pira 5watt pll

[Albert H]

Exactly. The first circuit is using that negative feedback network to broaden the frequency response. The second circuit wont work well for several reasons, particularly because the bias voltages for the two FETs are tied to the same source, and the FETs will need differing bias voltages.

[XXL]

pcb pira 5W

Have you built one yet ?

[sinus trouble]

Hello.I tried to make the mods for variable rf power and rf power shutdown when unlocked.Parts that need to be added are in blue rectangles.Hope it works and its not an overkill of parts

Sadly, your variable power modification won't work, because the final stage isn't linear - varying the drive power might make a very small change to the power output, but when the drive drops below the point at which the final will conduct, you'll just get no output at all! It's usually best to run the Class C final with just the right amount of drive so that it doesn't get saturated (which gives rise to harmonics).

Some "engineers" don't understand the problem, and try to make their exciters "no tune" by overdriving their final stage to get a "flat" output across the band (look at the A****f effort for an example of this). This practice invariably makes for a really nasty, dirty output spectrum - especially if the output filter is inadequate or poorly laid out (A****f again!).

Variable power out isn't trivial to achieve, and most design engineers don't bother to try to do this. You have to bear in mind that if you change the supply voltage, you'll also change the output match impedance, so the values needed in the final match to convert the lower output impedance of the transistor to the 50Ω or 75Ω that you want to feed into your coax. This is true both in exciter stages and in PAs.

Stephen Moss was lucky with the PLL Pro III (after a hell of a lot of experimentation) - he was able to get the final transistor to deliver ~4 Watts with its emitter grounded, and 1 Watt with the emitter sitting on top of a couple of power resistors. He managed to achieve a "conjugate" match in the lower power setting. If you try different emitter resistors (in an effort to get 2 Watts, perhaps), you'll find that you have a mismatch at the collector.... There are only the two power settings that work! In reality, the output match is even cleverer than it looks - the PLL Pro III gives 4 Watts at 88 MHz, and about 4.3 Watts at 108 MHz - this "sloping" output power was done to account for the gain droop with increasing frequency of the transistors he used in his PA modules. The 40 Watt BLW60C PA would give exactly 40 Watts out at any Band II frequency when driven by the PLL Pro III board.

There is little point in trying to make a variable power transmitter. For example: DB Elettronica make one which gives 300 Watts out at full power. They specify the harmonic and spurious content at 300 Watts out, but don't give any figures for when the power is turned down!

If you want your rig to deliver 10 Watts, design for 10 Watts, not 1 - 10 Watts! If you want 1 Watt, just design for 1 Watt. Incidentally, with a 12.65 Volt supply (allowing for 0.65V transistor loss) and 1 Watt, you'll get a natural match of 72Ω - ideal for feeding into "75Ω" coax (it's not - it's usually closer to 72Ω). Most of my Band IV link rigs would do 1 Watt, and would naturally match into TV coax and a TV Yagi. This makes for efficient, cheap link gear!

Albert is spot on!!

Even "No Tune" designs are a compromise!

When the voltage amplitude changes! The current flow through the matching circuit changes too! This leads to an impedance mismatch!

Whilst it may not seem apparent? It is still there!

[rigmo]

pcb pira 5W

Have you built one yet ?

long time ago with friends.. now i build new one.. after long time period of abstinence

[Albert H]

Exactly, Sinus. As the output match changes, the harmonic content changes. A couple of manufacturers used servo tuning, with motor-driven variable capacitors to correct the match as the power was varied. This was an insanely expensive approach, and was soon abandoned.

In the days when I worked on "Regional" FM transmitters - powers in the 100s of kilowatts with megawatt ERP - power output adjustment was done at the installation and calibration stage. These were big valved beasts, with PAs that you could walk inside and water-cooled valves that were inserted with a forklift truck!

The design power of the transmitter would be (say) 120 kW to 300 kW. Once the aerial array had been designed and tested, the design engineers would decide on the power level needed to achieve the required signal strength contours. The anode voltages necessary would be calculated, and the tune-up would begin.... Setting up one of those bigger rigs could take a month or two, and once aligned, they would be run at exactly the specified power level whilst field strength measurements were taken all over the projected service area. Only ever once did I see a re-calculation of the power requirement after the field measurements and a re-tune of the rig. The trick was to use the minimum power level that we could so as to minimise power consumption and keep the aging of the bottles to a minimum - the harder you run a valve, the more metal migration happens, and the shorter its life.

In the case of the little transistor and FET rigs we all run, it's best to design for a power output well within the SOAR of the output device(s) to allow for possible antenna and feedline mismatches and other misoperation. For example - the MRF101 is capable of (almost) 150 Watts output, but I'd never run one that hard. I'd design for (say) 100 Watts. The difference in field strength between 100 and 150 Watts is negligible and won't ever be noticed by the listeners, so why run the risk of frying expensive components when you don't have to?

Way back, the BLW60 was rated to 50 Watts, but we always designed for 20% less, and ran them at 40 Watts. Again this gave a useful margin for stupidity! Similarly, my SD1019 rigs always ran at 95 Watts, and always continued to run right up until they were collected by the "authorities".

When installing your rigs, it's always worth remembering that OFCOM limit the coverage of RSL rigs by specifying a maximum HAAT for the transmission site. Band II signals always go to the horizon, so the higher your aerial, the greater your coverage. The only difference that increasing power makes is to improve the signal to noise ratio away from the site. Also remember, you'll need at least 4 times the power for a stereo transmission because of the wider bandwidth which ruins S/N ratio!

[rigmo]

Exactly, Sinus. As the output match changes, the harmonic content changes. A couple of manufacturers used servo tuning, with motor-driven variable capacitors to correct the match as the power was varied. This was an insanely expensive approach, and was soon abandoned.

In the days when I worked on "Regional" FM transmitters - powers in the 100s of kilowatts with megawatt ERP - power output adjustment was done at the installation and calibration stage. These were big valved beasts, with PAs that you could walk inside and water-cooled valves that were inserted with a forklift truck!

The design power of the transmitter would be (say) 120 kW to 300 kW. Once the aerial array had been designed and tested, the design engineers would decide on the power level needed to achieve the required signal strength contours. The anode voltages necessary would be calculated, and the tune-up would begin.... Setting up one of those bigger rigs could take a month or two, and once aligned, they would be run at exactly the specified power level whilst field strength measurements were taken all over the projected service area. Only ever once did I see a re-calculation of the power requirement after the field measurements and a re-tune of the rig. The trick was to use the minimum power level that we could so as to minimise power consumption and keep the aging of the bottles to a minimum - the harder you run a valve, the more metal migration happens, and the shorter its life.

In the case of the little transistor and FET rigs we all run, it's best to design for a power output well within the SOAR of the output device(s) to allow for possible antenna and feedline mismatches and other misoperation. For example - the MRF101 is capable of (almost) 150 Watts output, but I'd never run one that hard. I'd design for (say) 100 Watts. The difference in field strength between 100 and 150 Watts is negligible and won't ever be noticed by the listeners, so why run the risk of frying expensive components when you don't have to?

Way back, the BLW60 was rated to 50 Watts, but we always designed for 20% less, and ran them at 40 Watts. Again this gave a useful margin for stupidity! Similarly, my SD1019 rigs always ran at 95 Watts, and always continued to run right up until they were collected by the "authorities".

When installing your rigs, it's always worth remembering that OFCOM limit the coverage of RSL rigs by specifying a maximum HAAT for the transmission site. Band II signals always go to the horizon, so the higher your aerial, the greater your coverage. The only difference that increasing power makes is to improve the signal to noise ratio away from the site. Also remember, you'll need at least 4 times the power for a stereo transmission because of the wider bandwidth which ruins S/N ratio!

Albert H you are a great poet... everything is said in just three stanzas

[Krakatoa]

pcb pira 5W

Have you built one yet ?

long time ago with friends.. now i build new one.. after long time period of abstinence

Congratulations, that board looks great!

Btw: the little 1uH choke at the output transistor collector, may give a problem or limit performance, I think it may hit its current specs limit.
I used a VK200 ferrite in there and was fine for either 1W or the 5W version.

[rigmo]

Btw: the little 1uH choke at the output transistor collector, may give a problem or limit performance, I think it may hit its current specs limit.
I used a VK200 ferrite in there and was fine for either 1W or the 5W version.

YES, at the very beginning I ask the same question. for now I stick to the original list only.. except for 10nF C15 and use 2k1nf

[rigmo]

almost done

[radium98]

Rigmo let us see the spectrum of this baby on a tinysa when it is finished.

[radium98]

Albert h isan open book .Thanks.

[XXL]

I’m worried that open air oscillator will be microphonic

[Albert H]

The oscillator shouldn't be microphonic if the coil winding wire is thick enough to be really rigid.

I remember one well known station in London that often used to broadcast fan noise behind their audio because their rigs were microphonic! They had a recording of DTI Inspectors talking in the lift room near to their rig, shortly before it was turned off!

[rigmo]

Frist result is not so bad.. at all... 3W @12V 5w @14V

[XXL]

Is that 3w @12v with c1971? And What is the trimmer for on the filter ?

[rigmo]

Is that 3w @12v with c1971? And What is the trimmer for on the filter ?

I find old-one cca 34pf resonance blau 50PF is the best 8mm

[rigmo]

VIDEO

[sinus trouble]

Nice job Rigmo! Looks neat!

[Albert H]

Shame about the 2nd harmonic, though.

[radium98]

Nice job Rigmo! Looks neat!

Agree .Happy new year 2023