Peak Reading Wattmeters

[K9ACT]

Peak watts seems more a curiosity on AM than a really useful tool but curiosity got this cat and I want one.

For about the same price I can get an MFJ stand-alone or a conversion kit for my Bird.

I would rather not have another box in my RF chain so I am leaning toward the Bird kit.

I found this on Ebay and unless someone knows something bad about it, it seems like a better deal and for some reason uses only one battery.

     http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=280260263383

Any thoughts?

One other point, I have seen a few discussions about legal limit relative to AM but it seems to me a peak meter takes all the discussion out of it.  PEP is PEP right?  We can use any DC input or modulation percentage we want as long the  PEP meter reading does not exceed 1500... right?

js

[Tom WA3KLR]

PEP is PEP; all modes the same PEP limit.

But all PEP-reading meters are not created equal, some are less accurate than others.  This is why I chose the Palstar PM2000A.  If I recall correctly, I was able to draw on a survey that was done on PEP meters; QST article?

[WA2ONK]

Jack,
   I have a Bird 43 with the Bird peak reading adapter. I've had that for about 8 years now. I got tired of buying elements for it, and the analog meter still didn't show true peak, it just can't respond quick enough.
   About 2yrs ago I bought a LP-100 watt meter. It's a little pricey even in the kit form that I got. I couldn't be happier! No more elements to change,covers up 1.8mhz thru 54mhz, 50mw to 2500w auto ranging plus a host of other features. It's worth a look.   

http://www.telepostinc.com/lp100.html

[k4kyv]

You are better off using the average reading meter for power measurement since it displays real, honest-to-god watts output from the transmitter.  You can use that figure to calculate peak power, based on the envelope or trapezoid pattern observed on an oscilloscope.

But if your open-wire tuned feeders are fed directly off the transmitter output tank circuit without an intermediate 50-ohm coax-fed link, where would you insert a wattmeter anyway?

Unless you have an accurate, calibrated meter right at the base of a series-fed vertical, or right at the physical point where the feedline is connected to a wire radiator, you will not be taking into account the losses in the feedline, which from my experience, will probably be substantial.

I once used about 140' of very expensive (to Uncle Sam) RG-144 doubly-shielded silver plated coax with special low-loss insulation that was supposed to be good up past 1 gHz, to feed my 160m vertical.  Using calibrated thermocouple rf ammeters for measurement, the best I could ever get was about 92% efficiency, with a 1:1 swr feeding a purely resistive dummy load.  After it had been in place for a few years, the efficiency dropped to under 80% due to deterioration of the protective vinyl jacket on the buried coax.

No matter how you try to do it, output power readings with a typical amateur radio setup will be an approximation at best.  The instruments required for power output measurement that meets FCC standards at a broadcast station, even given the precisely calculated antenna base impedance at a single licensed frequency, would be too expensive to suit most hams.

[AB2EZ]

I agree with Don on this one.

A year or so ago I decided that it would be nice to have a peak reading watt meter. Somehow, this adds new meaning to something I heard at a technical meeting years ago:

"The difference between theory and practice is larger in practice than it is in theory"

Putting aside the issue of accuracy... it would be nice if the peak power reading was equal to 4x the average power reading when the scope shows 100% modulation. Why this is so difficult to achieve eludes me to this day... but here (attached) is a circuit that should get you close. The last (4th) op-amp with a gain of 41 followed by a 56k ohm series resistor is intended to protect the Bird meter (at least to some extent) in case the output of the 3rd op-amp rises to the full 5 volt power supply rail. Separately, by keeping the signal prior to this final op-amp to a relatively low value... it is easier for the second op-amp to quickly charge up the peak detector capacitor.

Stu

[K9ACT]

You are better off using the average reading meter for power measurement since it displays real, honest-to-god watts output from the transmitter.  You can use that figure to calculate peak power, based on the envelope or trapezoid pattern observed on an oscilloscope.

Funny you should mention that because it is the driving force behind my desire for a peak reading meter.

For one thing, I just want to see it happen, i.e. 1600W pep when my carrier is 400W.

More importantly, in the two years I have been putzing with two separate transmitters, I have never seen either one display 100% modulation on a wave envelope pattern. Both have adequate audio power to do so at full power but can't even do it at greatly reduced power.  The peak always stops a bit short of twice the carrier peak.

This is true both with my bench scope and a Heath SB210 monitor.

On the other hand, I have made a divider and pickup for trapezoid monitoring and can do the same with the 610 and always get a clean perfect triangle long before any hint of baselining of the wave envelop pattern.

I think the 610 fakes it because I have never seen the trap exceed 100% no matter how much I over modulate and the home made divider pinches off and squirts out the base line with over modulation.

So, what's going on here?  Something to do with that "theory vs practice" bit?

js

[k4kyv]

I have an old Mirage wattmeter.  In peak mode, it indicates about double the carrier power on voice peaks, when the scope shows 100% on negative peaks and about 130% on positive.  With 100% modulated sine wave audio the peak reading shows about 3X the carrier power.

I mainly use the thing as a reflected power meter with the Gates BC1-T to tune the L-network that the transmitter works into, because the Gates is almost as finicky about working into a 50-ohm load with low SWR as a ricebox would be.

[K9ACT]

I have an old Mirage wattmeter.  In peak mode, it indicates about double the carrier power on voice peaks, when the scope shows 100% on negative peaks and about 130% on positive.  With 100% modulated sine wave audio the peak reading shows about 3X the carrier power.

Are you saying that it does not work right or that I do not understand the math?

I thought PEP was supposed to be 4 times the carrier level at 100%.

I suppose I am confusing voltage peaks with power but I would like to get this straight.

When I set my carrier height at 2 CM on the scope, it should talk up to 4 CM.  The scope is reading voltage and this is 2 X the carrier level.

If that carrier level was 100W on an av reading meter, what will a peak reading meter read at rest and talked up to the 4 CM level.

I would also think that the Peak level would be the same with either sine input of voice peaks. 

js

[ka3zlr]

Hey that LP-100 is NICE....I was looking at that website..Kewell....

[AB2EZ]

Most traditional peak reading wattmeters... including a Bird with a Bird peak reading adapter... do not give an accurate reading of peak power. This is a characteristic of how those peak reading adapters were designed (i.e., op amps don't have a fast enough slew rate, time constants are too short to hold the peak reading long enough for the meter to respond, etc.).

The circuit I posted is giving me a very accurate peak reading (based on what my scope shows and what my modulation monitor shows). It took a while to get it right.

Until I designed a circuit taking into account the issues I mentioned above, the peak power reading with 100% modulation was typically around 3x the average power reading (or less) ... even though the actual peak power was 4x the average (carrier level) power.

Stu

[k4kyv]

I thought PEP was supposed to be 4 times the carrier level at 100%.

I suppose I am confusing voltage peaks with power but I would like to get this straight.

None of these "wattmeters" are real power meters.  They are either rf ammeters or rf voltmeters, usually the latter.  The scale calibrated in watts, based on Ohm's law and the assumption that the load is 50 ohms non-reactive.  If the load is anything different, the power reading will be erroneous.

The most accurate way to estimate your power output with a typical amateur radio antenna set-up is to use the indirect method of measurement. Use a good quality meter like a Bird 43, working into an accurately calibrated, purely resistive 50-ohm dummy load.  Load the transmitter up to normal power input, take a power output reading, and note the DC plate current to the final. Then put the transmitter on the air, and load up to that same plate current.  The power output from the PA stage of the transmitter should then be pretty close to the same as the reading you got with the dummy load. Assuming you don't have a tremendous loss in your feedline, the actual radiated power should be close, but still a little less, than that figure.  As I said previously, with allegedly highly efficient microwave-rated coax I still had 8% loss on 160 when I temporarily placed the dummy load at the antenna end of the transmission line.

Use the formula voltage squared/resistance to calculate the instantaneous peak output power, comparing the amplitude of the unmodulated carrier to that of the crest of the modulation peaks.  If your scope deflection is not precisely calibrated in volts, just measure the number of cm of deflection with unmodulated carrier, and assign that reading a voltage value of 1.  Then measure the deflection at the modulation peak, and divide it by the carrier deflection you previously measured.  At 100% modulation, that should reach a value of 2 (twice the deflection of the unmodulated carrier).  Now, using voltage squared/R that should give a value of 4.  For other percentages of modulation the V squared figure will be  different.  Multiply the calculated peak factor by the previously determined carrier output measurement, and that should give you peak output.

You will get a more accurate measurement with the scope by feeding the rf sample directly to the deflection plates of the CRT, than by using the vertical amplifier, which may not be perfectly linear in response.  Good monitor scopes should have a receptacle that feeds directly to the deflection plates, as well as the normal vertical input through the amplifier.

But again, your actual output will still be a little less because of feedline loss.  If you can temporarily attach the dummy load right at the end of the feedline and get a reading, and compare that to the reading you  get with the meter and dummy load right at the transmitter output receptacle, to determine the feedline efficiency, you can use those figures to calculate the actual power feeding the radiating antenna, assuming your SWR is not far enough away from 1:1 and the  load is not reactive enough, to affect the feedline loss.

[AB2EZ]

 [/quote]

None of these "wattmeters" are real power meters.  They are either rf ammeters or rf voltmeters, usually the latter.  The scale calibrated in watts, based on Ohm's law and the assumption that the load is 50 ohms non-reactive.  If the load is anything different, the power reading will be erroneous.

[/quote]

Don

I'm not sure that this is correct. Most of the "power meters", including the Bird, employ a directional coupler... so that they can measure both forward power and reverse power. If the "power meter" employs a directional coupler... and if the directional coupler is doing its job properly... then the meter will accurately measure forward and reverse power (for an unmodulated carrier)... even if the load is not 50 ohms. It is, however, important that the transmission line between the watt meter's directional coupler and the load (e.g. the tuner or the antenna) is a 50 ohm impedance transmission line.

These meters respond to (at the point where the directional coupler is attached): the envelope of the r.f. waveform: V+50I  if the coupler is set to read power flowing in the forward direction; and the envelope of the r.f. waveform: V-50I if the coupler is set to read power flowing in the reverse direction. The fact that they respond to V +/- 50I is what necessitates their use with a 50 ohm transmission line.

I agree that the meter is calibrated to convert the envelope of the r.f  waveform it responds to (e.g. V+50I) into watts, assuming that it is connected to a 50 ohm transmission line.

This is why, for example, a regular (i.e. no add-on peak power reading board) Bird watt meter will read the same average forward power if you have no modulation or 100% modulation. We know that with 100% modulation (by a sine wave) the actual average power is 50% larger. But since the Bird (and most other wattmeters) respond to envelope of V+50I... rather than (V+50I) x (V+50I)*, the reading doesn't change when modulation is applied.

Stu

[k4kyv]

Isn't a directional  coupler a voltage device? 

SWR is usually defined as a voltage ratio called the VSWR, for voltage standing wave ratio. For example, the VSWR value 1.2:1 denotes a maximum standing wave amplitude that is 1.2 times greater than the minimum standing wave value. It is also possible to define the SWR in terms of current, resulting in the ISWR, which has the same numerical value. The power standing wave ratio (PSWR) is defined as the square of the VSWR.  http://en.wikipedia.org/wiki/VSWR

A real power meter would have to simultaneously measure both voltage and current and integrate the results so that it reads real watts, not volt-amps.  I have seen 60~ a.c. power meters that had 4 terminals, two that went in series with the load to measure current, and two more that went in parallel with the load to measure voltage.  The meter is physically configured so that the deflection of the mechanical movement is a product of both, reading real watts of power.  I'm not sure if such a device designed for rf exists.

The fact that the Bird's average power reading doesn't increase with modulation tells you it is actually reading average rf voltage, not power.  The average envelope voltage of a properly modulated AM signal is the same as the dead carrier.  That's why the plate current and plate voltage meters are supposed to stand still under modulation.  Any  movement is, by definition, evidence of carrier shift.  A thermocouple rf ammeter should kick up 22.5 percent when the carrier is 100% modulated with a sine wave tone, indicating a 50% increase in power output.  My crappy Mirage wattmeter actually does kick up with modulation when in the average power mode.

[AB2EZ]

Don

A directional coupler is a combination of a voltage device and a current device... but it adds the two responses. As a result (as I think we agree), the response of a directional coupler is proportional to the amplitude (not the power) of the r.f. signal  it measures.

At any point along a transmission line (regardless of how it is terminated), there is a wave traveling in the forward direction which has the following properties

1. The voltage produced across the transmission line by the forward wave is in phase with the current produced in the transmission line by the forward wave (if you define the current as in the direction of the forward wave, as measured in one of the conductors). The voltage is equal to the current multiplied by the characteristic impedance (e.g. 50 ohms) of the transmission line

At any point along a transmission line (regardless of how it is terminated), there is a wave traveling in the reverse direction which has the following properties

2. The voltage produced across the transmission line by the reverse wave is 180 degrees out of phase with the current produced in the transmission line by the reverse wave (if you define the current as in the direction of the forward wave, as measured in one of the conductors). The voltage is equal to -1 x the current multiplied by the characteristic impedance (e.g. 50 ohms) of the transmission line

Thus at any point along the transmission line

The total voltage, V,  across the line is equal to V(forward) + V(reverse)
The total current, I, in one of the conductors is I(forward) + I(reverse)

where

V(forward) = 50I (forward)

and

V(reverse) = -50I (reverse)

[Note, from the above equation, it follows that the total voltage is, in general, not in phase with the total current...unless the line is terminated in its characteristic impedance... but that is not an issue for what follows]

Now, if you measure the voltage (V) across the line... for example, with a transformer across the line, or a capacitive voltage divider (for an unbalanced line),

and,

if you also (separately) measure the current passing through one of the conductors... for example, with a transformer in series with one of the conductors or an inductive pickup,

Then, from a little algebra (remember, using simultaneous equations... to solve for the unknowns  ), one can derive (using both the total voltage and total current measurements) the voltage associated with the forward wave:

V(forward) =  [V (measured) + 50I (measured)]/2

One can also derive the voltage associated with the reverse wave

V(reverse) = [(V (measured - 50I (measured)]/2

The Bird element (for example) contains both a capacitive voltage divider to measure the total voltage across the line, and an inductive pickup to measure the total current in the center conductor. If you study the diagrams of a Bird element and a Bird line unit... you will find these cleverly implemented into the structure. By turning the Bird element 180 degrees in its line unit, you reverse the direction of the inductive pickup. The element includes a resistor that is part of a simple circuit that adds the measured voltage to 50 times the measured current. The element also includes a diode peak detector that responds to the envelope of the derived signal.

In summary

I agree with you, Don, that a power meter like a Bird (and most others) responds to the short term average (because of the meter response time) of the amplitude (actually the envelope) of the signal on the transmission line... and that this in converted into a calculated estimate of the power by the meter scale.

However, I do not agree that this type of meter requires a matched termination at the end of the transmission line in order to produce an accurate measurement.

Best regards
Stu

[flintstone mop]

I was also disappointed with the Peak reading function of my BIRD. And I was also interested seeing that my signal was as fully modulated as possible. 1500PEP!!! Never happen GI
Some of the others here have had better experiences actually seeing accurate PEP. Probably invovling electronic circuits and L.E.D. readout that can show the peaks faster than a meter movement.

Fred

[WQ9E]

The price of the used Tektronix and HP scopes seen at hamfests compare very favorably with trying to achieve a "pseudo peak" function from a meter.  The older 500 series scopes are going begging at most fests and most have a bandpass of at least 15 megahertz with most going to 30 or 50.  I "rescued" a Tektronix Type 545B at Peoria last year to keep it from being cannibalized for tubes; total for scope, scopemobile, and 3 plug-ins was $20 and it was working.  Replacing a few black beauties and a quick calibration turned it into another useful scope.   Even the later 7000 series solid state units are often available for under $100 with suitable plug-ins since they are obsolete as far as industry is concerned.

The modulation information provided by any meter pales in comparison to what can be done simply with a good quality scope.  My 2 cents anyway from the home of unwanted and orphaned Tek scopes.

Rodger WQ9E

[k4kyv]

If a meter reads the r.m.s. voltage (or current) from the rf output of a modulated AM transmitter, the reading will increase with modulation.  R.m.s = the square root of the mean (average) of the squares of all the the values as the varying waveform proceeds through a complete cycle.  Also called the quadratic mean, it is a statistical measure of the magnitude of a varying quantity, regardless of waveform.  The 0.707 and 1.414 figures apply only to a pure sine wave.  A thermocouple rf ammeter, for example, measures the r.m.s. value of the rf current passing through it, and kicks upwards with modulation.

Some rf voltmeters read average voltage while others read r.m.s. voltage.  The average voltage does not vary with modulation if the modulated amplifier has good modulation linearity.  The average voltage and/or current will go up or down if there is carrier shift (upward or downward modulation).  Unless the transmitter is running controlled carrier, or perhaps with the case of ultramodulation, carrier shift is undesirable, as it indicates modulation non-linearity in the final, or poor voltage regulation of the power supply.

Most rf voltmeters use diode rectifiers to produce d.c. current to operate a d.c. meter movement.  I need to dig out some of my old radio engineering handbooks and read up on measuring instruments, since I am not sure why some rectifier type rf voltmeters read average voltage while others read r.m.s.  My Mirage wattmeter reads r.m.s., since the reading kicks upwards with modulation, but the Bird 43 is said to read average, if the pointer indeed does not move under modulation.

The directional coupler reads forward vs reflected power, by comparing the voltage coupled through a  capacitor and the voltage induced by the current in a wire.  If the  current passes through the wire in opposing directions the induced voltage will be opposite in phase, while the capacitive coupled voltage is unaffected by the direction of the current through the wire.  I don't think this has anything to do with power; I think we are dealing with voltages here in both cases.  If the SWR is 1:1, there is no reflected power to modify the voltage reading through the capacitor, therefore the meter is strictly a voltage-reading instrument.  When reading reflected power, the meter becomes a current-reading instrument.

I have seen non-thermocouple rf ammeters that used a diode rectifier to read the rf voltage induced into a toroidal coil surrounding a wire carrying the current to be measured, and the pointer did not kick upwards with modulation.  That's how the old WW2-era ARC-5 rf ammeter with the arbitrary 0-10 scale works, except that the current passes through a one-turn loop serving as primary of a ferrite core transformer, instead of a toroid.  But IIRC, the meter in those instruments does kick up with modulation.

A field strength meter, using a short receiving antenna, diode detector and dc meter does not kick up with modulation if the transmitter is functioning properly.  It reads the average voltage induced into the antenna by the signal.

Perhaps someone here could explain off the top of their head the difference between the design of a rectifier type rf meter that reads average voltage versus one that reads r.m.s.?

[K9ACT]

As far as the so called “PEP limit” is concerned read the attached article Don wrote several years ago. It will explain things about that myth much better.

That article is precisely what prompted me to start this thread.

It seemed that all one needed to do was get a Peak Reading meter and not worry about trying to wade through all the arguments and theory.

Nothing is ever as simple as it seems.

js

[Opcom]

---===---

You will get a more accurate measurement with the scope by feeding the rf sample directly to the deflection plates of the CRT, than by using the vertical amplifier, which may not be perfectly linear in response.  Good monitor scopes should have a receptacle that feeds directly to the deflection plates, as well as the normal vertical input through the amplifier.

I could not agree more on the use of a CRT to judge carrier vs. peak. This is why every TX should have a scope built in, even feeding through suitable resistive dividers off the RF PA plate as well as the modulated HV. 60Hz sweep off the power transformer is still the good standby for those not into trapezoids.

It's not practical for some transmitters due to space limitations and no one wants to cut up/hambone a fine relic, but with a rack, a person can usually devote a panel right atop the PA stage with a 3" or 5" CRT in it along with the CRT PS without blocking airflow.

The CRT power can be taken from the transmitter if necessary, but the deflection sensitivity depends on the accelerating voltages and will be more linear during modulation if a separate supply is used. Every old scope I'm forced to scrap gets the power transformer or separate HV supply saved. The XFMRs are kind of large, so a 5" or 7" panel works. Don't bother with less than 3KV unless you have an older CRT with no PDA designed to work well at low voltages. 5KV will make a decently bright trace but for something that looks really good, a 10KV or more setup is better, ripped off an old Tektronix or HP scope. Used scopes with 'other' problems are dirt cheap because no one knows how to fix them and they are gutless wonders today, so they are not too hard to build-in or add if it's planned out.

p.s. I -hate- LCD displays for a scope whether laboratory or built into the $5000 plastic radio. It's an abomination. OK so back to the thread, did not mean to change the subject.