Author Topic: Siglent SDG1025 Arbitrary/function generator under some tests  (Read 85745 times)

0 Members and 1 Guest are viewing this topic.

Offline kwass

  • Frequent Contributor
  • **
  • Posts: 347
  • Country: us
Re: Siglent SDG1025 Arbitrary/function generator under some tests
« Reply #100 on: November 25, 2015, 05:12:47 pm »
Wonder what the factory does to cal the 25 MHz clock, or do they make the assumption it is "close enough" from the start?

I'm sure that they do.  The spec is within 50ppm, which is almost impossible not to achieve using a $0.15 crystal
-katie
 

Offline NivagSwerdna

  • Super Contributor
  • ***
  • Posts: 2495
  • Country: gb
Re: Siglent SDG1025 Arbitrary/function generator under some tests
« Reply #101 on: January 27, 2016, 10:54:45 am »
If you need higher clock accuracy without using an external 10MHz source, you can pretty easily modify the SDG1025 to use an internal 25MHz TCXO.  I've done this and it works great for most stuff so that I only occasionally need to use the 10MHz input.  Here's how to do it: https://youtu.be/RxCDfNCLj08

I was considering doing this... The video was made some time ago so I was wondering if anyone has done some further investigation in the meantime... specifically I'm wondering if it is possible to install the TCXO but leave the existing XTAL in place and avoid the fly lead to TP1.... it sort of looks like the TCXO might be a legitimate option and there might be firmware to switch between them?

Just a thought.
 

Offline kwass

  • Frequent Contributor
  • **
  • Posts: 347
  • Country: us
Re: Siglent SDG1025 Arbitrary/function generator under some tests
« Reply #102 on: January 27, 2016, 02:31:15 pm »


I was considering doing this... The video was made some time ago so I was wondering if anyone has done some further investigation in the meantime... specifically I'm wondering if it is possible to install the TCXO but leave the existing XTAL in place and avoid the fly lead to TP1.... it sort of looks like the TCXO might be a legitimate option and there might be firmware to switch between them?


I've changed mine to use a TCXO.  No firmware change is needed nor any other setting changes.  If you don't add the extra jumper wire the TCXO will not work.  If you leave in the old XTAL, I suspect that the TCXO output will swamp the XTAL oscillation and it will be ok, but that might not happen and you'll get some mixing of frequencies.  Why leave it in?
-katie
 
The following users thanked this post: electrongeek

Offline uncle_bob

  • Supporter
  • ****
  • Posts: 2441
  • Country: us
Re: Siglent SDG1025 Arbitrary/function generator under some tests
« Reply #103 on: January 27, 2016, 05:45:23 pm »


I was considering doing this... The video was made some time ago so I was wondering if anyone has done some further investigation in the meantime... specifically I'm wondering if it is possible to install the TCXO but leave the existing XTAL in place and avoid the fly lead to TP1.... it sort of looks like the TCXO might be a legitimate option and there might be firmware to switch between them?


I've changed mine to use a TCXO.  No firmware change is needed nor any other setting changes.  If you don't add the extra jumper wire the TCXO will not work.  If you leave in the old XTAL, I suspect that the TCXO output will swamp the XTAL oscillation and it will be ok, but that might not happen and you'll get some mixing of frequencies.  Why leave it in?


The bigger problem is that the feedback *through* the crystal will give you a really wacked out load impedance on the TCXO. That will not help it's stability at all.

Bob
 

Offline Muxr

  • Super Contributor
  • ***
  • Posts: 1369
  • Country: us
Re: Siglent SDG1025 Arbitrary/function generator under some tests
« Reply #104 on: January 27, 2016, 06:55:45 pm »
If you need higher clock accuracy without using an external 10MHz source, you can pretty easily modify the SDG1025 to use an internal 25MHz TCXO.  I've done this and it works great for most stuff so that I only occasionally need to use the 10MHz input.  Here's how to do it: https://youtu.be/RxCDfNCLj08

I was considering doing this... The video was made some time ago so I was wondering if anyone has done some further investigation in the meantime... specifically I'm wondering if it is possible to install the TCXO but leave the existing XTAL in place and avoid the fly lead to TP1.... it sort of looks like the TCXO might be a legitimate option and there might be firmware to switch between them?

