Author Topic: OWON XSA1032-TG 3.2G Spectrum Analyzer  (Read 7035 times)

0 Members and 1 Guest are viewing this topic.

Offline TomCTopic starter

  • Frequent Contributor
  • **
  • Posts: 670
  • Country: us
OWON XSA1032-TG 3.2G Spectrum Analyzer
« on: May 04, 2023, 10:49:43 pm »
These went on sale a few months ago at Walmart's marketplace from several Chinese vendors. I've been wanting a spectrum analyzer to play with for years, but couldn't afford the $2500 and up price just for my electronics hobby. So when I saw these I snatched a brand new one for under $600, delivered. Its been a couple months but I haven't had much time to play with it, and truthfully, I'm very new to this devices since I've never owned one before. If any one is familiar with these any comments will be appreciated! I may post more of my experiences with this new toy in the future if there is any interest.
 

Offline nenea dani

  • Regular Contributor
  • *
  • Posts: 74
  • Country: ro
    • electrounivers
Re: OWON XSA1032-TG 3.2G Spectrum Analyzer
« Reply #1 on: May 05, 2023, 07:01:12 am »
    Hi, I don't have this device, only a similar one from the competition. I want to give you a piece of advice, namely to start by purchasing a set of attenuators at least 10db and 6db, but 3db and 1db would also be good. You always have to have an external attenuator when testing something new or something you don't know everything about. Similarly, after any modification in the DUT, you must start with an external attenuator of 10db or 6db. Any connection or disconnection from the power supply or a sudden abnormal regime can exceed the strength of the signal entering the analyzer. You have to be very careful with the first mixer, even if it has some protection that displays a warning message on some analyzers.
    After everything is under control, you can remove the external attenuator and continue the measurements.
    An old military technician told me that whoever has an analyzer and his first mixer didn't burn means that either he didn't work enough or he was very lucky. From my not very big experience, I say that he is right. I recommend attenuators from Minicircuits FW-10+ and so on which are at 1W and 12gHz. And one more thing, do not connect the input of the analyzer to the DUT until the latter is not in a stable regime. :)
 

Offline TomCTopic starter

  • Frequent Contributor
  • **
  • Posts: 670
  • Country: us
Re: OWON XSA1032-TG 3.2G Spectrum Analyzer
« Reply #2 on: May 05, 2023, 08:53:06 am »
Hi nenea dani

Thanks for the hint! I did read about the mishaps with mixers and bought a 20dB 25 W at the same time I got the analyzer. I also have an old variable that has 1, 3, 6, 10, 10, 20, 20 dB switches.

Currently I'm comparing the specs of the Owon XSA1032-TG that I have with what appears to be a similar unit from Rigol, the DSA832E-TG. It's taking longer than I expected because they use different names for the same spec in some cases.
 

Offline pdenisowski

  • Frequent Contributor
  • **
  • Posts: 929
  • Country: us
  • Product Management Engineer, Rohde & Schwarz
    • Test and Measurement Fundamentals Playlist on the R&S YouTube channel
Re: OWON XSA1032-TG 3.2G Spectrum Analyzer
« Reply #3 on: May 05, 2023, 10:21:23 am »
An old military technician told me that whoever has an analyzer and his first mixer didn't burn means that either he didn't work enough or he was very lucky.

Don't tell my manager that - I've been using lab and portable spectrum analyzers for over 15 years and have yet to damage one (knock on wood) :)

Our service center does occasionally get spec ans where someone has burned out the front end by connecting to something they shouldn't have, but it's not as common as one might think.

