Author Topic: Help: PC based mixed signal oscilloscope, Hantek 3254(A) VS Picoscope 2206(B)  (Read 13360 times)

0 Members and 1 Guest are viewing this topic.

Offline NewtoElectronicsTopic starter

  • Newbie
  • Posts: 4
  • Country: gb
Hello All,

I need some help and advice on buying an oscilloscope for my embedded work and can't decide on which PC based oscilloscope to buy.

yes pc based oscilloscope due to restricted space. I need a mixed signal oscilloscope with logic analyser.
Need a decent oscilloscope with at least 50Mhz bandwidth and at least 500Ms/s real-time sample rate and my budget is less than £500 (including VAT).
In my research I have come across two manufactures one is Hantek and the other is Pico scope. Pico scope is a known brand and Hantek not so much.
The Hantek DSO3000(A) (Full Speck - http://www.hantek.com/en/ProductDetail_2_134.html)range has surprisingly good (in some cases better ) speck for the money. Even the 250Mhz bandwidth scope is over £100 cheaper  vs a Pico scpoe's 2206B MSO - 50Mhz MSO(Full Speck -  https://docs-emea.rs-online.com/webdocs/14db/0900766b814dbad1.pdf) which is little over my budget.
 
The Pico scope software looks really well built and also has capabilities to decode up to 15 serial protocols which is a built in software feature. The Pico scope also give 5 years warranty and life time software support/update for the product.
The Hantek software one on the other hand can only do basic functions according to what I can understand. The product is only available through Chinese supplier with at most 1year warranty and if at all any software update/ support rather than what comes in the CD.
I had contacted Hantek about the product and they were very helpful. They replied to my messages within 24hrs. 
 
Has anyone used a Hantek DSO3104A - 100Mhz, DSO3204A200Mhz  or DSO3254A - 250Mhz oscilloscopes before.
Could some one look at the speck and give me their expert advice on this please, what am I missing.
It  a big investment and I won't be able to afford another one for a very long time if I get this wrong.
So I am all ears to different ideas, suggestions, opinions and alternatives.
 
PS - I have already looked into Analogy discovery 2 and Red Pitaya which would have been great if they were below £200. Also they don’t meet my bandwidth and sample rate requirements.
 
 
Really appreciate your help.
 
Thank you.
Bruce
« Last Edit: March 19, 2018, 02:25:32 pm by NewtoElectronics »
 

Offline JPortici

  • Super Contributor
  • ***
  • Posts: 3452
  • Country: it
Get the picoscope. While in hardware they may be simillar the software is many times better.

You have extensive signal analysis capabilities, advanced math, you can add as many decoders as you want and new protocols are added from time to time.

There is a history mode so you can go back to previous acquisitions in the buffer (depends on memory settings

user MrWolf made a very good review here on the forum
 
The following users thanked this post: NewtoElectronics

Offline NewtoElectronicsTopic starter

  • Newbie
  • Posts: 4
  • Country: gb
Thank you JPortici
 

Offline JPortici

  • Super Contributor
  • ***
  • Posts: 3452
  • Country: it
simple and stupid examples,
i was looking at an Acceleration Pedal.
Channel one has position expressed as absolute voltage from 0 to 5V. (Channel A on the scope. 10V/div)
Channel two has position expressed as the duty cycle. (Channel B on the scope)
If the pedal is working correctly the two informations have to match, so i set up a math channel with this formula: 5 * duty(B), manual scaling (-10 to +10)
The math channel matched the analog channel. Pedal was working.

Being able to plot the measurements is much better than the single measurement information, if you are looking at signals that evolve with time.

For another task i had a sensor that output the signal encoded as frequency. I had to be sure that the difference in frequency of the two signals was the same.
I acquire the signals, add two math traces displaying the frequency and scale/offset them so they have to match.

The other scopes i've used that let you plot measurements (and operate on them) are Keysight 3000x (and up) and Lecroy scopes
 
The following users thanked this post: NewtoElectronics, werediver

Offline 2N3055

  • Super Contributor
  • ***
  • Posts: 6447
  • Country: hr
Pico, hands down...
I have old ADC212/100 parallel port one, bought in 2003... They stopped supporting it last year, because it was paralel port... It was supported until software version PicoScope® 6.11.12.1692 in 2016. It gained all software upgrades in between, including adding 17 serial protocol decoders...
And yes, that old guy still works.. And it is a 12 bit, low noise one... I keep one old laptop with XP and paraport just for it. Every now and then it's very useful..