Just a thought.
Found I think that same part on aliexpress. I will give it a try once it slow boats to me. http://www.aliexpress.com/item/ONE-PIECE-PRECISION-0-3ppm-25MHz-25-000MHz-25M-DIP-TCXO-Support-bulk-orders/32297054283.html
 

Offline OE2WHP

  • Regular Contributor
  • *
  • Posts: 94
  • Country: at
Re: Siglent SDG1025 Arbitrary/function generator under some tests
« Reply #105 on: August 05, 2016, 09:50:07 am »
Any update here? Did the Aliexpress Part fit?
If so, I would order it as well.

thanks
73

 

Offline electrongeek

  • Contributor
  • Posts: 20
  • Country: us
Re: Siglent SDG1025 Arbitrary/function generator under some tests
« Reply #106 on: September 20, 2016, 02:59:03 pm »
Just finished installing a TCXO. Getting at or better than .04 ppm accuracy at 25 MHz on initial testing with a GPSDO referenced counter at room temperature. Stable output. is listed on ebay as:
1PCS TCXO 0.3PPM 25MHz 25.000MHZ Ultra precision Oscillator

Very nice upgrade!
« Last Edit: September 20, 2016, 03:00:44 pm by electrongeek »
 
The following users thanked this post: tautech, OE2WHP

Offline OE2WHP

  • Regular Contributor
  • *
  • Posts: 94
  • Country: at
Re: Siglent SDG1025 Arbitrary/function generator under some tests
« Reply #107 on: February 01, 2017, 10:56:58 am »
Hi,

does anyone know a source for this 25MHz 0.3ppm TCXO?

Thanks
 

Offline SouthPark

  • Contributor
  • Posts: 28
  • Country: au
Re: Siglent SDG1025 Arbitrary/function generator under some tests
« Reply #108 on: March 15, 2017, 09:07:28 am »
Hi everybody. I'm assuming that a lot of people here have a SDG-1025 arbitrary function generator, or similar model.

I got a small question about the Amplitude Modulation (AM) generation mode. I notice that - as soon as I switch from normal sine-wave generator mode to the AM generation mode, the AM carrier amplitude measures (using an oscilloscope with the SDG-1025 set to Hi-Z mode) to be half the peak-to-peak value of the LCD display setting.

So, if the LCD display says '2 V peak to peak' for the AM carrier amplitude (with ZERO modulation), the oscilloscope measures half of that, ie. measures 1 V peak to peak.

Anybody else get the same thing on their SDG-1025 in AM generation mode?

Everything is fine in the usual function generator sinewave mode. But once I flick over to AM generation, the measured carrier amplitude always turns out to be half of the LCD display setting for carrier amplitude (in peak-to-peak values).

I notice that I'm getting the same behaviour for a GW-instek AFG-2105, so I must be really missing something here heheheh.


Kenny.
« Last Edit: March 15, 2017, 09:22:50 am by SouthPark »
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 28142
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: Siglent SDG1025 Arbitrary/function generator under some tests
« Reply #109 on: March 15, 2017, 07:56:02 pm »
Hi everybody. I'm assuming that a lot of people here have a SDG-1025 arbitrary function generator, or similar model.

I got a small question about the Amplitude Modulation (AM) generation mode. I notice that - as soon as I switch from normal sine-wave generator mode to the AM generation mode, the AM carrier amplitude measures (using an oscilloscope with the SDG-1025 set to Hi-Z mode) to be half the peak-to-peak value of the LCD display setting.

So, if the LCD display says '2 V peak to peak' for the AM carrier amplitude (with ZERO modulation), the oscilloscope measures half of that, ie. measures 1 V peak to peak.

Anybody else get the same thing on their SDG-1025 in AM generation mode?

Everything is fine in the usual function generator sinewave mode. But once I flick over to AM generation, the measured carrier amplitude always turns out to be half of the LCD display setting for carrier amplitude (in peak-to-peak values).

I notice that I'm getting the same behaviour for a GW-instek AFG-2105, so I must be really missing something here heheheh.


Kenny.
If you've got the latest firmware installed I'm suspecting you have a termination mismatch.
Scope input set to 50 \$\Omega\$ ?
In which case if you change the AWG output to 50 \$\Omega\$ displayed amplitude should then be correct.
Avid Rabid Hobbyist
Siglent Youtube channel: https://www.youtube.com/@SiglentVideo/videos
 
The following users thanked this post: SouthPark

Offline SouthPark

  • Contributor
  • Posts: 28
  • Country: au
Re: Siglent SDG1025 Arbitrary/function generator under some tests
« Reply #110 on: March 15, 2017, 10:13:02 pm »
Quote
If you've got the latest firmware installed I'm suspecting you have a termination mismatch.
Scope input set to 50 \$\Omega\$ ?
In which case if you change the AWG output to 50 \$\Omega\$ displayed amplitude should then be correct.

Hi tautech! Thanks for the note about firmware. I haven't yet installed the latest. I will try that.

The oscilloscope's input impedance is 1 MOhm,  so I've set the SDG1025 load impedance to hi-Z to tell it that the load impedance is high impedance. Everything is perfect in regular sine-wave generation mode (ie. not in modulation/AM etc mode). So in regular function gen mode, if I enter 2 V p-p on the panel, the oscilloscope will measure 2 V p-p. But, once I push the MOD button for AM generation, the amplitude on the scope's display immediately becomes one-half of the front panel value.

Also, if I  set the SDG1025 load impedance setting to 50 Ohm, and if I add a 50 Ohm termination across the input of the scope, the same behaviour occurs. The measured amplitude of the sinusoid becomes halved as soon as I hit the MOD button (to get into the AM mode).

What really surprised me a lot was - I noticed this halving thing on a GW-instek AFG-2105. So I then decided to see what I get for a SDG1025, and I get the same behaviour, which I wasn't expecting at all.

So I thought I'd post my observations here to see if anybody else with a SDG1025 observes the same thing. And to see whether the issue is me not understanding the function generator's 'AM Carrier Amplitude' setting definition, or whether it is a firmware/software issue. Will try a firmware update!

There is a work-around for this halving issue though. ie keep the scope input at 1 MegaOhm (without 50 Ohm termination), and then deliberately set the SDG-1025 load impedance setting to '50 Ohm' (even though the actual load impedance is 1 MegaOhm). That at least makes the front panel value match the measurement value.

Thanks again Tautech!

Kenny
« Last Edit: March 16, 2017, 04:43:38 pm by SouthPark »
 

Offline SouthPark

  • Contributor
  • Posts: 28
  • Country: au
Re: Siglent SDG1025 Arbitrary/function generator under some tests
« Reply #111 on: March 16, 2017, 04:03:19 pm »
Hi again tautech! Back again. Just checked the firmware and noticed it matches the latest version. I might have a chat to a Siglent representative to discuss this observation about the SDG1025 measured AM carrier amplitude (with zero modulation) being half of the front panel LCD setting. I'll give an update on this thread if I get some info about this. Thanks once again!
« Last Edit: March 16, 2017, 04:42:58 pm by SouthPark »
 

Offline rf-loop

  • Super Contributor
  • ***
  • Posts: 4064
  • Country: fi
  • Born in Finland with DLL21 in hand
Re: Siglent SDG1025 Arbitrary/function generator under some tests
« Reply #112 on: March 16, 2017, 06:19:42 pm »
Hi again tautech! Back again. Just checked the firmware and noticed it matches the latest version. I might have a chat to a siglent representative to discuss this observation about the SDG-1025 AM carrier amplitude (with zero modulation) being half of the front panel LCD setting. Thanks once again.

This is how it works. If you compare to real RF generator. Typically  they keep carrier level.


SDG5000 (and if I remember right also SDG1000).
When you turn AM mod on carrier drop 6dBm. Independent of modulation sidebands levels.
If you select (no mod) sinewave level 1Vpp. Then turn AM on and depth 0% amplitude is 6dB less 500mVpp
Now if you change mod depth from 0 to 120% carrier level stay still constant and sidebands are how they are.
Just "know your equipment".

So, if you want 0 dBm  carrier and AM mod. Just set +6dBm level.
Now if you have 0% mod you get only carrier and level 0dBm
If you now set 100% mod. You have 0dBm carrier and two -6dBm side bands.
If you set 50% mod you have now 0dBm carrier and two -12dBm side bands

It may feel strange but it may have some explanation.
If we take oscilloscope  and look.

Lets think we have  set  sine wave and AM mod off and level 1Vpp.
We can see level is 1Vpp.  Now keep level setting.
Turn AM on but keep 0%. We can see 500mVpp. 
What we see if turn mod 100%.  We see now that modulated signal and level is 1Vpp.
It is made so that with 100% mod it do not exceed set level.
If user want set SDG maximal level what it can output. Then user turn AM on and 0% level is half. If he turn AM mod 100% output is now this max peak-peak. 
If it works different and it keep carrier level as user have set with AM off, what user think if it can not modulate.

Of course it is possible to do different but who then define how it need show exactly in all circumstances and how to message and prevent possible situations when mod depth need reject or prevent some level settings depending if user turn AM mod on..etc..  Specially when user use dBm level setting and if AM is on is it better show carrier level dBm - perhaps.  (now it show what is peak level IF mod depth is 100% (as it also show Vpp level IF  modulation is 100%)  Wrong or right for Function generator - but this is how it works.





« Last Edit: March 16, 2017, 06:29:41 pm by rf-loop »
I drive a LEC (low el. consumption) BEV car. Smoke exhaust pipes - go to museum. In Finland quite all electric power is made using nuclear, wind, solar and water.

Wises must compel the mad barbarians to stop their crimes against humanity. Where have the wises gone?
 
The following users thanked this post: SouthPark

Offline SouthPark

  • Contributor
  • Posts: 28
  • Country: au
Re: Siglent SDG1025 Arbitrary/function generator under some tests
« Reply #113 on: March 16, 2017, 07:41:57 pm »
This is how it works. If you compare to real RF generator. Typically  they keep carrier level.

