help get more power from TX

[RX5]

hello,

can someone help me how to "Add" another RFAMP to my existing transmitter?? I would like to add a C1971 5W transistor.. but dont have an idea HOW to...

[vintage1]

Just replace what is there, and remove its 1 ohm resistor.

You probably don't need the 10 ohm base resistor either.

More supply voltage may net you more power as well.

[RX5]

vintage1,

hmmmm.... about the base resistor, I tried replacing 10R with 100R.. and low output, I guess removing the resistor(shorting) WILL improve output..

the C1971 Vce=17V and Ice=2A(max specs)

could I really get 6Watts of radiated RF power with supplies of 13.5V -FROM- the/my existing schematic??

[Rodney]

Depending on your location you may be limited by law as to how much power you can radiate. In the USA, on the RC frequencies, it is well below 1 watt plus, if you make any changes to a type accepted unit, you must go through the mandated type acceptance tests which are both time consumming and expensive.

[RX5]

in my place, such laws ( I dunno if applicable) :D are non-existent.. hehe SO I could radiate more power...

so ... how much RF power could I get out of using 2SC1971 tranny with 7.2V supply??

[vintage1]

I got around 5W out of a very similar circuit using 18v supply and a BFY51 output transistor.

The limiting factors are the heat in the output transistor and the way the antenna is coupled in..you have with a tuned class C about 2xsupply voltage swing on the output collector ..that will essentially PROBABLY be seeing a quarter wave type tuned output at 75 ohm impedance..

Unless there is some transformer matching.

RMS voltage will be 1.4 x supply, so on say 7.5v into 75 ohm that is 1.5W if my maths is right.

To get more on low supply voltage requires a step up output transformer. Easier to use a bigger battery.

You SHOULD get about 80% 3efficiency with it tuned up right, so take out the emitter resistor, put in a big fat output transistor, and tune it up..and measure the current. Should be around 200mA.

[XJet]

Leave the emitter resistor in but add a bypass cap. This will ensure that you don't drive the RF stage out of class C and into class B where it will be far less efficient and generate more heat.

[RX5]

guys, got rid of the 10R and used jumper.. and removed also the 1R emmiter resistor... used C1971... and DEAD or NO output... how come??

edit:

oh heck... transistor pinout is different from conventional To220 BCE.. twas BEC and fin = E.... lol... hahaha 1st time I manage to get hold of a power RF tranny... :p

[vintage1]

So, how much current is she pulling NOW?

[RX5]

still have not measured yet , I have to make a newer board to accomodate the C1971... because of BEC pins... dont want ANY bending...

I was currently using C2053( i think possible of 200mW of RF).... removed the emitter resistor AND base resistor.. man did it get HOT.. with supplies of 7.2V...

and I know its radiating MUCH power now than before.... my rx seems a bit bad?? if the Tx is in CLOSE proximity, it will go haywire(the rx output controls)... If I back out a little, it WILL WORK.. seems my current rx has bad AGC or has none??? dunno.. or maybe need s a bit tuning on the RF coil(rx) because of higher output power tx? dunno... maybe AGC or NO AGC or maybe just a de-tunes rx(again)?? need to experiment..

[Ron van Sommeren]

What about bandwidth after modification? About the width of an interstate highway :D ? In that case your fellowflyers will not be amused. If there are no fellowflyers there's probably no reason to jack up the power.

Vriendleijke groeten ;) Ron

[vintage1]

Bandwidth is mostly a function of the output coils and suchlike.

I get the impression RX5 is fairly 'isolated' geographically.

[Cirip]

Bandwidth is mostly a function of the output coils and suchlike.
Vintage1,
I have to disagree with you... for the first time, though :)
The ouput coils will have an impact on the harmonics, but I guess what Ron was talking about were the adjacent channels. The modulating signal bandwidth has the biggest impact on the side bands. If the signal is rectangular, unfiltered, it will generate a lot of splatter in the neigboring channels.

Cirip

[RX5]

still have not measured yet , I have to make a newer board to accomodate the C1971... because of BEC pins... dont want ANY bending...

I was currently using C2053( i think possible of 200mW of RF).... removed the emitter resistor AND base resistor.. man did it get HOT.. with supplies of 7.2V...

and I know its radiating MUCH power now than before.... my rx seems a bit bad?? if the Tx is in CLOSE proximity, it will go haywire(the rx output controls)... If I back out a little, it WILL WORK.. seems my current rx has bad AGC or has none??? dunno.. or maybe need s a bit tuning on the RF coil(rx) because of higher output power tx? dunno... maybe AGC or NO AGC or maybe just a de-tunes rx(again)?? need to experiment..

problem solved!! :D

just re-tuned a bit the rx(RF and 1st IF coil only) and the TX output L filter.. ALL is ok now.. AGC schmuck..twas just de-tuned.... hehehe if I got time, ill make a newer -RF baord- for the C1971 with the provision of adding an heatsink....

vintage1, regarding -isolated-, uhmmm yeah I live in a somewhat?? isolated part of leyte, philippines :D seriously, there are hardly ANY Rcers here.. so I could boost my RF output to 5W w/o getting to bother others.. :)

I could "experiment" to the max... and these C1971 is kinda costly... I better make a NEW RF board for it... and a new 7ch encoder board to go along with it.. :)

[vintage1]

Vintage1,
I have to disagree with you... for the first time, though :)
The ouput coils will have an impact on the harmonics, but I guess what Ron was talking about were the adjacent channels. The modulating signal bandwidth has the biggest impact on the side bands. If the signal is rectangular, unfiltered, it will generate a lot of splatter in the neigboring channels.

Cirip

That's what the output coils are there to remove. ;)

The modulator is keying the oscialator anyway, and that won;t be isntant switch on..Xtals take a few microseconds to 'spin up' so the modulation will be fairly soft.

And its on 27Mhz so anything goes..


And its in the Phillipines so lord knows WHAT else is around..

[RX5]

does this mean I got a lousy tx? :)

[Cirip]

If the signal is rectangular, unfiltered, it will generate a lot of splatter in the neigboring channels.
That's what the output coils are there to remove. ;)
If that were true, you would not be able to change crystals/channels on a transmitter without retuning the output. The output filter is much wider than the occupied bandwidth of the RF signal, therefore the splatter can't be cleaned by the output filter.
What Q should the output filter have in order to be as narrow as one channel, therefore filtering out the adjacent channels? Q=27MHz/20KHz=1350(!!!). Ain't going to happen :) Actually, Q=1350 will only attenuate by 3db the adjacent channel.

I agree though that maybe nobody would care if the OP's transmitter covers 5 channels at a time :)

Respectfully,
Cirip

[vintage1]

If that were true, you would not be able to change crystals/channels on a transmitter without retuning the output. The output filter is much wider than the occupied bandwidth of the RF signal, therefore the splatter can't be cleaned by the output filter.
What Q should the output filter have in order to be as narrow as one channel, therefore filtering out the adjacent channels? Q=27MHz/20KHz=1350(!!!). Ain't going to happen :) Actually, Q=1350 will only attenuate by 3db the adjacent channel.

I agree though that maybe nobody would care if the OP's transmitter covers 5 channels at a time :)

Respectfully,
Cirip

Well we just used to chop the power stage..on tone stuff. Ghastly. And you do normally need to retune a TX with a crystal change if you want the best output.

If you look at the transmitter there you will see there are actually 3 tuned circuits..in the power stage..the collector paralleltuned one, then a peaking low pass (tuneable) and then the base loading on the antenna. Not hard to get a fairly sharp response with that lot.

I agree the modulation has to be a little 'soft'..but I think thats why they are modulating the oscillator. The crystal itself has an enormously high 'Q' and will not instantly start to oscillate though it looks like it might stop quite smartly with power removed.

I am almost certain this is why everybody went FM anyway..controlling AM sideband modulation is trickier.

[RX5]

Hi again,

done with new RFamp board.. with C1971.... supply for RF side, 12VDC... AND.. RF output is VERYYYY low.....

I looked at datasheet, compared to what I was currently using(C2053 w/c outputs 150mW @ 4mW Rf input)..... the C1971 needs -at least- 600mW of input to be able to output 6W.... whew... this means a TOTAL re-design of the RF amp side...

[vintage1]

Try a better power transistor then.

[RX5]

I dont know of any other #s.. care to mention some?? but I doubt it would be available here locally...

and IS THIS LOGICAL?

with tx and rx BOTH detuned, start tuning 1st the TX for high RF output ... and 2ndly the rx??

[Cirip]

Try this too and see what happens. You can play with the 22K resistor a bit, but don't go too low. Say 5K minimum. Try to maximize the output RF level. With the proposed mod, max current does not mean max output power.

Also, unless you are absolutely sure that (150pF+transistor output capacitance) resonates with the "blue" inductor at your operating frequency, 150pF might do more harm than good. Do you know what is the inductance of the "blue" inductor?

Cirip

[Miami Mike]

Someone should point out that you'll get a much better return for your trouble if you just leave the transmitter alone and use a bigger antenna instead, assuming that's practical for you. If you can connect a coaxial feedline to your radio, you can use any of a variety of high-gain antennas made for 27 mhz.

[RX5]

hello all.... here is the updated scheme... I forgot to draw the 30pF from base-gnd of PNP tr...

and I have also measured the BLUE and grey coils... with an LC meter...

the original scheme used 1815 as the final RF.... I replaced it with C2053... heats up a bit (suppy of +7.2V).. going to change the 1R emitter res to 4.7R later...

is the 150pF in parallel with 5uH BLue coil OK? or not?"

[vintage1]

well I make the resonance about 14Mhz..which seems odd.

Try 39pF or 33pF

[Cirip]

well I make the resonance about 14Mhz..which seems odd.

According to my calculation, 150pF with 5uH in parallel resonates at 5.8MHz. In your circuit, due to additional capacitance, the resonance may be even lower, which makes the short circuit effect of the 150pF even worst.

Try the schematic I've suggested. The 22K resistor will pre-bias the transistor and you should be able to increase the conduction angle and squeeze more gain.

Cirip

[RX5]

According to my calculation, 150pF with 5uH in parallel resonates at 5.8MHz. In your circuit, due to additional capacitance, the resonance may be even lower, which makes the short circuit effect of the 150pF even worst.

Try the schematic I've suggested. The 22K resistor will pre-bias the transistor and you should be able to increase the conduction angle and squeeze more gain.

Cirip

Hi Cirip,

and what about the 150pF? should I just leave it in or change to -other- value?

[Cirip]

Leave it there for now and try to tune the output for max RF out. Then remove it and retune. See which case you get the better output. Also, compare with the original, no 22K res case.
Cirip

[ghoti]

I'm pretty sure output power goes up with square of supply voltage. I once raised osc supply from 12 to 15 volts and got a LOT more RF out. Bill

[Miami Mike]

I'm pretty sure output power goes up with square of supply voltage. I once raised osc supply from 12 to 15 volts and got a LOT more RF out. BillIt works out to less than 2 db.