But you're absolutely right, it doesn't hurt to connect an attenuator if you're not sure what you're looking at.  I sometimes even start with the bigger one :)

Test and Measurement Fundamentals video series on the Rohde & Schwarz YouTube channel:  https://www.youtube.com/playlist?list=PLKxVoO5jUTlvsVtDcqrVn0ybqBVlLj2z8
 
The following users thanked this post: nenea dani

Offline nenea dani

  • Regular Contributor
  • *
  • Posts: 74
  • Country: ro
    • electrounivers
Re: OWON XSA1032-TG 3.2G Spectrum Analyzer
« Reply #4 on: May 05, 2023, 02:47:58 pm »
I worked for a while with a Systron Donner 809-1 which is a heavy tank in the field and I never protect it. However, with a small amplifier I managed to fry his first mixer. It took 30 seconds to change the mixer diode, but in some analyzers the mixer is molded together with many other parts. All evil leads to good because I later managed to build a cartridge type mixer with Schottky diode. What can I say, each with his own stories. :)
 

Offline Anthocyanina

  • Frequent Contributor
  • **
  • Posts: 395
  • Country: 00
  • The Sara
Re: OWON XSA1032-TG 3.2G Spectrum Analyzer
« Reply #5 on: May 05, 2023, 03:15:09 pm »
An old military technician told me that whoever has an analyzer and his first mixer didn't burn means that either he didn't work enough or he was very lucky.

Don't tell my manager that - I've been using lab and portable spectrum analyzers for over 15 years and have yet to damage one (knock on wood) :)

Our service center does occasionally get spec ans where someone has burned out the front end by connecting to something they shouldn't have, but it's not as common as one might think.

But you're absolutely right, it doesn't hurt to connect an attenuator if you're not sure what you're looking at.  I sometimes even start with the bigger one :)



I don't have a spectrum analyser or ever interacted with a bench one. I'm curious about the mechanical aspect of connecting that chonky boi to the analyser's input, do you need to use a cable between the analyser and the attenuator? o.o or are the connectors sturdy enough to just plug it in and not make you worry about mechanical damage?
 

Offline pdenisowski

  • Frequent Contributor
  • **
  • Posts: 929
  • Country: us
  • Product Management Engineer, Rohde & Schwarz
    • Test and Measurement Fundamentals Playlist on the R&S YouTube channel
Re: OWON XSA1032-TG 3.2G Spectrum Analyzer
« Reply #6 on: May 05, 2023, 05:24:40 pm »
I don't have a spectrum analyser or ever interacted with a bench one. I'm curious about the mechanical aspect of connecting that chonky boi to the analyser's input, do you need to use a cable between the analyser and the attenuator? o.o or are the connectors sturdy enough to just plug it in and not make you worry about mechanical damage?

Oh, definitely a cable - that attenuator is even heavier and bulkier than it looks in the picture. 

I never hang any weight (other than a few centimeters of a standard cable) off a spectrum analyzer input, even sometime as light as a typical noise source.  Most of the labs I've been in have small lift stands like this one for holding things like power sensors, DUTs, filters, preamps, etc.  Trust me, it's worth the money.  You can also just prop things up ad hoc with a few books, etc. in a pinch.

But yeah, always a cable when I'm using the big one :)

Test and Measurement Fundamentals video series on the Rohde & Schwarz YouTube channel:  https://www.youtube.com/playlist?list=PLKxVoO5jUTlvsVtDcqrVn0ybqBVlLj2z8
 
The following users thanked this post: Anthocyanina

Offline Anthocyanina

  • Frequent Contributor
  • **
  • Posts: 395
  • Country: 00
  • The Sara
Re: OWON XSA1032-TG 3.2G Spectrum Analyzer
« Reply #7 on: May 05, 2023, 06:54:33 pm »
I don't have a spectrum analyser or ever interacted with a bench one. I'm curious about the mechanical aspect of connecting that chonky boi to the analyser's input, do you need to use a cable between the analyser and the attenuator? o.o or are the connectors sturdy enough to just plug it in and not make you worry about mechanical damage?

Oh, definitely a cable - that attenuator is even heavier and bulkier than it looks in the picture. 

I never hang any weight (other than a few centimeters of a standard cable) off a spectrum analyzer input, even sometime as light as a typical noise source.  Most of the labs I've been in have small lift stands like this one for holding things like power sensors, DUTs, filters, preamps, etc.  Trust me, it's worth the money.  You can also just prop things up ad hoc with a few books, etc. in a pinch.

But yeah, always a cable when I'm using the big one :)

ooh, cool, thanks, that makes sense. I really like the tiny lift!  :-/O
 

Offline TomCTopic starter

  • Frequent Contributor
  • **
  • Posts: 670
  • Country: us
Re: OWON XSA1032-TG 3.2G Spectrum Analyzer
« Reply #8 on: May 07, 2023, 07:29:25 am »
XSA1032-TG vs DSA832E Specifications.

I finally finished compiling the published specifications for these two units and creating a document where they can be compared side by side. They still look similar to me. The Rigol has better numbers in a few important areas, but the OWON has a few features that are either optional or not available for the Rigol. So all in all, I still think that for the price I paid for the OWON, if it actually performs as indicated in the specs, I did pretty good!

To get some hands on with the new toy, and to try to verify some of the specs, I'm going to try to test some of the specs with the equipment that I have available. Currently I'm working on the "SSB Phase Noise" specification. So as soon as I'm finished I'll post the results including the pertinent screen images from the analyzer.

@pdenisowski
I also like the little lift, I saw a very similar one at Amazon for $16, so I ordered it. Thanks for the tip :)

Note: I've recently discovered some errors in the attachment, particularly the 2nd Harmonic Distortion section. I'm working to try and get that corrected and will replace the attachment soon :palm:.

Attached corrected specs as of 5/17/2023
« Last Edit: May 17, 2023, 11:26:35 pm by TomC »
 

Offline TomCTopic starter

  • Frequent Contributor
  • **
  • Posts: 670
  • Country: us
Re: OWON XSA1032-TG 3.2G Spectrum Analyzer
« Reply #9 on: May 07, 2023, 09:03:12 am »
SSB Phase Noise Test part 1

To perform this test I need a signal generator that can output a 1GHz signal, and has better Phase Noise specs than the Spectrum Analyzer. The only signal generator capable of 1GHz that I have available is my SynthUSBII that can output from 34MHz to 4.4GHz. The manufacturer didn't provide specs for the assembled unit, but since it is based on the ADF4351, I'm looking at the noise specs for that IC and they seem to fit the bill. They are using 2.2GHz instead of 1GHz for the noise test, but I'm going to assume that that's good enough.

The signal generator is only 1" x 2", so I have it directly connected to the Spectrum Analyzer's RF input. Attachments 1 - 4 show the pertinent ADF4351 specs, the SynthUSBII settings, and the spectrum of the generator's output at full Span. These signal generators don't have a calibrated output, but as can be seen it comes pretty close to 0dBm with the settings at max power. The prominent third harmonic is due to the ADF4351 based generators design, their output is a square wave, but I don't think that will affect the test results, since the called for offsets are at 10kHz, 100kHz, and 1MHz.

I'll post the actual spectrums of the test next.
 

Offline TomCTopic starter

  • Frequent Contributor
  • **
  • Posts: 670
  • Country: us
Re: OWON XSA1032-TG 3.2G Spectrum Analyzer
« Reply #10 on: May 07, 2023, 11:06:28 pm »
SSB Phase Noise Test part 2

To setup and execute the test I used the following procedure:

1. Turn on the analyzer and allow it to warmup for at least 30 minutes.
2. Connect the generator to the RF input and set it up to 1GHz Max Pwr.
3. Use the analyzer's Auto-Tune to get a basic display.
4. Change the Detector to Sample, the Display to Ampt Graticule On, the Span to 25kHz, RBW to 1kHz, and turn Average On.
5. Set Marker to Delta 10kHz to place the marker at the 10kHz offset, set Marker Fctn to Marker Noise On to normalize the noise reading to 1Hz.

The Spec calls for "10 kHz < -82 dBc/Hz (Typical)" and as shown by the Step_5_Spectrum attachment we got a -83.15 dB/Hz noise reading. Since in this case the carrier power is pretty close to 0dBm, then I think -83dB/Hz is equivalent to -83dBc/Hz. In fact, since the carrier is actually slightly above 0dBm, it's probably closer to -84dBc/Hz.

The procedure for the 100kHz, and 1MHz offsets is basically the same except for the Span and Delta Marker, which need to be set to larger values. For the attached spectrums I used Span = 250kHz, 2.5MHz, and Delta = 100kHz, 1MHz.