Thanks RF Loop! Thanks for commenting about that AM feature. It is somewhat confusing - not your explanation - but rather the definition of AM Carrier Amplitude as the way that they seem to re-define it. They should really define it as exactly that ... AM Carrier Amplitude (bare carrier, no modulation). But at least we can easily compensate, by just doubling the value that we type into the front panel for AM mode.

Hopefully they define their setting properly (in future). It should be something like 'Maximum Peak-to-Peak AM Waveform Amplitude at 100 percent modulation depth' or 'Maximum Peak-to-Peak AM Signal Amplitude at 100 percent modulation depth'. I reckon it's technically wrong for them (all) to define it as 'AM Carrier Amplitude'. The carrier is the carrier. While the combination of carrier multiplied with a adequately DC-shifted message is an AM waveform - which is no longer 'carrier', unless the message drops to zero.

Thanks again RFLoop.

Kenny
« Last Edit: March 17, 2017, 10:24:29 pm by SouthPark »
 

Offline rf-loop

  • Super Contributor
  • ***
  • Posts: 4064
  • Country: fi
  • Born in Finland with DLL21 in hand
Re: Siglent SDG1025 Arbitrary/function generator under some tests
« Reply #114 on: March 17, 2017, 10:34:15 am »
Here is how it is checked. (SDG5k  but it works with this function very similar with SDG1k)



Set sinewave, level -4dBm
(AM modulation with 10Hz sinewave , carrier 3.7MHz)
SSA RBW 1Hz





Trace A  AM off  (carrier level -4dBm) (image)

Turn Mod on, AM, mod frequency 10Hz, MOD 0%, check carrier level is now -10dBm
Also checked that AM 0% same carrier level, in this case -10dBm (these checks not in image because SSA have only 4 traces)
After then Set AM 100%, 50% and 3.1% (image)
Trace B  AM on 100%  (carrier -10dBm, sideband -6dBc)
Trace C  AM on 50%    (carrier -10dBm, sideband -12dBc)
Trace D  AM on 3.1%   (carrier -10dBm, sideband -36dBc)





As first image but vertical scale 5dB/div and marker table off.

Note: Traces priority draw order when overlay: D, C, B, A   so carriers with AM on all same and there only D visible, B and C exactly behind and not visible.


« Last Edit: March 17, 2017, 10:46:00 am by rf-loop »
I drive a LEC (low el. consumption) BEV car. Smoke exhaust pipes - go to museum. In Finland quite all electric power is made using nuclear, wind, solar and water.

Wises must compel the mad barbarians to stop their crimes against humanity. Where have the wises gone?
 
The following users thanked this post: SouthPark

Offline SouthPark

  • Contributor
  • Posts: 28
  • Country: au
Re: Siglent SDG1025 Arbitrary/function generator under some tests
« Reply #115 on: March 17, 2017, 08:32:50 pm »
Very very nice results and work RFloop! Thanks for showing those nice measurements. I didn't think of using the spectrum to check on the carrier level like that. It's excellent as it shows in a very clear way that they really dropped the carrier power by 6 dBm when the AM mode is turned on - producing half the voltage setting entered by the user on the front panel.

The easy fix would be for the Siglent team to create a firmware update so that the SDG1025 voltage (in AM generation mode only) becomes TWICE the value that is being "currently" generated (in both Hi-Z and 50 Ohm load setting modes).

I think the issue is that the manufacturers are riding on the same train of thought, and incorrectly saying that 'carrier  Ac.cos(wc.t)' is the same as 'modulated carrier (ie. AM signal)'. Once AM modulation occurs, the waveform is no longer 'the carrier Ac.cos(wc.t)'. Instead, 'an AM signal' results. Their erroneous definition of AM carrier amplitude appears to be "maximum peak to peak amplitude of AM signal at 100 percent modulation depth" or "max peak to peak amplitude of modulated carrier at 100 percent mod depth".

Thanks again for putting up those measurements RFloop. Appreciated.
« Last Edit: March 19, 2017, 07:06:03 pm by SouthPark »
 

Offline SouthPark

  • Contributor
  • Posts: 28
  • Country: au
Re: Siglent SDG1025 Arbitrary/function generator under some tests
« Reply #116 on: March 29, 2017, 09:38:01 pm »
Hello RFloop! A Siglent representative very kindly passed on my question to the Siglent factory about the voltage halving of the unmodulated carrier (in AM generation mode). It was kindly indicated by the Siglent representative that the factory will be working on the fix. The fix will NOT be in the upcoming firmware, since that one is in pre-release testing stages. But the fix will likely be in the firmware that comes after that one. Siglent seem to be top notch in assessing the questions and doing something about issues where appropriate. My rating of them  ---->   :-+
« Last Edit: March 30, 2017, 01:05:57 am by SouthPark »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf