UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Specific Vintage Equipment > Vintage Test Gear and Workshop Equipment

Notices

Vintage Test Gear and Workshop Equipment For discussions about vintage test gear and workshop equipment such as coil winders.

Closed Thread
 
Thread Tools
Old 23rd Jul 2017, 2:45 pm   #1
John G8MWF
Hexode
 
Join Date: Aug 2014
Location: Dartford, Kent, UK.
Posts: 435
Default MHS-5200A (AWG) Amplitude modulated waveform

Has anyone, successfully, used a MHS-5200A or similar (AWG) to generate a Amplitude modulated waveform?

I would like to use my AWG to provide 455khZ modulated with a 1KHz to to line up the IF stages - can it be done?

Kind regards


John
John G8MWF is offline  
Old 24th Jul 2017, 1:20 am   #2
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: MHS-5200A (AWG) Amplitude modulated waveform

I've not used that device but I did buy an Analog Discovery a week or so ago. I've not had much chance to play with it but it can accept CSV files for waveform data. Can your arb do the same? If so you could generate the data in Excel although this is going to be a bit clunky. Probably better to do it in something like VB using an equation for AM and export the CSV data file for direct loading into the arb?

I tried loading CSV file data into my Analog Discovery and it worked first time. I also tried generating 455kHz AM with 50% mod at 1kHz using excel and loaded the resulting CSV data into the Analog Discovery this worked as well when looked at on a spectrum analyser. But it did generate some artefacts maybe 40-50dB down on what would be adjacent channels. I think my excel sheet may need some fine tuning to improve on this.

The plot below is my Analog Discovery generating 455kHz AM via its arb feature. The plot is taken with a Tek spectrum analyser that can demodulate AM and you can see it is very close to 50% AM
Attached Thumbnails
Click image for larger version

Name:	AM.gif
Views:	135
Size:	24.2 KB
ID:	146710  
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 24th Jul 2017, 7:56 am   #3
John G8MWF
Hexode
 
Join Date: Aug 2014
Location: Dartford, Kent, UK.
Posts: 435
Default Re: MHS-5200A (AWG) Amplitude modulated waveform

Good morning Jeremy ,
Thank you the information it looks very comprehensive and it does look like I will be able to do 'something' with MHS5200A
My chinglish manual says that I can import and export .CSV files but I haven't tried this - yet. The manual also has a very brief section on generating a AM waveform but it doesn't say how to setup specific frequencies and modulation.
In fairness I am still very new to AWG and I am picking it up slowly as I experiment with different options.

Don't forget this model only generates a wave shape with a maximum value of 1028 so it is a relatively simple device compared to some that are now on the market.

I would be interested in having a look at your CSV files if possible?

Kind regards


John
John G8MWF is offline  
Old 24th Jul 2017, 8:05 pm   #4
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: MHS-5200A (AWG) Amplitude modulated waveform

If your device only allows 1024 samples then sadly my CSV file won't be any use. However, I've attached my csv file below. I did this in excel and I aimed for a sample rate of 4.096MHz with 4096 samples. The idea was that this would grab a 1ms frame (for the 1kHz modulation) and this frame would play back fairly cleanly when repeated continuously by the Analog Discovery.

However, it does produce quite a few artefacts along with the wanted signal at 455kHz and would need a BPF to remove them. I'll have a play with mine to see how it performs with just 1024 samples but I think I'll need to generate the 455kHz signal as an alias term and then filter it.

Maybe someone else has been here already with your generator and maybe they can offer better advice?

Note that I've had to rename the CSV file below to a TXT file to get it to be accepted by the forum file uploader.
Attached Files
File Type: txt am455_4096.txt (53.7 KB, 72 views)
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 24th Jul 2017, 8:41 pm   #5
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: MHS-5200A (AWG) Amplitude modulated waveform

I had a go at creating a 1024 point waveform (with 1.024MHz sample rate) at 455kHz with 1kHz modulation at 50% and it works but it produces a lot of digital artefacts nearby. Also a huge alias term at 569kHz where 569 = 1024-455. This alias term looks like an identical copy of the 1kHz modulated signal at 455kHz.

However, I was able to exploit the alias term because it allowed me to listen to the signal on a MW receiver at 569kHz. This allowed me to listen to hear if there were any nasty noises caused by the repeating waveform. It did seem to sound OK to my ears...

Ideally it would really need to be fed through a narrow filter at 455kHz to get rid of the nearby artefacts on adjacent channels. Maybe something 12-15kHz wide if you want to get rid of the artefacts nearby?

I'll have a play to see if it's possible to produce something cleaner than this but a 1024 point waveform is a significant limitation I think...
Attached Files
File Type: txt am455k_1024_455kHz.txt (13.4 KB, 64 views)
__________________
Regards, Jeremy G0HZU

Last edited by G0HZU_JMR; 24th Jul 2017 at 8:47 pm.
G0HZU_JMR is offline  
Old 24th Jul 2017, 11:45 pm   #6
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: MHS-5200A (AWG) Amplitude modulated waveform

I googled a few things and I think I know a bit more about your AWG model now John.

I found this website

https://github.com/wd5gnr/mhs5200a

Some useful stuff here including a nifty excel sheet in case you haven't spotted it already?

I now think your AWG expects 8 bit data in the CSV file and maybe it also needs to be offset. So I had a go at converting my 1024 point CSV file into the correct format. However, this first attempt will almost certainly be wrong. But I think it needs to be clocked out at 1.024MHz to get the waveform at 455kHz.

Even if this does work it will look horrible on a scope unless you filter away the unwanted artefacts with a 455kHz BPF. So probably best tested into a receiver or spectrum analyser. Try feeding it to a medium wave receiver and tuning to 569kHz. Does it give a 1kHz modulated signal? Even if it doesn't work as intended, this file should do 'something' valid once renamed as a CSV file.
Attached Files
File Type: txt 5200_AM_455k.txt (4.6 KB, 440 views)
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 29th Jul 2017, 12:19 am   #7
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: MHS-5200A (AWG) Amplitude modulated waveform

If it didn't work out, I could try and modify the 5200 AWG excel sheet on that website to do AM for you if that helps? It just needs to have a classic AM equation in one of the columns.

See below for a screenshot how I generated AM with the Analog Discovery using Excel. It has user inputs for output frequency, Sample rate, AM mod depth and AM mod frequency. You can see in the example below that I changed the mod depth to 20% and the mod frequency to 5kHz. This makes it look nicer on an excel FFT as it has enough info for the sidebands. You can see that the data produces AM at 455kHz in the FFT plot.

The Excel spreadsheet is a bit clunky because I generated each sideband and the carrier with individual equations and then summed or beat them together to get AM. It could have all been done in one column but I can check each tone of the FFT this way.

Obviously, the excel sheet FFT graph is an FFT that isn't limited by bit depth so it will not look this good in reality. But see the 'real' plot of this data when output from the Analog Discovery to the spectrum analyser at 455kHz. It is spot on for 20% AM at 455kHz.
Attached Thumbnails
Click image for larger version

Name:	455k_AM_20.gif
Views:	95
Size:	21.1 KB
ID:	146996   Click image for larger version

Name:	455k_am20gif.jpg
Views:	96
Size:	84.0 KB
ID:	146997  
__________________
Regards, Jeremy G0HZU

Last edited by G0HZU_JMR; 29th Jul 2017 at 12:28 am.
G0HZU_JMR is offline  
Old 29th Jul 2017, 2:23 pm   #8
John G8MWF
Hexode
 
Join Date: Aug 2014
Location: Dartford, Kent, UK.
Posts: 435
Default Re: MHS-5200A (AWG) Amplitude modulated waveform

Hello Jeremy, I will give that a try.
Another read of the chinglish manual does, as previously stated, reveal that an AM waveform can be generated using the software supplied and I am enclosing the 3 relevant pages. This is VERY tempting to say the least.

I simply, still, cannot work out from their brief explanation of this setup how the frequency, modulation frequency and index are determined!

Can you enlighten me?

Kind regards


John
Attached Files
File Type: pdf MHS5200A_manual P24.pdf (1.76 MB, 108 views)
John G8MWF is offline  
Old 29th Jul 2017, 2:27 pm   #9
John G8MWF
Hexode
 
Join Date: Aug 2014
Location: Dartford, Kent, UK.
Posts: 435
Default Re: MHS-5200A (AWG) Amplitude modulated waveform

Here are the other 2 attachments that went AWOL
John G8MWF is offline  
Old 29th Jul 2017, 8:47 pm   #10
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: MHS-5200A (AWG) Amplitude modulated waveform

Your last two attachments haven't uploaded but I managed to find the user manual online. Obviously it doesn't help that the user interface is in Chinese! I couldn't find the installer download for your version but I did find the English version of the 5200P version. Sadly, their download rar file is corrupted (so it fails to install anything) but I managed to fix it such that it installs and runs on an old laptop.

Having played with this for a few minutes I think you are meant to create the AM waveform in stages. The idea is to create a 'fast' waveform (the carrier) in the first step and save it. This is the 0 to 80 waveform. Then create the modulation waveform with one complete cycle with the 0 to 2047 waveform. Presumably this would be 0 to 1024 with your version. Then use their math function tab and button to multiply the two waveforms together. Then I guess you save it to one of the 16 memory slots?

See the hasty youtube video below to see this in action on the 5200P version. I know this isn't the same GUI as you are using but maybe the process is the same and you can reverse engineer what the buttons do on your version. I suspect there will be a way to make your version run with English text on the GUI.

https://www.youtube.com/watch?v=-W0fg-ht7DY

If you point me in the direction of the installer for your version I think I could get it to run and maybe even run with English text on the functions?
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 29th Jul 2017, 10:15 pm   #11
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: MHS-5200A (AWG) Amplitude modulated waveform

For example, if this version of the MHS5200A user interface runs with your hardware then it runs in English. It supports the version with the 1024 sample limit like yours?

http://www.mhinstek.com/rar/MHS-5200A_en.rar


I did another hasty youtube video with my old laptop. I really am just guessing if this is the right way to do it...

https://youtu.be/zhHOP_DQbpk

Hope it helps...
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 30th Jul 2017, 5:53 pm   #12
John G8MWF
Hexode
 
Join Date: Aug 2014
Location: Dartford, Kent, UK.
Posts: 435
Default Re: MHS-5200A (AWG) Amplitude modulated waveform

Hello Jeremy, that is brilliant!

I noticed that you high lighted the 'Address' bar button. Does this select which ARB button the waveform is assigned to?
John G8MWF is offline  
Old 30th Jul 2017, 8:03 pm   #13
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: MHS-5200A (AWG) Amplitude modulated waveform

I'm not sure... I think you can just press the 'Write' Button when the AM waveform appears at about 1:30 in the latest video. This might send the waveform to the selected memory address inside the actual instrument. This probably becomes a stored memory that you can recall using the 0-15 memory under the AWG menu on the instrument. So once you have written it to the instrument at Address 2 for example it will be in AWG memory 2 on the front panel of the instrument. So you wouldn't need the PC once you write it into memory?

But this is just a guess. Assuming this works, the next step is to get it to produce AM at 455kHz with 1kHz modulation. I think this could be a problem with that user interface.

But you could try using that interface to load a custom 1024 point CSV file with the OPEN button and then select all files *.* to get it to find CSV files and then load the custom file and then write it to memory address 2. I could produce this custom file for you. See if that works?
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Old 31st Jul 2017, 1:00 am   #14
John G8MWF
Hexode
 
Join Date: Aug 2014
Location: Dartford, Kent, UK.
Posts: 435
Default Re: MHS-5200A (AWG) Amplitude modulated waveform

Hello Jeremy, I now have the test gear and patterns setup and, unfortunately, got the results I expected.

I used a 10KHz sine as a starting point and then changed the wave shapes using the default ones and the ARB functions defined by me.

The periodic times, 100uS, for standard waveforms looked good. The AM modulated waveforms looked the right shape but it was obvious that the periodic times are way out and it does not look like a 10KHz carrier modulated with audio but something higher.

I still do not see how the maths calculates this pattern for one cycle and then allows to specify a 'carrier' frequency giving a modulation tone.
It uses 'Mult' as the math option but I always thought that heterodyne was a sum and not a product i.e. f1+f2 and f1-f2.

It is probably me trying to ask for more than is possible out of a basic box such as this but it is fun trying!
John G8MWF is offline  
Old 31st Jul 2017, 2:20 am   #15
G0HZU_JMR
Dekatron
 
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
Default Re: MHS-5200A (AWG) Amplitude modulated waveform

Quote:
It uses 'Mult' as the math option but I always thought that heterodyne was a sum and not a product i.e. f1+f2 and f1-f2.
The classic trig identities for a multiplication do produce sum and difference terms just as you have typed.

However for AM you also need to include the carrier term. So I assume as they call this an AM function it also includes the carrier term in their 'multiply' function. Without the carrier you would just get double sideband but no carrier = DSBSC or double sideband suppressed carrier.


So I assumed they include a DC/offset term to make it AM and not DSB when the multiply routine is run? But looking closer at the 'AM' waveform on their user interface it looks more like DSBSC to me. So it isn't regular AM after all. So maybe there is a way to add an offset for AM?

I still think it's going to be easier to create the correct AM waveform in Excel and upload it as a CSV file to one of the AWG slots. It worked on my Analog Discovery using this method.
__________________
Regards, Jeremy G0HZU
G0HZU_JMR is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 2:52 am.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.