Regards,
Sinisa
 
The following users thanked this post: faraday, NewtoElectronics

Offline hussamaldean

  • Supporter
  • ****
  • Posts: 266
  • Country: iq
why don't you go with MSO1074Z which offers 70MHz and 1GSps and you could hack it to obtain 100MHz with advanced triggers and decoders ?
the price is 699$
it might little bit out of your budget, but it worths the price
 
The following users thanked this post: NewtoElectronics

Offline JPortici

  • Super Contributor
  • ***
  • Posts: 3452
  • Country: it
Not at all if you need more decoders than those included in the scope, need analysis on the ACQUISITION and not on what's displayed on screen, which is heavily undersampled.
 
The following users thanked this post: NewtoElectronics

Offline TopLoser

  • Supporter
  • ****
  • Posts: 1922
  • Country: fr
why don't you go with MSO1074Z which offers 70MHz and 1GSps and you could hack it to obtain 100MHz with advanced triggers and decoders ?

"PC based mixed signal oscilloscope"

"Can't decide on which PC based oscilloscope to buy"

"Yes PC based oscilloscope due to restricted space"

But yes, PICO all the way. the only bad thing is that the inputs are a bit sensitive and blow up if you shove more than 30V into them. Always use a x10 probe whenever possible. If you do blow it up (and lots of customers do...) PICO take it back and repair it under warranty no questions asked, they have excellent customer support.
« Last Edit: March 19, 2018, 06:49:29 pm by TopLoser »
 
The following users thanked this post: NewtoElectronics, derree

Online egonotto

  • Frequent Contributor
  • **
  • Posts: 648
Hello,

I understand your decision problem. The spec of the Hantek looks good if it is true. 1mV/div and 128MB/CH would be great. But what if they are not true?
As is not other specified the 1GS/s is for all 4 channels shared.
If you use 4 channels you have only 250MS/s much to less for 200 MHz.

They have no equivalent sampling as the pico has.

And in practice I had several issue with my Hantek DSO 2074G.
Look at https://www.eevblog.com/forum/testgear/experience-with-dso-2074g/msg1332616/#msg1332616

On the other side the specs of the 2206B MSO looks not so promising:
4mV/div (no fine adjustment of the sensivity) and 32MS (but only in single trigger mode) for all channels.
But what you get is stable as other have already written. The software has much feature. The customer service is very good. They have a Forum where you can ask question.

Overall I think with a picoscope you have pleasure.
Along the way with the Hantek I am afraid you have some hassle.

Best regards
egonotto


 
 
The following users thanked this post: NewtoElectronics

Offline NewtoElectronicsTopic starter

  • Newbie
  • Posts: 4
  • Country: gb
Thank you Egonotto that was helpful.
 

Offline Performa01

  • Super Contributor
  • ***
  • Posts: 1594
  • Country: at
The max. vertical gain is not as important as the noise performance, because this ultimately defines the sensitivity. The PicoScopes are excellent in this regard and on many scopes the higher gain settings below 5mV/div are just fake simulated by digital zoom. This is of course possible to do with a PicoScope as well, just that it is actually called “zoom” then and not a vertical gain setting.

32Mpts of memory is quite a lot – compare this with many benchtop scopes, even the much more expensive ones from e.g. Tek or Keysight. This is segmented memory as a standard and the actual record length in repetitive trigger modes depends on the number of segments set.
 
The following users thanked this post: NewtoElectronics

Offline JPortici

  • Super Contributor
  • ***
  • Posts: 3452
  • Country: it
 
The following users thanked this post: NewtoElectronics, derree

Offline CustomEngineerer

  • Frequent Contributor
  • **
  • Posts: 463
  • Country: us
Hantek is never the correct choice.
 
The following users thanked this post: NewtoElectronics

Offline NewtoElectronicsTopic starter

  • Newbie
  • Posts: 4
  • Country: gb
Hello all,
        Thank you for all your valuable opinion.
I think I am convinced for a Picoscope.  Will go for a normal oscilloscope and buy a separate 8 channel logic analyser.

Has any one wished they have had the oscilloscope and logic analyser together??


Thanks
Bruce
« Last Edit: March 20, 2018, 08:48:33 pm by NewtoElectronics »
 

Offline JPortici

  • Super Contributor
  • ***
  • Posts: 3452
  • Country: it
they have (or had) MSOs
however one big limit of the picoscope (which is probably common to most if not all usb scopes. it is for all that i've looked at) is that it can only trigger on edge conditions (simple, window, alternating, .. there's the list on the software. besides, you can try the sotware with a simulated device) or on logic match.
since the decodes are applied on the pc-side, the scope can't trigger on message,message condition or message error.
 
The following users thanked this post: NewtoElectronics

Offline Elektronik

  • Newbie
  • !
  • Posts: 8
  • Country: de
Hantek? I had one, but after two days, it went back to sender. No AC coupling, many bugs and a manufacturer that is not interested at all.

Not sure about Picoscope? Have a look at this review. The video tells a story.

http://www.afug-info.de/Testberichte/PicoScope-5444D-MSO/

I am considering purchasing a Peaktech USB oscilloscope. Anyone who can tell his experiences with it?

 

Offline MrW0lf

  • Frequent Contributor
  • **
  • Posts: 922
  • Country: ee
    • lab!fyi
Not sure about Picoscope? Have a look at this review. The video tells a story.

Did not watch the video but diagonal read with Google Translate did reveal this (besides complaints about lack of EXT trigger on MSO):

Quote
Mathematical functions
...
The mathematical functions make sense only if at least two channels can be triggered simultaneously.

:scared: :palm:
 

Offline 2N3055

  • Super Contributor
  • ***
  • Posts: 6447
  • Country: hr
Hantek? I had one, but after two days, it went back to sender. No AC coupling, many bugs and a manufacturer that is not interested at all.

Not sure about Picoscope? Have a look at this review. The video tells a story.

http://www.afug-info.de/Testberichte/PicoScope-5444D-MSO/

I am considering purchasing a Peaktech USB oscilloscope. Anyone who can tell his experiences with it?


That review is full of misunderstandings of how digital scope works, and prejudices that any new tool has to be exact like old ones.
You cannot give new functionality and keep old way of use.
Reviewer keeps repeating about "simultaneous trigger on all channels" which doesn't exist on any digital scope in existence.
And Picos can do triggers based on several channels at the same time, and have very advanced triggering.
Whoever did that review didn't understand half of functions of the scope.


Only real complaint I have for Picos are that they don't have triggering on serial protocols.. That is real problem.


Picoscope are KINGS of USB scopes.. Top quality, solid specifications.
Problem is that with all the changes in scope world, they are EXPENSIVE, unless you get really basic ones.
When 2000 € buys you a LOT of standalone scopes that are very good, giving 2000 € for a MSO 3406D is not such a good deal..
Unless you need 500 MS sample depth, bunch of decodes, arbitrary math etc etc.



 

Offline Performa01

  • Super Contributor
  • ***
  • Posts: 1594
  • Country: at
Not sure about Picoscope? Have a look at this review. The video tells a story.

Not everyone who knows how to upload a video is automatically qualified to review modern T&M gear.

I have 3 PicoScopes for private use (2 x 3000 series, 1 x 4000 series) and for professional use I have made sure my whole R&D department got equipped with PicoScope 3000 series MSO. You bet if there were any serious issues with these instruments, I would know it.

The only justified complaints might be the limited input sensitivity and only 1MHz/600 ohms wavegen on the 3000 series, but 5000 series has taken care of both these shortcomings.

The general lack of serial triggers has already been mentioned. But there is a bunch of serial decoders and with super deep segmented memory, we should be able to get away without serial triggers most of the time. The engineers in my department certainly do.

I am considering purchasing a Peaktech USB oscilloscope. Anyone who can tell his experiences with it?

Seriously? Peaktech isn't even a manufacturer, just a trading company that licenses outdated/obsolete designs from various B-C brands and sells them under their own name. Sorry, but comparing Peaktech with PicoTech is just ridiculous (even when they might sound similar).

But yes, PicoTech gear is anything but cheap.
 

Online egonotto

  • Frequent Contributor
  • **
  • Posts: 648
Hello,

2N3055 wrote:
"Reviewer keeps repeating about "simultaneous trigger on all channels" which doesn't exist on any digital scope in existence"

The DSO 2074G can trigger from all active channels with individual timebase. (https://www.eevblog.com/forum/testgear/experience-with-dso-2074g/)
But for what is this useful?

From review:
"Speichertiefe

Die Speichertiefe des Oszilloskops ist ausreichend. "

In Germany  this means the memory of the 5444D MSO is only just ok :(
The 5444D MSO has 512MB memory. How much memory would be than good?

The review is mostly a review of the software with little info about the 5444D MSO

Best regards
egonotto


 
The following users thanked this post: 2N3055

Offline nctnico

  • Super Contributor
  • ***
  • Posts: 26751
  • Country: nl
    • NCT Developments
2N3055 wrote:
"Reviewer keeps repeating about "simultaneous trigger on all channels" which doesn't exist on any digital scope in existence"

The DSO 2074G can trigger from all active channels with individual timebase. (https://www.eevblog.com/forum/testgear/experience-with-dso-2074g/)
But for what is this useful?
A different timebase will make it hard to correlate the signals unless the timebase isn't shared in the same window. There are DSOs which can trigger on each channel individually. Usually this works using a round robin method so the signals aren't related. However this does allow to get signals with unrelated frequencies stable on the screen. In some cases this can be useful.

To the OP: when buying equipment think of some tests you can do to make sure it fits your needs AND that you can return a piece of equipment if you don't like it. Also: equipment should work to your satisfaction with the newest firmware available at the time of delivery. Don't fall into the trap where new firmware gets promised but never delivered because when you get fed up with waiting it is too late for a return.
« Last Edit: October 11, 2018, 01:16:18 am by nctnico »
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline 2N3055

  • Super Contributor
  • ***
  • Posts: 6447
  • Country: hr
The DSO 2074G can trigger from all active channels with individual timebase. (https://www.eevblog.com/forum/testgear/experience-with-dso-2074g/)
But for what is this useful?
I agree with nctnico, alternate trigger is sometimes useful. Having signals that are time uncorrelated I never had need for.
On Pico, you can have logic trigger (works on analog and digital ch) where you show signals when several CH have trigger event at the same time. It works same as alternate with larger blind time. But you get several stable waveforms on a screen.

The general lack of serial triggers has already been mentioned. But there is a bunch of serial decoders and with super deep segmented memory, we should be able to get away without serial triggers most of the time. The engineers in my department certainly do.
Pico software is not ideal. I have a list of my gripes with it. Some of them are:
1. No timestamps on segmented captures. I checked, and verified with Pico, my 3000D has them in hardware. Software just doesn't show them. From API it's available.
2. Math is not always stable. But they fix bugs.
3. In general, I would like if the software was more like what LeCroy does, I just think that one is better organized. Pico does work on a new version of software, but dates are not know yet...

In general, very reputable company, great support, quality products.
But, I agree that for decoding large amount of data, you generally just capture the lot and then look at them on PC. This is how I use it. I use standalone scope for general work and decoding odd packets now and then, and for decoding larger amount of data, or capturing longer sequences, or documenting stuff on PC I use Pico. I actually use it a lot. And I have a old 212/100 12bit one for measuring noise and audio. That old one from 2003 was still supported on software from 2016. It gained all new features and protocol decodes in meantime, that weren't even planned when I bought one. For free.
So yes, expensive is relative term..
Regards,
Siniša
 

Offline MrW0lf

  • Frequent Contributor
  • **
  • Posts: 922
  • Country: ee
    • lab!fyi
On Pico, you can have logic trigger (works on analog and digital ch) where you show signals when several CH have trigger event at the same time. It works same as alternate with larger blind time. But you get several stable waveforms on a screen.

Exactly, heres old thread with some related discussion and examples:
https://www.eevblog.com/forum/testgear/out-of-curiosity-are-there-dsos-with-alternate-trigger/

As for roll mode in "slow sampling" mode (with configurable timebase threshold) you just set trigger to Repeat and move trigger dot to upper right corner of the screen. Once trace gets there is starts "rolling".

Now if to talk actual drawbacks then:
https://www.picotech.com/library/oscilloscopes/streaming-mode
at the end in section Aggregation there is talk about Peak Detection eg Min-Max envelope which is said to be in hardware "on some products" and could be indeed useful but at least in end user software it does no seem to work like this by default and no special mode in sight also :-// Maybe they just mean downscaling memory data to screen in Min-Max manner (in hardware, w/o full memory transmission over USB), which indeed works.



 

Offline Performa01

  • Super Contributor
  • ***
  • Posts: 1594
  • Country: at
I agree with nctnico, alternate trigger is sometimes useful. Having signals that are time uncorrelated I never had need for.
Yes, this is a feature from old analog scopes. Not many of us will have a need for that, since the whole point of using a multi-channel scope is checking/measuring the relative timing of related signals. The monitoring of several independent (unrelated) signals is a completely different and rather specialized application which I don’t think needs to be supported at all costs by general purpose scopes.

On Pico, you can have logic trigger (works on analog and digital ch) where you show signals when several CH have trigger event at the same time. It works same as alternate with larger blind time. But you get several stable waveforms on a screen.
Logic (Pattern) triggers might be a workaround in some situations, but you still have the same timebase for all traces, which defies the purpose for monitoring multiple unrelated signals with significantly different frequencies.

Pico software is not ideal. I have a list of my gripes with it. Some of them are:
1. No timestamps on segmented captures. I checked, and verified with Pico, my 3000D has them in hardware. Software just doesn't show them. From API it's available.
2. Math is not always stable. But they fix bugs.
3. In general, I would like if the software was more like what LeCroy does, I just think that one is better organized. Pico does work on a new version of software, but dates are not know yet...
There is certainly always room for improvement, but then again, what is ideal in this world? 😉
I think most of the times when I had troubles with math, it was user error, i.e. some misunderstanding what a certain constant means or how a specific function actually works.

My biggest grieve is the limited count of only 1000 max. for the measurement statistics. This means you cannot have a long-term average and – even worse – not run some experiment for hours and then determine the correct min/max values for some measurements for the entire timespan.

I use standalone scope for general work and decoding odd packets now and then, and for decoding larger amount of data, or capturing longer sequences, or documenting stuff on PC I use Pico. I actually use it a lot.
Same here…

Plus, I use the Pico whenever really advanced measurements and/or math are required. Especially the FFT is a joy to use with its easy setup, high speed, high resolution and comprehensive measurements.

And I have a old 212/100 12bit one for measuring noise and audio.
You should consider a 4262 – this is truly fantastic!

That old one from 2003 was still supported on software from 2016. It gained all new features and protocol decodes in meantime, that weren't even planned when I bought one. For free.
So yes, expensive is relative term..
I fully agree – price is relative. Considering that you get a full featured, well maintained package without the expensive option game, with good chances that even older devices get significant enhancements anytime for free, is a major benefit of headless USB gear from manufacturers like Pico technology or Signal Hound.
 

Offline 2N3055

  • Super Contributor
  • ***
  • Posts: 6447
  • Country: hr
Logic (Pattern) triggers might be a workaround in some situations, but you still have the same timebase for all traces, which defies the purpose for monitoring multiple unrelated signals with significantly different frequencies.
I meant alternative trigger on a same time scale..  But yes you are correct.

I think most of the times when I had troubles with math, it was user error, i.e. some misunderstanding what a certain constant means or how a specific function actually works.
I'm certainly guilty of lot of that.. But I was talking specifically about formula delimiter bug (that I worked with Pico and they fixed quickly and well) and fact that sometimes, when I go a little crazy with math, it can occasionally crash. But that is not often and you have to push it a bit.

My biggest grieve is the limited count of only 1000 max. for the measurement statistics. This means you cannot have a long-term average and – even worse – not run some experiment for hours and then determine the correct min/max values for some measurements for the entire timespan.
Yes you are absolutely correct.

Same here…

Plus, I use the Pico whenever really advanced measurements and/or math are required. Especially the FFT is a joy to use with its easy setup, high speed, high resolution and comprehensive measurements.
FFT is superb (but I do miss automatic peak list, nothing is perfect like you say ), and DeepMeasure is very nice too.

You should consider a 4262 – this is truly fantastic!

Thank you for rubbing it in  :-DD ... I'm looking for a project to justify buying it. I just spent my budget for this year on a better standalone scope. 4262 will have to wait just little more ..

I fully agree – price is relative. Considering that you get a full featured, well maintained package without the expensive option game, with good chances that even older devices get significant enhancements anytime for free, is a major benefit of headless USB gear from manufacturers like Pico technology or Signal Hound.
+ 1 for SignalHound  too..

Regards,
Siniša
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf