Author Topic: FeelTech FY3224S 24MHz 2-Channel DDS AW Function Signal Generator  (Read 259880 times)

0 Members and 2 Guests are viewing this topic.

Offline pdereus

  • Contributor
  • Posts: 6
  • Country: nl
Re: FeelTech FY3224S 24MHz 2-Channel DDS AW Function Signal Generator
« Reply #25 on: July 14, 2015, 09:38:03 pm »
Hi GBY,

I have to disappoint you here, as I did analyze the protocol at the software level (so at the Windows driver). I did not do any hardware level sniffing for this. I guess it's pretty easy to find out though: just put a scope to the pins and fire up the USB based software.
Fire commands that don't get a response (like changing the frequency) and you will be able to distinguish TxD from RxD (also with a multi meter  :bullshit:)

Paul
 

Offline pascal_sweden

  • Super Contributor
  • ***
  • Posts: 1519
  • Country: no
Re: FeelTech FY3224S 24MHz 2-Channel DDS AW Function Signal Generator
« Reply #26 on: July 24, 2015, 11:47:37 pm »
Are the 2 outputs completely floating (negative is not ground) and completely separate from each other (negatives not common)?

How easy is it to use arbitrary waveforms on the FeelTech FY3224S?
 

Offline dadler

  • Supporter
  • ****
  • Posts: 848
  • Country: us
Re: FeelTech FY3224S 24MHz 2-Channel DDS AW Function Signal Generator
« Reply #27 on: July 24, 2015, 11:56:15 pm »
I think all generators share a common potential reference between the channels (I am not aware of any multi-channel generators that have entirely isolated outputs). However, since  (I think?) these units are powered by a wall-wart (or even a battery, if you wish), they must be floating relative to earth ground (as long as not connected via USB-usb may earth reference the device).

Some of these cheap units in this form factor have an internal power supply, and I do not know if those models are floating.
« Last Edit: July 25, 2015, 12:02:25 am by dadler »
 

Offline gby

  • Regular Contributor
  • *
  • Posts: 184
  • Country: us
Re: FeelTech FY3224S 24MHz 2-Channel DDS AW Function Signal Generator
« Reply #28 on: July 25, 2015, 01:40:45 pm »
The FY3200S series of generators have an internal supply and do not use a wall wart type supply.  The power cord is only 2 wire and so the instrument does not get third prong ground/Protective Earth.  This means that by definition the entire instrument floats with respect to earth.  Like the previous poster listed all of the returns/negative on the BNC's for most instruments are shorted together.  This includes this generator...to be sure I just used an Ohm meter and the BNC returns/negative terminals all are shorted together.

I would think that it is safe to float this generator 40-50 volts with respect to earth ground.  Above that you will need to be very, very careful since you could get badly shocked by touching the buttons/knob and would certainly get shocked touching the return/negative of the BNC's.  If you float the generator I would NOT connect the USB port since that port is not isolated.  I would go with my idea of using a Bluetooth to serial module mounted inside the generator and then only adjusting/setting the generator via the PC software while the generator is floated.

gby
 

Offline gby

  • Regular Contributor
  • *
  • Posts: 184
  • Country: us
Re: FeelTech FY3224S 24MHz 2-Channel DDS AW Function Signal Generator
« Reply #29 on: July 25, 2015, 01:50:42 pm »
pascal_sweden,

I find using the ver 1.3 software makes doing arbitrary waveforms very easy.  The utility's built graphical screen to draw/create waveforms is functional but can not really be used for point by point precision.  It is quite functional for a couple of standard waveform types but to get a precise waveform you need to calculate the points or use a data record recorded from a scope.  I made a spreadsheet to make the 2048 point waveforms.  I then just cut and pasted the points from the spreadsheet into the utility's "Text Window" tab and downloaded.  This "Text Window" tab even has a button to plot the points to check that it looks right before downloading.  Easy as can be.

gby
 

Offline pascal_sweden

  • Super Contributor
  • ***
  • Posts: 1519
  • Country: no
Re: FeelTech FY3224S 24MHz 2-Channel DDS AW Function Signal Generator
« Reply #30 on: July 26, 2015, 06:58:57 pm »
What is the benefit of the MHS-5200P model with broadband-type option?

http://www.benl.ebay.be/itm/6-25MHZ-Power-Amplifier-Dual-Channel-DDS-Signal-Generator-NC-Arbitrary-Waveform-/281749051899?pt=LH_DefaultDomain_3&var=&hash=item41998d59fb

I am confused about the difference between MHS-5200 (normal) and MHS-5200P (broadband-type).
 

Offline dadler

  • Supporter
  • ****
  • Posts: 848
  • Country: us
Re: FeelTech FY3224S 24MHz 2-Channel DDS AW Function Signal Generator
« Reply #31 on: July 26, 2015, 08:54:04 pm »
What is the benefit of the MHS-5200P model with broadband-type option?

http://www.benl.ebay.be/itm/6-25MHZ-Power-Amplifier-Dual-Channel-DDS-Signal-Generator-NC-Arbitrary-Waveform-/281749051899?pt=LH_DefaultDomain_3&var=&hash=item41998d59fb

I am confused about the difference between MHS-5200 (normal) and MHS-5200P (broadband-type).

These P models seem to have power amplifiers at the outputs, to increase the power/current they can supply. The non-broadband model (claims it) can do 1A @ 80kHz, whereas the "broadband model" can do 300mA@5MHz. So it matters what you want to drive and at what frequency.
« Last Edit: July 26, 2015, 08:57:35 pm by dadler »
 

Offline dadler

  • Supporter
  • ****
  • Posts: 848
  • Country: us
Re: FeelTech FY3224S 24MHz 2-Channel DDS AW Function Signal Generator
« Reply #32 on: July 26, 2015, 08:56:38 pm »
If you float the generator I would NOT connect the USB port since that port is not isolated.  I would go with my idea of using a Bluetooth to serial module mounted inside the generator and then only adjusting/setting the generator via the PC software while the generator is floated.

gby

I use this USB isolator in these sorts of situations, and it works quite well:

https://www.adafruit.com/product/2107
« Last Edit: July 26, 2015, 08:58:31 pm by dadler »
 

Offline pascal_sweden

  • Super Contributor
  • ***
  • Posts: 1519
  • Country: no
Re: FeelTech FY3224S 24MHz 2-Channel DDS AW Function Signal Generator
« Reply #33 on: July 26, 2015, 09:06:07 pm »
What is the benefit of the MHS-5200P model with broadband-type option?

http://www.benl.ebay.be/itm/6-25MHZ-Power-Amplifier-Dual-Channel-DDS-Signal-Generator-NC-Arbitrary-Waveform-/281749051899?pt=LH_DefaultDomain_3&var=&hash=item41998d59fb

I am confused about the difference between MHS-5200 (normal) and MHS-5200P (broadband-type).

These P models seem to have power amplifiers at the outputs, to increase the power/current they can supply. The non-broadband model (claims it) can do 1A @ 80kHz, whereas the "broadband model" can do 300mA@5MHz. So it matters what you want to drive and at what frequency.

Can you highlight the application area for these frequency power amplifiers?
 

Offline pascal_sweden

  • Super Contributor
  • ***
  • Posts: 1519
  • Country: no
Re: FeelTech FY3224S 24MHz 2-Channel DDS AW Function Signal Generator
« Reply #34 on: August 06, 2015, 08:09:06 am »
Can someone do a similar teardown on the FeelTech FY3224S, and identify if the same issues are present regarding harmonic distortion, or if the output stage design is better, and results in better waveform signals?

https://www.eevblog.com/forum/testgear/mhs-5200a-function-generator-teardown-review-reverse-engineering/

In this video they reverse engineered the analog front end of the MHS-5200A to track down the source of the harmonic distortion that happens at higher output amplitudes and higher frequencies.

They identified that the supply voltage to the final output stage (an AD812 dual op-amp) is not high enough (backed by the datasheet of the AD812) and that this is causing clipping/distortion.

When the supply voltage is increased from +9V,-9V to +15V,-15V, there is still some harmonic distortion, but the output waveform is looking better.

Maybe the FeelTech FY3224S and the MHS-5200A are the same underneath, or very similar. Just that on its own would be good to know.
« Last Edit: August 06, 2015, 08:12:53 am by pascal_sweden »
 

Offline gby

  • Regular Contributor
  • *
  • Posts: 184
  • Country: us
Re: FeelTech FY3224S 24MHz 2-Channel DDS AW Function Signal Generator
« Reply #35 on: August 07, 2015, 12:15:08 am »
Pascal_Sweden,

I know that the supply is not 9V on the FS3200xx.  The internal pictures clearly show the power supply labeled +/- 12V for the analog supplies.  Also, when open circuited the output easily goes to 20 Vp-p or +/-10V which means that the supply rails on the analog output stage must be > than 10V.

Based on my measurements it is pretty clear that the output analog buffer is current limited at just about 70 mA 0-p.  Into a 50 Ohm load this max current limits the output to about 7.5 Vp-p matching my measurements.  I highly suspect they limited the current to only 70 mA to limit the power consumed and dissipated in the analog output buffer.  Could also just be a limit of the analog buffer IC they chose.

