Well here she is, 10 watts of pure AM 100+% modulated...
Front of it so far, no labels yet but the controls are from left to right.. modulation level, modulation % bias,set frequency (base frequency set inside), bandpass filter tune, and of course power switch
You can see I even added plugs to the transistors rather than soldering them in because you just never know when something bad may happen I'm not worried though, this rig runs cool, just wanted to be a bit anal about things.
I plan on adding a SPOT switch yet to turn off the PA and modulator to spot the VFO on frequency before going on air. Also I plan to add an RF meter possibly.
Puts out a solid and I do mean SOLID 10 watts completely 100% modulated signal. It's possible to get up to 125% modulation with much less power and 15+ watts with less modulation which is all variable via the % control on the panel. Best modulation is right around 8 watts to get that extra punch. Modulation out of this thing sounds like a dream. I used a series modulator. The modulator is the two unseen transistors under the heat sink on the back. They need to be fanned down more than the RF stages does! But it's worth it because the audio can push the limits of AM. It almost sounds like FM on classic radios that don't cut the bandwidth but to keep within limits I cut the audio at around 10kHz on an EQ outside the unit.
I still have to build the antenna jack and that loop of wire dangling is the audio input. I still need to solder that into the volume pot and put an audio jack somewhere. The audio in is around 10ohms but very sensitive. Line level works but I need to get a 600ohm to 8ohm transformer or build some kind of a preamp that will get the job done.
The whole circuit is linear from the first stage. Very inefficient but I don't care because the quality modulated signal is worth it and that pure sign wave causes NO spurs or bogus signals. This thing is as pure as I could build a transmitter from front to back.
In the future I may add a good 100 watt linear amp to it but that may be overkill. This thing will easily cover my whole town as is and punch through the static with a fist of quality AM.
Schematics so far, not including power supply yet..
I will paint the schematics on the computer when I get a chance, but this gives an idea behind the circuits. Some of the parts in this rig came from my previous one which was dismantled to make way for this better one. The 4w output was just wrote down in the design stage. After adding the heat sinks and power supply guts I easily got 10 watts out carrier and 40 watts at 100% peaks. The modulator is two identical fat PA transistors connected in series and then that connected to another smaller transistor as a darlington pair. Can't wait to put this beast on the air. More to come.
I am actually not sure as they were scrapped from some old computer power supplies. The small ones work exactly the same as the standard T-50-2 so my guess is they are almost the same. The larger one I have no clue but it works better than good so I used it. I was going to use another one of the small ones for the output but they just couldn't handle the full 10 watt load w/o getting really hot. I'll be honest I don't know much about different toroids, I just use what works in the junk box a lot of the time I will try to find more info on them and reply back with info.
Finally got around to professionally drawing up the schematics. I built a new nice modulator board for it the other night. Frequency response is damn good for this unit. Almost flat from 25Hz-15KHz. Of course US stations cut everything off at 10KHz but I could go hifi if I wanted Now time to setup my EQ for NRSC standards and use the breakaway audio processing software. This unit sounds professional! Best of all no harmonics, no issues, tunes up like a charm, best audio I ever heard from an AM transmitter..
In this new picture you can see the audio preamplifier board with the circular gray transformer on it. The transformer came out of an old Gates radio console. Things were made for HIFI. Next to that is the internal audio phase switch.
I didn't draw the power supply schematic because it's pretty basic. I plan on removing the RF filter at the end because I don't think it's even necessary. The signal is so pure that I don't hear any harmonics up or down the radio bands. I may install a dummy load toggle switch in place so as to "tune up" first before flipping it on air. That way it will give the oscillator some time to drift up to par and parts to get to correct temperature. Once it's on for 5-10 mins the oscillator stays on frequency with no noticeable drift at all.
It may only be 5 watts (I originally said 10 but was mistaking), but it's a damn solid 5. No mismatch will effect the RF amplifier. I can short the RF out and it doesn't hurt it! The thing is built like a brick and should be able to take a hell of a beating. Would be interesting to pair it up with a linear amplifier one of these days.
Finalized. Only thing missing is the hood. Built a headphone modulation monitor into it along with a tune up/dummy load switch. Added second capacitor and transformer filter to power supply to reduce all trace of AC hum.
Kind of cool to be able to monitor the modulation through headphones while working into a dummy load at the flip of a switch.
Wow long time update. Hard to believe it has been 5 years now! Since my last post I added in a negative peak modulation indicator. The red lamp above the modulation dial simply blinks when reaching <0% modulation and will stay on for a few hundred milliseconds so tuning up the modulation level is much easier.. just pump audio in while the transmitter is on "Monitor" and dial down modulation level until the red light stops flashing with music/talk peaks, then VOILA you have her set proper. No need for a spare oscilloscope anymore to monitor modulation!
Put the modulation jack on front, where the red jack is going into. Modulation preamp was replaced with a new circuit that is far more hifi and accurate. I will have to update the schematic soon when I get the time.
Carrier power is dialed down to 2.5 watts or 10 watts PEP AM at 100% modulation. Easily settable to positive modulation swing if wanted via the carrier level control.
I purposely am running this at lower power now because the amp here...
That amp has been rebuilt slightly onto a larger heat sink, fan to cool it, beefier power supply and filtering and now I am pushing 10 watts carrier with at least 100% modulation with perfect modulation linearity. I have enough headroom to play with higher power when I am feeling daring.
I really need to document my project(s) better. New schematics to come, along with negative modulation indicator circuit. Even my FM gear has had a few upgrades since my last posts. Soon to release a ton of info I know soon along with new links on other threads for great informative stuff. Keeped tuned!
BTW I want to say this.. I know the amp seems like overkill running a fan and huge heat sink, maybe even overkill with running the actual AM exciter transmitter underpowered, but I found out the hard way that there is a HUGE difference between keying up PTT to talk to friends on a ham radio or the likes compared to continuous power output. People forget how much heat PAs actually make when running continuous operation. It is actually quite mind blowing even to me as a electronics designer how much parts can be pressed into service when proper cooling is involved and no commercial gear outside that which is designed just for these purposes can handle hours of broadcasting without something going pop. I really believe this is the problem with the import FM type transmitters, they are designed by people who think some output FET can run full throttle for broadcast usage, then people wonder why the PA tranny dies hours after usage even with good SWR.
Pfking photobucket ugh. I will repost them later, though no one should rebuild upon this old experiment. It was fun but I was young in RF electronics and it was a highly inefficient transmitter with a sloppy tuned VFO. It sure was fun and got my feet wet but some things are better left to the past lol. I'll see if I can edit the posts to reuploaded images. Photobucket I swear.. screw them SO HARD. So many schematics and pictures I was unable to see because of them too that service makes me so angry.
Yeah, I think Kage is right about that. Building a class D is not really any harder and it is much better in terms of efficiency. You can build something with a few hundred watts easy and very cheap. The hardest part is the modulator imo.
I have a class D design and you can see it on my website linked in my signature. I recommend the version 1.0 for AM broadcast band. The V2 and greater is better for shortwave, but harder to build. And quite honestly, I haven't really been able to make it work very well. The 1.0 design is great though.
It's pretty much as simple as getting some mosfet drivers (TC4452 or IXDD614) and some mosfets. You can use SiC mosfets or even just the traditional IRFPxxx which are only $2 a piece. And you connect them together. You may need a couple resistors and bypass caps, but that's about it. For the output, just use a simple toroidal balun with FB-43-1020 and make 1 turn of wire on the input side and 2-3 turns on the output. The power input to the amp is filtered with T157-3. Once you get the hang of it, this part is very easy. The hard part is the modulator which should be PWM for best results. You can find kits here: www.s9plus.com/kits.html
Those kits are designed for shortwave band, but they should work much easier on the AM broadcast band. Ignore their complex output match and just use a balun instead like I said.