Forums :: Resources :: Features :: Photo Gallery :: Vintage Radio Shows :: Archives
Support This Site: Contributors :: Advertise


It is currently Jul Tue 07, 2020 6:52 pm


All times are UTC [ DST ]





Post New Topic Post Reply  [ 214 posts ]  Go to page Previous  1 ... 7, 8, 9, 10, 11  Next
Author Message
 Post subject: Re: Microcontroller Based Signal Generator
PostPosted: Jul Sun 02, 2017 3:29 pm 
Member
User avatar

Joined: Aug Wed 26, 2009 1:50 am
Posts: 1482
Location: Mesquite NV 89027 (from Coventry, UK)
Bill - I did breath a sigh of relief, that's for sure!

Steve - Thanks for the photos of the board in the case. That is good news, for sure.

A couple of things to note in regards to loading the software:

The Arduino Nano clone probably has a CH340 chip on it. If so, you have to download a driver for it, as the latest version of the Arduino development software doesn't support it. I found one for Windows, which worked just fine. It allows the creation of a virtual COM port to connect to the Nano. If you use a MAC or Linux I have no information.

When I opened Lou's README file, which is the instructions on how to load his software, I found that for whatever reason it had lost its formatting, so all of the numbered steps ran together. I went through and separated each of the nine steps, which made it a lot easier to understand, and everything worked just fine.

Colin


Top
 Profile  
 
 Post subject: Re: Microcontroller Based Signal Generator
PostPosted: Jul Sun 02, 2017 3:41 pm 
Member
User avatar

Joined: Jan Mon 18, 2010 2:13 am
Posts: 20580
Location: Dayton Ohio
Hmm, sounds like this is going to be interesting.

I usually use a Mac, but am trying to get the hang of Linux on my old PC.
I do have another PC with Windows 10 and a netbook with Windows XP. I'll see what I can do. :)

Thanks for the info!

-Steve

_________________
Radio Interests
-Zenith
-Sparton
-Pre-War FM
Consoles and floor models, the bigger, the better!


Top
 Profile  
 
 Post subject: Re: Microcontroller Based Signal Generator
PostPosted: Jul Sun 02, 2017 4:12 pm 
Member
User avatar

Joined: Feb Fri 25, 2011 8:00 pm
Posts: 963
Location: Dublin, Ohio
Yes, there may be a driver problem. I had no issue when I tested with windows but I needed a driver for my Mac. I am out today but will try to find that info and post later.

_________________
Website - http://www.louhaskell.com/


Top
 Profile  
 
 Post subject: Re: Microcontroller Based Signal Generator
PostPosted: Jul Mon 03, 2017 12:14 am 
Member
User avatar

Joined: Aug Wed 26, 2009 1:50 am
Posts: 1482
Location: Mesquite NV 89027 (from Coventry, UK)
Well, my elation was short lived. I went to check the outputs, to adjust the duty cycle of the DDS modules, and I have nothing. I hooked up the 'scope directly to the outputs using a BNC-BNC cable. Nothing :cry: . The 'scope is working, as I connected the cable via a jumper to the calibration output and I got the 5 volt square wave. I checked 5 volt and 3.3 volt supplies, both are good. Checked for shorts and bad joints with a magnifier and couldn't see anything wrong. I touched up all solder joints, still nothing. I don't know where to look next.


Top
 Profile  
 
 Post subject: Re: Microcontroller Based Signal Generator
PostPosted: Jul Mon 03, 2017 12:48 pm 
Member
User avatar

Joined: Feb Fri 25, 2011 8:00 pm
Posts: 963
Location: Dublin, Ohio
Colin, sorry to hear you are having problems. Possibly a few suggestions. First off, work on the DDS 1 module that supplies the square wave and sine wave. Use the square wave output that is set by the FN settings. The square wave, as you see from the schematic, comes straight off of the chip. I noted on my module that the duty cycle as set by the little blue pot was not fully settable but had a narrow range of operation. So it is possible, you have it set now to give nothing which I initially saw, i.e. a duty cycle of zero! Once the square wave is good, check the sine wave which goes through an addition amplifier and amplitude pot. For the RF, try to get a scope right on the top of R5 or pot R6. Again, try to set the blue pot for 50% duty cycle. If you get the signal there, trace out to the BNC, going a step at a time. Notice you can clip the scope ground lead onto one of the outer BNC connectors to get a ground. If there is nothing coming through the transformer, you can get a scope probe (carefully) onto pin 19 or 20 on top of the DDS module. It can be done but delicately .
Let us know if this works.

_________________
Website - http://www.louhaskell.com/


Top
 Profile  
 
 Post subject: Re: Microcontroller Based Signal Generator
PostPosted: Jul Mon 03, 2017 4:14 pm 
Member
User avatar

Joined: Aug Wed 26, 2009 1:50 am
Posts: 1482
Location: Mesquite NV 89027 (from Coventry, UK)
Hi Lou,
Thanks for your advice. For DDS1 you were absolutely correct. There is a very small window in which it works, and obviously I missed it when adjusting it before. Now I have sine and square waves out of their respective ports, and with the correct amplitudes.

The RF from DDS2 is a different matter. I have an output whose frequency follows the programmed values, and it can be modulated. However, the maximum amplitude varies between 300mV to 500mV depending on the level of modulation. I thought it should have a max value of 1Volt peak to peak. In addition, the potentiometer on the DDS seems to have no effect, so I am not sure what I am supposed to see.

Thanks again,
Colin


Top
 Profile  
 
 Post subject: Re: Microcontroller Based Signal Generator
PostPosted: Jul Mon 03, 2017 4:56 pm 
Member
User avatar

Joined: Feb Fri 25, 2011 8:00 pm
Posts: 963
Location: Dublin, Ohio
Good to hear that progress is made. As for the level from the RF stage, I have seen similar results. The higher the frequency, the lower the output level. But I have had enough amplitude to do my alignments in every case. I have wondered if the modulation section with the MOSFET is not providing enough signal, even at zero % modulation to provide a maximum output level, i.e. zero attenuation. I used the values of the components from the article referenced by the other author who built a similar device. I do not know if changing the value of R4 would have any effect, for example. Maybe something to try if you or we collectively feel a need for more output amplitude.

_________________
Website - http://www.louhaskell.com/


Top
 Profile  
 
 Post subject: Re: Microcontroller Based Signal Generator
PostPosted: Jul Mon 03, 2017 7:29 pm 
Member
User avatar

Joined: Aug Wed 26, 2009 1:50 am
Posts: 1482
Location: Mesquite NV 89027 (from Coventry, UK)
At this point I think I need to use the system to do an alignment. If you have had no issues then I suspect my outcome will be the same. It will be a couple of days before I can get around to that, so I will post the results.

As for the trimmer on DDS2, does it even do anything?

Thansk,
Colin


Top
 Profile  
 
 Post subject: Re: Microcontroller Based Signal Generator
PostPosted: Jul Mon 03, 2017 7:47 pm 
Member
User avatar

Joined: Feb Fri 25, 2011 8:00 pm
Posts: 963
Location: Dublin, Ohio
The trimmer pot on DDDS2 should perform the same as the pot on DDS1 since they are the same part. Let know what level you get on the RF port when the freq is set to 455KHz and we can all compare.

_________________
Website - http://www.louhaskell.com/


Top
 Profile  
 
 Post subject: Re: Microcontroller Based Signal Generator
PostPosted: Jul Mon 03, 2017 9:14 pm 
Member
User avatar

Joined: Aug Wed 26, 2009 1:50 am
Posts: 1482
Location: Mesquite NV 89027 (from Coventry, UK)
Here are three photos showing waveforms. Sorry for the poor quality. The frequency is 455KHz, modulated with 400Hz. The first photo is with R1 set to minimum, R6 at maximum. Amplitude is 250mV, and the time base is set to 1uS /div. The second is with R1 set to max. The amplitude increased to 500mV. The third shows the modulated waveform. Time base is 1mS/div
Attachment:
R1 at MIN.JPG
R1 at MIN.JPG [ 93.84 KiB | Viewed 3232 times ]

Attachment:
R1 at MAX.JPG
R1 at MAX.JPG [ 82.35 KiB | Viewed 3232 times ]

Attachment:
Modulated waveform.JPG
Modulated waveform.JPG [ 87.9 KiB | Viewed 3232 times ]

The 'scope probe calibration is correct. Still don't get any change in waveforms when adjusting the trimmer. I turn it very slowly from one extreme to the other, nothing changes. I have set it to the same position as DDS1.


Top
 Profile  
 
 Post subject: Re: Microcontroller Based Signal Generator
PostPosted: Jul Mon 03, 2017 9:44 pm 
Member
User avatar

Joined: Feb Fri 25, 2011 8:00 pm
Posts: 963
Location: Dublin, Ohio
I am getting similar results - 300mv no modulation and 600mv with modulation. As I sit here and think about it, this does not make sense to me. The FET circuit should modulate by decreasing signal in the troughs of the signal with the P-P staying the same. This is how I would think the circuit works. However, If the no modulation signal has the gain somewhat reduced , then the modulated signal would indeed increase P-P. I saw this before but never questioned it, but now I do. Something tells me that changing R4 and perhaps the bias resistors R2 and R3, some improvement may be attainable. Of course, we need to stay in the linear region and not clip either up or down but perhaps some experimenting would be in order.

_________________
Website - http://www.louhaskell.com/


Top
 Profile  
 
 Post subject: Re: Microcontroller Based Signal Generator
PostPosted: Jul Tue 04, 2017 1:51 am 
Member
User avatar

Joined: Aug Wed 26, 2009 1:50 am
Posts: 1482
Location: Mesquite NV 89027 (from Coventry, UK)
I'm afraid I am not well versed enough in theory to offer suggestions, but I would be happy to try anything that you suggest.


Top
 Profile  
 
 Post subject: Re: Microcontroller Based Signal Generator
PostPosted: Jul Wed 05, 2017 5:29 pm 
Member
User avatar

Joined: Jan Mon 18, 2010 2:13 am
Posts: 20580
Location: Dayton Ohio
Well, I have yet to check the output signal, but I got the firmware uploaded to the Nano last night and powered it up.

I placed it in the case, but I'm not happy with the hardware I chose.
The screws seem to be unplated and easily rust. I will look for some button head screws that are at least nickel plated (not Zinc or Tin) or maybe Stainless.

Colin, Lou, Thanks for the info on the signal strength and tweaking the pots on the DDS boards. I bet I'll have to do the same.

-Steve


Attachments:
compulign.jpg
compulign.jpg [ 191.16 KiB | Viewed 3188 times ]

_________________
Radio Interests
-Zenith
-Sparton
-Pre-War FM
Consoles and floor models, the bigger, the better!
Top
 Profile  
 
 Post subject: Re: Microcontroller Based Signal Generator
PostPosted: Jul Wed 05, 2017 5:49 pm 
Member
User avatar

Joined: Aug Wed 26, 2009 1:50 am
Posts: 1482
Location: Mesquite NV 89027 (from Coventry, UK)
Steve, good to see that you got the software loaded. It will be interesting to see the level of your output signals.

Colin


Top
 Profile  
 
 Post subject: Re: Microcontroller Based Signal Generator
PostPosted: Jul Wed 05, 2017 11:44 pm 
Member
User avatar

Joined: Aug Wed 26, 2009 1:50 am
Posts: 1482
Location: Mesquite NV 89027 (from Coventry, UK)
Tried to use the generator on a radio, an Emerson 888 transistor portable. Using the FIX function, it works just like my old tube signal generator. No problem. When I tried the sweep functions, I saw results different to the ones in the user guide. I am not saying the results are wrong, just different. Specifically, the curves in both SWi and SWe were inverted with respect to the ones in the guide. Also, the level on the internal display was very low, meaning the curve was almost flat. On external, my 'scope display didn't look anywhere as nice as Lou's, but that is probably a function of the 'scope or the operator. In both cases it made it difficult to peak. These results are quite possibly due to the radio, but I thought I would post them as food for thought. I will try to find the time to take another radio off the shelf and see what happens with it.

One more thing I noticed was that the device doesn't remember its settings. No matter what settings are active when it is powered down, upon power up it always defaults to FIX, with FN 400Hz, RF 455KHz, BW 20KHz, Rate 9, Mark 0 and Bias 0.0. This might be by design, I don't know.

Colin


Top
 Profile  
 
 Post subject: Re: Microcontroller Based Signal Generator
PostPosted: Jul Thu 06, 2017 11:22 am 
Member
User avatar

Joined: Feb Fri 25, 2011 8:00 pm
Posts: 963
Location: Dublin, Ohio
I have not used my CompuLign on a transistor radio yet so interesting to see how that works. Any photos may be good if you are looking for ideas on how to best use this thing. In reality, we are all now in the "beta test phase" as more diversity of use occurs. I kind of see this as a journey so I hope all of you are in that mood. I think we can help each other on this. And I hope any modifications we all think about will not require a board change. ;-)

You are correct that the settings return to an initial state on power-up. Designed that way. I had not thought about storing the previous settings and returning to that on the next power-up. That may be possible but I would have to research. I believe that chip has some non-volatile storage.

_________________
Website - http://www.louhaskell.com/


Top
 Profile  
 
 Post subject: Re: Microcontroller Based Signal Generator
PostPosted: Jul Thu 06, 2017 1:09 pm 
Member
User avatar

Joined: Jan Mon 18, 2010 2:13 am
Posts: 20580
Location: Dayton Ohio
I had a chance to put a scope on it this morning.

Initially, I had a nice sine out of BNC 2, but nothing out of the square output. After moving the scope cable back to sine, I had nothing.

So I tried the blue pot on DDS1 and as you both said, it has a narrow range of adjustment. I set it to 50% no problem.

I scoped DDS2 at pin 17 and found it does have an output but only at the far counterclockwise rotation of the blue pot. I could not change the duty cycle, so I left it at the far counterclockwise setting.

Sinewave output is 0-1 VPP and RF is a little less than 300mV to about 500 mV fully modulated.
So it seems I'm getting similar results.

However, I'm not sure why I lose the sinewave output as well as modulation when I move the coax cable between ports?
It requires a reboot to get it back.

I'm not using a standard scope probe. I'm just connecting a 50 ohm coax cable between the generator and the scope.
I could not find a scope probe here at work, which is pathetic. :?

-Steve

_________________
Radio Interests
-Zenith
-Sparton
-Pre-War FM
Consoles and floor models, the bigger, the better!


Top
 Profile  
 
 Post subject: Re: Microcontroller Based Signal Generator
PostPosted: Jul Thu 06, 2017 1:41 pm 
Member
User avatar

Joined: Aug Wed 26, 2009 1:50 am
Posts: 1482
Location: Mesquite NV 89027 (from Coventry, UK)
Lou - I will post photos, hopefully some time today. As for the default settings at power on, they are well chosen and don't cause a problem, I just made the observation. Not worth reengineering.

Steve - I think I too noticed the need for a reboot after swapping ports. I will do further checks to confirm.


Top
 Profile  
 
 Post subject: Re: Microcontroller Based Signal Generator
PostPosted: Jul Thu 06, 2017 2:19 pm 
Member
User avatar

Joined: Jan Mon 18, 2010 2:13 am
Posts: 20580
Location: Dayton Ohio
I noticed a bit of hash and noise on the signals so I may try running mine on a battery. I'm not sure a standard 9 volt would be sufficient, but a 12 volt gel type would do the job.
I'll probably add 5 1N400x in series to drop the voltage.

The wall wart I'm using came from an old GPX portable DVD player which has a built-in screen. and outputs 9 volts at 1 amp. It weighs next to nothing so its clearly an SMPS type.

This is going to be a very useful piece of equipment!
Lou, thank you for putting all the effort into designing it!

-Steve

_________________
Radio Interests
-Zenith
-Sparton
-Pre-War FM
Consoles and floor models, the bigger, the better!


Top
 Profile  
 
 Post subject: Re: Microcontroller Based Signal Generator
PostPosted: Jul Thu 06, 2017 11:04 pm 
Member
User avatar

Joined: Aug Wed 26, 2009 1:50 am
Posts: 1482
Location: Mesquite NV 89027 (from Coventry, UK)
Some photos. The first and second show internal and external on an Emerson 888, and the third and fourth are the same on a GE P-875.
Attachment:
Emerson int small.jpg
Emerson int small.jpg [ 180.68 KiB | Viewed 3148 times ]

Attachment:
Emerson ext small.jpg
Emerson ext small.jpg [ 169.74 KiB | Viewed 3148 times ]

As you can see, the internal trace is very flat, so peaking is not easy with this radio.
Attachment:
GE int small.jpg
GE int small.jpg [ 204.12 KiB | Viewed 3148 times ]

Attachment:
GE ext small.jpg
GE ext small.jpg [ 104.81 KiB | Viewed 3148 times ]

You can see the internal trace has a nice shape to it, so peaking was easier.

One thing I noticed was that the instructions for the GE call for all alignment to be done by forming a loop of a few turns of wire and coupling the generator in this manner. This did not work out well, as with a frequency of 455KHz there is so much noise that the tone could not be heard. At higher frequencies the tone could be heard perfectly. The Emerson calls for injecting the 455 KHz, so the problem did not arise. Maybe the wall wart is generating some noise that affects the lower frequency, or maybe it is a strange quirk of the circuitry. If I can figure out a way to rig up a battery supply, even if it's for just a couple of minutes, it will prove if the power supply is the cause. Not sure when this will happen.


Last edited by Colin Ames on Jul Fri 07, 2017 5:16 am, edited 2 times in total.

Top
 Profile  
 
Post New Topic Post Reply  [ 214 posts ]  Go to page Previous  1 ... 7, 8, 9, 10, 11  Next

All times are UTC [ DST ]


Who is online

Users browsing this forum: No registered users and 6 guests



Search for:
Jump to:  


































-->


Privacy Policy :: Powered by phpBB