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

   Home   Help Search  
Pages: Prev 1 2 3 4 5 [6] 7 Next   Go Down
  Print  
Author Topic: Amplifier PEP power ratings...  (Read 19912 times)
VE7RF
Member

Posts: 212




Ignore
« Reply #75 on: January 02, 2012, 03:27:01 AM »

Another marketing orientated wrong with intermod is comparing it to PEP rather than to a single tone of a two tone test. This of course makes the IM spec look 6 db better than it really is.
There are some that say that is ok as long as you define how you are looking at it. But in reality it is not the proper way to look at IM product levels.

The proper way is to compare the IM level with a two tone test to the level of one of the two tones. Comparing to PEP gives a 6 db better IM figure. The 6 db increase for PEP comes from the addition of the two tone power levels that make up the PEP level.

If you are going to compare IM to PEP you need to also sum the two IM signals (upper and lower) which will add 6 db to the IM figure. Then it will come out the same as if you were comparing IM levels to one tone of a two tone signal level as seen on a spectrum analyzer.

This is referenced in the Eimac “care and feeding” book.

When looking at a spectrum analyzer you don’t really know what the PEP level is as you are looking at each individual tone level.


Hope this posts ok. My first post on Eham.

73
Gary  K4FMX


##  dunno if I agree with this.  S meter's are peak reading devices to start with. [fast attack, slow decay].  Each tone of a 2 x tone signal should be 6db below pep output, [and a little less, if you factor in IMD.]  If a fellow is 30 db over S-9 on your [peak reading] receiver's S meter....and off freq, his splatter is S-9, then I would say his splatter is "30 db down from pep".  But in that case...I'm either well below his freq...OR well above his freq..but not both, since I'm only listening to one sideband at a time. Dunno if this is all valid or not.

##  at one time, the arrl used db below one tone..then switched to db below pep.  I think they gave some kind of reason at the time, but I don't know what it was.

## The problem with 2 x tone tests is the spacing used.  If u used say 100-200 hz tone spacing, like 1400 hz and 1500hz, then most of your imd is IN band.  Now if you used say 100 hz + 3100 hz, then all your  imd will be OUT of band.

##  With normal ssb, even if we used imd below pep S meter readings, unless we specify at what offset freq we are listening at...and how many db down below pep,it's meaningless. Big difference between noisy low bands...and quiet upper bands.

Later...Jim  VE7RF
Logged
K4FMX
Member

Posts: 17




Ignore
« Reply #76 on: January 02, 2012, 08:12:37 AM »

Hi Jim,

It is difficult to really measure IMD with a receiver by just listening. If we are going to do comparisons of amplifiers then a narrow filter of a few hundred Hertz is required.
Agree that the S meter will respond to PEP but doing that with a regular SSB filter not only gets the PEP value of the two tones but it may also include some of the IM products added in too which will increase the apparent PEP.

I contend that there is no easy way to measure PEP on a spectrum analyzer because the bandwidth would have to be wide enough for the modulation envelope and that could include some IM products.

by using a narrow band filter and looking at each tone individually and comparing to each IM product individually you will get the true IM level difference.

When measuring IMD it is important to use tones with enough spacing so that you can see each individual IM product separately. If the tones are too close together the IM products may fall too close in and you will not be able to accurately measure them.

For voice modulation which contains many tones spaced close to each other, many of the IM products will fall within the band pass of the receiver while listening to the wanted signal. As I have mentioned earlier and I think that you have too, switching side bands on the receiver will reveal close in IMD products.
On the wanted side band those IMD products add to the wanted signal and slightly increase PEP of the signal as observed.

My point being if we are to accurately measure IMD products a good standard needs to be followed in order to compare one amplifier to another.

73
Gary  K4FMX
Logged
W8JI
Member

Posts: 9304


WWW

Ignore
« Reply #77 on: January 02, 2012, 04:02:45 PM »

To use a spectrum analyzer, you use peak storage mode and multiple slow sweeps, with a very narrow filter.

The problem with doing it over the air is QRM and noise masking signals. It is actually better to use a narrow filter and tune around slowly on a receiver.

Over the air IMD is difficult to measure. The law, and the general rule also, is if you hear it on the next channel loud enough to bother things, and it is not the receiver, it is too strong.

The letter of the law is we are not supposed to bother things outside the normal channel width.

73 Tom
Logged
VE7RF
Member

Posts: 212




Ignore
« Reply #78 on: January 02, 2012, 11:54:01 PM »

To use a spectrum analyzer, you use peak storage mode and multiple slow sweeps, with a very narrow filter.

The problem with doing it over the air is QRM and noise masking signals. It is actually better to use a narrow filter and tune around slowly on a receiver.

Over the air IMD is difficult to measure. The law, and the general rule also, is if you hear it on the next channel loud enough to bother things, and it is not the receiver, it is too strong.

The letter of the law is we are not supposed to bother things outside the normal channel width.

73 Tom

## agreed on using narrow filters to search for imd.  I use a pair of 500hz. [a pair of 250's is too narrow].

I tune around, up/down with the pair of 500 hz filters on the main RX, [on the left ear]...while listening to the main channel on the sub rx, with the usual ssb filter.[on the right ear].  Then I know who is talking at any given time.

Typ, hams will add their rx bandwidth, to the other fellow's  TX BW, to arrive at total BW, which is incorrect.  IE:, station A is on 3820 khz, LSB, and station B is listening on 3823..and hears a tiny bit of splatter.  Station B then tunes down to 3817, and also hears a tiny bit of splatter.  Then station B concludes, that since he can hear a tiny bit of splatter up and down 3 khz, that Station A must be "6 khz wide" .  Now if station B had used a pair of 500 hz filters, he would have heard that station A is only 3 khz wide.

When I TX on my MK-V...and listen to myself on a 2nd MK-V, tuned way off freq, then toggle between Class A  and class AB... the difference is like night and day for imd, huger diff.  [that's  75 watts in either mode on tx] . When I toggle between analog ssb  and DSP ssb, on tx, their is another big difference too.  Hit the PTT, and mic gain on zero, the noise floor drops 20 db, when using analog SSB.  When using dsp ssb, the noise floor increases 20db, and I hear this broadband hash on either side of the vfo tx freq. George at Inrad warned me of this effect, several years ago.  The MK-V is unique, since it's the last of the xcvr's that will generate either an analog or dsp ssb signal.

On a similar note, the mic pre-amp in the mk-v is more junk imo. With mic gain at 9 oclock or higher, you can easily hear this hiss, and crap, on a 2nd mk-v...3' away, and loosley coupled.  I gave up on the internal mic pre-amp. Instead, I installed a new rca jack on rear apron..and fed the semi line level audio directly into the Analog balanced modulator..[ via  a 220 uf cap]. Ok, that works good,  BUT I lose my vox.  The fix for that is to feed the audio 1st into a mixer, then split it into 2 x paths, one into the new BM input..and the 2nd path into the phone patch input.  With mk-v mic gain on zero, just enough level is fed to the patch input, to make the vox work.

I tossed all the heil mics, hand mics, and everything else i had..and went to a RE-27 broadcast mic.  I also use a sennheiser boom headset.   The difference in imd/harmonic distortion etc, between a heil mic and a re-27 is like 3 magnitudes apart.  Of course none of this will do much for TX imd..except for the class A.  And the class A  will cook a mk-v, since there is NO air between heatsinks 1+2.

Later... Jim  VE7RF
Logged
K4FMX
Member

Posts: 17




Ignore
« Reply #79 on: January 03, 2012, 08:43:56 AM »




To use a spectrum analyzer, you use peak storage mode and multiple slow sweeps, with a very narrow filter.



Yes I agree that you need a narrow filter and slow sweep to see IM products.
My point was that you can not easily see PEP and IM products on a spectrum analyzer. If using a narrow filter as needed to see the IM products it will also only show the individual frequencies of a two tone test. It will not show the PEP. So you really don't know what the PEP is when trying to see how far down the IMD products are from PEP.

73
Gary  K4FMX
Logged
VE7RF
Member

Posts: 212




Ignore
« Reply #80 on: January 03, 2012, 11:37:39 PM »




To use a spectrum analyzer, you use peak storage mode and multiple slow sweeps, with a very narrow filter.



Yes I agree that you need a narrow filter and slow sweep to see IM products.
My point was that you can not easily see PEP and IM products on a spectrum analyzer. If using a narrow filter as needed to see the IM products it will also only show the individual frequencies of a two tone test. It will not show the PEP. So you really don't know what the PEP is when trying to see how far down the IMD products are from PEP.

73
Gary  K4FMX

## Then try it with regular voice on ssb !    It will resemble a dogs breakfast. The 2 x tone won't exercise the various Hv /bias/screen supplies etc. And it won't show alc induced imd distortion, nor  overshoots.

## The SA  won't show pep at all, just good for looking at various im products..but that's it's main function anyway..and it's limitation.

later... Jim  VE7RF
Logged
N4RSS
Member

Posts: 258




Ignore
« Reply #81 on: January 04, 2012, 12:39:29 PM »


"RMS power" is new physics also.

RMS is the square root of the mean value or average value squared, or quadradic mean. It cannot be applied to power in any meaningful way.

We have RMS voltage
We have RMS current
We cannot have, as a useful quantity, RMS power

We have continuous power, we have peak power, we have average power.

RMS power is as fictitious and silly as peak-to-peak power. It has no meaning in power. It is something the audio industry cooked up to sell things to audio-phools.




All correct exept that RMS is defined as the square root of the arithmetic mean of the squares of the original values, not the square root of the mean squared. Averaging the squares versus squaring the average, big difference
Logged
N4RSS
Member

Posts: 258




Ignore
« Reply #82 on: January 05, 2012, 07:43:13 AM »


 Heck,  any 100watt light bulb is 200 watts peak....and 400w  peak to peak. 


There is no such thing as "peak to peak" with power since there is no such thing as negative power
Logged
KB8E
Member

Posts: 24




Ignore
« Reply #83 on: January 05, 2012, 08:45:18 AM »

A 100-watt light bulb powered by 120 V rms sinewave is 100 watts average, 200 watts peak, no such thing as peak-to-peak since power doesn't flow backwards (negative), 122.5 watts rms.

Sam
Logged
K8LV
Member

Posts: 14




Ignore
« Reply #84 on: January 09, 2012, 10:25:12 AM »


By the way, none of it is RMS power. There is no such thing as RMS power. There is peak

I am tired of hearing this expression. To those of us who work in the world of engineering the term "RMS power" has a very real meaning and is used frequently by many who fully understand all of the basic concepts of power, energy, averaging, etc. It simply means power that is computed/measured using some variation of the simplest formula P=VI where the values of the variables are RMS values, and not average values. Up to fairly recently in electronics history, instruments that could measure RMS values were not very common and most standard instruments measured only long-term averages. Hence, there are important contexts in which the distinction is necessary - especially in cases where waveforms are not periodic (at leats, over the averaging period).

Until fairly recently, to my knowledge there were never any ham radio devices capable of measuring RMS power. They have always been average-reading and relied on the fact that for periodic waveforms the ratio between avg. and RMS is a fixed, known quantity which is just boiled into the overall calibration factor. This has changed starting recently as advanced IC technology has brought to market chips that can properly perform RMS measurements in real time - both analog and digitally.

Eric K8LV
 

   
Logged
VE7RF
Member

Posts: 212




Ignore
« Reply #85 on: January 10, 2012, 02:48:19 AM »


By the way, none of it is RMS power. There is no such thing as RMS power. There is peak

I am tired of hearing this expression. To those of us who work in the world of engineering the term "RMS power" has a very real meaning and is used frequently by many who fully understand all of the basic concepts of power, energy, averaging, etc. It simply means power that is computed/measured using some variation of the simplest formula P=VI where the values of the variables are RMS values, and not average values. Up to fairly recently in electronics history, instruments that could measure RMS values were not very common and most standard instruments measured only long-term averages. Hence, there are important contexts in which the distinction is necessary - especially in cases where waveforms are not periodic (at least, over the averaging period).

Until fairly recently, to my knowledge there were never any ham radio devices capable of measuring RMS power. They have always been average-reading and relied on the fact that for periodic waveforms the ratio between avg. and RMS is a fixed, known quantity which is just boiled into the overall calibration factor. This has changed starting recently as advanced IC technology has brought to market chips that can properly perform RMS measurements in real time - both analog and digitally.

Eric K8LV
 

   


## OK, so RMS V x RMS current = RMS power...or does it = average power ??

##or put another way. Lets use my fluke .."true rms dvm" [model 87]..and use the dvm to measure the 120 vac out of the wall outlet.  Lets say it's 120 v rms.   lets put a 120 ohm resistor across the outlet.  RMS V  / 120 ohms = 1 A RMS current.  Power is 120 watts.  OK, now if 120 watts rms is the heating eq, then why can't we call it rms watts ??

## If we had used 120 VDC /120 ohms..we would  end up with the same 120 watts.

##  rms v = .707 x peak.   Average V = .637 peak.   So what is the deal on average = .637 peak ?   I think that's where the confusion lays.

Later...Jim  VE7RF
Logged
KB8E
Member

Posts: 24




Ignore
« Reply #86 on: January 10, 2012, 07:53:12 AM »

Jim,

A lot of the confusion with regards to AC power is the fact that the power waveform (instantaneous power) is not a constant i.e. DC, but in fact, with a sine voltage waveform and resistive load, is actually a sinusoid at twice the frequency of the voltage waveform with an offset equal to the peak such that the power waveform never goes negative. This power waveform has a peak value, an average value, and even an rms value. In most cases, the peak and average values are the most useful. The average value would produce the same average power (heat) in a load as would a DC power waveform (DC power waveform is constant and equal to the average value of the AC example). It turns out with sinusoidal waveforms for voltage (and current) that when the rms values for these are used, the result of the usual power equations is the average power. Therefore, using rms values for the voltage and/or current waveforms is convenient to get average power. I don't know about the Average V = .637 peak since the average voltage should be zero. Maybe the average voltage in a rectified sinewave.

At the risk of causing even more confusion, in the audio world, the term "rms power" doesn't really mean rms power at all, even though such a number exists, but rather, "mean sine power" which is average power with a sinusoidal voltage waveform.

It really is possible to calculate rms power for any any waveform, but its usefullness is limited in the real world.

Sam
Logged
N4RSS
Member

Posts: 258




Ignore
« Reply #87 on: January 10, 2012, 09:14:32 AM »

The answer to what constitutes an RMS calculated value is readily available on the internet.  If one inspects the formula by which an RMS value is calculated, it can easily be seen why VRMS and IRMS multiplied together yield P average and why P average is the heating value.  Likewise it can be seen why a power number calculated using the RMS approach is not the same thing as P average.

The .636 number is actually 2 divided by pi and is computed from the standard approach for determining the average value of a function.  In this case, the integral of sin over 180 degrees, which is 2, divided by the interval length, pi, yielding 2 over pi.
« Last Edit: January 10, 2012, 09:21:03 AM by N4RSS » Logged
W8JI
Member

Posts: 9304


WWW

Ignore
« Reply #88 on: January 10, 2012, 03:06:47 PM »


By the way, none of it is RMS power. There is no such thing as RMS power. There is peak

I am tired of hearing this expression. To those of us who work in the world of engineering the term "RMS power" has a very real meaning and is used frequently by many who fully understand all of the basic concepts of power, energy, averaging, etc. It simply means power that is computed/measured using some variation of the simplest formula P=VI where the values of the variables are RMS values, and not average values. Up to fairly recently in electronics history, instruments that could measure RMS values were not very common and most standard instruments measured only long-term averages. Hence, there are important contexts in which the distinction is necessary - especially in cases where waveforms are not periodic (at leats, over the averaging period).

Until fairly recently, to my knowledge there were never any ham radio devices capable of measuring RMS power. They have always been average-reading and relied on the fact that for periodic waveforms the ratio between avg. and RMS is a fixed, known quantity which is just boiled into the overall calibration factor. This has changed starting recently as advanced IC technology has brought to market chips that can properly perform RMS measurements in real time - both analog and digitally.

Eric K8LV
 

   


Well, you better get all the other engineers in the world on page with you:

http://www.eznec.com/Amateur/RMS_Power.pdf

http://www.n4lcd.com/RMS.pdf

http://www.hifi-writer.com/he/misc/rmspower.htm

Every textbook I have that mentions power in any detail says RMS voltage times RMS current is average or heating power.

I haven't seen anything, outside of Hams and Audiophools, who use RMS to describe average or heating power. Please give a reference.  Even the Communications Standard Dictionary only uses RMS to describe voltages or currents.

So where does it appear as applied to power?
Logged
KG0DB
Member

Posts: 52




Ignore
« Reply #89 on: January 29, 2012, 08:00:49 PM »

Okay, okay.  You guys are miles over my head technically speaking.  The question is..............the question is.................what company, what amp isn't "overstating" their specs?
Logged
Pages: Prev 1 2 3 4 5 [6] 7 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!