40m class E CW transmitter kit NS-40 with an audio modulator

[ka1tdq]

Here's another idea I was thinking about that might work.  4 States QRP sells an inexpensive 40 meter CW Class E transmitter kit.  I put mine together and it works fine with 5 watts output/12 volts regulated input.  It is sitting on the right side of the desk in the picture.  I connected it to my 3-500 amp and I'm getting 92 watts output on 40 meters.

How about audio modulating it with a simple analog modulator?  Then use another crystal at 7.290 MHz.  I've attached a picture of the schematic.  I think you would just need to break the 12 volt line going to the choke on the output MOSFET.  Set the carrier at around 8 volts and my amp would put the output around 50 watts carrier.  Not screaming, but it's simple and inexpensive.  

Steve gave me a schematic for a simple audio modulator for a previous project.  I think I could just use that here.

Jon
KA1TDQ

[AB2EZ]

Jon

This is a neat little transmitter, and it should be fun/educational to try to AM modulate it.

A couple of important issues:

The output circuit of a class E transmitter has to be tuned to the frequency of operation; and the loading capacitor has to be properly set to produce a proper class E drain voltage waveform (or adjusted for maximum efficiency. The tuning and loading capacitors in this transmitter have fixed values (450pF and 1000pF). If you change frequency to 7290kHz, they will have to be changed... and really should be replaced with variable capacitors if you want the transmitter to operate at high efficiency (i.e. to actually operate in class E).

You may have problems on negative peaks of the modulated drain voltage if it falls below around 2V. You will have to try various combinations of carrier level drain voltage and modulation depth to obtain a decent sounding AM signal. You may have to settle for less than 100% modulation.

Good luck!

Stu

[steve_qix]

The IRF510 is a small, relatively low voltage MOSFET (100V max).

So, depending on how well designed the RF amplifier is, you can expect approximately 4x the applied DC (or peak audio) voltage to appear across the MOSFET under the most ideal conditions. If something is tuned wrong, etc.  the voltage will be higher.  This is why I use a 900V MOSFET in my designs, even though the calculated peak drain voltage is about 450V.  It can get a lot higher if there's an anomaly with the antenna, tuning, etc.

So, 12V to 14V is a good maximum voltage for an IRF510.  If you're running a 14V supply, the carrier voltage would be more like 5 or 6VDC to leave plenty of headroom for positive peaks.  This is very low voltage.

The oscillator may also drop out of oscillation (as Stu pointed out above) when the voltage falls too far.  In the case of this transmitter, it would be a modulated oscillator.

Regards,

Steve

[AB2EZ]

Steve
Jon

As a clarification:

I believe that Jon proposes to only modulate the drain voltage of the amplifier's FET (not the oscillator's FET).

Regarding my comment about low values of the drain voltage (on negative modulation peaks):

I was thinking that the FET would require some minimum value of (modulated) drain-to-source supply voltage to function properly in this application.

However, thinking more about this... perhaps it is okay if the modulated drain-to-source supply voltage drops within a few hundred millivolts of 0V on negative modulation peaks. The modulation linearity (RF envelope amplitude vs. audio voltage amplitude) in the region of the negative modulation peaks might suffer a bit... but the effect might not be noticeable.

Stu

[ka1tdq]

That's true, I only planned on modulating the final and leaving the oscillator at 12 volts. My previous project's modulator did have a negative peak limiter too. I think I'm going to leave this as a CW transmitter though. I just added an FT243 crystal socket for more frequency options.

What I'm trying to do is move my operating activity to 40 meters. I think I got spoiled in New England listening to megawatt carriers with exceptional audio. Everybody seems to be pw in the southeast. I think what I will do is convert my single tube AM transmitter from 75 to 40 meters and use it with my dual 3-500 amp. That gives me both AM and CW options. I just need to come up with a 2 watt 40 meter driver or use a 5763 tube as a crystal oscillator.

jon
KA1TDQ

[ka1tdq]

Southwest... potato potaaahto

[DMOD]

Hi Jon,

Here is a paper by David Cripe on ClassE QRP transmitters.

I worked with him at RC until last year.

Phil - AC0OB

[ka1tdq]

I'm going to print that off and read it when I can totally concentrate on the math to appreciate the theory.  My boy was just born on Thursday, Wyatt, and I'm just trying to get in sleep here and there now. I have enough homebrew stuff in my shack now to spark some curiosity about electronics and radio.

For now though I think I will just build a small wattage driver for my single tube rig.  I will borrow the oscillator circuit from the NS-40 to get enough gate drive.  I will use an 11N90 for the final FET in order to handle the larger voltage that I will put into it (up to 50 volts). 

Then I just need to get a crystal for 7.290 MHz and I'm on 40 meters. 

Jon
KA1TDQ

[DMOD]

Congratulations on your son's birth Jon.

Our two children were born 11 months apart so I feel your need for sleep. 

Phil - AC0OB

[flintstone mop]

Slightly off-topic...BUT I love the desk and chair...From an executive's office of a now-defunct business?
Nice layout there. A very nice OFFICE ham shack. Nice work getting the amplifier to look so classy.
Now radio-wise question.
Those nice big bottles and 90 watts or 50?...You put a lot of work into the amp and I thought you were going for nirvana. Just kidding
Fred

[ka1tdq]

Yeah, what was I thinking?  CW?  

I just finished making the 40 meter band mods to my single tube AM transmitter and the stability problem I had before is gone.  I'm using the NS-40 to drive the grid and I'm using a 3db attenuator pad to bring grid power down to about 2+ watts.  I ordered a $12 Chinese audio amp from eBay to drive the Heising circuit.  I'm skeptical at $12, but it could work okay.  I'm backwards feeding an 8 ohm audio transformer in the modulator.  It sounded great back in Massachusetts on 75 meters, now I just need to get a working audio modulator again.

Now I have a working 40 meter antenna, transmitter and amp.

It's funny.  When I moved down here and was looking for work, I sold off lots of radio stuff to get things like tires and food.  Now I'm building all that stuff back again.

On the office furniture, I got the desk for free from Craigslist from a local business and the leather chair is $14 from a Goodwill thrift store.  

Jon
KA1TDQ

[N2DTS]

What are you using for a receiver?
Yes, nice looking shack.

I will listen for you on 40, I get on weekend mornings in the winter.

[ka1tdq]

I'm using an Alinco DX-R8 with a separate long wire antenna for receive.

I just need to make sure my Chinese audio amp works and then get a crystal for 7.290 MHz from AF4K.

Jon
KA1TDQ

[N2DTS]

Lots of nice audio amps on ebay cheap, and this one looks very interesting!

http://www.ebay.com/itm/Large-Power-Vehicle-Amplifier-With-Card-Reader-USB-And-FM-Radio-Digital-Display-/400666137695?pt=US_Ham_Radio_Amplifiers&hash=item5d4990145f

[ka1tdq]

The one I ordered is supposedly 100 watts per channel and runs off of 12 volts. I will just use 1 channel and I don't need nearly that much power, so it should be fine. I should have it in a few days.

I just ordered my 7.290 MHz crystal so that's taken care of too.

Chinese stuff really isn't all that bad either. I've found that it's sometimes cheaper to order stuff from the slow boat than domestic. I bought a Baofeng 2m/440 dual bander with the desk charger, programming cable and speaker/mic for $54 delivered. I got it in a week and it works great!

[ka1tdq]

I finally have everything working on 40 meters.  I ended up using a car audio amplifier for the modulator.  I used a dummy load (because also I was transmitting on 7.030 MHz) and listened to myself on the receiver.  It sounds pretty darn good, if I may say so myself! 

I'm getting 300 watts carrier using the amplifier.  Voice peaks may not be as high as I would like as I notice a little distortion if I drive it really hard.  I just cut the audio back to where it sounds nice and linear and full. 

I'm just waiting for my 7.290 MHz crystal to arrive in the mail and I am on - dah - ayuh!

Jon
KA1TDQ

[steve_qix]

Looks good!  Can't say sounds good yet - haven't heard it    (yet)

[K1JJ]

I'm getting 300 watts carrier using the amplifier.  Voice peaks may not be as high as I would like as I notice a little distortion if I drive it really hard.  I just cut the audio back to where it sounds nice and linear and full.  

Jon
KA1TDQ

Hola Jon -

Cool looking white panel - my favorite!

To set the carrier/ peak power -  first find out how much the rig will put out maximum.  If this is 1000w (1000 w pep) then your carrier should be about 200 watts or 1/5th the peak power.   This will allow for a normal asymmetrical voice.

By doing the opposite - IE, starting with a fat carrier and trying to match proper peak power invites distortion or low positive peak modulation %.


T

[ka1tdq]

Ok, because I just took a stab at it. My tube rig will put out more or less by varying the loading capacitor. I have it set right now between 15 or 20 watts going into the amp. I can make the carrier between 200 or 400 watts-ish too.

I'm hoping that my crystal comes in Monday.  A few nights ago in the early evening I heard a guy from the east coast running a Valiant on 40 meters. I couldn't make out too much since signals were still down with the sun being up. I think this is going to prove to be a fun little rig.

Jon
KA1TDQ

[AB2EZ]

Jon

I think that you already know this, but just to be sure:

The proper way to tune and load a linear amplifier (a separate issue from how to operate it on AM after it is tuned and loaded) is as follows

Adjust the RF input level, and tune/load the amplifier to produce the maximum output power that the amplifier can safely deliver. I.e. you want the tuning/loading to be correct for the peak power output you will produce on AM.

Then, as Tom has pointed out... reduce the RF input so that the amplifier is putting out about 1/5 the peak power that it was putting out when you tuned/load it.

If you don't tune/load the amplifier this way (i.e. if your tune/load it a carrier level output, instead of peak power output), then you will run out of "headroom" on positive AM peaks... and your modulation will be distorted.

Separately... running a linear in AM mode is hard on output tubes, because the efficiency of the output stage is very low at carrier. For example, if the amplifier is properly adjusted (as above), then the efficiency of the amplifier at carrier will be only around 30%. This is normal for a linear amplifier in AM operation... and follows from the theory for a linear amplifier operating in class AB.  This means that if your AM carrier level is 300W, then the plates of the tubes will be dissipating 700W or more when you are at carrier level.This will make the tubes turn red in a hurry. Even if you run the amplifier (properly tuned and loaded) at 200W carrier, the tubes will be dissipating 467W (or more). The pair 3-500Zs can handle 467W of plate dissipation... but they will get very hot on long transmissions. Then when you "turn it over" to the other party, the tubes will cool back down.

This cycle of heating and cooling will take its toll on the 3-500Zs. The plate structure may separate from the plate pin due to metal fatigue or the glass envelope may develop a crack around the plate cap.

When running at carrier, keep an eye on the plates of the tubes. If they are glowing red, then they are not going to last very long. Make sure that you have good air flow over the tubes.

Separately, for a grounded grid amplifier, the load presented to the driver increases (higher impedance) and decreases (lower impedance) as the input (driver) power goes down and up.
Since the load (impedance) that the driver is looking into is changing, it would not be surprising if the output of the amplifier does not faithfully track the modulation that you see on the output of the driver when it is looking into a fixed load. You could mitigate this in a number of ways, but that is a topic for another day.

Stu

[ka1tdq]

Thanks for the info. Yes, I do need to reduce carrier power to get the level of positive peaks I want without distortion.

I did let this go key down for 1 minute at 300 watts and the tubes didn't get red. I have made them red with my previous amp, so I will definitely watch out for that.

A good resonant antenna (which a dummy load is) helps too. I also need to trim my dipole to resonate at the upper side of the band rather than the CW portion.

I have separate antennas for RX and TX too. I also need to install a relay on the receiver antenna input so that I don't kill the front end during TX. I have that in my parts box though and that's a quick fix.

Jon
KA1TDQ

[AB2EZ]

Jon

Keep in mind the following:

If you tune/load a class AB amplifier for maximum output, with the input drive level adjusted so that the output is 300W, then the amplifier will be operating at around 60% efficiency (i.e. 300W RF output power, 200W total plate dissipation, 500W total input power). Therefore the plates of the 3-500Z's will be dissipating around 200W total. However, you won't have any headroom for positive modulation.

If you tune/load the amplifier for maximum output, with the input drive level adjusted so that the output is 1200W ... and then reduce the input drive level to back down the output to 300W ... then you will have enough headroom to produce 100% positive modulation peaks (i.e. 1200W peaks). On 100% modulation peaks, the amplifier will (as above) be operating at around 60% efficiency (i.e. 1200W RF output power, 800W total plate dissipation, 2000W total input power). But, when at carrier level (300W) output, the amplifier will be operating at around 30% efficiency (i.e. 1000W total input power, 300W RF output power, 700W plate dissipation)

Bottom line:

The tubes will dissipate a lot more power sitting at 300W carrier, with the amplifier tuned/loaded to accommodate 1200W positive peaks vs. the power the tubes would dissipate sitting at 300W output if the amplifier is tuned/loaded for maximum output at an output level of 300W.

Stu

[ka1tdq]

Ah, ok! I never knew that. I will do just that and then crank the audio as much as I can.

Back in Mass, I was tuning for 300 watts on 75 meters and left it that way. I apparently had more room to go for audio if I had tuned for maximum power first. I do have a negative peak limiter installed on the transmitter too.

Jon
KA1TDQ

[ka1tdq]

I tuned the transmitter for maximum output, and it's not a whole lot higher at 410 watts with the amp.  The transmitter seems to modulate best a little lower at 25 watts (300 watts with the amp) and I get around 900 watts peak with undistorted audio.  I can lay into it and get 1200 watts but I get distortion.

Also, it's a tradeoff between the amount of bass I put in as well.  More bass equals more distortion at higher power levels.  I could crank it down, but then I'd end up with audio that makes the receiving station want to hurl.

Jon
KA1TDQ

[K1JJ]

I get around 900 watts peak with undistorted audio.  I can lay into it and get 1200 watts but I get distortion.

Jon
KA1TDQ

That's the bottom line... the way the rig stands right now, you are able to get 900 watts peak out with undistorted audio.  This means you should run the carrier at 180 watts out and everything will be happy at a clean 1: 5 carrier to peak power ratio.  You will have plenty of headroom and have a clean signal.   (180 watts carrier and 900 w peak)

BTW, play around with audio tones and the plate variable loading cap (C2) trying to get cleaner, higher peaks on the scope. And maybe the L/C ratio of the tank needs a tap change. A pair of 3-500Zs should be good for a clean 1400 watts peak using 3KV anyway.

T