering for a few years now and reading as much as I can online so many different opinions. I keep remembering extra information I having to go back to the original post and put it in.

In reply to your question. Both Analog and Digital Supplies. My budget is around the $500 Aud Mark.
But I will go higher if I have to. “Buy Once Cry Once” right?
 

Offline jlamTopic starter

  • Contributor
  • Posts: 11
  • Country: au
Re: Oscilloscope Bandwidth Question
« Reply #9 on: August 02, 2019, 05:17:38 am »
Thank you to everybody for the help I have moved this over to another topic with a poll.

https://www.eevblog.com/forum/beginners/help-me-choose-my-first-scope/

I am not sure if I need a 300Mhz Scope of checking some thing like a THS4012 for Oscillation. Can I do this with a 200Mhz Scope.
« Last Edit: August 02, 2019, 05:48:56 am by jlam »
 

Online tggzzz

  • Super Contributor
  • ***
  • Posts: 21227
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: Oscilloscope Bandwidth Question
« Reply #10 on: August 02, 2019, 07:07:43 am »
Using 300MHz op amps for audio is a poor engineering decision: the best way to avoid oscillation is to avoid using excessively high bandwidth components. Besides, more important specs may be the slew rate into resistive load and/or the output current capability.

If you are looking at "power supplies", you may need to understand where you can put a scope probe's shield. Getting that wrong could wreck the scope, the device under test, or even you.

To understand more about which type of probe to use, and how to use them safely, see the references in https://entertaininghacks.wordpress.com/library-2/scope-probe-reference-material/

Switch-on transients are a good use case for requiring a storage scope. Simple transient response could be measured without a storage scope by making the transients repetitve.

If you are interested in audio, a key scope specification might be the input sensitivity, and be aware of the consequences of having a scope with an 8-bit ADC in the front end.
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 
The following users thanked this post: jlam

Online magic

  • Super Contributor
  • ***
  • Posts: 7454
  • Country: pl
Re: Oscilloscope Bandwidth Question
« Reply #11 on: August 02, 2019, 08:14:53 am »
This is EEVblog. Buy a 500MHz scope, hack it to 1GHz, build a rise time tester and post screenshots :popcorn:

55MHz is way above audio signal bandwidth, but you will encounter such bandwidths in opamps used for audio these days like LM4562, OPA16xx or the good old '637 because Miller compensation is brutal on them - 55MHz is what it takes to get 55k gain at 1kHz and one needs that gain for specs, particularly if driving external buffers.

A 300MHz scope seems like an overkill, though.

Transient response testing of supplies above a few hundred kHz will really be testing of transient response of bypass capacitors. I doubt you need to test at hundreds of MHz unless you work with RF or digital circuitry which will actually draw such ripple currents.
 

Offline jlamTopic starter

  • Contributor
  • Posts: 11
  • Country: au
Re: Oscilloscope Bandwidth Question
« Reply #12 on: August 02, 2019, 08:39:49 am »
LM4562/LME49720 and LM6172 are my personal favourites. How much bandwidth do I need to check those Opamps for Stability.

Haha I knew about the Miller Compensation but never realised it was call that.  :clap:

Now Provided those Opamps compensated with say a 22pf capacitor in the Feedback path and 0.1uF directly on the underside of the PCB, can they still oscillate?


That should have been the original question. :palm:
« Last Edit: August 02, 2019, 08:50:23 am by jlam »
 

Online tggzzz

  • Super Contributor
  • ***
  • Posts: 21227
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: Oscilloscope Bandwidth Question
« Reply #13 on: August 02, 2019, 08:44:46 am »
LM4562/LME49720 and LM6172 are my personal favourites. How much bandwidth do I need to check those Opamps for Stability.

That should have been the original question. :palm:

Working out the right question is often more important and difficult than answering the question. See https://entertaininghacks.wordpress.com/library-2/good-questions-pique-our-interest-and-dont-waste-our-time-2/  for some hints :)

Checking for stability is the last thing you do, and there should be no surprises. The first thing is to design for stability.
« Last Edit: August 02, 2019, 08:46:25 am by tggzzz »
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Offline jlamTopic starter

  • Contributor
  • Posts: 11
  • Country: au
Re: Oscilloscope Bandwidth Question
« Reply #14 on: August 02, 2019, 09:24:04 am »
Thanks to everyone.

It seems like I can get by with a lot less than I thought. What would be the bare minimum I need to do this kind of work ?
 

Online tggzzz

  • Super Contributor
  • ***
  • Posts: 21227
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: Oscilloscope Bandwidth Question
« Reply #15 on: August 02, 2019, 09:51:26 am »
Thanks to everyone.

It seems like I can get by with a lot less than I thought. What would be the bare minimum I need to do this kind of work ?

Have a look at the Digilent Analog Discovery; it is pleasingly small and capable. Being connected to a computer means you have to be careful about grond connections (especially when looking at power supplies). Using an isolating probe is the correct way to solve that issue.
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Online magic

  • Super Contributor
  • ***
  • Posts: 7454
  • Country: pl
Re: Oscilloscope Bandwidth Question
« Reply #16 on: August 02, 2019, 11:01:17 am »
I would like a device like the AD but with some 100MHz IN, 50MHz OUT bandwidth to better cover the bandwidth of new opamps and jellybean BJTs.

Also, for signals within audio bandwidth, you can't beat the resolution and distortion of a good 24b/192kHz soundcard (ESI Juli@, ASUS Xonar STX, whatever else is hot these days) without going all out and getting a dedicated audio analyzer. Buying the AD for THD measurement at 1kHz makes as much sense as using it to test stability of fast opamps or discrete stages :P

You may also go to diyaudio.com and ask people there what they use. It's a more audio-oriented forum than here, go figure.
 

Offline jlamTopic starter

  • Contributor
  • Posts: 11
  • Country: au
Re: Oscilloscope Bandwidth Question
« Reply #17 on: August 02, 2019, 11:29:41 am »
I have the audio range covered I already have multichannel 192Khz interface with 16 In and 16 Out.
Two Mytek 8x192’s
« Last Edit: August 02, 2019, 11:42:52 am by jlam »
 

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 10174
  • Country: gb
Re: Oscilloscope Bandwidth Question
« Reply #18 on: August 02, 2019, 11:34:50 am »
That would have been worth mentioning earlier.   :-\

Time to ask a mod to lock your other thread.
« Last Edit: August 02, 2019, 11:38:06 am by Gyro »
Best Regards, Chris
 

Online tggzzz

  • Super Contributor
  • ***
  • Posts: 21227
  • Country: gb
  • Numbers, not adjectives
    • Having fun doing more, with less
Re: Oscilloscope Bandwidth Question
« Reply #19 on: August 02, 2019, 11:50:18 am »
I have the audio range covered I already have multichannel 192Khz interface with 16 In and 16 Out.
Two Mytek 8x192’s

Read this for some hints.
There are lies, damned lies, statistics - and ADC/DAC specs.
Glider pilot's aphorism: "there is no substitute for span". Retort: "There is a substitute: skill+imagination. But you can buy span".
Having fun doing more, with less
 

Offline rstofer

  • Super Contributor
  • ***
  • Posts: 9964
  • Country: us
Re: Oscilloscope Bandwidth Question
« Reply #20 on: August 03, 2019, 01:31:29 am »
If you are interested in audio, a key scope specification might be the input sensitivity, and be aware of the consequences of having a scope with an 8-bit ADC in the front end.

The 8 bit ADC is also used in some very expensive scopes, not just the entry level.  Scopes are qualitative kinds of things, not quantitative.  If you need an accurate measurement, use something else, like a DMM.  If you want to see squiggles, don't use a DMM, use a scope.

It is only the very recent scopes that have moved beyond 8 bits and they tend to be pricey.

Oddly, I can't find where Siglent has disclosed the number of ADC bits for the 1204X-E or any of the other members of the 1000 series.  Probably just my search deficiency.  But I'll bet that even the new 1204X-E is only 8 bits.  Could be wrong...

 

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
Re: Oscilloscope Bandwidth Question
« Reply #21 on: August 03, 2019, 02:27:06 am »
Thanks to everyone.

It seems like I can get by with a lot less than I thought. What would be the bare minimum I need to do this kind of work ?

The bare minimum? You'd probably be just fine with a 30 year old 15MHz analog scope, people have used much more primitive instruments than that to design and debug excellent audio gear.

With your budget I would be looking at one of the popular midrange hobbyist DSOs out there, Rigol, Siglent, that sort of thing. They will all do what you need, ask several people and you'll get as many different responses on what to get. Don't forget to leave some budget for probes and other accessories.
 

Online tautech

  • Super Contributor
  • ***
  • Posts: 29811
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: Oscilloscope Bandwidth Question
« Reply #22 on: August 03, 2019, 02:28:05 am »
Oddly, I can't find where Siglent has disclosed the number of ADC bits for the 1204X-E or any of the other members of the 1000 series.  Probably just my search deficiency.  But I'll bet that even the new 1204X-E is only 8 bits.  Could be wrong...
::)
In every datasheet under Vertical Resolution.
https://www.siglentamerica.com/wp-content/uploads/dlm_uploads/2018/02/SDS1000X-E_DataSheet_DS0101E-E03C.pdf

At this time all Siglent DSO's are 8 bit and some models allow for Eres bit enhancement of a further 3 bits at the expense of sampling rate.....all X and X-E models.
Avid Rabid Hobbyist.
Some stuff seen @ Siglent HQ cannot be shared.
 

Offline 0culus

  • Super Contributor
  • ***
  • Posts: 3032
  • Country: us
  • Electronics, RF, and TEA Hobbyist
Re: Oscilloscope Bandwidth Question
« Reply #23 on: August 03, 2019, 02:45:29 am »
Thanks to everyone.

It seems like I can get by with a lot less than I thought. What would be the bare minimum I need to do this kind of work ?

The bare minimum? You'd probably be just fine with a 30 year old 15MHz analog scope, people have used much more primitive instruments than that to design and debug excellent audio gear.

With your budget I would be looking at one of the popular midrange hobbyist DSOs out there, Rigol, Siglent, that sort of thing. They will all do what you need, ask several people and you'll get as many different responses on what to get. Don't forget to leave some budget for probes and other accessories.

Jim Williams Application Note 47 is a great read for virtually everything related to these problems. Granted his focus is CROs but he covers a lot of important territory.
 

Offline rstofer

  • Super Contributor
  • ***
  • Posts: 9964
  • Country: us
Re: Oscilloscope Bandwidth Question
« Reply #24 on: August 03, 2019, 03:01:35 am »
Oddly, I can't find where Siglent has disclosed the number of ADC bits for the 1204X-E or any of the other members of the 1000 series.  Probably just my search deficiency.  But I'll bet that even the new 1204X-E is only 8 bits.  Could be wrong...
::)
In every datasheet under Vertical Resolution.
https://www.siglentamerica.com/wp-content/uploads/dlm_uploads/2018/02/SDS1000X-E_DataSheet_DS0101E-E03C.pdf

At this time all Siglent DSO's are 8 bit and some models allow for Eres bit enhancement of a further 3 bits at the expense of sampling rate.....all X and X-E models.

I knew I missed it, it's too important to omit.  I searched for 'bits' and 'ADC' but didn't give Vertical Resolution a thought.  Thanks!
« Last Edit: August 03, 2019, 03:04:35 am by rstofer »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf