Saw this on social media and thought I'd post it here for commentary.

It's especially easy for newcomers to get wrapped up in Watts when considering losses, and in the lure of amplifiers to increase those Watts, but signal works in dBs, not Watts.

The typical S meter is supposed to be something in the range of 6dB per S unit.

Folks often squabble over fractions of a dB when it comes to things like transmission line loss. Not that it doesn't matter, and it matters more if you're adding a bunch of them up, but if you're comparing A to B and there's just a couple tenths of a dB in difference, will anyone notice?

As an example, if 100W gets you an S5 signal, what's the over and under?

Attached File


QRP to 100W is more of a difference than 100W to full legal power.

100W to 195W isn't nearly as much improvement as 5 to 100W.

By contrast, if you have a 10 or 15dB null in an antenna pattern, might someone notice that?

(ETA: this is only an example for comparison, this is not to imply that any particular wattage will have you being received at any particular level at the other end)

Attached File


I have been having fun with the G90 mobile. 20W is doing more than I thought.

That chart right there should be posted on the wall of every shack in the world. As a ham of a long time, I struggle to understand it all the time, and I am sure many others do as well.

Excellent thread

decibels are a great way to measure gain, and losses

Prob needs some more explanation.  I imagine some people might look at that and think “hey lots of people give me 59 signal reports and I’m not running 25,000 watts so how does that graph relate”

I agree 100% with the theory behind that picture, however real world can be quite different.  Many times I and the station I'm talking to will turn down power from 1KW+ to 1 watt or less with less than an S unit of difference on both S meters.  Other times I have observed that turning down from 1KW+ to 100 watts makes a signal go from S9+20 to less than S5, then restoring the power brings the S meters back to S9+20.  The ionosphere is an extremely complex RF refletor and models are a guestimate at best.

That definitely puts it in a way that's easy to understand. Kinda puts the legal limit guys to bed.

I also have always heard that sound isn't linear, so sometimes little can be a lot and a lot can be a little.

How big of a reactor do I need to power a 26BW amp?

As some stated, if you don't bear in mind that this is logarithmic you can easily be sold a bill of goods.

For power measurements:
dB gain = 10 log (P1/P2)
So double the power = 10 log 2 = 3dB  

For voltage measurements, since power goes as the square of voltage:
dB gain = 10 log (V1/V2)^2 = 20 log (V1/V2)
So double the voltage = 20 log 2 = 6dB

S-meters measure voltage, so as you can see in this chart, going from S1 to S2 is double the voltage, or 6dB increase in power.

Attached File

When the signal is near the noise level, doubling the power can make a big difference in the readability (the "R" in RST) of the signal.

If you can't HEAR them, you can't work them.

I went thru this in a casual way in my last presentation a couple of weeks ago, when I went over setting up and interpretation of WSPR, and how to declare you power level and what that meant. As well as the signal levels received on PSK reporter for FT8 signals.

I will think this over and maybe utilize a greater understanding of this as my next presentation.  I am not sure it could carry a 2 hour presentation as that is how long they have been running but I will think it thru. Thanks for the idea. I am sure many in my club do not have a very good handle on these concepts.

It gets even more complicated with FT8, because what is exchanged in FT8 is not signal levels, but signal to noise ratio. If you have a quiet receive site, you'll be giving good reports to everyone, and wondering why your reports are rarely if ever as good. And also that you can hear a lot of people that you can't work. And vice versa if you have a noisy receive site.


Sound is definitely not linear, at least as far as the human perception of sound.

It varies somewhat between individuals and environments, but in general humans perceive about 10dB of difference in sound level to be double or half as loud. 10dB would be 10 times the actual sound energy. So four times the perceived sound level takes 100 times as much power.


Let's see. Chernobyl, one reactor was 3200MW? So all four at full power would be 12.8BW? So you'd probably need... 12 reactors unless your amp is 100% efficiency.

I'd be thinking more in terms of what you'd use for a feedline and antenna that wouldn't be vaporized.

Which Caribbean island are we taking over for this contest station

I guess my island that would let us build 4 Chernobyls…

Attached File


While the math on that is all right and tight and perfectly correct, the fact is most of us have experienced that QRO does make a difference. A single S-unit is more valuable than that image would appear to make it.

For instance:

5W to 20W = one 1 S-unit (6dB). Many portable op's have said that 20W works much better than 5W, and that 20W is the sweet spot for portable op's.

20W to 100W = a little over 1 S-unit (7dB). Nobody would say that operations at 20W are equivalently easy as they are at 100W.

100W to 500W = a little over 1 S-unit (7dB). Nobody has ever said that when they bought their first 500W amplifier that things did not get easier and more enjoyable for them.

500W to 1500W = a little over 0.5 S-unit (4.8dB). Yeah, this is gilding the lily a bit. Owning both a 500W and a 1500W amplifier, of course I run the 1500W unit preferentially, but it was not a giant leap in performance by any means. Clearly I need a 2KW amplifier to get another S-unit

One measly S-unit is a significant operational difference in amateur radio.

And before people start posting the usual "Well I worked a billion countries on 5W!", yes, of course you did, anyone could. But how easy was it and how long did it take?

antenna gain trumps power gain, as antenna gain works on TX and RX

good propagation trumps power also
 

True to all of the above.  But there are times when propagation is fair to poor that adding 10 dB of amplifier power will make all the difference in the world Between "I hear you but you are difficult copy" and change it to "yeah, now I hear you well."

Of course when the propagation falls out, all the amp in the world ain't gonna fix it.

I have one wire antenna (ZS6BKW) and no way to improve it, so for me 500 to 600 watts has put many DX contacts in my logbook. YMMV


Attached File

yep

especially on the lower frequencies

Right. My point wasn't to disparage amplifiers, just for folks to have a realistic understanding of the relationship of power to signal.

The generally accepted figure for the minimum amount of signal difference that can be detected by ear on SSB is 1dB. That's about the width of the needle on an S meter,  and that's in stable conditions, as most know on HF signals are often fluctuating quite a bit more than that.

On FM the generally accepted figure is 3dB. If you're right on the fringe of reception maybe a little less.

You can get the dBs with power, antenna, efficiency (reducing losses)... and all have their points of diminishing returns.

Yep I think the best bang for the buck is the most efficient antenna you can put up. This is going to help you hear and help your signal. That 10db of gain from the amp will not allow you to get better at hearing the other signals. But 10db of gain helps you tremendously if it is done at the antenna.

I have run most of my life as a ham Barefoot, but there are times when an amp is called for. Now I work a lot of weak signal VHF and up modes, and that is where it shines. But doing EME and Meteor scatter, building out the best system you can to get the lowest noise levels and strongest signal go hand and hand. A good example is my 4x15 element EME station on 432 with 100 watts I could work just the biggest stations out there. But add an LNA at the mast and bump up to 500W and you can hear yourself off the moon. EIRP, Noise Floor, and noise rejection are the keys with weak signals.

For HF low band, its all about getting the signal high enough above the noise floor. I have rarely found a contact I worked who could not hear me if I kicked the amp off. But the quality of the signal was changed dramatically. I think your point is spot on, do you NEED an amp to rag chew on 40m, not really. But I seldom run on 40m without it anymore. I think the high bands it is less of an issue, but after spending money to build out the system and make sure it is capable of handling the power I am running, it never hurts.

My old days of chasing rare DX I would spend hours trying to get through a pileup. Nowadays I may not be the first one to bust through, but it never takes more than a couple of minutes. I think there is a tradeoff though, jumping from 100 watts to 500 is a big difference, but most of the guys running legal limit seldom know I am running a SB-200 at around 500 watts.

I have one of these I've been wanting to try. How do you like it? Will it tune on 60m?

In the photo it shows a 1:1 balun. Coax is 50 ohm and ladder line is either 300 or 450 ohm, so wouldn't a 4:1 balun be called for? Just asking, maybe the length of ladder line completes the transformation?

Every ladder line fed antenna I have put up performs better with a 1:1 balun instead in place of the recommended 4:1.

From what I've read about G5RV and ZS6BKW antennas the ladder line is tuned to a specific length and the balun should be a 1:1 choke balun to keep RF out of the coax.

I get what they are trying to convey but chart seems misleading in regards to RX. If you TX with 1w beside another transmitter you'll get +20 on the RX

please read paragraph 5 again

You misunderstood what the annotated (with power levels) S meter is trying to say.

The S Meter should be taken as an "if-then" statement.  So the premise was that IF an 100 watt station could produce an S 5 signal at the other end of the circuit provided by the current propagation THEN, it would take for instance 1600 watts to produce an S 7 signal at the other end. As long as the propagation circuit stayed the same for the length of the test between the two stations. Likewise the lower signal strengths illustrated by the lower powers.

You have to understand it is NOT a literal measurement of power emitted on one end and signal strength on the other end exactly. It assumes that the propagation path and mode will stay the same. So your illustration of putting a 1 watt transmitter near a receiver and getting a +20 dB over S9 signal is NOT what we are trying to say here.

Again, IF this does this, THEN you can expect that to do that, providing all means to get the signal there in the first place remain the same. There is NO assurance that 100 watts will deliver an S5 signal on the other end at any time frequency or place.

And another way to parse your statement that a 1 watt transmitter will produce an S9+20 dB signal at some nearby receiver THEN you would expect that if you increased your power to the 10 watt level you should expect your receiver to show S9+30 dB.  Likewise if you decreased your power by 10 dB to 100 milliwatts then you should expect to see just an S9 + 10 dB signal and reducing even further to 10 milliwatts should produce just an S9 signal.

You did not state nor explain the loss path from your transmitter to your receiver. We can make assumptions to how your set up might be, but you did not tell us this.

Again I am seriously thinking of going over this as a presentation to my club.

I can see this is not well understood by some here, and I think some of our club members may have even less understanding of these concepts.

There is another technical presenter in our club, and he may hit on this in greater detail. I need to see what he comes up with but we shall see.

Being a technical presenter is not the easiest thing to do, because you want to talk about subjects that will inform and entertain your audience, not necessarily what you want to talk about.

This thread happens to be a subject that interest me, but may not strike a chord with others. Not that they would not profit by learning more about this subject, but they may not care to.

Another subject I would like to present, IF I can figure out how the the acquired skill of working DX in the split "pileup" mode. Either SSB or CW.

Demonstrating this in a classroom setting might be difficult. I can put it into words, but they may seem "dry and boring". Beside the fact that not everyone is "in" to working DX or learning how to not only succeed, but how to actually "beat the pants off" of your fellow ham trying to work this guy at the same time you are. It is a game that does not appeal to everyone.  You have to be skilled, and a little mean and ruthless does not hurt. It is about "winning" it is not about how great of a guy you  are. You have to get there the "firstest with the mostest".

Unless it's Bird Watts it's all BS...

How ya figure gunny?

I don't remember spouting any BS.

Bird watts is a CBer's term.

Looks like a kind of stub match.

Here is a YT that talks in general about matching stubs.
https://www.youtube.com/watch?v=EKPgt45Yg98

Fixed a little. Even the rank-and-file 11M op I chat with occasionally doesn't use the term.

But in the words of Prime Minister:



(One of my ham buddies personally knows Anthony; they hit their area car and radio meets on occasion. That gent put a lot of time and money into his station. It's powerful but very clean.)

LOL yea he builds amps and is big in that scene. I have heard his name before, I have to admit that some of the 11m guys are interesting to watch, couple of them are hams as well. I really enjoy watching BBI on youtube, he is an amazing tech and he is quick to call out bad builders and puts out a clean signal, one of the only ones checking his builds out on a spectrum analyzer and O scope out there. He speaks very highly of PM.

A couple tales of PM (from radio reception). One was circa 2010; the other, last summer.

The 2010 encounter involves a Yaesu FT-901DM. 2023, an Icom R-71A (sub-receiver of the station next to where I'm typing this). Both pieces of gear have stock SSB/AM (OEM) filtering, so the blocking dynamic range figures on Sherwood's site are relevant if one wishes to look them up.

I keep an ear (or a receiver) on the Bowl to spot propagation openings. Like the elderly lady in Airplane, I "speak jive" (along with a lot of other societal and hobby colloquialisms). Upon listening to the participants, I may find an opening to an area of interest and it's up to 10 or down to 12M to solicit contacts. The '901 was monitoring Ch6 when I heard a station whose audio and signal stood head and shoulders above the rest. Broadcast quality. No hum, distortion, anything of the sort. And it put the S meter almost to the pin. Dialing the VFO either side of the frequency, the signal simply disappeared once it was out of the filter passband.

One of the members of a forum I admin was involved in a discussion of 11M and related topic. He responded to some comments I posted about the reception report with a description of who the gentleman actually was - and filled me in on lots of details.

Fast forward to last spring:

Band opening to the PNW (where he and Anthony live) and PM is on Ch6 again. He is the only non-local person to ever put the R-71's S meter to the +60dB/S9 mark. I switched over to the main rig (an IC-751A) and got identical results. Preamps on both rigs are usually kept off. Same clean signal which disappears if you tune away from it.

Contrast this with the asshats who get the Golden Screwdriver into the typical RCI 29xx series radio and proceed to turn the rig into something which will wipe out a good portion of 10M if you're anywhere near them. One such local hat has a Technician license, has been cited by the FCC at least once for emissions standards yet continues to it. He's been offered the services of a number of 2-way radio shops to correct the problem yet refuses.

Back to PM:

I wouldn't be surprised at all if the Timtron (WA1HLR) may have spoken to him at some point about station building. Perhaps more.

Only on ARFCOM can we turn a logical informative thread on dB
….into a fucking CB idiots thread

ERP is ERP. Doesn't matter which service or operator is involved; the laws of physics and electromagnetic wave propagation apply universally.

An exercise, should the rest of the participants wish to perform some calculations:

All of the receivers mentioned above have S meters which are calibrated to the 50uV/S9 "Collins" standard. Linearity has been verified via generator from S-5 through full scale, and a delta figure of approximately 6dB per S Unit exists across the range. Calculate the generator power necessary to give a 60dB/S9 reading. This is the received power level (in mV) of the station in question.

Said station is using a 7el wide-spaced Yagi. Any of the Antenna Handbooks (or the NEC modeling tools) should be able to provide a fairly close forward gain figure for a construct of this sort. Path length is 2000mi, give or take.

I'll be generous and assume zero path loss and a major lobe that's precisely aligned with the receiving antenna (mine) but it could be upwards of 10dB depending on skewing.

Take these figures, do the math (keeping Square Law in mind) and tell me approximately how much that amplifier is putting out. And yes, the topic is germane to the overall thread.

Good try OP, but you just can't fix stupid.

The OP's depicted power scale doesn't fly too well in the physical world, in terms of finites.

If redrawn with typical amateur power levels (ex: 100w TX output for S9 at the receiving end) and set up as a free-space, single-point reference (i.e., the ONLY observed signal attenuation being a function of distance per Square Law) then the illustration would be somewhat meaningful in terms of how much delta-P it takes to move the model S meter to full indication.

As Jay Ferguson said years ago "Real life ain't this way".

Propagation modes and intensity, wave addition/cancellation/refraction/diffraction and a host of other things such as degree of ionosphere excitation or absorption and so on can and do determine the intensity of the E and H plane beam fronts at the receiving end of the path. Any one or more of these variables can and will skew the scale.

The model also assumes a perfectly linear S meter. Rest assured, no such thing exists. If you ever had a transistor on a curve tracer, you'll observe three regions: The "knee", where gain increases rapidly from zero to the point it levels out. The region of linearity, wherein it's desirable to operate the device for most faithful waveform reproduction. And the area of gain compression, where the reverse of the "knee" is evidenced. An AGC circuit uses active devices (tubes, transistors, an ADC, etc)  to develop a voltage (or value) proportional to the strength of the signal being received. The receiver's S meter samples that voltage and indicates the strength.

All well and good if both the AGC amp AND S meter driver are being operated within their linear areas. The only problem...small signal devices aren't set up to process large amplitude signal changes (in excess of 100dB dynamic range). Nor are ADCs of lower resolution (as found in the cheaper SDRs). Many designers simply didn't take this dynamic into account when designing the AGC sections of their company's offerings. Anyone who's spent any time in an RF lab or on a repair bench working with the stuff knows what I'm talking about.

How this applies to the guy sitting in his shack running his radio: There will be a region on the S meter where the OP's scale will reflect a sliver of reality in terms of power increase. Above or below that - depending on the design of the AGC circuit and the accuracy of calibration - all bets are off.  

This is not always true. Sure, no measurement is ever "perfect", but in this case the issue is gross linearity errors. That problem does not exist on Flex, openHPSDR-based (e.g. Apache, Hermes Lite, etc.), or Elecraft radios.

In the case of the big three, Yaesu, Kenwood and Icom, it is widely known that the S-meters are wildly non-linear. The cynic in me believes that this is because the marketing departments told the engineering departments to make certain that radios have the appearance of having a "quiet receiver", i.e. a true S5 noise floor appears as, say, an S2 noise floor, and only around S9 are they at all accurate. People coming over into Flex/Apache/Elecraft often have a heart attack. "I spent $4000 on this radio and is as noisy as shit!" No, it's probably got a lot better sensitivity and an S-meter that is accurate.

Nevertheless, I agree with you in spirit. There is no substitute for cubic Watts. Saying otherwise is just a case of die hard QRP love or a case of the poors  I hated the hobby when I tried QRP. It was an OK hobby at 100W. Buying a 500W amp was literally a life-changing event. The hobby became truly enjoyable. And this was with a set of very reasonable wire antennas on tap, all up higher than 50ft. Not much more antenna gain left to obtain without a tower and a Yagi or two.

Note that unlike many here I only own two radios because I spent all the rest of my money on an amplifier

Funny that some are quick to bash 11M. If you stay off the places which attract idiots (same with 75/40/20M) there are opportunities to polish one's radio craft.

I learned a LOT about QRP operation on that band, courtesy of weak-signal SSB work. Never did own an amplifier back in the boom years, and one which was acquired in trade years later was converted to 80-15M operation - for use with an HW-8.

Eventually I'll have a couple towers up at our new place. Not many trees around in the far areas of the yard so phased vertical arrays in the pasture for DX work and wires in the nearby trees for stateside contacts will do for now.

Most of the amateur amps I have are 500-600w class solid state stuff, though a few Drake L7/L75s and Cubic 1500Zs made their way into the collection. As did an FL-2100Z for use with my FT-90x gear. I've never had an amp in operation from this shack yet, but the majority of folks I've talked to state I don't need one. However, nice to have when conditions are bad.

Not sure why I got quoted for bashing 11 meters? I didn't even mention that band?

This is why i only run 12dB gain antennas on 40m

I would sure like to be let in on those designs.

See posts a couple above yours.

You're good. I'm just making some general observations.

Ref: Duga-3 ("Steelyard") array in Gomel, Belarus. 22dBi gain, electrically steerable through ~70deg azimuth and 30deg elevation. B/W was 3-26ish MHz.

Decontaminating the structure from fallout produced from the Chernobyl incident would be challenging. Then there's the matter of disassembly and transport.

I have looked longingly at that structure for years. I remember when I first heard the woodpecker come on the air. With my Yaesu FT101 back in the early 70's. Of course I did not know what it was. and at first no one else seemed to know either. By beam heading most guys put is somewhere in the Soviet Union. It sure was a PITA.

Most assuredly.

Power output of the transmitter was rumored to be anywhere from 2-20MW. Supposedly sited as close to Pripyat as it was to take advantage of the nearby nuclear power generating facilities.

Newcomers to the hobby might wonder why the noise blankers on a lot of gear produced from the late 70s through the early 90s had a noise blanker width adjustment or control. This was the reason...to handle the Woodpecker, as its pulses were wider than the common power-line or ignition noise pulse.

20M seemed especially prone to interference from it. I remember trying to copy W1AW code practice bulletins on the band when the thing would start up with an honest meter-pinning signal. This with random-wire antennas. One could imagine the signal levels a 5 or 6el directional antenna up a couple wavelengths would get.

There is a youtube video of a guy base jumping off the woodpecker antenna. His chute does not open, and it looks like a roadrunner cartoon when he hits the snow.