I have thought about changing the output buffer to have more output current and more capability.  That probably would require changing the buffer IC and adding more heat sink.  We might get lucky that the analog buffer IC under the heat sink can do more current with more heat sinking but until someone takes the unit completely apart, removes the heat sink we don't know what IC they are using.

For a cheap generator the harmonic distortion is actually pretty low below 5Vp-p amplitude into 50 Ohm.  At 100 kHz I measured only 0.56% THD at 20Vp-p output with an open circuited load.

gby
 

Offline pascal_sweden

  • Super Contributor
  • ***
  • Posts: 1519
  • Country: no
Re: FeelTech FY3224S 24MHz 2-Channel DDS AW Function Signal Generator
« Reply #36 on: August 07, 2015, 12:25:57 am »
A friend of me has a Rigol SDG1025. He paid about 300 EUR for it!

I was showing him the link to the FeelTech FY3224S signal generator, but he was not impressed about the build quality, compared with his nice looking Rigol SDG1025. Of course I fully understand why :)

As the design could not convince my friend, I told that the specs seem quite good, given that it only costs 100 EUR.

Then I was thinking. Is the Rigol SDG1025 really that much better? And how much better?
Here is the data sheet from the website:
http://www.siglent.com/2014EnglishWebsite/Documents/Datasheet/SDG1000_DataSheet_DS02010-E07A.pdf

I don't have the full specifications from the FeelTech FY3224S at hand, so can't do the comparison.
And then who knows if the specs from FeelTech really tell the truth.

But if we assume that the specs from Siglent are correct, then it should be possible to do a comparison between the two tables, and summarize the weaknesses/strengths/equalities among the Siglent SDG1025 and the FeelTech FY3224S.

Would you mind taking up this homework for the community? :)
 

Offline dadler

  • Supporter
  • ****
  • Posts: 848
  • Country: us
Re: FeelTech FY3224S 24MHz 2-Channel DDS AW Function Signal Generator
« Reply #37 on: August 07, 2015, 02:12:36 am »
Rigol SDG1025


Huh???
 

Offline pascal_sweden

  • Super Contributor
  • ***
  • Posts: 1519
  • Country: no
Re: FeelTech FY3224S 24MHz 2-Channel DDS AW Function Signal Generator
« Reply #38 on: August 07, 2015, 02:27:58 am »
Yes, both are around 25MHz.

To know how good the Siglent is, and how bad the FeelTech is, relatively :)
 

Offline jcrubin

  • Frequent Contributor
  • **
  • Posts: 322
  • Country: us
    • Technocoma
Re: FeelTech FY3224S 24MHz 2-Channel DDS AW Function Signal Generator
« Reply #39 on: August 09, 2015, 09:28:49 pm »
I have this temptation to write a perl module to interface with this and put it on CPAN.  If there would be a practicality for a command line interface to the unit on a computer or over a network.......
CVM CATAPVLTAE PROSCRIPTAE ERVNT TVM SOLI PROSCRIPTI CATAPVLTAS HABEBVNT

https://www.youtube.com/c/jordanrubin6502
 

Offline jcrubin

  • Frequent Contributor
  • **
  • Posts: 322
  • Country: us
    • Technocoma
Re: FeelTech FY3224S 24MHz 2-Channel DDS AW Function Signal Generator
« Reply #40 on: August 17, 2015, 12:59:51 am »
I started making a few movies doing some basic tests with this device.  Ive noticed that for the frequency counter the reliable reading seems to be just above 1.5 volts at least at 455khz using a sine wave.



Also, I was curious, as it was mentioned that the device has an amplitude modulated waveform.  I may be a bit confused here, however I ran the FY3200 at 455KHZ 2V AM waveform in channel 2 and my RF generator using the same setting in channel 1.  See the video and perhaps explain to me what im missing.   Im thinking that the carrier is 455KHZ but the signal injected into the carrier is an entirely different frequency all together


« Last Edit: August 18, 2015, 12:38:49 am by jcrubin »
CVM CATAPVLTAE PROSCRIPTAE ERVNT TVM SOLI PROSCRIPTI CATAPVLTAS HABEBVNT

https://www.youtube.com/c/jordanrubin6502
 

Offline quantalume

  • Supporter
  • ****
  • Posts: 26
  • Country: 00
Re: FeelTech FY3224S 24MHz 2-Channel DDS AW Function Signal Generator
« Reply #41 on: August 17, 2015, 05:35:24 pm »
The lower trace on your 'scope looks like a properly-modulated waveform. What is the modulating frequency? I'm guessing it's around 57 KHz as there are about 8 carrier cycles inside of each modulation cycle.

What you're seeing in the top trace when you decrease the sweep frequency is just an artifact of sweep frequency vs. carrier, not actual modulation.

EDIT: On second thought, the top waveform might actually be AM, just not 100% modulation like the bottom.  What was the initial horizontal sweep speed before you changed it in the video?  It looks like the modulation frequency on the DDS generator is the same as the carrier frequency on the analog generator.  What do you have the other (SF?) frequency set to on the DDS?
« Last Edit: August 17, 2015, 08:02:45 pm by quantalume »
 

Offline jcrubin

  • Frequent Contributor
  • **
  • Posts: 322
  • Country: us
    • Technocoma
Re: FeelTech FY3224S 24MHz 2-Channel DDS AW Function Signal Generator
« Reply #42 on: August 18, 2015, 12:39:52 am »
Im wondering, has anyone used the software to define a waveform that replicates the injection of a waveform onto a modulated AM carrier?
CVM CATAPVLTAE PROSCRIPTAE ERVNT TVM SOLI PROSCRIPTI CATAPVLTAS HABEBVNT

https://www.youtube.com/c/jordanrubin6502
 

Offline jcrubin

  • Frequent Contributor
  • **
  • Posts: 322
  • Country: us
    • Technocoma
Re: FeelTech FY3224S 24MHz 2-Channel DDS AW Function Signal Generator
« Reply #43 on: August 18, 2015, 12:57:49 am »
The lower trace on your 'scope looks like a properly-modulated waveform. What is the modulating frequency? I'm guessing it's around 57 KHz as there are about 8 carrier cycles inside of each modulation cycle.

What you're seeing in the top trace when you decrease the sweep frequency is just an artifact of sweep frequency vs. carrier, not actual modulation.

EDIT: On second thought, the top waveform might actually be AM, just not 100% modulation like the bottom.  What was the initial horizontal sweep speed before you changed it in the video?  It looks like the modulation frequency on the DDS generator is the same as the carrier frequency on the analog generator.  What do you have the other (SF?) frequency set to on the DDS?

Yes the top is from my RF generator which is AM.   The frequency knob was set to 455khz and checked on the frequency counter, the voltage was set so both channels would have the same amplitide.  As for the injected signal produced by the RF generator, I cant be sure what the frequency is as the amplitude and frequency variation are set by pots with a screw driver.  I will check to see if I can remove the 455khz from the output and see what the injected frequency is

UPDATE:


It seems that as per the manual, when set to internal modulation, the audio input jack will have the modulated frequency on it for measurement and setting.  I will retest tomorrow and provide the figures

http://site.ldbepurchase.com/Manuals/TSG17.pdf
« Last Edit: August 18, 2015, 01:05:46 am by jcrubin »
CVM CATAPVLTAE PROSCRIPTAE ERVNT TVM SOLI PROSCRIPTI CATAPVLTAS HABEBVNT

https://www.youtube.com/c/jordanrubin6502
 

Offline jcrubin

  • Frequent Contributor
  • **
  • Posts: 322
  • Country: us
    • Technocoma
Re: FeelTech FY3224S 24MHz 2-Channel DDS AW Function Signal Generator
« Reply #44 on: August 19, 2015, 12:41:46 am »
I was able to use the audio input jack to set the AF to 500hz.  The peak to peak amplitude showed about 6 volts on the auto output.  The mod-level POT had little effect on the amplitude.
CVM CATAPVLTAE PROSCRIPTAE ERVNT TVM SOLI PROSCRIPTI CATAPVLTAS HABEBVNT

https://www.youtube.com/c/jordanrubin6502
 

Offline gby

  • Regular Contributor
  • *
  • Posts: 184
  • Country: us
Re: FeelTech FY3224S 24MHz 2-Channel DDS AW Function Signal Generator
« Reply #45 on: August 24, 2015, 07:35:50 pm »
I have been working with my FY3206S generator some more and noticed that it does not change frequency smoothly.  When using the knob or commanding over the USB connection when you ask for a new frequency the generator output pauses/holds the voltage and then continues on with the new frequency.  It seems to take about 1.35 mSec for the frequency to update and during that time the output values seem to be held at DC values.  See attached scope photo "Step Changes To Frequency Detail.png" for an example of what I am seeing.

If you use the built in sweep function the frequency changes smoothly as you would expect.  Unfortunately, I am trying to make 0 and 90 Degree synchronized sine waves sweeping up in frequency and the FY3200 built in sweep mode only works on CH1 and CH2 stays fixed/not sweeping.  In sweep mode it does not seem to matter matter whether the System menu setting "Freq CH1=CH2" is set or not CH2 does not sweep.

I did use a Python script to command the frequency to step every 12 mSec and the result is that I get 0, 90 sines with the frequency sweeping up but with a 2.7 mSec glitch at each step (one 1.35 mSec of CH1 and one 1.35 mSec for CH2).  So, whatever mechanism the FY3200 uses to change set point frequency it does not do it in a step manner.  It seems to set frequencies to 0 and then to the new set point.  See attached scope photo "Sine Ch2 Sync Ch1 Serial Sweep Freq.png" to see the Python script stepping the frequency with two channels synchronized in frequency.

Does anyone know if more expensive $300 generators such as Rigol DG1022/Siglent SDG1025 or the next step up $600 Rigol DG1032Z/Siglent SDG5082 would be able to sweep two channels while staying locked at 0, 90 Degree with no glitches???  That is, the frequency changes smoothly and the 90 phase relationship stays in synch at each frequency change?

gby
 

Offline SaabFAN

  • Frequent Contributor
  • **
  • Posts: 733
  • Country: de
Re: FeelTech FY3224S 24MHz 2-Channel DDS AW Function Signal Generator
« Reply #46 on: August 24, 2015, 10:35:46 pm »
I just checked for this behavior on my MHS-5200A (60 - 80€ 6-25Mhz 8bit Signal-Gen from ebay) and got an immediate response to changing the frequency as you can see in the attached screenshots.
How did you measure this btw.? I used Roll-Mode and then switched between 1 and 2 khz on the generator.

So if you can live with the distortion the MHS-5200A introduces into the signal at higher amplitudes, you don't necessarily have to pay several hundred bucks^^

Offline usagi

  • Frequent Contributor
  • **
  • Posts: 368
  • Country: us
Re: FeelTech FY3224S 24MHz 2-Channel DDS AW Function Signal Generator
« Reply #47 on: August 24, 2015, 11:29:14 pm »
have anyone else confirmed the mhs-2300a bug does/does not exist in the FeelTech FY3224S or MHS-5200A ?

set CH1 to 10khz sine, CH2 to 10khz sine. Phase should be perfectly locked, and lissajous should be stable. Both are integer ratio of the master clock.

set CH1 to 10khz sine, CH2 to 20khz sine. Phase should be perfectly locked, and lissajous should be stable. Both are integer ratio of the master clock. But in mhs-2300a, CH2 has phase shift errors and CH2 is not stable. <- bug!!
« Last Edit: August 24, 2015, 11:31:16 pm by usagi »
 

Offline gby

  • Regular Contributor
  • *
  • Posts: 184
  • Country: us
Re: FeelTech FY3224S 24MHz 2-Channel DDS AW Function Signal Generator
« Reply #48 on: August 25, 2015, 03:17:59 am »
SaabFAN, to measure the frequency change from the knob or from the PC adjustment I just set the scope to take lots of points and then randomly hit the stop button on the scope while spinning the knob/stepping the frequency via a program.  I then zoomed into the shot to find the frequency change moment.  Nasty bug/deficiency to be sure.

Usagi, when CH1 and CH2 on the FY3200 are set to be locked they are frequency and phased locked.  If I set the phase to be 90 Deg apart they are locked 90 Degree apart.  Since you mentioned lissajous I just set up the scope with 1 kHz, equal amplitudes and got a perfect circle with the scope set to X versus Y.  I then set slightly unequal amplitudes and the oval from the un-equal amplitudes did not precess at all.  Whatever the bug you mention that the MHS-2300a has the FY3200 does not have that problem.
« Last Edit: August 25, 2015, 04:34:36 am by gby »
 

Offline usagi

  • Frequent Contributor
  • **
  • Posts: 368
  • Country: us
Re: FeelTech FY3224S 24MHz 2-Channel DDS AW Function Signal Generator
« Reply #49 on: August 25, 2015, 08:24:08 am »
Usagi, when CH1 and CH2 on the FY3200 are set to be locked they are frequency and phased locked.  If I set the phase to be 90 Deg apart they are locked 90 Degree apart.  Since you mentioned lissajous I just set up the scope with 1 kHz, equal amplitudes and got a perfect circle with the scope set to X versus Y.  I then set slightly unequal amplitudes and the oval from the un-equal amplitudes did not precess at all.  Whatever the bug you mention that the MHS-2300a has the FY3200 does not have that problem.

try CH1 1khz CH2 2khz and see if it precesses.


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf