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


It is currently Feb Sat 04, 2023 12:11 pm


All times are UTC





Post New Topic Post Reply  [ 11 posts ] 
Author Message
 Post subject: 2N2222 based AM modulator
PostPosted: Jan Tue 24, 2023 11:27 am 
Member

Joined: Mar Thu 01, 2018 12:30 am
Posts: 1027
Location: India
From my old diary. It's a simple AM modulator where the audio signal is applied to the base and works from 800KHz to 16MHz. Any garden variety transistor can be used but 2N2222 is preferred.


Attachments:
IMG_20230124_165456.jpg
IMG_20230124_165456.jpg [ 608.19 KiB | Viewed 692 times ]
Top
 Profile  
 
 Post subject: Re: 2N2222 based AM modulator
PostPosted: Jan Tue 24, 2023 6:19 pm 
Member
User avatar

Joined: Jan Thu 01, 1970 12:00 am
Posts: 6075
Location: Austin TX
This reminds me of some modulator simulations I did using Bob Weaver's gain cell. I tried it two ways, switching the carrier and audio inputs. One modulates the carrier, the other modulates the audio at an RF frequency. Here are the circuits and graphs. They both produce similar results.

The second one is presented as a slow ramp that helped me see the range of input voltage. Neither reaches 100% modulation.


Attachments:
Image5.png
Image5.png [ 398.66 KiB | Viewed 646 times ]
Image4.png
Image4.png [ 503.98 KiB | Viewed 646 times ]
Top
 Profile  
 
 Post subject: Re: 2N2222 based AM modulator
PostPosted: Jan Tue 24, 2023 9:36 pm 
Member

Joined: Jan Wed 15, 2020 11:29 pm
Posts: 2477
Location: Queensland, AU
The problem with this sort of modulator, is that a large component of the output signal is signal addition and not signal multiplication. Multiplication does occur, just as it does in a radio mixer-oscillator, because of non linear, or square law amplification due to the transistor bias. Ideally the amplitude modulator is a pure multiplier where you can regard for example, in this case, the amplitude of the audio A, controlling the amplitude of the RF carrier R, where there is some scale factor k, for example the output equals kAR. This is why multiplier IC's or voltage controlled gain blocks of various types make great amplitude modulators for AM transmitters.

It is easy to see how you get multiplication in a non linear system for example if you add two signals A and R and amplify them in a non linear or square law manner:

(A + R)^2 = A^2 + 2AR + R^2, and the product component 2AR appears in the output. The non- linear mixer/modulator has a lot of components in its output. In the radio mixer application the product component of the received carrier and the local oscillator wave generates the sum and difference frequencies at the IF frequency.

A transistor's gain can be controlled by manipulating its emitter current and that works well when it is done in a long tail pair arrangement by another transistor. I'll post a simple circuit later that works very well as an amplitude modulator for an AM Transmitter, just have to dig it out.

It seems these days that many designers are reluctant to use transformers, but sometimes they are a very good solution. The old fashioned way of amplitude modulating a carrier was to simply amplitude modulate the power supply to the RF output stage with a secondary winding on the audio amplifier transformer. It has the advantage that the DC output to the stage is kept high, near the power supply rail because of the relatively low DC resistance of the transformer winding, and then the voltage gets modulated above and below the power supply voltage. So the modulator power also contributes to the output power. Also, it is easy to prevent modulating the carrier out, by making sure the output stage of the audio amplifier clips before the carrier voltage hits zero. Even then that is seldom necessary, because when the supply to the RF stage gets very low, it just tends to level off to a fixed residual carrier level. On the other hand using a 4 quadrant multiplier IC you can modulate the carrier out and then get a phase reverse carrier after that with excessive modulation.

The attached photo shows what happens with the vintage transformer modulator described above when the modulation level is excessive and before the audio amp has gone into clipping. The self limiting effect is not unhelpful, especially say if you are using audio source material for your AM transmitter and you have not processed that via a compressor & limiter circuit.


Attachments:
oervmod.jpg
oervmod.jpg [ 700.3 KiB | Viewed 617 times ]
Top
 Profile  
 
 Post subject: Re: 2N2222 based AM modulator
PostPosted: Jan Wed 25, 2023 12:38 am 
Member
User avatar

Joined: Jan Thu 01, 1970 12:00 am
Posts: 6075
Location: Austin TX
Charles Wenzel uses a long tail pair in the modulator of his famous Wenzel transmitter.

The transformer discussion reminds me of an asymmetrical modulation circuit I saw that uses a transformer with a resistor in one side of it so that positive peaks could be at 110% or 120% modulation while the negative peaks were suppressed by about the same amount. Supposedly this produces the fattest signal possible short of added compression, but I've never heard a recording of a 110% or120% asymmetrical modulation.

Transformers seem to have been largely forgotten in designs these days.


Top
 Profile  
 
 Post subject: Re: 2N2222 based AM modulator
PostPosted: Jan Wed 25, 2023 6:21 am 
Member

Joined: Mar Thu 01, 2018 12:30 am
Posts: 1027
Location: India
This was meant for the 40m ham band. I did achieve 100% modulation without clipping. Air band communication like those used in airplanes use positive modulation of 125% and limit the negative side to 90% for loud and clear audio. In my experiments once I was able to get 300% positive modulation using two LM386 modulators modulating both the oscillator and final stage but output power was very low, less than 100mW. The drive to the final had to be reduced greatly in order to achieve this effect. I'll post the schematic here. These days AM is just a novelty with very little practical use.


Attachments:
crystal-controlled-am-transmitter-with-100-modulation-v0-jqalvyrwig491.jpg
crystal-controlled-am-transmitter-with-100-modulation-v0-jqalvyrwig491.jpg [ 637.73 KiB | Viewed 574 times ]
Top
 Profile  
 
 Post subject: Re: 2N2222 based AM modulator
PostPosted: Jan Wed 25, 2023 7:51 am 
Member

Joined: Jan Thu 01, 1970 12:00 am
Posts: 3033
Location: Saskatoon, Canada
Macrohenry wrote:
I've never heard a recording of a 110% or120% asymmetrical modulation.

My one tube controlled carrier transmitter has asymmetrical modulation. From rough measurements off my scope, positive modulation is approximately 218% and negative modulation is approximately 91%. A video clip is on my site near the bottom of this page:
http://electronbunker.ca/eb/OneTubeXMTR_2.html
It shows the trapezoidal trace on the scope, and the audio track is as received on a nearby receiver. Note, in addition to the asymmetrical modulation, that percussive transients are inherently soft limited (as can be seen on the scope), which makes the transmitter fairly forgiving of different program material having different audio levels. I've never needed to use any audio preprocessing with this transmitter. As Ron Popeil would say, "Set it and forget it." :)


Top
 Profile  
 
 Post subject: Re: 2N2222 based AM modulator
PostPosted: Jan Wed 25, 2023 8:38 am 
Member

Joined: Mar Thu 01, 2018 12:30 am
Posts: 1027
Location: India
BobWeaver wrote:
Macrohenry wrote:
I've never heard a recording of a 110% or120% asymmetrical modulation.

My one tube controlled carrier transmitter has asymmetrical modulation. From rough measurements off my scope, positive modulation is approximately 218% and negative modulation is approximately 91%. A video clip is on my site near the bottom of this page:
http://electronbunker.ca/eb/OneTubeXMTR_2.html
It shows the trapezoidal trace on the scope, and the audio track is as received on a nearby receiver. Note, in addition to the asymmetrical modulation, that percussive transients are inherently soft limited (as can be seen on the scope), which makes the transmitter fairly forgiving of different program material having different audio levels. I've never needed to use any audio preprocessing with this transmitter. As Ron Popeil would say, "Set it and forget it." :)


Wow. The sound quality is very good!

Here's one more page that I found with dozen or more AM transmitter circuits.

http://www.g4kfk.co.uk/html/top_band.html


Top
 Profile  
 
 Post subject: Re: 2N2222 based AM modulator
PostPosted: Jan Thu 26, 2023 2:21 am 
Member
User avatar

Joined: Jan Thu 01, 1970 12:00 am
Posts: 6075
Location: Austin TX
BobWeaver wrote:
Macrohenry wrote:
I've never heard a recording of a 110% or120% asymmetrical modulation.

My one tube controlled carrier transmitter has asymmetrical modulation. From rough measurements off my scope, positive modulation is approximately 218% and negative modulation is approximately 91%. A video clip is on my site near the bottom of this page:
http://electronbunker.ca/eb/OneTubeXMTR_2.html
It shows the trapezoidal trace on the scope, and the audio track is as received on a nearby receiver. Note, in addition to the asymmetrical modulation, that percussive transients are inherently soft limited (as can be seen on the scope), which makes the transmitter fairly forgiving of different program material having different audio levels. I've never needed to use any audio preprocessing with this transmitter. As Ron Popeil would say, "Set it and forget it." :)


Thanks for reminding me of this, Bob. It does sound good.

How is asymmetrical modulation implemented?


Top
 Profile  
 
 Post subject: Re: 2N2222 based AM modulator
PostPosted: Jan Thu 26, 2023 8:07 am 
Member

Joined: Jan Thu 01, 1970 12:00 am
Posts: 3033
Location: Saskatoon, Canada
Macrohenry wrote:
How is asymmetrical modulation implemented?

In mine it's largely an artifact of the way the carrier quiescent level shifts with the audio modulation as well as the way that the soft limiting cuts in sooner for the negative excursions of modulation.

For commercial transmitters, essentially, the audio is passed through a nonlinear processor that increases the level of the positive half cycles, and decreases the level of the negative half cycles. This processed audio then modulates the carrier.


Top
 Profile  
 
 Post subject: Re: 2N2222 based AM modulator
PostPosted: Jan Thu 26, 2023 8:32 am 
Member

Joined: Mar Thu 01, 2018 12:30 am
Posts: 1027
Location: India
Set VR1 where the waveform just clips. Then set VR2 for 300% positive modulation. The 3.3K drive pot should be adjusted for 10 to 20mW output power after the push-pull final.

The circuit is a bit of an overkill as the oscillator buffer and driver transistors need to be glued to the copper clad board for heatsinking, they are passing around 40ma of standing current. The transistor biasing resistor is 2.2K and not 22K, drawing error in the schematic.

The transmitter waa tested with a 7.2MHz crystal. Audio was loud and clear. Just one of my many curious experiments in the past.


Attachments:
IMG_20230126_132707.jpg
IMG_20230126_132707.jpg [ 508.51 KiB | Viewed 436 times ]


Last edited by Dare4444 on Jan Thu 26, 2023 6:22 pm, edited 6 times in total.
Top
 Profile  
 
 Post subject: Re: 2N2222 based AM modulator
PostPosted: Jan Thu 26, 2023 3:51 pm 
Member
User avatar

Joined: Aug Tue 02, 2022 6:37 pm
Posts: 40
Location: Burke, VA 22015
BobWeaver wrote:
Macrohenry wrote:
How is asymmetrical modulation implemented?

In mine it's largely an artifact of the way the carrier quiescent level shifts with the audio modulation as well as the way that the soft limiting cuts in sooner for the negative excursions of modulation.

For commercial transmitters, essentially, the audio is passed through a nonlinear processor that increases the level of the positive half cycles, and decreases the level of the negative half cycles. This processed audio then modulates the carrier.

Your automatic carrier level control is very neat. It's based on the same principle as the grid-leak detector. I don't think this could be implemented with a single solid-state device.

Since the carrier level is shifted with the average modulation level, there is more headroom for the negative part of the modulation signal to swing, resulting in less clipping and splatter. The maximum headroom for the positive part is limited by the plate load and supply voltage.

_________________
Binh aka bb.odin


Top
 Profile  
 
Post New Topic Post Reply  [ 11 posts ] 

All times are UTC


Who is online

Users browsing this forum: Dare4444 and 2 guests



Search for:
Jump to:  
Privacy Policy :: Powered by phpBB