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

[YvesMalherbe]

Hello,

I have seen that some of us have developped a control program for the Feeltech FY3200 generator.

This is also something that I have done in VB .NET for Windows, as I was not so satisfied with the program provided by Feeltech.

My program supports remote controlling of CH1/CH2 and internal memories.  It is also able to upload custom arbitrary waveforms.

I am glad to share it to the community.  You can download it following this link :
https://drive.google.com/drive/folders/1hqqHqvfk1FKhzYSjJ-S6_SUTr_zx0fhp?usp=sharing

The ZIP file behind the link above contains a PDF document with all needed instructions.

Greetings,


Yves

[marco_c]

Yves,

A quick technical question about your software. When I use it directly connected via USB there is no issue and it all seems to work fine.

I am trying to replace the USB with a Bluetooth connection (HC-05 bluetooth module) in the Feeltech unit that, when paired, looks like a com port on the Windows computer. When an application on the PC tries to connect to the COM port the paired BT connection starts up and establishes a serial link to the device. At the BT receiver you can see that a blinking LED transitions to steady ON when this happens.

If I connect to the FY3224S using a terminal program I can send protocol commands (like bw[n]) and the unit responds on the display with a changed wave form. When I use your software and open the com port it sends 'a' but then reports that there is no device attached. Do you know what might be the problem or if there is something I can do to work out what the issue may be?

Marco

UPDATE: I have got this working but I have to ignore all the error messages from the application. I set ComPortAutoOpen and StartWithoutConnexion to be enabled. This generates the error messages on startup about the bad port but works when I then set parameters. I suspect that the BT startup delay is not tolerated by the software - is that probable? If you can make the source code available I can test what is going on and correct this.

[YvesMalherbe]

Hi Marco,

Your suspicions are likely right... I did not figure out that someone could connect the appliance otherwise than with a cable, so I hardcoded the delay of the COM port to 1200 milliseconds    Very probably this timeout is too short when connecting via your BT device.

I introduced two new timeout parameters (in GeneralParameters.txt file), which I gave the default value 1200

• ComPortReadTimeout=1200
• ComPortWriteTimeout=1200

By changing this to other larger values, the situation will probalbly be solved.

Just download the new program by using the same link I gave on september 2.  I also attached the source code in the new download, should you need to modify something else.

Greetings,

[marco_c]

Thanks for the source code.

The timeouts did not fix the problem but I managed to change the OpenComPort Try/Catch statement so that it at least works on my setup. For the record the changed code is below.

What seems to happen is that the FY unit does not respond to the 'a' message when it is on BT (this works on on cable). The mod to the code avoids the software thinking it is in error, but it is probably not recommended for wider implementation.

Code: [Select]

    Try
      ' Configure COM port parameters
      sp.PortName = comPortName
      sp.BaudRate = 9600
      sp.Parity = Parity.None
      sp.DataBits = 8
      sp.StopBits = StopBits.One
      sp.NewLine = vbLf
      sp.ReadTimeout = prmComPortReadTimeout
      sp.WriteTimeout = prmComPortWriteTimeout
      ' Check if Feeltech 32xx is answering on the selected port
      sp.Open()
      SendFeeltech("a")
      answerFromFY32 = sp.ReadLine

    Catch ex As Exception
      answerFromFY32 = "Unknown"

    Finally
      LBMainStatusInfo.Text = answerFromFY32 & " on " & comPortName & " - Buffers (R/W) : " & sp.ReadBufferSize.ToString & "/" & sp.WriteBufferSize.ToString
      LBMainStatusInfo.ForeColor = Color.Green
      ' Connection is OK -> Show the tabs
      TABTabControl.Visible = True
      If prmShowTestTab <> True Then
        ' Do not show test tab if configured so
        TABTabControl.TabPages.RemoveAt(4)
      End If
    End Try

Anyway, thanks for the code and the quick response.

Marco

[soundtec]

Hi Yves,
I have the model FY6600 ,I did try it with your software but no luck , would it be possible to put the FY6600 serial command set into the software you created ?

[YvesMalherbe]

What seems to happen is that the FY unit does not respond to the 'a' message when it is on BT (this works on on cable). The mod to the code avoids the software thinking it is in error, but it is probably not recommended for wider implementation.

Hi Marco,

Thanks for keeping me posted.

The 'a' command is the only way I have found to check the connectivity, as the feeltech is supposed to answer by returning the model name.  Perhaps the BT connexion does not support this two way communication.

As you said, it is probably not a good idea to remove/bypass connectivity check.

But, OK, if you are satisfied with the modification you have done, all is well that ends well.

I have the model FY6600 ,I did try it with your software but no luck , would it be possible to put the FY6600 serial command set into the software you created ?

Hello soundtec,

I would like to please you, but I do not think it would be reasonnable to try to adapt the software for a FY6600.

This has been wrote in a kind of quick and dirty way, just to cover my needs regarding the device I own.

FY32 and FY66 are quite different, so that's not only about replacing the serial commands.  For instance, FY66 has the ability to sweep voltage and frequency while FY32 is limited to frequency sweep.  So there is nothing in the GUI which I can recuperate to implement the extra commands needed for the FY66.

Another issue is that, due to the poorly documentation provided by the chineese manufacturer, a lot of programing requires discovering how it works by trial and error and/or reverse engineering (This said, the FY66 documentation looks better than those of the FY32).  As a consequence, it is nearly mandatory to have a device available to develop/test; which is not my case.

But, who knows, perhaps I will buy one FY66 and develop a new software for it...

I see there is also another group on this blog which aim to improve the FY66.   Perhaps you can find what you are looking for there :

• https://github.com/DerKammi/FY6600-15-30-50-60M
• https://www.eevblog.com/forum/testgear/feeltech-fy6600-60mhz-2-ch-vco-function-arbitrary-waveform-signal-generator/

Wish you the best.

Yves

[soundtec]

Thanks Yves,
Yes Ive been contributing to the FY6600 page from way back ,
Its become such a gigantic amount of information  Ive forgotten bits along the way ,

There is another link page here on the Fy6600 Im sure you will like if you havent found it already ,

https://www.eevblog.com/forum/rf-microwave/multiband-wspr-on-the-cheap/msg2352564/#msg2352564

[Calambres]

Sorry to refresh this old thread but I think this is the correct place to write this request.

I've got a Feeltech FY3224S and I need to realign a late 50's AM tube radio receiver. This unit lacks AM modulation AFAIK. I wonder if there's a way to AM modulate one channel set to the receiver's IF frequency with the other channel set to a, say, 1KHz audio tone. I suppose I'll need some kind of AM modulator circuit for this.

Is it possible?

Any ideas on how to do it?

It seems the newer units like the FY6900 can do this. I'm not 100% positive as I have not read anything about this, but I rather keep my old FY3224S and make a modulating circuit for it if it is feasible.

Thanks a lot in advance.

[cdev]

You could perhaps import the mathematically calculated values as a custom waveform, in CSV format? Ive never done that personally but maybe somebody here has. I would try it, why not?

It does have some custom waveforms, One may be an AM modulated carrier. It looks a bit like a DSB-modulated carrier, but I remember it had some issues.

My Feeltech is upstairs in a broiling attic and I should bring it downstairs.

 This may have been discussed in this thread a long time ago.

Another alternative might be to use its signal and modulate it externally perhaps with a unity gain amplifier you modulate with its own B channel? (that is what they should have implemented, but didn't, but for the money, i'm not complaining. )

Maybe in your junk box you have a suitable device you could use?  When I was a kid I made a balanced ring modulator.

Here is a simple circuit to modulate a carrier with a second signal.

https://www.engineersgarage.com/circuit_design/circuit-design-how-to-make-an-amplitude-modulated-wave/
using an n-channel FET.

[Calambres]

You could perhaps import the mathematically calculated values as a custom waveform, in CSV format? Ive never done that personally but maybe somebody here has. I would try it, why not?

I have no idea how to do that 

Here is a simple circuit to modulate a carrier with a second signal.

https://www.engineersgarage.com/circuit_design/circuit-design-how-to-make-an-amplitude-modulated-wave/
using an n-channel FET.

Interesting!

In think that circuit may deliver much more signal than needed (for IF tuning) when it is injected enough signal to make the FET work but I suppose I can put a voltage divider (i.e.: portentiometer) in the output.

Thanks a lot!

[cdev]

Cant you just estimate it, ear out the adjustment?

[ultrablog]

There are litteraly 2 small caps: one from neutral to the output and one from the life to the the output. These 2 caps form a capacitive divider and thus one see's half the voltage.  For safety reasons these caps are Y-rates types that a guarantied not to fail short. Usual capacitance is about 4.7 nF.

In principle a transformer power supply will show a similar voltage. Here it is from capacitive coupling in the transformer. However the capacitance is usually much smaller (e.g. maybe 1%) than the Y-caps. A linear supply does not need the extra caps because there are essential not RF emissions. Even a SMPS behind the conventional transformer can usually get away without.

PE is the short for protective earth, so the safty ground from the 3 rd. pin. If available one could have the capacitors from L to PE and N to PE and than either a direct connection from PE to the BNC or just a capacitor.

I opted for adding a 30VA isolation transformer (Conrad), but I added no PE to the mains connection, so everything is still floating. I prefer to keep it floating, if I can.
I tested the voltages from BNC ground to mains ground, I see 45V AC and 6 µA AC. So the current value is quite okay, but the 45V still worries me (a bit). Do I have options to reduce it apart from adding PE ground to the box? Could I safely remove (or decrease value) of those two capacitors? And even then, won't that just lower the current further (but not the voltage)? Am I doing this right?

[ledtester]

Sorry to refresh this old thread but I think this is the correct place to write this request.

I've got a Feeltech FY3224S and I need to realign a late 50's AM tube radio receiver. This unit lacks AM modulation AFAIK. I wonder if there's a way to AM modulate one channel set to the receiver's IF frequency with the other channel set to a, say, 1KHz audio tone. I suppose I'll need some kind of AM modulator circuit for this.

Is it possible?

Any ideas on how to do it?

It seems the newer units like the FY6900 can do this. I'm not 100% positive as I have not read anything about this, but I rather keep my old FY3224S and make a modulating circuit for it if it is feasible.

Thanks a lot in advance.

This page says you can do AM modulation with a single FET:

https://www.engineersgarage.com/circuit_design/circuit-design-how-to-make-an-amplitude-modulated-wave/

[Calambres]

Very interesting. Thanks! 

[elwood]

I have seen that some of us have developped a control program for the Feeltech FY3200 generator.

Hi, YvesMalherbe,

Thank you, great program. Can you ask for a couple of improvements?

1. Make the same settings and presets for the second channel.
2. In presets, add the ability to change the frequency after a certain time, for example:

preset 1:
100 Hz - 5 minutes
2000 Hz - 20 minutes
3000 Hz - 2 minutes
... 10-20-30 different frequencies

and loop this work. There are an unlimited number of presets, but only the selected one works. That would be amazing  Thank you!

Good luck! Elwood

[ledtester]

For controlling the FeelTech in complex ways, there's a Python library which is easy to use:

https://github.com/atx/python-feeltech

[dazwin]

Has anyone noticed the FY3224S can go up to 2^32Hz (about 40MHz)?

A bug in the firmware allows this to be set on the unit when sweeping, but the fixed frequency can be changed via USB (fytool if you change the max)

[YvesMalherbe]

Hello Elwood,

Glad to hear that my software helps you.

I have seen that some of us have developped a control program for the Feeltech FY3200 generator.

Thank you, great program. Can you ask for a couple of improvements?

1. Make the same settings and presets for the second channel.
2. In presets, add the ability to change the frequency after a certain time.

I will have a look to your requests, but don't expect a fast reply, as I am currently quite busy on other things professionnally.

Regards,

Yves

[cdev]

Is there any way to either use the FY3224S with an external 10 MHz reference, or calibrate it to improve the frequency accuracy?

[Calambres]

Sorry to revive an old topic but my FY3224S just developed an issue. Somehow the sine wave is no longer working. It outputs absolutely nothing when the sine wave is selected, neither CH1 nor CH2. All the other waves work OK, even ARB1 which is indeed a sine wave.

I have already tried the "reset" explained previuosly in this very topic:
https://www.eevblog.com/forum/testgear/feeltech-fy3224s-24mhz-2-channel-dds-aw-function-signal-generator/msg1175563/#msg1175563

...unfortunately to no avail. Any Ideas?... anything I could do to restore the native sine wave?

Thanls a lot in advance.

[Szasza]

Please help I have just started learning about fy3224 dds.
I tried several things, including PC connection, everything worked.
The sinus signal of ch1 and ch2 changed unexpectedly.
I don't know what happened and how to fix it?
Does it have factory reset?

[ledtester]

You might try the "reset" procedure mentioned in the message by Calambres just before yours.

[Calambres]

I keep investigating this strange and sudden behaviour. Somehow now CH2 displays a "somewhat" sinusoidal wave that do not correspond to the paramaters given, neither frequency nor amplitude. CH1 still no SINE wave but a 213mV DC offset. I cannot understand why SINE is not working but ALL other waves work perfectly OK. What's different in SINE to the others?

I've opened the case and found nothing apparently wrong there. Also extracted the generator panel and all seems OK. No bulging or leaking capacitors, no overheating components, no open traces, no rusting leads, no nothing!

I'm amazed to see that my FY3200S is not the only one developing these sympthoms    The "reset" I mentioned before is not a real reset but a frequency calibration mode. The other working waves are spot on in frequency... wish I knew of a real reset mode.

[Calambres]

I give up!

I, for the life of me, cannot see any physical problems anywhere. I strongly suspect it's a data corruption of some sort. Nobody knows of a REAL reset for this AWG?

[Calambres]

This thing is driving me nuts!

It has developed another annoying sympthom: the selected frequency is increased by no less than a 43% 

100Hz selected  -> 143.050Hz measured
10KHz selected  -> 14.3050KHz measured
100KHz selected -> 143.050KHz measured

The measures are made with a DSO scope and an external freq. counter. At least, the increasings are consistent!
On the other hand, if I read the counts with the FY3224S' own counter, the readings correspond with what I selected, i.e.: 100Hz reads 100Hz, etc.

No idea what is happening here but still I wish I had a way to reset this thing  (SINE wave is still not working at all)

Of course, the frequency calibration routine I cited above works well and brings it back to normal. now 100Hz really is 100Hz, etc, but I have no idea whay it suddenly changed by such a great percentage.

[EDIT] It seems to be related to a FW bug regarding the sweep function. I used that function a couple of days ago... 

[EDIT2] Added a Feeltech official document on this procedure I got from somewhere...