Who posted this MRF300 amp PCB ?

[XXL]

Cant remember who posted this mrf300 amplifier now but iv just taken a look at it, sadly its for 70mhz only in its current form. Is there any way to bump it up to FM frequency's without expensive RF gear to work out the values ?

This Ldmos is definitely the way forward for cheap gear at £33 each but there's just no good working designs out there that are only using the 1 fet like the mrf101 amp.

https://70mhz.org/equipment/amplifiers/ ... ransistor/

[XXL]

Failing that does anyone know how I would get the matching right on a copper board without expensive RF gear ? I can just guess and play around with it but iv popped many devices this way.

[Albert H]

You can't just "poke and hope" - you need the proper tools. If you're going to design the matching networks, there are a set of formulae that you can apply, but that's only going to get you part of the way there. RF circuits never actually behave in the Real World™ as they are drawn, even with the best CAD software. Every PCB track, every component and even every component lead is going to have a characteristic that doesn't appear in their spec sheets!

You need a network analyser at the very least, and then a spectrum analyser, RF probes, dummy loads, attenuators and so on to actually get the design right. There's no other way - especially if you don't want to cause interference or destroy expensive components!

[rigmo]

You can't just "poke and hope" - you need the proper tools. If you're going to design the matching networks, there are a set of formulae that you can apply, but that's only going to get you part of the way there. RF circuits never actually behave in the Real World™ as they are drawn, even with the best CAD software. Every PCB track, every component and even every component lead is going to have a characteristic that doesn't appear in their spec sheets!

You need a network analyser at the very least, and then a spectrum analyser, RF probes, dummy loads, attenuators and so on to actually get the design right. There's no other way - especially if you don't want to cause interference or destroy expensive components!

RF MAGIC

[XXL]

You can't just "poke and hope" - you need the proper tools. If you're going to design the matching networks, there are a set of formulae that you can apply, but that's only going to get you part of the way there. RF circuits never actually behave in the Real World™ as they are drawn, even with the best CAD software. Every PCB track, every component and even every component lead is going to have a characteristic that doesn't appear in their spec sheets!

You need a network analyser at the very least, and then a spectrum analyser, RF probes, dummy loads, attenuators and so on to actually get the design right. There's no other way - especially if you don't want to cause interference or destroy expensive components!

Yes unfortunately the pirate way is to just poke and hope tbh. adding trimmers everywhere and changing out coils until your getting the most output is the way iv done it for years. Obviously this is a crap way to build stuff and a lot of devices have gone bang. But I'm now at the stage where I'm trying to learn how to do it properly.

For example, guessing would be to add a LPF with 33pf 4T, 66pf 4T, 66pf 4T, 33pf. <<< literally no idea what that is filtering but that's what iv used and never had any issues with the Com.

Its a super nasty way of building stuff but I want to build real working gear.

[XXL]

Could I use a nanoVNA to build a matching network ? Iv been reading about the smith chart to get the 50ohm output from the driver to match the input of the transistor. (I think this is what's happening ???)

I just cant work out what the input/output impedance of whatever transistor I'm using is atm.

[yellowbeard]

I posted that link in reply to a question about using a thermistor to turn own the bias under heating Sénor XXL.
viewtopic.php?f=11&t=3088

[sinus trouble]

Please forgive me for asking a silly question?

Why are experimenting with 300W RF amplifiers when you have very little knowledge or test equipment?

To be honest i have moderate amounts of both of the above, Yet i still would not feel confident doing what you are trying to achieve!

Anyways, Its your choice and i wont be contributing more to the subject! Good luck! Im Out!

[XXL]

I dunno. Can’t be that hard can it ? It’s because I have little knowledge that I’m asking on here.

[Albert H]

It really IS difficult! That's why I was so highly paid when I designed stuff!

[Marcel]

You can buy this empty PCB from him and that looks very good despite the low buget equipment he used.

Also ask him about the schematic for 88-108 MHz he also has.

Regards,

Marcel

[Marcel]

Here is already the schematic he designed for 88-108 MHz.

The only difficult how he mounts the teflon coaxes on the PCB (not really clear how to connect this things) for me.

[thewisepranker]

Cant remember who posted this mrf300 amplifier now but iv just taken a look at it, sadly its for 70mhz only in its current form. Is there any way to bump it up to FM frequency's without expensive RF gear to work out the values ?

Failing that does anyone know how I would get the matching right on a copper board without expensive RF gear ?

That's not what the expensive RF gear is for. It's there to verify that what you get in reality matches what you expect to see using the values that you've calculated.

[XXL]

I see a lot of ham radio users building their own amps and some haven't got a clue about the mathematical and physics side of things, but they just know by looking at the circuit and what band/frequency's they're using.

@marcel. Yeh I saw that on YouTube. Ill have a deeper look. Id still rather learn and understand how to create matching networks instead of just copying someone else's stuff though.

[Marcel]

XXL better stolen well than poorly designed.

He also just sent me the schematic of the low pass filter he designed.

The empty PCBs can be ordered from him for the mail address, search between the comments on his video.

Regards,
Marcel (from the Netherlands)

[XXL]

Thanks Marcel

[RF-Head]

Marcel it's a copy of my disgn you have tested a year ago (had only a hairpin instead of a stripline)
He only mounted the Drain coil
The stripline on the gate and drain is about 10-15nH
also the LPF that he use is realy good loss is about 0,10dB only Harm reduction is not mutch on second harmonic because of the -3db point is way to high

[RF-Head]

There is also a other thing
I see a realy good Bias control for the gate and is realy good filterd
My old design used a 78L05

[Marcel]

Hi Henk,

All designs are very similar only with your old design you still used 2 coils at the input.
A while ago I saw that you now also use a teflon coax (transformer) at the input.

The bias control for the gate lacks a temperature compensation in his design, something I wil compensate with a diode sensor like a 1N4148 or something.

The coils in the LPF also seem a bit small for the frequency used, but I will rebuild the filter with ceramic capacitors so that I get a reasonable picture and since the network analyzer can only make a maximum of +10db that is no problem.

(When you have the Enima coders back in stock, l"LL drive by with Dennis and we talk about it.)

Regards,
Marcel

[XXL]

Is this some other design your talking about ?