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


It is currently Jun Wed 20, 2018 12:45 am


All times are UTC [ DST ]





Post New Topic Post Reply  [ 24 posts ]  Go to page 1, 2  Next
Author Message
 Post subject: attempt to modulate rf with an opamp
PostPosted: Mar Tue 13, 2018 2:41 am 
Member
User avatar

Joined: Jan Fri 02, 2009 11:32 pm
Posts: 512
Location: Equinunk PA 18417
A month or two ago, Macrohenry suggested am modulation using an opamp controlled by a light dependent resistor
http://antiqueradios.com/forums/viewtopic.php?f=12&t=335809

Here's a similar approach, albeit not a wholly successful one. (My actual circuit has a few more components to accommodate a unipolar supply and some DC offset adjustment.)
Attachment:
fet-opamp modulator.jpg
fet-opamp modulator.jpg [ 21.63 KiB | Viewed 809 times ]


The circuit can only provide gain. Modulation pinch-off is not really possible, at least without further circuitry. For these traces I am using a relatively small rf sinewave. The opamp can provide plenty of gain for positive modulation and reduce gain to near-unity at maximum negative modulation. This opamp has a very wide bandwidth that goes far above the am band. Even with 100kHz rf you can see modulation is non-symmetrical, which is one of two major problems caused by the fet I used. The other major problem is more obvious in the second trace, with 300kHz rf. Severe non-summetry. That may be partially my fet, which is a relatively slow device, and most likely made worse by stray capacitance of my quick breadboard mess. Audio was 1kHz in both.
Attachment:
fet control rf modulator 1st try.jpg
fet control rf modulator 1st try.jpg [ 97.82 KiB | Viewed 809 times ]


Here's a graph of input control voltage vs. AC output voltage, both converted to a DB scale for clarity. This is the cause of some of this non-symmetry.
Attachment:
fet-opamp Cv to gain.JPG
fet-opamp Cv to gain.JPG [ 23.86 KiB | Viewed 809 times ]


Personally, I don't think I'm going to pursue this further but I thought it is interesting and worth posting about.

_________________
-Richard Fairbanks


Top
 Profile  
 
 Post subject: Re: attempt to modulate rf with an opamp
PostPosted: Mar Tue 13, 2018 4:28 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 3453
Hi, Richard, Thanks for posting this!

This circuit is on my to-do list, so I'm very interested in your results. The original circuit provided to us uses a J271 FET, which is obsolete except for SMD packaging. Its curves are very different from any FETs I have. That may be a reason your configuration isn't optimized.

I've been reluctant to order one, as I loathe working with SMD. I was going to start with biasing a bipolar transistor. When I get a round tuit I'll see if I can find your posting here so we can compare notes.


Top
 Profile  
 
 Post subject: Re: attempt to modulate rf with an opamp
PostPosted: Mar Tue 13, 2018 12:44 pm 
Member

Joined: Aug Thu 06, 2015 2:20 am
Posts: 124
Just a comment.

This FET is a depletion mode JFET, but in the circuit it is biased as a MOSFET or enhancement mode FET. The result is that the FET is saturated and can only decrease current, not increase it.

To provide linearity it should have a bypassed Source resistor to ground to make the Gate more negative than the Source so that it is in linear mode just like a cathode biased tube. Its input should also be capacitively coupled.

It should be possible to replace the 2N5457 FET with a 2N7000 enhancement MOSFET directly into the existing circuit.

An NPN transistor may also be able to be directly inserted into the existng circuit.


Top
 Profile  
 
 Post subject: Re: attempt to modulate rf with an opamp
PostPosted: Mar Tue 13, 2018 4:17 pm 
Member
User avatar

Joined: Jan Fri 02, 2009 11:32 pm
Posts: 512
Location: Equinunk PA 18417
LM386 wrote:
Just a comment.

This FET is a depletion mode JFET, but in the circuit it is biased as a MOSFET or enhancement mode FET. The result is that the FET is saturated and can only decrease current, not increase it.
Right. I mentioned I am using more components than the basic circuit I posted so as to deal with a unipolar supply as well as fet bias. I thought a more complete schematic would be confusing but here it is. I've noted the two important DC voltages, you can see my fet's gate is a couple of volts below drain/source, with audio imposed on it. Will a 2n7000 perform better here (assuming appropriate bias)? This idea is borrowed from a TI article where a p-channel jfet was suggested. I only have a few n-channel type on hand so I flipped the bias around. Note that at some point yesterday I swapped in a MPF102 in place of a 2n5457, but did not notate it on my first schematic. (MPF102 and 2n5457 both have interchangeable drain and source.)
Attachment:
fet control rf modulator 1st test1.jpg
fet control rf modulator 1st test1.jpg [ 89.66 KiB | Viewed 746 times ]


Quote:
It should be possible to replace the 2N5457 FET with a 2N7000 enhancement MOSFET directly into the existing circuit. An NPN transistor may also be able to be directly inserted into the existng circuit.
I appreciate these suggestions. I am concerned about the extremely poor modulation with higher rf frequency. I don't understand what is causing it and so I don't know if 2n7000 will give an improvement. What do you think?

Macrohenry wrote:
The original circuit provided to us uses a J271 FET, which is obsolete except for SMD packaging. Its curves are very different from any FETs I have. That may be a reason your configuration isn't optimized.
Maybe, I need a better selection of parts in my junk boxes!

Quote:
I was going to start with biasing a bipolar transistor. When I get a round tuit I'll see if I can find your posting here so we can compare notes.
That will be interesting!

The reason I am not enthusiastic about pursuing this approach is because of the built-in limitation to negative modulation depth. The opamp cannot attenuate, only amplify.

_________________
-Richard Fairbanks


Top
 Profile  
 
 Post subject: Re: attempt to modulate rf with an opamp
PostPosted: Mar Tue 13, 2018 5:25 pm 
Member

Joined: Aug Thu 06, 2015 2:20 am
Posts: 124
I am curious why the input capacitor C4 is not connected directly to the Gate of the FET?

Also, what is the purpose of R8?

The types of arrangement below are frequently used in audio compressor circuits. It might be worthwhile shunting or putting the FET in series with the RF input rather than adjusting the RF feedback gain. Shunt and series examples are shown below.


Attachments:
vca-f7.gif
vca-f7.gif [ 5.45 KiB | Viewed 734 times ]
TL084-Audio-Compressor.gif
TL084-Audio-Compressor.gif [ 18.18 KiB | Viewed 734 times ]
Top
 Profile  
 
 Post subject: Re: attempt to modulate rf with an opamp
PostPosted: Mar Tue 13, 2018 7:30 pm 
Member
User avatar

Joined: Jan Fri 02, 2009 11:32 pm
Posts: 512
Location: Equinunk PA 18417
Quote:
This type of arrangement is frequently used in audio compressor circuits.

You often see a pair of resistors like these (R7, R8) in audio compressors, as in example "A". They are intended to improve a fet's linearity when used as a variable resistor. Usual rule of thumb is to make the resistors equal value, which places onto the gate about 1/2 of the AC signal present at the drain. Your example "B" accomplishes the same thing but allows a lower impedence DC control signal onto the gate. The actual ratio of source-gate AC signal varies from one fet to another and so can be tweaked a bit for lowest distortion. Non-linearity increases with increasing AC level, particularly above 100mv. That is why in my test circuit there are two feedback resistors (R4, R5) around the opamp. The fet "sees" a much reduced voltage and, in theory, its linearity benefits from the relatively small voltage it has to control. Or, so the story goes.

Quote:
It might be worthwhile shunting or putting the FET in series with the RF input rather than adjusting the RF feedback gain.

My thinking is that stray rf leakage should be less of a problem this way.

I re-read the AD844's datasheet and realized that I might have better success feeding rf into the non-inverting input. This reconfigured circuit has good RF bandwidth up to around 8mHz, drops off above that. Modulated audio bandwidth is flat to 100kHz, the limit of my generator. Of course, the antenna system will limit audio bandwidth far below that number. I adjusted R6 by measuring unmodulated maximum rf voltage output and setting R6 so the output drops to 1/2 of that.

I borrowed my antenna and matching network from our little modified ECS transmitter, and tuned up at 1Mhz. Sinetones at low levels sound pretty good up to about 50% modulation. As feared, distortion picks up considerably as modulation nears (but does not achieve) 100%. Stray rf is an issue and, for purposes of this test, I put a 100pf cap from gate to ground, which helped. Also, having the antenna tuned affected the adjustment of R6.

I'm sure some optimizing can be done, if one really wanted to.

<schematic deleted... see my next post>

_________________
-Richard Fairbanks


Last edited by richfair on Mar Tue 13, 2018 10:07 pm, edited 1 time in total.

Top
 Profile  
 
 Post subject: Re: attempt to modulate rf with an opamp
PostPosted: Mar Tue 13, 2018 7:40 pm 
Member

Joined: Aug Thu 06, 2015 2:20 am
Posts: 124
Clever guy!

I learned something about linearizing compressor circuits today.


Thanks


Top
 Profile  
 
 Post subject: Re: attempt to modulate rf with an opamp
PostPosted: Mar Tue 13, 2018 10:35 pm 
Member
User avatar

Joined: Jan Fri 02, 2009 11:32 pm
Posts: 512
Location: Equinunk PA 18417
The schematic in my first post is practically a lift from Ron Mancini in a TI article "Op amps for everyone". Thank him!

I didn't plan to spend any more time on this but I did and so here is my last version. It works well enough to prove the idea works. A higher supply voltage would allow higher output voltage, this chip is rated -+4.5 to -+18, with 20mHz bandwidth at 20v p-p output (you'll need -+15 to achieve that). I do not know what current draw is I did not measure. My 9v single supply is the bottom end of this opamp's supply needs and my broadcast range is limited. The opamp is a linear device, rf sinewaves are best. It is happy with the miniscule rf output from my heathkit generator but hated the square waves from our ECS oscillators no matter what I tried. My matching circuit (well, actually, 35Z5's matching circuit!) develops about 30v p-p at the antenna when driven by this circuit with 9v. That's not a lot but its enough to satisfy a radio in the same room.

I replaced the MPF102 fet with a 2n5457, which works better here. It requires a bit more audio drive but gave me lower distortion in return. The LM386 would drive it very easily, but in that case the 1uf cap should be flipped around so + is connected to the LM386 output.

edit: add info to 2nd paragraph, corrected schematic mistake


Attachments:
fet-opamp modulation.jpg
fet-opamp modulation.jpg [ 55.31 KiB | Viewed 699 times ]
fet control rf modulator test3.jpg
fet control rf modulator test3.jpg [ 84.21 KiB | Viewed 658 times ]

_________________
-Richard Fairbanks


Last edited by richfair on Mar Wed 14, 2018 1:12 pm, edited 4 times in total.
Top
 Profile  
 
 Post subject: Re: attempt to modulate rf with an opamp
PostPosted: Mar Wed 14, 2018 4:01 am 
Member
User avatar

Joined: Jan Thu 01, 1970 1:00 am
Posts: 3453
Excellent work!


Top
 Profile  
 
 Post subject: Re: attempt to modulate rf with an opamp
PostPosted: Mar Wed 14, 2018 4:04 pm 
Member
User avatar

Joined: Aug Sun 02, 2015 7:34 pm
Posts: 875
Location: Yukon, OK USA
Glad to see you guys working on this. I like the idea of a sine wave from a signal generator as the source and not needing to build an oscillator. Interesting approach to modulation as well.

I try to learn from circuits I see and have a question. The FET has its source through capacitor C4 to ground. Can you describe the operation of this design for me? I see the effect on the gain of the op amp but don't know enough about FETs to understand how it functions.

_________________
Rick
AB5ON

Forest Gumping through life


Top
 Profile  
 
 Post subject: Re: attempt to modulate rf with an opamp
PostPosted: Mar Wed 14, 2018 6:27 pm 
Member

Joined: Aug Thu 06, 2015 2:20 am
Posts: 124
The FET is acting as a variable resistor.

Many JFETs may be used with their Source and Drain interchanged.

In this circuit the Source (Drain) is at about 4.3V as a result of the opamp biasing.

The potentiometer is adjusted so that the Gate is at a lower voltage (3V) than the Source, like the cathode biasing in a tube circuit. This increases the FET resistance and sets the opamp carrier output to 50%.

The only trick in the circuit which was a learning experience for me are the 2 100K resistors connected to the FET. Apparently, they help linearize the resistive transfer curve of the FET which is normally logarithmic. The audio input varies the FET's resistance which modulates the carrier by changing the opamp gain.

The graph below is a normalized view of FET resistance vs. Gate to Source voltage with zero Drain to Source voltage. The FET resistance can vary more than 1000 to 1, but in compressors is usually used at lower ratios for linearity reasons.


Attachments:
JFET Resistance Ratio.jpeg
JFET Resistance Ratio.jpeg [ 53.67 KiB | Viewed 648 times ]
Top
 Profile  
 
 Post subject: Re: attempt to modulate rf with an opamp
PostPosted: Mar Wed 14, 2018 7:41 pm 
Member
User avatar

Joined: Jan Fri 02, 2009 11:32 pm
Posts: 512
Location: Equinunk PA 18417
Article about fets as voltage controller resistors that explains why there is non-linearity and how those two resistors improve it: https://www.vishay.com/docs/70598/70598.pdf

The above circuit should function with other opamps at lower rf frequencies, if you want to play with it. Something like an NE5534/5532 or TL082 will have trouble with rf above 100kHz but that's still good enough to see how it works. The value of an AD844 is an extremely large slew rate and bandwidth at full output, but OMG I just looked it up on Mouser. My heart skipped a beat when I saw the price of a ceramic 8-dip version. Plastic 8-dip is less, heart only races, but is EOL. I happen to have a few left over from a project years ago.

_________________
-Richard Fairbanks


Top
 Profile  
 
 Post subject: Re: attempt to modulate rf with an opamp
PostPosted: Mar Wed 14, 2018 8:12 pm 
Member
User avatar

Joined: Aug Sun 02, 2015 7:34 pm
Posts: 875
Location: Yukon, OK USA
Thanks for the link. Even with the explanation I am still not grasping why the source does not need a D.C. connection. Is it suppling the R component of an RC circuit along with C4?

_________________
Rick
AB5ON

Forest Gumping through life


Top
 Profile  
 
 Post subject: Re: attempt to modulate rf with an opamp
PostPosted: Mar Wed 14, 2018 8:36 pm 
Member

Joined: Aug Thu 06, 2015 2:20 am
Posts: 124
The Source does have a DC connection. It is set to about 4.3V. The Source in this circuit is labelled as the Drain. They are interchangeable.

It is the Drain (labelled as the Source) which doesn't have a DC connection.

The graph shows the normalized resistance of a FET vs. Vgs when the Drain to Source voltage is zero volts, which it is in this case because a capacitor is connected in series with the Drain (labelled as the Source).

The FET may be physically picked up on the circuit board, rotated and resoldered so that the Source and Drain are interchanged and the circuit will behave the same.

It might be easier to understand if the schematic were labelled with the Source and Drain labels interchanged.


Top
 Profile  
 
 Post subject: Re: attempt to modulate rf with an opamp
PostPosted: Mar Wed 14, 2018 8:49 pm 
Member
User avatar

Joined: Jan Fri 02, 2009 11:32 pm
Posts: 512
Location: Equinunk PA 18417
Quote:
Can you describe the operation of this design for me?

Quote:
Even with the explanation I am still not grasping why the source does not need a D.C. connection. Is it suppling the R component of an RC circuit along with C4
I was working up a more complete explanation for anyone who is interested and I'll answer this question at the end.

I'd like to go deeper into fet's operation. First, the opamp is there to amplify the rf input signal. Like most opamp circuits, this one uses negative feedback (through resistors R4 and R5). Negative feedback reduces the opamp's gain. More feedback=lower gain. The fet, as a variable resistor, can shunt to ground a portion of the feedback signal so less of it appears at the opamp's inverting input. So, as the fet's resistance is lowered, the feedback is also reduced, which increases the opamp's gain. The audio signal is applied to a steady dc voltage on the fet's gate, and as the gate's voltage rises/falls with the audio signal, the fet's resistance changes, which changes the amount of negative feedback, which changes the amount of ampllification of the rf signal.

As mentioned before, fets have some linearity problems. Those problems are reduced when the drain-source voltage is kept below a couple hundred millivolts. The most clever part of this circuit is that R4, R5 form a voltage divider between the output's full AC voltage and the inverting input's near-zero AC volts. Only millivolts appear at the connection between R4 and R5 so the fet is not stressed badly into non-linearity, yet the fet can easily shunt that AC signal to ground and is thus able to control the amount of negative feedback.

So, to answer WHY there is no DC connection from fet's source to ground. There are two reasons. First, I only want the fet to alter AC signals in the feedback loop. There is a DC component of the feedback signal that I wish to leave untouched, since doing so will greatly help circuit stability. To leave DC alone, a capacitor must be placed somewhere in the connection of FB loop-fet-ground. The other reason? I specifically choose to put the cap between fet and ground so that the fet's source (and drain) will be at the same DC voltage as the opamp's feedback loop (which is determined by the opamp's output), which is about 4.5 volts half way between 0 and 9 volts. Why is that desireable? It is because the fet's operation is controlled by gate-source voltage, and for this circuit to work right, the gate needs to be more negative than the the source. As long as the source (and drain) remain above 0 volts, I can make the gate voltage lower without a separate negative supply. Does that make sense? Although my trimmer can make the gate anywhere between 0 and +9, the useful area is below 4 .5 volts.

A DC connection would be possible, but not required, if I had built this using bipolar supplies.

I think circuit layout will be an important consideration, to minimize stray rf leakage into the fet and around the opamp. The opamp's datasheet strongly recommends decoupling the power supply pins and mentions a couple of ways to do it.

I'm not real good at these explanations so please let me clarify if it is confusing!

_________________
-Richard Fairbanks


Top
 Profile  
 
 Post subject: Re: attempt to modulate rf with an opamp
PostPosted: Mar Wed 14, 2018 9:05 pm 
Member
User avatar

Joined: Aug Sun 02, 2015 7:34 pm
Posts: 875
Location: Yukon, OK USA
Explanation is perfect. It was my lack of knowledge that was the problem. The key for me was shunting AC signal to ground without changing the DC component. Then the light bulb came on. :oops:

Thanks!

_________________
Rick
AB5ON

Forest Gumping through life


Top
 Profile  
 
 Post subject: Re: attempt to modulate rf with an opamp
PostPosted: Mar Thu 15, 2018 3:20 am 
Member
User avatar

Joined: Jun Sun 22, 2008 2:03 am
Posts: 1258
Location: Burke, VA
That is a nice circuit!

I worked out some equations to verify that the gain of the op-amp is a linear function of the audio modulation voltage (Vgs).
The gain is defined by R4 and Rds (the JFET drain-source resistance). It's linearly proportional to R4 and inversely proportional to Rds:

Gain = 1 + R4/Rds

Since Rds is inversely proportional to Vgs, the op-amp gain is linearly proportional to Vgs. As long as Vds is kept under 0.1V, the nonlinearity of Rds works in our favor and I don't think there's a need to linearize Rds by adding the negative feedback R7 and R8.

I made some modifications to the circuit. The mid-voltage divider formed by R1 and R2 creates a virtual ground for the op-amp. This virtual ground is bypassed by a ceramic capacitor C4 for RF signal and an electrolytic capacitor C5 for AF signal. The op-amp's inputs and the JFET source are all referenced to this virtual ground. R3 connects the op-amp's noninverting input to the virtual ground and its value should be equal to the output impedance of the RF source. R5 has the same value as R3. The JFET source is directly connected to the virtual ground (without going through a capacitor). R2 is changed to a variable resistor which is used to adjust the negative bias for the JFET gate to set the RF output level with no modulation.


Attachments:
Op-Amp AM Modulator.png
Op-Amp AM Modulator.png [ 34.19 KiB | Viewed 593 times ]

_________________
Binh
Top
 Profile  
 
 Post subject: Re: attempt to modulate rf with an opamp
PostPosted: Mar Thu 15, 2018 10:46 am 
Member

Joined: Aug Thu 06, 2015 2:20 am
Posts: 124
Just a question triggered by the clever analysis above.

Over the years I have built numerous regenerative receivers in the medium wave band using only an LM386 audio amplifier or a TL431 shunt regulator as the oscillator.

Some manufacturer's LM386s can be made to oscillate at frequencies higher than 6 MHz and I have been successful using an LM386 as an 80 meter regenerative receiver with the TL431 used as an audio amplifier.

I don't have the resources to try the experiment, but I was hoping others interested in the topic might chime in.

Instead of using an expensive opamp as the transmitter, what are the chances of using a circuit similar to that used in this thread to modulate an LM386 audio amplifer? It is not truly an opamp, but might provide enough output at 1 MHz to become a useful device.

The LM386 gain may be directly modulated in a number of different ways. One method would be to increase the gain by shunting part of the built in gain determining negative feedback path on Pin 1 to ground.

The LM386 low frequency gain may be increased to over 70dB by removing one end of the gain enhancing capacitor going to to Pin 8 and adding a series resistor to ground.

For interest I have include a schematic of the 80 meter regenerative receiver below which uses the LM386 as the oscillator and a TL431 shunt regulator as an audio amplifier.

In the graph below it may be seen that the LM386 has gain above 1 MHz, but the Samsung versions can be made to oscillate at over 6 MHz.

The TL431 shunt regulator with its high gain, high input impedance and low output impedance may also hold some promise as a medium wave transmitter modulated using the ideas in this thread. I have had them oscillating over 2 MHz. They are basically an opamp with its output buffered by a transistor and a 2.5V internal reference going to the hidden inverting input.

The TL431 does not have an active pull up and would require a pullup load which could be part of the antenna matching circuit.


Attachments:
lm386 80m regen 2.png
lm386 80m regen 2.png [ 82.89 KiB | Viewed 574 times ]
lm386 gain.jpeg
lm386 gain.jpeg [ 13.11 KiB | Viewed 574 times ]
TL431.png
TL431.png [ 4.68 KiB | Viewed 574 times ]
lm386 schematic 2.png
lm386 schematic 2.png [ 25.96 KiB | Viewed 571 times ]
Top
 Profile  
 
 Post subject: Re: attempt to modulate rf with an opamp
PostPosted: Mar Fri 16, 2018 1:45 am 
Member
User avatar

Joined: Jan Fri 02, 2009 11:32 pm
Posts: 512
Location: Equinunk PA 18417
Thank you both for your interest in this little sideshow. Funny, I thought I had brushed my hands of it. Nope.

Binh, your rework is very smart especially taking fet gate bias from the lower half of the virtual ground divider. Much better for both bias range and not loading the audio source as my first attempt did. I worked on a different project today but soon I'll re-build your circuit because there are several things I want to check on. Earlier, I saw no need for a large cap between fet and ground. It is rf that is being shunted to ground. The other thing I am not sure of is whether a pair of resistors around the fet will increase fet linearity. I believe they will. Even though rf voltage at the fet is quite small (well below 200mv in my circuit), the audio voltage is much higher and therefore I believe fet non-linearity is still a strong consideration. Testing will answer the question. Third, I experimented with R5's value a bit and had poor results using values much lower than 200. I dont' know why. I like your effort to balance input currents using the terminating resistor value, which is supposed to be 50 ohms for this generator (and many others). I often don't bother to terminate properly out of laziness and the fact that my generator is a hobby grade device anyway, so thanks for reminding me!

LM386, opamps tend to run out of useable gain as the signal frequency increases. Many audio opamps have bandwidth that far exceeds the audio range so long as gain is kept low, so spec sheets often boast about bandwidth at unity gain. Run of the mill audio opamps can keep up at1 mHz and some more, as long as they don't have to amplify the signal much, but this circuit requires lots of linear gain. Regenerative feedback needs far less gain, and although an opamp can oscillate at 6mHz it doesn't mean it can perform as it was meant to at such frequencies. Do you have an rf generator? If you do, hook up this circuit using whatever you have on hand instead of the AD844. It'll be fun to play with but I'll be surprised if you get useable results in the broadcast band. But, I may well be wrong.

_________________
-Richard Fairbanks


Top
 Profile  
 
 Post subject: Re: attempt to modulate rf with an opamp
PostPosted: Mar Sat 17, 2018 3:55 am 
Member
User avatar

Joined: Jan Fri 02, 2009 11:32 pm
Posts: 512
Location: Equinunk PA 18417
bb.odin wrote:
Since Rds is inversely proportional to Vgs, the op-amp gain is linearly proportional to Vgs. As long as Vds is kept under 0.1V, the nonlinearity of Rds works in our favor and I don't think there's a need to linearize Rds by adding the negative feedback R7 and R8.

I made some modifications to the circuit.

Hello Binh, Your changes, all of them, work great! Thank you! Regarding the pair of feedback resistors around the fet, I was surprised to see no useful difference with or without (once bias was readjusted to compensate). I need to study more to figure out what you already know about that. I had thought the voltage swing of audio on the gate would be a factor. It is not and so there is something basic about fet operation that I don't understand.

C5, the 100uf cap, makes no practical difference but it does remove a trace of audio (couple of millivolts) that would otherwise be present at the opamp's output (but not radiated by an antenna), so I agree it should be included. The wire from fet source to virtual ground is, I feel, optional as long as there is a cap from the fet's source lead to ground (as in my original attempt). Modulation works equally well with or without that DC connection.

This circuit draws nearly 6ma quiescent current from a 9v supply, without an antenna. With the antenna network from our ECS transmitter connected and tuned, current draw rises to 11ma.

_________________
-Richard Fairbanks


Top
 Profile  
 
Post New Topic Post Reply  [ 24 posts ]  Go to page 1, 2  Next

All times are UTC [ DST ]


Who is online

Users browsing this forum: fifties, Olden Broak and 6 guests



Search for:
Jump to:  
























Privacy Policy :: Powered by phpBB