As can be seen, the 10kHz, and 100kHz offsets appear to be within specs. However, the 1MHz offset is slightly off, but in my opinion, not by much.
 
The following users thanked this post: nenea dani

Offline DaneLaw

  • Frequent Contributor
  • **
  • Posts: 660
  • Country: dk
Re: OWON XSA1032-TG 3.2G Spectrum Analyzer
« Reply #11 on: May 08, 2023, 11:44:13 pm »
That's quite a sale - if you'r are able to pick them up at 600 bucks..like 1/4 of Wallmarts msrp.  congrats..
cheers, from one envious puppy here  ;D
 

Offline TomCTopic starter

  • Frequent Contributor
  • **
  • Posts: 670
  • Country: us
Re: OWON XSA1032-TG 3.2G Spectrum Analyzer
« Reply #12 on: May 09, 2023, 04:22:14 am »
To tell you the truth, when I first saw it, I thought it was some kind of scam. But since Walmart generally stands behind anything bought from their website, I decided to take my chances and order it right away anyway, I didn't even research it. Although, since I own other OWON products (2 DSOs and 2 ARBs), I remembered seeing it on their website while looking for updates.

First I ordered from the vendor with the lowest price, about $425, but that fell thru with the package lost in transit, it never made it out of China. I got my money back and the price had gone up a little, so I ordered again for less than $600 and got my prize this time.
 

Offline TomCTopic starter

  • Frequent Contributor
  • **
  • Posts: 670
  • Country: us
Re: OWON XSA1032-TG 3.2G Spectrum Analyzer
« Reply #13 on: May 09, 2023, 07:50:21 am »
Displayed Average Noise Level (DANL) Test part 1

To setup and execute the test I used the following procedure:

1. Turn on the analyzer and allow it to warmup for at least 30 minutes.
2. Connect a 50Ω terminator to the RF input.
3. The steps that follow assume you are starting from the factory preset settings, press Preset if needed.
4. Change the Detector to Sample, the Display to Ampt Graticule On, turn Average On, set Ref Level to -70dBm, set Attenuation to 0dB, the Start Freq to 1MHz, the Stop Freq to 1GHz, and Press Peak (Max Search) to find the noisiest spot in this range.
5. To get better resolution press Mkr→CF, and change Span to 1MHz. Note that the noise reading is lower because RBW & VBW were automatically changed to 10kHz.
6. Set Marker Fctn to Marker Noise On to normalize the noise reading to 1Hz. This is the DANL for PA off: 1MHz to 1GHz, the spec calls for -140 dBm (Typical), <-130 dBm.
7. Turn the Preamplifier to On. This is the DANL for PA on: 1MHz to 1GHz, the spec calls for -160 dBm (Typical), <-150 dBm.

The procedure for the 1GHz to 3.2GHz ranges is basically the same except for the Start and Stop frequencies on step 4. The attachments for steps 4-7 are on the next post.
 
The following users thanked this post: nenea dani

Offline TomCTopic starter

  • Frequent Contributor
  • **
  • Posts: 670
  • Country: us
Re: OWON XSA1032-TG 3.2G Spectrum Analyzer
« Reply #14 on: May 09, 2023, 07:53:46 am »
Displayed Average Noise Level (DANL) Test part 2

Here are the attachments for steps 4-7 of the 1GHz to 3.2GHz range.
« Last Edit: May 09, 2023, 08:17:43 am by TomC »
 
The following users thanked this post: nenea dani

Offline nenea dani

  • Regular Contributor
  • *
  • Posts: 74
  • Country: ro
    • electrounivers
Re: OWON XSA1032-TG 3.2G Spectrum Analyzer
« Reply #15 on: May 09, 2023, 09:16:15 am »
How does it handle immediately after 9kHz? Did you see the signal generated by TG? Does it have sufficient amplitude and the correct shape in 9Khz -50Khz? How is the sensitivity in this range?
 

Offline TomCTopic starter

  • Frequent Contributor
  • **
  • Posts: 670
  • Country: us
Re: OWON XSA1032-TG 3.2G Spectrum Analyzer
« Reply #16 on: May 10, 2023, 05:03:00 am »
I wonder if you are thinking audio work?

I attached some spectrums that include the range you are thinking about, but as you can see the noise floor is quite uneven in that region. Probably means response is going to be uneven too. All I had handy right now was the 50 ohm terminator, but I'll experiment with a low frequency generator later on, maybe one of my ARBs, and post some spectrums.

On the last attachment I had the Start Frequency set to 9kHz before I turned on the amplifier, but if you look at the labels you can see that it automatically jumped to 99.95kHz. There for a second I thought I had found a way to flatten the noise floor, but it looks like they limited the frequency range you can use with the amplifier turned on.

From the little bit I've been able to research so far, I think that I read that SAs that start in the kHz range are AC coupled, which probably wouldn't be too good for audio.

I did look at the TG output a while back, but don't remember all the details. On this SA you have to set Span to 0 to use it as a generator. From what I recall it's a sine wave, and I was looking at about 0dBm with one of my DSOs that goes to 200MHz. I'll hook that up again and post some DSO screens and some FFT screens later on.
« Last Edit: May 10, 2023, 05:09:59 am by TomC »
 
The following users thanked this post: nenea dani

Offline mawyatt

  • Super Contributor
  • ***
  • Posts: 4117
  • Country: us
Re: OWON XSA1032-TG 3.2G Spectrum Analyzer
« Reply #17 on: May 10, 2023, 12:47:09 pm »
What you are likely seeing at these lower frequencies is the phase noise of the SA sweeping oscillator which has a 1/f^n characteristic. The better the oscillator 1/f PN the better noise floor in this low frequency region.

BTW great buy for the OP, ~$600  :-+

Best,
Curiosity killed the cat, also depleted my wallet!
~Wyatt Labs by Mike~
 
The following users thanked this post: TomC

Offline nenea dani

  • Regular Contributor
  • *
  • Posts: 74
  • Country: ro
    • electrounivers
Re: OWON XSA1032-TG 3.2G Spectrum Analyzer
« Reply #18 on: May 10, 2023, 03:45:18 pm »
I'm really looking to find a way to use a 9Khz analyzer in the audio field. My Siglent SSA3021X tool has a weak signal between 9Khz and 50Khz, and in addition it is not really sinusoidal. I would somehow go with the normalization function. Now I'm experimenting with lowering the frequency with the help of an H-mixer. I take the spectrum between 20Hz and 100Khz to 1Mhz, obviously only with a very clean external sine signal.
 

Offline mawyatt

  • Super Contributor
  • ***
  • Posts: 4117
  • Country: us
Re: OWON XSA1032-TG 3.2G Spectrum Analyzer
« Reply #19 on: May 10, 2023, 05:19:45 pm »
The PicoScope 4262 might be worthwhile to look into.

Best,
Curiosity killed the cat, also depleted my wallet!
~Wyatt Labs by Mike~
 
The following users thanked this post: egonotto, nenea dani

Offline TomCTopic starter

  • Frequent Contributor
  • **
  • Posts: 670
  • Country: us
Re: OWON XSA1032-TG 3.2G Spectrum Analyzer
« Reply #20 on: May 10, 2023, 09:51:34 pm »
What you are likely seeing at these lower frequencies is the phase noise of the SA sweeping oscillator which has a 1/f^n characteristic. The better the oscillator 1/f PN the better noise floor in this low frequency region.

BTW great buy for the OP, ~$600  :-+

Best,

Thanks for the tip!

Looks like you nailed it! I see at least a 10dB difference by just changing the Sweep from Auto (about 29ms in this case) to Manual 500ms. So at least at the very beginning of the sweep that made a big difference. I set the reference level to -30 in this set of attachments for better accuracy. Thanks again! :)
 

Offline TomCTopic starter

  • Frequent Contributor
  • **
  • Posts: 670
  • Country: us
Re: OWON XSA1032-TG 3.2G Spectrum Analyzer
« Reply #21 on: May 11, 2023, 08:30:02 am »
TG Output at Different Frequencies Part 1

Some of the XSA100 series SAs can use the TG output as an independent CW generator. However, the XSA1032-TG is not one of them. The soft menu selections for this feature are grayed out as can be seen on attachment 2. But as is the case with other TG SAs, it's still possible to set the Span to Zero, turn on the TG output, and use the FREQ soft menu to set the output frequency (see attachment 1). The output level can then be set via the Source soft menu (attachment 2). 0dBm is the highest level and what I'm using for all the screens I captured from my SDS8202 DSO.

Attachment 3 is a screen of one of my favorite dB calculator websites showing some conversions for 0dBm. The website is at:

https://www.analog.com/en/design-center/interactive-design-tools/dbconvert.html
« Last Edit: May 11, 2023, 09:10:43 am by TomC »
 
The following users thanked this post: nenea dani

Offline TomCTopic starter

  • Frequent Contributor
  • **
  • Posts: 670
  • Country: us
Re: OWON XSA1032-TG 3.2G Spectrum Analyzer
« Reply #22 on: May 11, 2023, 08:58:30 am »
TG Output at Different Frequencies Part 2

The bottom left of the OWON SDS8202 screens shows a few automatic measures, but the names may be confusing, Vk is what they use to identify the RMS measurement, and Vp is what they use to identify volts peak to peak (Vpp). One other thing that may be confusing is that on the FFT screens the scale is on dBV, which is not mentioned anyplace as far as I know. The SDS8202 is rated as a 200MHz DSO, but on attachment 14 I think the drop in signal strength is due to the DSO being at the limit of its bandwidth.

Note: I had to replace the attachments because they were hard to read.
« Last Edit: May 11, 2023, 09:15:17 am by TomC »
 

Offline TomCTopic starter

  • Frequent Contributor
  • **
  • Posts: 670
  • Country: us
Re: OWON XSA1032-TG 3.2G Spectrum Analyzer
« Reply #23 on: May 12, 2023, 06:45:15 am »
9kHz to 100kHz Spectrums and Frequency Response

For these tests I'm using an OWON AG1022F Arbitrary Waveform Generator (AWG or ARB). The relevant specs for the Sine output and Sweep function are attached. Attachment 3 shows the AWG setup to output a 25kHz Sine at 633mVpp, since I'm using 50Ω that's around 0dBm. There's just a short 20" cable between the AWG and the SA. Attachments 4-7 show a few spectrums for AWG outputs in this range, including the 25kHz shown in attachment 3. They look pretty clean to me, don't even see any notable harmonics, and the dB levels seem fine.

Attachment 8 shows the AWG's setup to output a Sweep from 9kHz to 100kHz. I set the sweep length to 220 seconds to try to capture the frequency response using the SA's Trace Max Hold function. The idea is to let the SA sweep several times before the AWG changes the frequency (SA's sweep is set to auto which in this case is 404ms). Attachment 9 shows the SA's screen halfway thru the process, and attachment 10 shows when the process is completed. The entire process takes about 4 minutes. I don't know if there is an easier way to do this! So any one with a better idea please chime in. :)
« Last Edit: May 12, 2023, 06:51:31 am by TomC »
 

Offline TomCTopic starter

  • Frequent Contributor
  • **
  • Posts: 670
  • Country: us
Re: OWON XSA1032-TG 3.2G Spectrum Analyzer
« Reply #24 on: May 13, 2023, 09:00:05 am »
Frequency Response Test

To perform this test, I need a signal generator with a range that at least covers 100kHz to 3.2GHz with a nearly flat response. Unfortunately, I don't have a generator that matches those requirements. My AG1022F has an adequate response, but maxes out at 25MHz. I also have the Windfreak SynthUSBII that has a 34MHz to 4.4GHz range but has an uncalibrated output and the response is around plus or minus 4dB. So, I tried to do the test in two sections, 100kHz to 25MHz, and 34MHz to 3.2GHz. The first section gave me what I think are fairly accurate results, but the second section is just showing the response of the generator, not the SA.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf