The RF path from TX output flows through the capacitor and then the resistor to ground. So far we have talked about using a 30pF 500V 5% mica capacitor and a 1/4W 5% carbon composition resistor. I propose we change the components to the following:
30pf 500V 5% ceramic C0G capacitor (NOT stocked by Mouser)
Substitute two 15pF capacitors in parallel.
2 x 15p 500V 5% ceramic C0G capacitor (Mouser # 140-500N5-150J-RC, in stock, $0.11 each)
1/4W 1% metal film resistor (Mouser # 660-MF1/4DC30R0F, in stock, $0.06 each)
The 30 ohm resistor is listed as a standard value.
Although mica capacitors have low ESR, C0G ceramics have even lower ESR
Mica ESR = 9.94718 ohms at 1.6MHz
C0G ESR = 3.31573 ohms at 1.6MHz
The actual load resistance seen by the TX output is the sum of the series connection of the capacitor ESR and the 30 ohm resistor. The capacitor ESR (Equivalent Series Resistance) will change the load seen by the TX from our desired 30 ohms to 30 + ESR ohms. ESR is frequency dependent and is minimum at the high end of the band and will be about double at the low end of the band, we should either specify 1.6 MHz as the standard frequency or standardize a small range of frequencies at the high end of the band.
As examples, at 1600kHz (a candidate for part15.us standard test frequency), ESRs are as follows:
Mica (DF=0.3%): 9.94718 ohms
C0G ceramic (DF=0.1%): 3.31573 ohms
The equation to calculate ESR from the published DF (dissipation factor) is
ESR = DF / (2*pi()*F*C*10^-6)
where:
DF is Dissipation Factor as a percentage or decimal equivalent
F in MHz
C in pF
Example ESR calculation for C0G capacitor: DF = .1% = .001:
ESR = .001 / (2*pi()*1.6*30*10^-6) = 3.31573
You can use the Windows calculator in scientific mode or a spreadsheet
NOTE: the ESR increases as frequency is lowered. ESR approximately triples from its 1.6MHz value at .55 MHz. This pretty much mandates that we define a standard test frequency like 1.6MHz, at the high end of the band.
I am working on simplifying an equation that factors in the ESR to yield the TX Pout value from voltage measured across the 30 ohm resistor. Such an equation would be included in the part15.us standard measurement procedure.
Phil
So what results did you get?
Out of curiosity I rounded up some parts from my stock and performed a test with a "close to" simulated antenna. I don't have the exact part values Phil spec'd but they are close and here's the report.
The transmitter I used was my vintage Knight Kit broadcaster, a tube unit from my early days of part 15 activity. This was designed to drive a 10 foot antenna with no coils or other tuning and with no mention of grounding in the manual. The KK has a "hot" circuit with the oscillator circuit ground connected to the power line while the preamplifier and chassis are connected to this ground via a 47K resistor. (this might explain the characteristic hum of this transmitter). So the question is "what to use for an output ground?". Not wanting to expose myself and my test equipment to a "hot" ground I chose the chassis for the simulated antenna ground. This is probably not the RF ground the unit "sees" in operation but it has to suffice.
The simulated antenna capacitor was 27 pF and the load resistor was 29 ohms which were the closest to Phil's values in my stock. The transmitter operating at a frequency of 1400 kHz delivers a power of 5 mW to the resistor. Maybe more power would result by using the actual circuit ground but since this is only a semi-serious measurement this was not pursued.
Though this report is not of general value due to the scarcity of the KK broadcasters it was an interesting excursion in using our new "standard simulated antenna".
Neil
The "First Light" experiment is very interesting and (I think) certainly demonstrates that we will now be able to discover the output strength from a typical part 15 transmitter.
Neil, can you find the input power to the final on your KK? Oh, wait, I think there is no final on those early models.... they run straight from the oscillator. Well that's something to discuss.... if there is no final stage, how can the 100mW to the final be applied as rules go?
I find it rather interesting to find a "standard" simulation antenna when every transmitter out there has a different output design and matching network, not to mention the variations in loading schemes and grounding systems. They are all not going to fall within the simulated antenna being presented here.
True that it is a starting point. But unless the TX's are built with a standard output, then this particular standard won't apply. Now if the TX's had a cross the board standard output such as 50 ohms, then we could develop a simulated antenna load that would work across all models of TX's.
I don't need to tell anyone here that a wire will present a certain impedance strung across the roof, but a 1/2 inch copper rod will have its own impedance, so will a 102 inch CB whip, or a wire inside a pvc pipe. Each one of those will present a different loading approach, in turn presenting a different impedance characteristic, thus how can you have a standard simulated antenna when it is based on just a simple wire alone and expect that to apply to all transmitter tests?
This is why I say bench testing cannot really be applicable to testing these transmitters on antenna systems. Forget about the simulations, go with the real deal and hook one up to an outdoor antenna system. If the TX is not really designed for that, then it's pointless to even go any further in reviewing that TX on an outdoor antenna system, or run tests on it against a TX sporting a 1/2 inch aluminum rod or 102 inch CB whip.
JMT
RFB
50 ohm transmitter output impedance does not usually apply to Part 15 AM because coax is not often used, in order to meet the antenna system length limits of Section 15.219(b).
In the case of the proposed standardized dummy antenna, 30 pF represents the capacitance of a typical 3 meter thin rod vertical monopole. The 30 ohm resistor in series represents the ground loss resistance for the case where the loading coil is internal to the transmitter assembly.
An actual antenna would have different capacitance and resistance. A large diameter monopole would have more than 30 pF of capacitance, and the actual ground resistance can have a minimum of about 10 ohms loss resistance for a really good ground to about 200 ohms for a poor ground.
Moreover, because of space constraints, Part 15 hobbyists all have their own individual limits to what sort of antennas they can use. We have heard about antennas built into window frames, and grounds that use the neutral wire in the house electric power wiring. Therefoe, the true output impadance seen by a Part 15 AM transmitter can be just about anything.
Using some standard dummy antenna for bench testing can provide a means of realistically making comparisons between different transmitters. Specifying a standard antenna would not be practical because actual antennas used by hobbyists vary greatly because of the large variability of individual circumstances.
I have personally used a 30 pF mica dipped capacitor in series with a nominally 47 ohm carbon composition resistor (measured to be 46 ohms) as a dummy antenna load for the last five years to test various transmiters.
"Using some standard dummy antenna for bench testing can provide a means of realistically making comparisons between different transmitters. Specifying a standard antenna would not be practical because actual antennas used by hobbyists vary greatly because of the large variability of individual circumstances."
I agree and this is my whole point. So in effect a simulated loading system for the bench is in of itself it's own category and would not apply at all to any outdoor antenna system, thus a standard for testing transmitters on an outdoor antenna system would be impractical unless a standard was developed for outdoor antenna systems, with a +/- tolerance range window to account for the variations in antenna systems being used.
Perhaps what is needed here is the development of specific categories of testing. Bench with simulated loads and outdoor systems with standardized configuration, preferably of the common type..ie 3 meter rod over a basic, and then elaborate grounding system. Carl is right to a point about things being universal in as far as the ground system, but even that won't have any universal to it because a ground system consisting of 4 radials 20 feet long will present a certain impedance to the TX vs one with 20 radials 40 feet long.
Would a Rangemaster plugged into the simulated load properly tune like that of an AMT5K? Would a Procaster do the same? What about a TH and it's auto tuner or even using it's ATU?
Where is the window of standard developed from when your dealing with different designs with different loading requirements...even for the bench?
RFB
Hi Guys
The only reason why I brought up the meter was That some of these transmitters
have a 50 ohm output.
Yes testing a system that has a built in antenna is done better by
the way you are talking about. Still the AMT 3000 ,Talking sign, an Ramsey
transmitters have a 50 ohm output.
This meter could show the output power into a 50 ohm dummy load.
Also if someone is buliding a antenna like the Phil has on his site
with the loading coil then it could show the reflected power back into
the transmitter.
Now it no going to do much for a Rangemaster,Procaster,and the AMT 5000
which really need field test and bench test like you are talking about.
Really the best way to test the strength of the Rangemaster,Procaster,and
AMT 5000 transmitters is to have a very good Field Strength meter that
measures in microvolts. This way you could stand 3 meters out from the
antenna and take readings from each transmitter. I would set each
transmitter up in the same location one at a time and take readings.
This way you are putting each transmitter under the same exact conditions.
This would really show which one performs the best under what conditions
a person chooses, but they all need to be test in the same location
under the same conditions. This way you will get the most accurate
results.
I did this when help setting up large broadcast stations.
Anyways the meter was really only intended to test for output power from
a transmitter that has a 50 ohm output. Also you could test for input
power to the final while the transmitter is under a 50 ohm load.
I have tested many transmitters and found a lot of them that claim 100mw
output would have near a 300 to 400 mw DCinput power to the final.
Phil was really right when he chose the type of biasing he choose. I do
not know if Phil remembers years ago when I ask him how many miliwatts did he
get out from the AMT 3000 verses the 100mw DCinput.
