FeelTech FY3224S 24MHz 2-Channel DDS AW Function Signal Generator

[Aldo22]

I know I'm a little late to the party, but the FY3224S is still being sold and is still the cheapest 2 channel AWG available as far as I know.
I got it yesterday.

Board version: 3.2
PSU version: 1.32
Firmware version: 4.1

So far everything works.
But I have some jitter at the square wave.
Strangely not at 6MHz, but for example at 5MHz and around.
I see two vertical traces (See fy3200s_5MHz_sqr.jpg).

Is this a known problem or is my device faulty?
There is also a possibility that it is partly due to the cheap Zeeweii DSO2512G, but normally it is still stable in these frequency ranges.

Does anyone know this phenomenon?

Thank you!

[Aldo22]

I'll answer my own question:
here you can see that it is not due to the oscilloscope. This jitter seems to be typical for the FY3200S.
The images there look partly worse than on my device.
Especially 5V/5MHz looks more like a square wave on my version. There was apparently a small improvement.

[Pygmalion]

So far everything works.
But I have some jitter at the square wave.
Strangely not at 6MHz, but for example at 5MHz and around.
I see two vertical traces (See fy3200s_5MHz_sqr.jpg).

Does anyone know this phenomenon?

This reminds me irresistibly of the Fourier series of a square signal. The frequency of the oscillator is fixed at 24 MHz, and the final signal is probably the sum of a series of sine waves with frequencies, multiples of which is 24 kHz. It will therefore be much more difficult to fit a 5MHz square wave than a 6MHz square wave into series of 6MHz, 12MHz, 18MHz and 24MHz.

See https://en.wikipedia.org/wiki/Fourier_series

[Aldo22]

This reminds me irresistibly of the Fourier series of a square signal. The frequency of the oscillator is fixed at 24 MHz, and the final signal is probably the sum of a series of sine waves with frequencies, multiples of which is 24 kHz. It will therefore be much more difficult to fit a 5MHz square wave than a 6MHz square wave into series of 6MHz, 12MHz, 18MHz and 24MHz.

Yes.
The good thing is that there are "sweet spots" at certain frequencies (multiples).
With an appropriate waveform, the device can also generate stable 36MHz or 48 MHz, which is twice the specified bandwidth.
Not particularly nice (forget the amplitude), but stable.

[Aldo22]

I know the FY3200s isn't the latest and greatest anymore, but it's still the cheapest AWG and it's not that bad.

However, the user interface is a bit annoying.
Fortunately, you can control the device remotely with a PC.
Since I couldn't find a GUI application that also runs on Linux and Raspberry Pi 4, I wrote one myself.
I have added function to function and now I'm almost finished. ;-)

Features are (besides the obvious).

- 50Ohm load resistor mode
- Attenuator mode (dB)
- Amplitude limitation for sensitive devices (e.g. TinySA).
- Amplitude correction table (for adjusting e.g. oscilloscopes).
- Display of Vrms for standard waveforms and dBm at 50 ohms.
- Software sweep with amplitude sweep.
- Unlimited named presets.
. Waveform upload (15 seconds)
- Waveform conversion from bitmaps (PNG etc.) with single and double scan for fun shapes.

I still have some tweaking to do.
When it's done I can share the source code if anyone is interested.

[sims070]

Hey, are you still recommending to get this function generator. It'll be my first. Thanks in advance

[Aldo22]

Hey, are you still recommending to get this function generator. It'll be my first. Thanks in advance

It's just the cheapest 2Ch AWG as far as I know.
More expensive ones are better (e.g. FY6900 etc. etc.).
The user interface is not very comfortable, but it is a "proper" function generator.  2Ch., 20Vpp, offset, phase, sweep, pulse, duty cycle, frequency counter etc...
It is OK for the price as a first AWG.

[nikbry]

Just got my FY3200S and it's working fine, except when I connect the USB cable to my MacMini (To run the Feeltech software in a Win10 virtual machine) even with the signal generator turned off, the Mac just crashes every time.
I did the grounding mod, don't think that's related though.

Any thoughts or suggestions would be greatly appreciated. Thanks.

[nikbry]

Just got my FY3200S and it's working fine, except when I connect the USB cable to my MacMini (To run the Feeltech software in a Win10 virtual machine) even with the signal generator turned off, the Mac just crashes every time.
I did the grounding mod, don't think that's related though.

Any thoughts or suggestions would be greatly appreciated. Thanks.

The problem was caused by outdated CH340 drivers. In case someone runs into the same problem, you can download updated drivers and read up on the procedure to delete the old ones and install the new ones here: https://learn.sparkfun.com/tutorials/how-to-install-ch340-drivers/all

[nikbry]

I found some time to fiddle with my new sig generator and found a strange behavior that makes me wonder if there's a calibration issue with it or if this is just what's to be expected from such an affordable device.
When generating a 5Vpp sine wave, the voltage remains at 5Vpp up to 1kHz and then reaches 4.76Vpp at 10kHz, and 4.65Vpp at 100kHz and 1Mhz. This happens to a lesser extent with the square wave 4.9Vpp at 1Mhz.

I tried the "calibration" button press sequence, but my unit runs firmware 4.1 and that doesn't seem to work.

Any suggestions or explanations to what I'm seeing would be greatly appreciated.

[Aldo22]

That should not be.

How did you measure?
Direct BNC connection or probe (1x, 10x?).
Bandwidth of the oscilloscope?
Any filters active?

I have not changed the amplitude in the attached screenshots (FY3224S, Hantek DSO2000).

[nikbry]

That should not be.

How did you measure?
Direct BNC connection or probe (1x, 10x?).
Bandwidth of the oscilloscope?
Any filters active?

I have not changed the amplitude in the attached screenshots (FY3224S, Hantek DSO2000).

Thanks for your quick reply.

I'm using the BNC to alligator cable at came with the scope and connecting the scope probes directly to that, with 10x settings.
Have setup my two scopes and both measure the same: OWON VDS1022i (25MHz) and Zotek ZT703s (40-50Mhz)
With filters you mean sampling mode averaging? Turned off

[Aldo22]

I don't know.
I can only say that with my FY3224S the amplitude is completely stable up to at least 5MHz.
So it is not due to the FY3224S in general.
Can you try a better/shorter connection? (Coax, BNC).

[nikbry]

OK.

Does the button push sequence calibration work on your device? Mine has FW 4.1 like yours and it never shows a calibration value.

[Aldo22]

Does the button push sequence calibration work on your device? Mine has FW 4.1 like yours and it never shows a calibration value.

Isn't that for frequency deviations?
I have never tried that.
I don't think it's for the amplitude, is it?

[nikbry]

Does the button push sequence calibration work on your device? Mine has FW 4.1 like yours and it never shows a calibration value.

Isn't that for frequency deviations?
I have never tried that.
I don't think it's for the amplitude, is it?

I tried again and it does work, and yes it is for frequency adjustments only. Was hoping by saving the value it may reset something but no luck.

Anyway, I'm not impressed so far

[nikbry]

I just retested with the probe set to 1x and what do you know!? I measure 5Vpp all the way up to 2MHz...

With my ignorance I get the feeling that I'm missing something fundamental here.

Hope someone can explain.

[Aldo22]

Anyway, I'm not impressed so far

I don't know what you mean.
I don't think your problem is with the FY3224S, unless your unit is broken.
The FY3224S is quite widely used and does its job.
Of course, it's not “impressive”, you'd have to pay more for that.

[nikbry]

Anyway, I'm not impressed so far

I don't know what you mean.
I don't think your problem is with the FY3224S, unless your unit is broken.
The FY3224S is quite widely used and does its job.
Of course, it's not “impressive”, you'd have to pay more for that.

I do have the XR2206 kit signal generator, and once I adjust the amplitude it stays solid ( 1-100kHz) over the range the FY3224S does not. Same oscilloscopes, same probes...

Anyway, thank you for your critical comment.

P.S.: just realized using the expression "I'm not impressed" may lead to confusion. I meant it the colloquial way expressing unhappiness.

[Aldo22]

I do have the XR2206 kit signal generator, and once I adjust the amplitude it stays solid ( 1-100kHz) over the range the FY3224S does not. Same oscilloscopes, same probes...

If you are sure that your FY3224S is defective, then I would report it to the dealer. It's still new, isn't it?
The amplitude should stay stable, as you can see from my screenshots above.

[nikbry]

Thanks for the suggestion. I am afraid it's too late for that, since I did the grounding mod.

Still puzzled why the 1x probe setting measures the 5Vpp up to almost 2MHz.

[Aldo22]

Still puzzled why the 1x probe setting measures the 5Vpp up to almost 2MHz.

Whatever that may be, it does not indicate to me that the FY3224S is responsible.

Cables, oscilloscope, settings...

[nikbry]

Still puzzled why the 1x probe setting measures the 5Vpp up to almost 2MHz.

Whatever that may be, it does not indicate to me that the FY3224S is responsible.

Cables, oscilloscope, settings...

Thanks Aldo for bearing with me.

Since I had nothing better at hand, I connected the two BNC to Crocodile cables that came with the Feeltech together, to connect the generator to the scope. With this arrangement Vpp is 5V up to 5Mhz, then it starts to rise peaking at 5.9Vpp at 15MHz to then decay to 3.7Vpp at 24MHz.

I'm not sure what to make of it all. First thought would be that all three of the probes that came with my Owon and Zotek have a similar attenuation error when used above 5kHz with the 10x setting. But at that low of a frequency this shouldn't be the case???

Anyway, interesting little problems to deal with.

[Atlan]

Use bnc cable without crocodile

[nikbry]

Use bnc cable without crocodile

Thanks for the tip,  I do not have a BNC to BNC one at hand.

I did some further testing and have come to the conclusion that the problem is not due to the signal generator. With one of my oscilloscopes (Owon VDS1022i) I did a scope and probe calibration and now the Vpp I measure stays at 5Vpp through the entire frequency range up to 21MHz. 

So sorry everyone for my rant and confusion.

Special thanks to Aldo, and I stand corrected about my unhappiness with the Feeltech.