MOSFET RF power amplifier design

[Buddhafollower]

In case of the AB class, the diode is there to set the bases at 0.7V or so. This is so that i won't need a voltage divider network. Probably made up of two or rather four resistors to give me 0.7V at the bases. It is in a way ''automatic biasing''.

[Buddhafollower]

I did more experiments this morning.
I added an additional diode in series with the first and increased the value of the resistor. Now the resistor is much larger, say 8.2kOhm and the idle current is much smaller, 280mA. Also in the 10MHz range the waveform appears blurred a bit for some reason... When i set the signal generator to 5Vpp, the sine wave is blurred in 10MHz. Also at 20Vpp in the same range the sine wave is blurred. Only at 10Vpp the sine wave appears normal or less blurred than before. From that point on there is a blurring of the waveform and because of this i don't think the amplifier can go any further.
Previously the idle current was way too great and the waveform was very good way after 10MHz. Now At 1MHz the sine wave is a bit bent towards right, an almost imperceptible bend, the idle current is much more acceptable and so the heating of the heat sink/.

[Buddhafollower]

Sorry, i wanted to say the bigger the value of the resistor the more distorted the waveform is. Basically the more you distance the base from the turn on voltage of the transistor, say 0.6V, the more distorted the waveform appears.

[spacecase0]

when you say smaller values, you mean lower resistance ?
like 20ohms gets you a worse waveform that 100 ohms ?

I don't really see the need for the diode at all,
unless it is meant to soak up all the extra current. but for that to work the transistor and diode need to be quite similar in forward voltage response.

the resistor is just there to turn on the transistors just a bit so they are in linear mode and not shut off entirely on the opposite side of the sine wave (it takes to much time to turn on.)
potentially
try removing the diode and see what your idle current is,
try removing the resistor and see how bad your wave form is.

[Buddhafollower]

The smaller in value the resistor the more distorted the waveform is. And i don't have the 0.7 or 0.65 volt at the bases. I'am thinking whether i should put two diodes in series and work with the resistor value a bit, because the bigger the resistor, the smaller the voltage at the bases irrespective of the diode's turn on voltage.

[spacecase0]

looking at that schematic, the resistor sets your idle current.
you only need enough current so your waveform looks good at the zero voltage crossing.
more current than that will cause you issue (as you have already noticed)
if 100 ohms gets you 0.8A, then try something higher value, maybe 200 ohms, that might get you 0.4A idle current.
you only need enough current to just turn on the transistor a bit. even 1K ohm might work,
try it and see what happens.

[Buddhafollower]

This is what i made. The amplifier in the schematic.
It works well from 1MHz up to many . To be more precise i tried it up to 14MHz and the sine wave is beautiful.
At 1MHz looks a bit distorted and the top and down peaks of the sine wave are a bit rounded up but it is good.
The ultra fast diode is MUR840 i believe and the four resistors are 100Ohm each. A very interesting effect i observed. When you don't feed it signal there still is an idle current that it takes from the power supply. At 15.5V i remember it takes about 0.8A, even with the bases being at 0.65V.
The heat sink get warmer and warmer... or so it seems.
The smaller the values of the resistors, the more they heat up and if i put a bigger resistor in there (bigger in value), then the amplifier is no more a class AB and severely distorts the wave form. So whether i want or not i have to put this value in there, 100Ohm or a bit less. And they will heat up like crazy. So i put four of them in series-parallel.
The same thing, heating up of the heat sink, happens in the case of a russian in origin amplifier also a Class AB one. Here is the schematic. The first two are class A and the final is Class AB. This thing heats up the heat sink so much that after about 30 minutes i'am forced to shut it down fearing that the transistors will destroy themselves.

[Buddhafollower]

I really looked for this book but couldn't find it. I looked for it for i think an hour by now... File share sites and not one of them has it. I'am sorry...
As you said, if you are so kind as to help me with a few pages daily i would be more than grateful.
As i said, the internet isn't a friendly place for people who want to learn science.
Look for einsteinian crap and there's plenty of that on the internet. It is sold as cheap as toilet paper. In fact I should buy them strictly to give them the proper use.

[spacecase0]

seems like you need this book
https://www.goodreads.com/book/show/...-experimenters
but at 1600$ on amazon, there has to be somewhere else to get it. (my copy was $5.95 when new)
I have a copy in my hand, and it is what taught me how to design with transistors.
if you can't download it somewhere I might be able to scan a few page a day and send it to you.

[spacecase0]

bipolar transistors do have capacitance internally, but it is often not a big deal when designing (but I am not sure about that for your parallel transistor part as it has many many base connections),
they are current devices,
so current on the base turns into more current at the collector, the ratio is set by the gain of the part.
gain is going to be higher at lower frequencies. and lower at higher frequencies.
so for that part you listed, I am pretty sure that the gain at 60MHz is going to be one. so over 60MHz you are not getting any amplification.
that is why the data sheet says that 40MHz is the real limit of the part.

when designing, set it all up for the max. frequency you will be using, then turn down your input power to get the desired output power on the lower frequencies.

by the way, most people don't make there own amplifiers...
not sure if you can get them where you live,
but this company is low priced and the hardware works well enough as long as you don't push the limits of what they say it can do (example, don't push a 300W amp more than 100W)
https://www.rmitaly.com/product-cate...ar-amplifiers/

and the last amplifier I got is a vacuum tube kind, and yes, they still use vacuum tubes for the higher power amplifiers.

[Buddhafollower]

You've asked me about the gain of the transistor at the frequencies I'am using it for. I don't know. I don't know where to find the values of the base to emitter capacitance and base to collector capacitance. They aren't in the datasheet. And i also don't know the formula explaining how gain is influenced by frequency. The internet is not a friendly place for us new to this and who want to learn more science.
I looked for such information as you asked and what i found.... trust me, you don wanna know. Complicated formulas and analysis of the CE amplifier... I asked a very simple question and the internet bombarded me with nonsense. That's why i said at the beginning that i want a website or a book written by a competent person such as a radioamateur. As smart as these ''scientists'' and ''engineers'' are, they write complicated books to show how educated they are, i know someone that talks to people at least two states away using no more than 1W RF power. Tell this to the engineers and you'll be laughed at. The man even has his own antenna, a piece of wire thrown into a tree and hanging from a branch. That's his antenna )
Nowadays ''scientists'' know and understand science just like i was born and raised on Mars.

[Buddhafollower]

Pretty much everything changed nowadays because of this digital crap we see everywhere. Nobody needed this but it is everywhere.
Thank you for your help. Much appreciated.

[Buddhafollower]

I have some mosfet amplifiers right now and i also have plenty of transistors of this kind, 2SC3264. It is strange how a transistor of this case can go so far in frequency. I'am not at all interested in 60MHz. It is Dr. Tesla's opinion that such frequencies are useless. At most, i want to cover the 40m band as well as the 30m band. This means that the amplifier has to be broadband and to cover from 7Mhz-7.3MHz , the 40mband, and from 10.1MHz to 10.15MHz, the 30m band. I would like for it to be able to go from 1MHz to the 30m band. I'am writing from the living room and I have here with me an amplifier with two IRFP250 which is broadband. It goes from about 400Hz to 10.15MHz very beautifully. Nice wave form. Sine wave. I had to pick the right ferrite for it. I didn't have material 43 so i used some from Kemet. ESR-25-R i believe. Very beautiful wave form but i haven't yet measured it or measure the distortion. It does distort a bit from i think 2200kHz up to 3000kHz. A bit distorted the sine wave is but still a sine wave.

[spacecase0]

designing with bipolar transistors is kind of annoying
the impedance is not really easy to calculate as it changes with current.
the gain for that transistor is zero at 60MHz...
so what is it at the frequency you are using ?
seems to me that there is a big reason why most transmitters meant to use with a power amplifier have adjustable power out,
and more often than not they have feedback to cut power out if the amplifier is putting to much power out.
so my suggestion is to just try it and see what it does.
and pretty sure that is how many electrical engineers do it when faced with some unknowns.
also, I am a somewhat qualified electrical engineer, but got out of the field before digital really took over most things.

edit:
why that transistor ?
do you have some already ?
why not the mosfets ?

[Buddhafollower]

If anything, i think that the 6:1 ratio of the secondary is going to work but this device may be perfected.
I await your answer and if this device has some chance of working properly i will build it and then measure it's characteristics.