Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

donate to eham
   Home   Help Search  
Pages: Prev 1 ... 16 17 18 19 20 [21] 22 Next   Go Down
  Print  
Author Topic: 1500 pep output max  (Read 94340 times)
QRP4U2
Member

Posts: 262




Ignore
« Reply #300 on: February 22, 2016, 06:14:56 PM »

AA6CJ, you have to read my earlier post, Reply #265 on: January 29, 2016, 01:47:24 PM

Quote from: QRP4U2

It indeed can be measured accurately with the proper detector circuits such as the IC's I discussed, but not with slow thermal or diode envelope detectors that have long thermal or electrical time constants, respectively.

However, you can determine it with a calibrated storage scope.

Here is another try at the math (forget the two-tone tests for the moment that the manf. have to make during testing).

1) Let's say you have a "T" connector connected to your 50 ohm non-reactive dummy load and a storage scope with a probe monitoring the load.

2) You input a constant, maxiumum level voice recording into your SSB rig. You measure a maximum voltage somewhere over the snapshot's time period. At some point in the snapshot, you zoom in to the waveform and observe a maximum of 200 Volts Peak-to-Peak across the load.

3) PEP = V^2/R = (200V P-P/2 X 0.707)^2 / 50 = (100V X 0.707)^2 / 50 = 5000/50 =
100 W PEP power.

Now, 200V Peak-to-Peak voltage is the same voltage you would see with a high amplitude sine wave (or any audio wave of any shape) inputted to your SSB rig and the PEP would still be 100 Watts PEP.

But here is the catch: In normal voice situations, this peak envelope power rarely occurs. In fact, with normal unprocessed voice, the Peak Envelope Power may only be 25-33 Watts PEP.

Phil


« Last Edit: February 22, 2016, 06:22:36 PM by QRP4U2 » Logged

AC0OB - A Place Where Thermionic Emitters Rule!
--------------------------------------------------------------------
Besides, when you're a Ham, you experiment with and improve boat anchors - that's what you do!. Smiley
AA6CJ
Member

Posts: 95




Ignore
« Reply #301 on: February 23, 2016, 04:33:02 PM »


Thank you W6RZ, it is very clear. 
Logged
AA6CJ
Member

Posts: 95




Ignore
« Reply #302 on: February 23, 2016, 04:51:00 PM »

Phil,
...
Part 97 definition says:
(6) PEP (peak envelope power).  The average power supplied to the transmission line by a transmitter during one RF cycle taken at the crest of the modulation envelope under normal operating condition.
...
You and w6rz have clarified the RF cycle.  Thanks.  "...under normal operating condition..." Is ambiguous. W6rz used a single word in his example. Is this normal?

At hf frequencies the RF cycle is so short, and the time period of the envelope so ambiguous, that I am having a hard time seeing how PEP is a good measure of Ssb power. It isn't setting well with me at all.  I don't see an alternative yet.  Do you?
« Last Edit: February 23, 2016, 04:56:11 PM by AA6CJ » Logged
N0GW
Member

Posts: 47




Ignore
« Reply #303 on: February 24, 2016, 07:07:46 AM »

"At hf frequencies the RF cycle is so short, and the time period of the envelope so ambiguous, that I am having a hard time seeing how PEP is a good measure of Ssb power. It isn't setting well with me at all.  I don't see an alternative yet.  Do you?"

Sorry it isn't setting well with you.  What you are describing is simply the difference between Peak Envelope Power and "Average Envelope Power".  (I just made that term up!)  Yes, the SSB envelope waveform shows many peaks of many different amplitudes.  Which peaks do you use to determine your PEP?  That is easy.  It is the it is the highest ones.  Nominally, with a typical 100 watts transceiver, those highest peaks will be at the 100 watts RMS peak-to-peak voltage.  That, of course means that those lower peaks are below 100 watts.  That is the basic physics of un-processed voice SSB.  The Average Envelope Power is usually 25% to 35% of Peak Envelope Power.

Various methods are employed to process audio and SSB envelopes to improve the average to peak power ratio and thus make a louder, easier to copy under poor propagation conditions, signal.  Some are quite successful.  Keep in mind that even with SSB average power only about one third its PEP value, a 100 watt PEP SSB signal usually performs better at a distant location than a 100 watt full carrier NBFM signal.  Even on VHF, DX stations run SSB.

You say you are having a hard time accepting that PEP is a good measure of SSB power.  In fact, it is the only practical way to define it.  Let's say you come of with some time constant for averaging voice SSB power.  For the sake of argument, let's say 1 second.  Now, you pick up the microphone and transmit a nice long "Hellooooooooo" and adjust your output power to 1500 indicated watts.  All good now?  What will you see on that meter when you say "one, two, three, four."  What happens if you whistle into the mic?  3000 to 4000 watts out is what you would see provided your amplifier would not go into peak clipping causing splatter or simply blow up.

As many folks have already said, measuring waveform peak power (thus PEP) is not difficult or complex.  A simple voltage detector with a slow voltage decay is all that is necessary.  The tricky part is not the circuitry to capture the peak voltage, it is coming up with a detector that you can leave in line without introducing a high loss or impedance bump.  That same detector would be needed for any kind of averaging detector, as is the case with those non-PEP RF wattmeters we use.

This is all stuff that has been worked out decades ago.
Logged
W1BR
Member

Posts: 4188




Ignore
« Reply #304 on: February 24, 2016, 08:39:09 AM »

The confusion for most is that every RF cycle should be a sinusoidal waveform due to the flywheel effect of a PA tank of reasonable Q.

Using peak voltage  (instead of RMS voltage, for any AC voltage produced by a RF generator delivering a sinusoidal waveform) to calculate power seems to be something audiofools would use to exaggerate actual meaningful power ratings.

Why peak power isn't determined by the maximum peak AC voltage times .707 is very confusing.

Pete
 
Logged
N0GW
Member

Posts: 47




Ignore
« Reply #305 on: February 24, 2016, 09:54:06 AM »

"Why peak power isn't determined by the maximum peak AC voltage times .707 is very confusing."

Pete,

That is exactly the way it is done.  In the case of the Oscilloscope trace, you multiply the peak-to-peak voltage by .3535 to get the RMS voltage to calculate the power for a given resistance. (or .707 of the peak voltage if you prefer - remember, peak is 1/2 peak-to-peak for a sine wave).  That RMS voltage of the sine wave at the peak of the SSB waveform is used to calculate the PEP power.  Though the diode in at RF power meter does not know RMS from PMS, there is a simple linear relationship between peak-to-peak (or peak) voltage and RMS voltage.  That is taken into account when the meter is calibrated to indicate RF power.

To restate it:  Peak Envelope Power is calculated based upon the RMS voltage at the highest peak or crest of the SSB waveform.  Both an oscilloscope and a diode detector respond to peak-to-peak or peak voltage depending upon the circuit.  We calculate the RF power of that waveform by calculating the RMS equivalent of that voltage, either by direct calculation or by how we mark the scale on a power meter.

That clarify it?

Gary - N0GW
Logged
N4RSS
Member

Posts: 346




Ignore
« Reply #306 on: February 24, 2016, 12:59:32 PM »

"Why peak power isn't determined by the maximum peak AC voltage times .707 is very confusing."

Pete,

That is exactly the way it is done.  In the case of the Oscilloscope trace, you multiply the peak-to-peak voltage by .3535 to get the RMS voltage to calculate the power for a given resistance. (or .707 of the peak voltage if you prefer - remember, peak is 1/2 peak-to-peak for a sine wave).  That RMS voltage of the sine wave at the peak of the SSB waveform is used to calculate the PEP power.  Though the diode in at RF power meter does not know RMS from PMS, there is a simple linear relationship between peak-to-peak (or peak) voltage and RMS voltage.  That is taken into account when the meter is calibrated to indicate RF power.

To restate it:  Peak Envelope Power is calculated based upon the RMS voltage at the highest peak or crest of the SSB waveform.  Both an oscilloscope and a diode detector respond to peak-to-peak or peak voltage depending upon the circuit.  We calculate the RF power of that waveform by calculating the RMS equivalent of that voltage, either by direct calculation or by how we mark the scale on a power meter.

That clarify it?

Gary - N0GW


So if you use RMS voltage as you describe above and divide by R, you've calculated average power, per its definition
Logged
N0GW
Member

Posts: 47




Ignore
« Reply #307 on: February 24, 2016, 04:21:07 PM »

"So if you use RMS voltage as you describe above and divide by R, you've calculated average power, per its definition"

Well, yes, sorta.  Given the amount of confusion on this thread I'd like to be a little more specific.  I think I would write that as:

"So if you use RMS voltage as you describe above squared and divide by R, you've calculated power for that voltage, per its definition"

Saying "average power" is more than a little ambiguous in this conversation.  It is correct if you are meaning the power averaged through one cycle of RF voltage and current.  It is not correct if you are meaning the average of the SSB speech waveform.  In general when we are discussing RF power and SSB, "average power" is usually meant to be the average over some significant time period, usually something like the length of a full sentence of speech.  We use the term Peak Envelope Power as the highest instantaneous (one RF cycle) power generated by a transmitter during that sentence.  (Let's not get wrapped up in the duration of the average - I'm presenting the general concept here.)

As a quick refresher, when comparing the 100 % modulated AM and the same Peak Envelope Power SSB signal, remember that only 33% of the AM transmitter's power is modulation information.  100 watts PEP 100% modulated AM produces 33 watts of sideband power.  66% is the carrier that does not convey information.  The 100 watt PEP SSB transmission is 100 percent information.  (OK, let's not get into a discussion of the conversations we hear on the air are conveying information - by information, we actually mean the desired transmission information - regardless of its value to the receiving parties ;-)  )
Logged
AA6CJ
Member

Posts: 95




Ignore
« Reply #308 on: February 24, 2016, 05:55:57 PM »

In general when we are discussing RF power and SSB, "average power" is usually meant to be the average over some significant time period, usually something like the length of a full sentence of speech.  We use the term Peak Envelope Power as the highest instantaneous (one RF cycle) power generated by a transmitter during that sentence.  (Let's not get wrapped up in the duration of the average - I'm presenting the general concept here.)

I take it from your comment that you are an RF engineer.  This thread was started asking the question whether average power would be better to measure vs PEP given that other modes like CW and RTTY average power is equal to PEP while Ssb average power is 25-33% of PEP.  Based on your discussions about average power you refer to above, and going along with using a "significant time period" as an averaging time constant, why is PEP preferable for our ham radio use? 

I see this as a discussion of power parity between modes...
Logged
N0GW
Member

Posts: 47




Ignore
« Reply #309 on: February 24, 2016, 07:38:29 PM »

"I see this as a discussion of power parity between modes..."

Interesting..  I kinda assumed that was the original question but folks were going off into the weeds figuring out what Peak Envelope Power means.

Now, you will have to define what you think parity between modes might be.  Do you mean that all modes should be given a power level that allows them the same communications capability at some distant point on the earth?  What is the base reference?  JT65, PSK31, SSB voice, Full Carrier AM, NBFM, Spread Spectrum?  I'm not trying to be silly here but what are you concerned about SSB achieving parity with?

Obviously each mode has different communications capability under different propagation and noise conditions.  I saw reference earlier to comparing the average power of RTTY with the average power of SSB.  RTTY is a 100% average power mode.  Under most conditions a 1500 watt PEP SSB signal gets through better than a 1500 watt RTTY signal.  That difference is why RTTY stations typically run high power.  It is necessary even with modern computer based decoding programs.  On the other hand, a 150 watt PSK31 signal would likely get through more reliably than a 1500 watt SSB signal.  Most PSK31 operation is with power levels below 50 watts and world wide DXing is possible.  Each mode has its advantages and disadvantages for different purposes and under different conditions.

So, the question is: Parity with what?

Gary - N0GW



Logged
K4KYV
Member

Posts: 116




Ignore
« Reply #310 on: February 25, 2016, 12:26:47 AM »

In general when we are discussing RF power and SSB, "average power" is usually meant to be the average over some significant time period, usually something like the length of a full sentence of speech.  We use the term Peak Envelope Power as the highest instantaneous (one RF cycle) power generated by a transmitter during that sentence.  (Let's not get wrapped up in the duration of the average - I'm presenting the general concept here.)

To be meaningful, the term "average" power must be integrated over a defined time period that is relevant to the quantity being measured.  For example,  with CW, we could mean the average power over an entire transmission, which would include spaces between characters, words and sentences.  We could even define it as the average power over a QSO, including periods of reception.  Obviously, those definitions would have little meaning in terms of effective transmitter output power.  The chosen time period must be relevant to the nature of the transmitted data.  With CW, the relevant time period would be the duration of the shortest character of the normally transmitted  signal, i.e. a single dit.  With SSB voice, the relevant factor would be a function of the syllabic rate of human speech.

With SSB voice under the 1 KW DC input rule, the FCC addressed the issue of "input power" in a letter to ARRL:

  The following ... may be considered as a presently acceptable method for determining the d.c. plate power input to the final r.f. stage of a single-sideband  amateur transmitter:
  The maximum d.c. plate power input to the (RF) tube(s) supplying power to the antenna system of a (SSB) suppressed-carrier transmitter, as indicated by the usual plate voltmeter and plate milliammeter, shall be considered as the "input power" insofar as ... the Commission's rules are concerned, provided the plate meters utilized have a time constant not in excess of approximately 0.25 second, and the linearity of the transmitter has been adjusted to prevent the generation of excessive sidebands.  The "input power" shall not exceed one kilowatt on peaks as indicated by the plate meter readings.


- Single Sideband for the Radio Amateur (ARRL), second edition 1958, page 14

Average output power from a transmitter is generally proportional to the DC input to the final stage, assuming a consistent efficiency factor. That was the basis for the amateur power limit for many decades and served the purpose well. The 0.25 second (250 ms) time constant prescribed in the FCC statement is consistent with the concept of the audio V-U meter used  in audio recording and broadcasting, in which by definition the mass of the meter movement effectively integrates the signal with a rise time of 300 ms.

Here is a treatise which may be of interest, from the BBC in 1963, on the subject of VU meters, peak-reading meters and loudness measurements:
http://www.mwigan.com/mrw/Publications__49_Edmund_Ramsay_Wigan_files/1963-29.pdf
Logged
W1BR
Member

Posts: 4188




Ignore
« Reply #311 on: February 25, 2016, 07:55:45 AM »

Which begs the question, if a CW transmitter has ALC overshoot, does the peak power of the transient define the peak power for that situation?

Pete
Logged
M0HCN
Member

Posts: 566




Ignore
« Reply #312 on: February 25, 2016, 09:25:30 AM »

Average output power from a transmitter is generally proportional to the DC input to the final stage, assuming a consistent efficiency factor. That was the basis for the amateur power limit for many decades and served the purpose well.
The problem is that if you throttle a PA back by say 10dB without changing the loading, the input power will only fall by a factor of about 3 (square root of 10).... The assumption of linearity does not hold here unless you are changing the load impedance or supply voltage to track the envelope, peak output may be generally proportional to peak input, but this does not in general follow for the average values.

VU and even PPM are largely history in broadcast these days, the BS1770-3 loudness meter being the weapon of choice, but it says little to nothing about instantaneous power, perceived loudness having little to do with peak power in typical program material. I would note that for SSB, the real valued audio envelope and the RF envelope are not the same thing, as the RF envelope is bounded by Sqrt (I^2 + Q^2), not simply by I.

PEP seems to me to be a perfectly reasonable way of slapping a limit on transmitters which can operate in many, many different modes, some having constellation diagrams that are almost constant power  (RTTY, PSK) and some having horrible peaks in the envelope (OFDM, SSB), the alternative is to have the regulator having to set different rules for each mow mode somebody comes up with.

Now personally a DC input limit holds a certain appeal, it puts the emphasis on getting clever with the design again, and we might start seeing things more interesting then the usual class AB stuff that everyone and their dog makes (Doherty, EER, harmonic tuning).

Silly thought but how about a limit in terms of power per Hz of occupied bandwidth (After all, that is how we measure the system noise), so CW (With maybe ~30Hz of occupied bandwidth) would have a limit of 15W, while SSB (3KHz occupied bandwidth) would have a limit of 1,500W... I can hear the screaming from here!

73 Dan.
Logged
G3RZP
Member

Posts: 1269




Ignore
« Reply #313 on: February 25, 2016, 11:51:05 AM »

Quote
Silly thought but how about a limit in terms of power per Hz of occupied bandwidth (After all, that is how we measure the system noise), so CW (With maybe ~30Hz of occupied bandwidth) would have a limit of 15W, while SSB (3KHz occupied bandwidth) would have a limit of 1,500W... I can hear the screaming from here!

That allows more occupied bandwidth. Better would be the inverse so 15 watts for 3kHz and 1500 for 30 Hz - and 3000 for 15 Hz!
Logged
M0HCN
Member

Posts: 566




Ignore
« Reply #314 on: February 25, 2016, 01:05:45 PM »

But the noise power in 3Khz bandwidth is inherently greater then that in 30Hz...

Maybe go with something creative, allowed power is exponential in how close you get to the limit in the telegraphers Eqn, after all, the more intelligence you manage to cram in per joule, the more efficiently you are using the spectrum?

Something clever with COFDM, FEC and viterbi decoders should be allowed more power then something only managing a few kb/s in 3KHz, because it is surely moving more data for its bandwidth? Kind of hard to build a simple meter for that however, so PEP (Or DC input) is probably a good choice.

73 Dan.
Logged
Pages: Prev 1 ... 16 17 18 19 20 [21] 22 Next   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!