MRF141G 300 watt FM Linear Amplifier - ADVICE ON BUILDING THIS KIT

[EFIALTISFM]

.
I'd like your advice and expertise before I start putting together this MRF141G amp.

I bought the parts 33 years ago (1989) but never got around to build this unit. Last week I remembered the kit and started to look through some old storage boxes and successfully found all the bits.


I have 3-4 questions:

1. IMPROVEMENT TO THE ORIGINAL DESIGN of 33 years.
Q: Can any improvements be made to the original design or should I build based on the original instructions?

2. POWER and BIAS /rest (instructions suggest adding a zener to the circuit)

The unit was designed to run on a 28V regulated PSU but I'd prefer to run it on a much lower voltage (as low as 12V) working it with lower output. To avoid BIAS rest, the instructions recommend adding a zener and setting idle current once (then run from PSU 12 to 28 V)

Q: Is there a better way to do this or should I follow the original instructions by adding a zener diode?
Q: If adding a zener is the best way, which zener will work best, a 6.8, 7.5 or 8.2 V ?

Quick excerpt from the Instruction document : (for full document, click on link at the end)

" The unit was originally designed to operate on 28 volt regulated power supply, for this reason the regulation of the gate bias voltage was omitted. lf the supply voltage is varied by more than 2 volts, the bias will have to be reset by R4 for a nominal 400-500 mA drain idle current. This can be avoided by connecting a 6.8-8.2 V zener diode (1N5921A-1N5923A) from the junction of R3 and R4 to ground. The idle current can then be set once, and would not change considerably from a supply of 12 to 28 V "

3. POWER INPUT - MIN / MAX
Typical power is listed as 15 watts. Can't find anything on lower input.
Q: Can I drive this unit with say 4 watt (NRG) or 6 watt (dutch), will it work at all and if it does will it cause and interference, harmonics / sproging ? OR do I need to make a intermediate smaller amp of 10-15 watt ?


LINKS

AR313 KIT - COMMUNICATIONS CONCEPTS INC
https://www.communication-concepts.com/ar313-1/

RF POWER AMPLIFIER - FULL DOCUMENT & INSTRUCTIONS
https://www.communication-concepts.com/ ... n_Note.PDF

PARTS PLACEMENT (IMAGES)
https://www.communication-concepts.com/ ... cement.pdf

MOTOROLA DATA SHEET - MRF141G
https://www.communication-concepts.com/ ... rf141g.pdf

ZENER DIODE MODIFICATION ON AR305 AND AR313 AMPLIFIER
https://www.communication-concepts.com/ ... _AR305.pdf

Ta

[sinus trouble]

Looks like an interesting build

I am sure others can add more? But just to be clear, Dont expect 300 Watt from this amplifier! Especially @ 12V supply!

There is good reason why high power amplifiers use substantial supply voltages! The efficiency is terrible at anything lower!

Sinking 22 Amps into a single device is never a good idea!

I dont think running at a lower input/outputs would produce problems?

But the gain is quite low? (12 - 14 times) which @ 4 Watt input? would equal to something like 50 Watt out?

[Albert H]

Oh wow! That really is a "golden oldie"!

A few pointers - Helge Granberg always claimed that his designs would work over a wide range of supply voltages, but extensive experience with this device shows that it's rather useless below about 25V supply! The gain diminishes, and the supply current rises - never a good combination, because it indicates that the efficiency is collapsing!

The first thing to point out is that the heatsink is crucial - at best, with 300 Watts of RF squirting out of this thing, you're going to find that the Real World™ efficiency will be around 60% at best, This means that you'll be getting rid of ~200 Watts of heat. That calls for a huge heatsink, and force-cooling it is the only way to keep the size manageable.

The next thing is that silly output matching trimmer. DO NOT use anything other than the specified part, or you'll just end up with a very expensive stripline fuse! The genuine part will probably cost about as much as the FET!

Biasing is crucial - don't let Helge tell you anything different. Zener regulation is unlikely to be stable enough - they wander about with temperature, and never in the direction you want! I always used an "active" bias regulator, usually using an LM317 variable regulator (that is internally temperature compensated) for each side, and the highest quality multiturn cermet trimmers I could find for setting the bias voltages. Be careful to RF bypass all three pins on each regulator (I used 100n to ground), since some species of LM317 can freak out if RF gets in! Also be certain to provide heatsinking for the regulators - They're not providing much current, but the voltage drop - from 28V down to 5V (for example) means that the heat has to go somewhere.....

That amplifier has no output filter, since it's designed to be "broadband" - you'll need a pretty beefy one to handle the power with really fat coils, and the capacitors to deck need to be really good quality "ARCO" or similar or you will have a very expensive firework display. This should be shielded from the rest of the PA and you will be getting some hot capacitors, so you'll have to heatsink that board too. If you don't put a good filter on the end of that PA, you'll have harmonics up into the TV band, and the worst of them (usually 3f) will only be around -30dBc - this is right in the military aircraft band and will get you a very quick "unfriendly visit" if you're jamming their comms!

You WILL need to build an intermediate stage amplifier to get enough drive for the FET to work with the slightest bit of efficiency. I used to use a BLF244 stage to get these things to go.

Be really careful with the mounting of the MRF141G - the legs should be completely horizontal to minimise internal stress, and the contact of the flange to the heatsink is crucial - do NOT overdo the heatsink paste (and don't miss it out altogether either)!

You MUST check this amplifier with a good quality dummy load and spectrum analyser. Don't even dream of connecting to an aerial until you've proved that it's clean. In the grand scale of things, 300W isn't a lot of power, but if you get it wrong, you can cause mayhem with it!

Final word of warning - test it on a good quality, current limited power supply. That way, if there's anything significantly wrong, you shouldn't kill your expensive FET!

[sinus trouble]

I think Albert has "Hit the nail on the head" with that comment!

Keep us updated with any build progress, Nice little project!

[EFIALTISFM]

Sinus and Albert,

First of all thank you both for your time and advice.


Sinus,
I wouldn't mind if this amp worked at lower voltage + low input and gave 50 watts to begin with, leaving me the option to increase voltage & input at a later date to get more output power. And yes, Albert has "Hit the nail on the head"


Albert,
Information overload!

VOLTAGE: Maybe best to stick with 28V - 22Amp PSU

COOLING: Hope my aluminum heatsink (L 180mm, W 124mm, H 35mm, 11 fins) will suffice. Also adding a copper bar (120x60x4mm) as heat spreader + fans

TRIMMERS: ARCO 402

BIAS: Many questions here, not ready to post. Hopefully in next message.

FILTER:
Thinking of the dutch one : https://dutchrfshop.nl/en/diy-kits-pcb- ... ilter.html
ATC100B 500V caps + 2mm enameled copper wire. Would Silver plated copper wire coils be better? thicker than 2mm?

INTERFERENCE : Definitely don't want any of this

INTERMEDIATE AMP: Need to build this one. (should have an old BLY88c somewhere, will look into it)

FET MOUNTING: Might visit a "lathe" to shave off a thin slice of the copper bar where the FET will sit (depth = difference between PCB height and Seating Plane - Legs height). Would Silver Polysynthetic paste.be better to sit FET on copper?

As for the rest (Dummy load, Analyzer, and Power supply availability) need to do some work in this area...


Finally, some pics of the "old time" components.

---
https://scontent.fath3-3.fna.fbcdn.net/ ... e=634D5438
---
https://scontent.fath3-3.fna.fbcdn.net/ ... e=634C4FCE
---

[Albert H]

That filter should be fine, but you will need to dissipate some heat from it, and be sure to screen it from the rest of the rig.

The silver paste might be slightly better for heat transfer - I can't recall if I've ever used it. I usually just use the usual silicone gunk - the kind that you can never quite get rid of - you can just move it from place to place!

The BLY88 would be OK, but will necessitate another supply rail in the box.....

[sinus trouble]

Nice!

Looking at your kit, It seems like the Power FET is solderless??

I dont know how to explain? Maybe "Sandwiched" between the upper and lower PCB??

[Albert H]

I really wouldn't trust a non-soldered PA device!

[EFIALTISFM]

Looking at your kit, It seems like the Power FET is solderless??
I dont know how to explain? Maybe "Sandwiched" between the upper and lower PCB??

Sinus,

The FETs legs look a bit short but from the info and images I have, it looks like it is to be soldered.
The two PCBs are part of one split unit (one left and one right of the FET).
See info + photo of an assembled kit: http://www.geocities.ws/allcircuits/lineal_300w.html

Also see Motorola datasheet: http://www.om3bc.com/datasheets/mrf141g.pdf
@ page 5, HANDLING CONSIDERATIONS (last sentence)
" As a final note, when placing the FET into the system it is designed for, soldering should be done with a grounded iron.

Haven't see any info or images of solder less install.

[sinus trouble]

Ignore my previous comment, Due to the fact it has two PCBs I jumped to the wrong conclusion!

My experience With Power FETs is pretty good? The precautions are correct but i think you will be ok if you handle with care and avoid the woolly jumpers and socks!

Also if you space the PCBs from the heatsink correctly the FET terminals should sit perfect on the surface of the PCB when bolted down!

Most UK soldering irons are grounded as standard! Im not sure about other Continents?