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


It is currently Aug Sat 18, 2018 1:49 pm


All times are UTC [ DST ]





Post New Topic Post Reply  [ 31 posts ]  Go to page 1, 2  Next
Author Message
 Post subject: AM transmitter prototype
PostPosted: Dec Tue 05, 2017 3:01 am 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 228
Location: Arlington, TX, USA
I finally got replacement DDS modules; took almost a month to get here. But I got my solderless breadboard prototype up and running again and I will show some scope screen shots of how it performs. This transmitter uses a MC1496 to generate amplitude modulation and is amplified by a low power linear amplifier. Carrier is generated by an AD9833 DDS generator mounted on a small PC board with the master clock and associated capacitors. For the purposes of development I used an Arduino Uno clone to send the data needed by the AD9833 to generate a sine wave at a chosen frequency. The scope pictures are all shown at a carrier frequency of 1500kHz. The modulated waveforms are at 2V/div.


Attachments:
File comment: 1kHz, ~100% modulation, 2V/div
1khz.JPG
1khz.JPG [ 221.21 KiB | Viewed 2456 times ]
File comment: 1kHs square wave
1ksquare.JPG
1ksquare.JPG [ 212.29 KiB | Viewed 2456 times ]
File comment: 20kHz sine
20khz.JPG
20khz.JPG [ 227.98 KiB | Viewed 2456 times ]
File comment: 20kHz square wave
20ksquare.JPG
20ksquare.JPG [ 233.32 KiB | Viewed 2456 times ]
File comment: 1500 kHz unmodulated carrier
carrier.JPG
carrier.JPG [ 221.58 KiB | Viewed 2456 times ]
File comment: X-Y trapezoid display
trapezoid.JPG
trapezoid.JPG [ 220.67 KiB | Viewed 2456 times ]
File comment: 20Hz sine
20hz.JPG
20hz.JPG [ 57.22 KiB | Viewed 2456 times ]
File comment: the circuit as it currently is
protoype.JPG
protoype.JPG [ 106.22 KiB | Viewed 2456 times ]
Top
 Profile  
 
 Post subject: Re: AM transmitter prototype
PostPosted: Dec Tue 05, 2017 7:06 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 3535
Lovely waveforms! I admire your know how. I briefly looked at the coding needed to send instructions to a DDS chip and decided it was too much of a learning curve for me at the time.

Do you think there will be a way to simplify the circuit by removing the Arduino and just using the preprogrammed chip with some inputs to change the frequency?


Top
 Profile  
 
 Post subject: Re: AM transmitter prototype
PostPosted: Dec Tue 05, 2017 4:39 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 228
Location: Arlington, TX, USA
Thank you for the kind words. My plan was to be able to use the microcontroller as a stand alone like you are suggesting with just the three input switches (up, down, select) and the LCD. I just need to incorporate the necessary support circuitry and use a stand alone programmer to load the software. A simple change in code will allow for 9kHz channel steps used in other countries.

The one problem I still need to work on is the output transformer and amplifier circuit. I'm not able to get much power out of it at this time (a few mW at most). I need to find out why. This circuit does not have any tuned circuits, rather it's designed as a wide band linear amplifier, so it's probably not as efficient as something made to work at a specific frequency. This design is made for best audio quality without tuning being necessary for operation.

I want to add a low pass audio filter at 20kHz to keep the output within a reasonable bandwidth, and a switchable NRSC preemphasis. I wish I had a wideband AM tuner here to see how the transmitted signal sounds with full 20kHz range audio.


Top
 Profile  
 
 Post subject: Re: AM transmitter prototype
PostPosted: Dec Tue 05, 2017 4:53 pm 
Member
User avatar

Joined: May Fri 04, 2012 10:39 pm
Posts: 1392
Location: Hidden Valley, AZ
That thing is very linear, excellent characteristics! :D

It ought to sound fine.

Please don't push the positive peak asymmetry past 105%, it just distorts in the detector stage of our old school receivers.

73DG

_________________
Workin' my Gramophone in a smartphone world.


Top
 Profile  
 
 Post subject: Re: AM transmitter prototype
PostPosted: Dec Tue 05, 2017 9:50 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 3535
Erich Loepke wrote:
I wish I had a wideband AM tuner here to see how the transmitted signal sounds with full 20kHz range audio.

I have different models of the GE Superradio with wideband tuners. With wideband sources they sound close to FM. Very similar audio can be recovered with a crystal radio using a high Q coil. Maybe you can make do with that for the time being.


Top
 Profile  
 
 Post subject: Re: AM transmitter prototype
PostPosted: Dec Wed 06, 2017 1:04 am 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 2179
Location: Saskatoon
Erich Loepke wrote:
The one problem I still need to work on is the output transformer and amplifier circuit. I'm not able to get much power out of it at this time (a few mW at most). I need to find out why. This circuit does not have any tuned circuits, rather it's designed as a wide band linear amplifier, so it's probably not as efficient as something made to work at a specific frequency. This design is made for best audio quality without tuning being necessary for operation.

Antenna matching is probably the most important aspect of homebrew Part 15 transmitter designs. It will be difficult, if not impossible, to properly couple from the output stage to the antenna without a resonant matching circuit. Without that, the range can be expected to be very short.


Top
 Profile  
 
 Post subject: Re: AM transmitter prototype
PostPosted: Dec Wed 06, 2017 2:27 am 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 228
Location: Arlington, TX, USA
BobWeaver wrote:
Erich Loepke wrote:
The one problem I still need to work on is the output transformer and amplifier circuit. I'm not able to get much power out of it at this time (a few mW at most). I need to find out why. This circuit does not have any tuned circuits, rather it's designed as a wide band linear amplifier, so it's probably not as efficient as something made to work at a specific frequency. This design is made for best audio quality without tuning being necessary for operation.

Antenna matching is probably the most important aspect of homebrew Part 15 transmitter designs. It will be difficult, if not impossible, to properly couple from the output stage to the antenna without a resonant matching circuit. Without that, the range can be expected to be very short.


You're right about the antenna matching, but the problem I have right now is even with just a resistor as load it doesn't put out much power. At most I can get 10mW out of it with ~80mW in. I'd expect around 40-50mW. The power supply is +/- 12V and the linear amplifier stage runs from rail to rail so it effectively runs at 24V. I did this to stay within as much of the linear operating area as possible. I'm guessing the reflected impedance to the output transistors isn't correct for maximum power transfer. Or maybe it's just because it's built on a solderless breadboard which isn't the best thing for high frequency circuits. Im going to try to build the linear on a perforated board next to see if that helps. I did some experimentation with a Class E stage a while ago on the solderless breadboard and got really lousy operation; lots of heat and low output. Doing it on a perfboard got much better results. Maybe the same will happen with this one? Getting the circuit right is the next step before doing a PC board layout and committing the design to a PC board.


Top
 Profile  
 
 Post subject: Re: AM transmitter prototype
PostPosted: Dec Wed 06, 2017 2:33 am 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 228
Location: Arlington, TX, USA
W7TFO wrote:
That thing is very linear, excellent characteristics! :D

It ought to sound fine.

Please don't push the positive peak asymmetry past 105%, it just distorts in the detector stage of our old school receivers.

73DG


While this circuit has no problem going past 100% I personally wouldn't do it because it is just distortion. Maybe a comparator and an LED added to show overmodulation? Just how widespread is the 125% positive modulation in the commercial AM broadcast industry? Some of the stations around here just sound plain lousy on the circa 1930 radios I have here and not much better in the car. The local gospel station KHVN 970 is the worst for this.


Top
 Profile  
 
 Post subject: Re: AM transmitter prototype
PostPosted: Dec Wed 06, 2017 9:43 am 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 2179
Location: Saskatoon
Macrohenry wrote:
Do you think there will be a way to simplify the circuit by removing the Arduino and just using the preprogrammed chip with some inputs to change the frequency?

Erich Loepke wrote:
My plan was to be able to use the microcontroller as a stand alone like you are suggesting with just the three input switches (up, down, select) and the LCD.

LCD displays are so cheap nowadays, there's hardly any reason not to use them. But this got me thinking that if you're going for minimum parts count, an option would be to use BCD coded rotary "DIP" switches, and then eliminate the display, since you can dial in the absolute frequency with the switches directly, eliminating the need for a display. An advantage is that you don't have to reenter the frequency every time you power it up. The example in this photo allows direct entry of frequencies from 0 to 1590 kHz in 10 kHz steps, using the two mini rotary switches.
Attachment:
Minimal_DDS.jpg
Minimal_DDS.jpg [ 61.64 KiB | Viewed 2317 times ]

The left switch is 100 kHz increments (0 to 1500 kHz) and the right switch is 10kHz increments (0 to 90 kHz) covering all but the top of the AM band. Interface between the switches and the frequency synthesizer (AD9851 in this case) is a PIC 16F630. If barebones is what you want, this is one option.


Top
 Profile  
 
 Post subject: Re: AM transmitter prototype
PostPosted: Dec Wed 06, 2017 6:04 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 3535
BobWeaver wrote:
...an option would be to use BCD coded rotary "DIP" switches, and then eliminate the display, since you can dial in the absolute frequency with the switches directly, eliminating the need for a display.


In the context of configuring a kit that would replace the SSTrans, this is an excellent option. Some users don't need to vary the frequency enough to need an LCD, and this might provide a lower cost option for them. The LCD could be offered as an upgrade and it offers an additional cool factor.

I take it that the PIC both interprets the BCD switches and sends instructions to the DDS to set the frequency. How did you choose this PIC? How do you program it? What's its PC interface? Would an Arduino be just as easy?


Top
 Profile  
 
 Post subject: Re: AM transmitter prototype
PostPosted: Dec Wed 06, 2017 7:06 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 228
Location: Arlington, TX, USA
The operating frequency is stored in EEPROM memory internal to the microcontroller (Atmega 328, IIRC), so it power up on the stored frequency every time. The LCD really only adds a few bucks to the overall cost by itself. BCD or DIP switches make more sense with the single chip PLL synthesizers of yore or circuits like the discrete synthesizer in the SSTRAN units. The DDS chips require data to be sent over a three wire serial bus (SPI), so if using switches as an input they still need to be interpreted by a microcontroller anyway. The single chip PLL stuff is largely no longer available from first tier suppliers, so I had to reject that as a design solution from the get go. If you want one of those parts it has to come from a specialist parts supplier like Lansdale (who is just a company who buys the rights to make discontinued parts and has them made in someone else's semiconductor plant), so who knows at what price or quantity. More useful for keeping existing equipment running than for new designs.


Top
 Profile  
 
 Post subject: Re: AM transmitter prototype
PostPosted: Dec Thu 07, 2017 7:41 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 2179
Location: Saskatoon
Right. I shouldn't have implied that the Arduino won't remember the last frequency when power is cycled. I was really trying to point out that if you don't have any output display, then the absolute nature of the DIP switches is a substitute form of indication (unlike an encoder or pushbutton).

To answer Macrohenry's question about the PIC microcontroller, the one shown in my photo is a very low end version with 12 digital I/O lines. They sell for about a dollar in single quantities. A notable difference between these and an Arduino, is that the Arduinos don't require any special programming hardware. Just plug them into your computer with a USB cable. PIC's require a programming unit (PICkit2 or PICkit3) in order to program them. The programmer is inexpensive ($8.00 on Ebay), but the need for this programming hardware does tend to discourage some people from using them. The development software is free for both PIC's and Arduinos.


Top
 Profile  
 
 Post subject: Re: AM transmitter prototype
PostPosted: Jan Mon 08, 2018 4:40 am 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 228
Location: Arlington, TX, USA
I'm still working on this circuit. I have the design pretty much where I'm happy with its performance, so I did a PC board layout with everything except the power transformer, the LCD and the up/down/select pushbuttons included since they are intended to be mounted off the board. Once I check it over well and am sure the circuit is laid out correctly I will have some prototype boards made. The board is 4.8" X 6.5". I know this is a real far cry from the circuits discussed here as far as complexity goes, but I was aiming for a very high quality and easy to use unit. Antenna tuning components will be external to this box.


Attachments:
am-1_pcb.png
am-1_pcb.png [ 112.65 KiB | Viewed 1884 times ]
Top
 Profile  
 
 Post subject: Re: AM transmitter prototype
PostPosted: Jan Mon 08, 2018 5:28 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 3535
For starters, that sure is a pretty layout! It looks like art--framable, even...

I look forward to hearing/seeing how it turns out.


Top
 Profile  
 
 Post subject: Re: AM transmitter prototype
PostPosted: Jan Tue 09, 2018 3:58 am 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 228
Location: Arlington, TX, USA
I did find a few errors on the initial layout and corrected them. Next step is to create the Gerber files that the board houses use to fabricate the boards. The PCB design software creates the Gerbers, but a separate Gerber viewer is needed to be able to see the result. Sometimes there are mistakes in the layout resulting in orphaned copper areas or things like that, so this is the last step before submitting the Gerber files for board fabrication. I went with split ground planes to keep digital and analog separate. Here are the top and bottom views:


Attachments:
File comment: Top side view
am-1 top.png
am-1 top.png [ 163.75 KiB | Viewed 1848 times ]
File comment: Bottom side view
am-1 bottom.png
am-1 bottom.png [ 100.35 KiB | Viewed 1848 times ]
Top
 Profile  
 
 Post subject: Re: AM transmitter prototype
PostPosted: Jan Wed 10, 2018 12:10 am 
Member
User avatar

Joined: Aug Wed 26, 2009 1:50 am
Posts: 1395
Location: Mount Pleasant UT 84647 (from Coventry, UK)
Looks interesting. I look forward to seeing how this turns out.


Top
 Profile  
 
 Post subject: Re: AM transmitter prototype
PostPosted: Jan Wed 24, 2018 2:55 am 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 228
Location: Arlington, TX, USA
Success, well at least somewhat anyway. I misinterpreted the data sheet for the 25MHz crystal oscillator and routed the traces to it mirror imaged. The datasheet showed the pinout from the bottom of the device, not the top, so I had to mount the oscillator in an unusual way to get it to work. That bottom view is what threw me off. Oh well, I can correct this for the next run of boards if need be. The transmitter is working as expected and has quite a bit less hum now that it is built on a proper PC board. The microcontroller from the Arduino now stands alone on this board. If I can believe my frequency counter, this transmitter is accurate to less than 1 Hz. The error was a bit higher but still within the commercial 20Hz tolerance with the Chinese evaluation board I was using on the solderless breadboard earlier. I don't have the case or correct power transformer yet, so it's not 100% complete, but I'm enjoying music and OTR programming here at home as of now using this transmitter. I'm going to have to build a suitable variometer to tune these short wire transmitting antennas for better range. Currently I can receive signals on a car in the driveway but not much more than that so Part 15 compliance should not be a problem. In fact I can't detect any hum on the car radios, just those within the house, so it might be just from AC wiring, and even that background hum isn't intrusive.


Attachments:
File comment: PC board
am-1_PCB.JPG
am-1_PCB.JPG [ 193.81 KiB | Viewed 1582 times ]
File comment: AD9833 and 25MHz oscillator
AD9833.JPG
AD9833.JPG [ 150.28 KiB | Viewed 1582 times ]
File comment: the LCD
AM-1_LCD.JPG
AM-1_LCD.JPG [ 146.47 KiB | Viewed 1582 times ]
File comment: 540kHz
540khz.JPG
540khz.JPG [ 165.61 KiB | Viewed 1582 times ]
File comment: 1500kHz
1500khz.JPG
1500khz.JPG [ 158.31 KiB | Viewed 1582 times ]
Top
 Profile  
 
 Post subject: Re: AM transmitter prototype
PostPosted: Jan Wed 24, 2018 3:01 am 
Member
User avatar

Joined: Aug Wed 26, 2009 1:50 am
Posts: 1395
Location: Mount Pleasant UT 84647 (from Coventry, UK)
Nice work Erich.


Top
 Profile  
 
 Post subject: Re: AM transmitter prototype
PostPosted: Jan Wed 24, 2018 6:24 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 3535
Bravo! Keep up the good work!


Top
 Profile  
 
 Post subject: Re: AM transmitter prototype
PostPosted: Jan Wed 24, 2018 6:18 pm 
Member

Joined: Jan Thu 01, 1970 1:00 am
Posts: 228
Location: Arlington, TX, USA
Thanks guys!


Top
 Profile  
 
Post New Topic Post Reply  [ 31 posts ]  Go to page 1, 2  Next

All times are UTC [ DST ]


Who is online

Users browsing this forum: No registered users and 2 guests



Search for:
Jump to:  
























Privacy Policy :: Powered by phpBB