Anyways I knew about the type of biasing he choose for the AMT 5000,and
really there is no better known way to bias the transistor in the final
circuit. Class E biasing is the very best way to do it. Most transmitters
are Bias class C and that ends up in a lower output power,and a very high
input power.
So my hands are off to phil for a well designed transmitter since Class
E biasing takes time to get it to work right.
I also do amateur Radio and I have designed and built all my own equipment.
I only have one receiver that not home made and that a Sangean ATS-803A
Shortwave receiver. All my HAM equipment is home made from receivers,transmitters,RF power amps. I thought being in Amateur radio should
be seeing what you could do with making your own equipment,and plus it fun
when you make contacts with things you designed and built.
I really think the AMT 5000 is going to give the Rangemaster a true challenge.
Who knows it may outperform the Rangemaster.
I really cannot waite to see the results.
SKW40
The certification of the Rangemaster or other transmitter with a particular antenna has nothing to do with testing its efficiency. The 30 pF capacitor of the dummy load approximately simulates the capacitance of the whip antenna, which should be removed for output power testing on the test bench. The 30 ohm resistor in series with the capacitor simulates the ground loss resistance. The proposed dummy load works with the Rangemaster et al. as well as with the SSTRAN or any other Part 15 AM transmitter.
"The 30 pF capacitor of the dummy load approximately simulates the capacitance of the whip antenna, which should be removed for output power testing on the test bench. The 30 ohm resistor in series with the capacitor simulates the ground loss resistance. The proposed dummy load works with the Rangemaster et al. as well as with the SSTRAN or any other Part 15 AM transmitter."
The question of 30pf simulating the whip is in of itself questionable considering that figure of 30pf stems from a much thinner and less rigid conductor as well as comprised of strands of wire and not solid such as the whip..leaving out also the ground difference from a static 30 ohms to a ground consisting of dirt and radials and ground rods...all of which the bench is going to simulate on a scale broad enough to cover outdoor situations of various styles of antenna systems?
Ok so if I understand this..the bench testing is enough to establish a baseline reference to any TX for just the bench..right?
Now..what about the real world outdoor setups, which I am quite sure is the majority of setups out there, be it using a Rangemaster or SSTran or any other Part 15 AM transmitter?
I do not think the bench baseline testing is going to tell us anything about real world outdoor results.
Even if the changes between a 10 foot wire and 102 inch long CB whip is minor, there is still that major change in everything from the test bench to the outdoor system by just the surroundings itself, then add in the ground system, loading system, lead in's for audio/power etc, all of which WILL have far more than minor changes in loading effects, thus an effect on tuning and system performance.
I doubt it very seriously the bench simulated setup is going to give us any ballpark area to build a baseline reference for reviewing transmitters when one primarily is designed to drive a piece of wire vs one that is designed to drive a 1/2 inch aluminum rod or whip over a ground system..even a very basic ground system of nothing more than a ground rod.
In honesty, and I think others would agree..especially those who do have an outdoor antenna system, do not care about bench testing or setting up a simulated load for the bench, but more interested in how the thing will perform on said outdoor system...if the TX in question is designed for that.
If it is not...end of review as far as the category of outdoor Part 15 AM systems. As to the bench...well anything can be tailored on the bench. This is all not saying that establishing a bench baseline reference isn't worth it, I just think it will not do any good when applying tests to outdoor systems.
ADDING: Let me expand on that point.
When initially testing one of my LPB's, which has a 50 ohm output, was connected to the internal 50 ohm load of my Motorola RD 2012 analyzer. Yes..the dummy load is spot on at 50 ohms as is the analyzer calibrated.
Now while on the bench peaking the low pass filter performance into this simulated antenna (dummy load) I tuned the filter for peak attenuation and minimum losses.
The next step was to connect to the coupler and sample the output for low pass filter performance.
Guess what kids...the notch curve of the filter changed. Oh and BTW, the coupler input is also spot on 50 ohms.
So why the change?
Reactive and resistive loading does different things at RF. And obviously the simulated load vs the actual coupling load changed things enough to cause the low pass filter to loose some of it's notching performance, thus required re-tuning into the actual coupler instead of the simulated load, even though both are of 50 ohms impedance.
RFB
Carl,
A while back I measured the DC input power for the Knight Kit AM tx. and posted but the search function doesn't work for me and I am going on memory that it was around 180 mW. Allied Radio "certified" that the unit complies with part 15.204 of the rules but this was in 1959. I suspect this was based on the field strength limit rather than on the input power but I can't be certain.
On the subject of bench testing, RFB makes a viable case regarding the predictive value and hence the utility of the bench test as proposed given the different designs of the transmitters. Any transmitter designed to use a 3 meter radiator and ground system needs to be able to drive this simulated load since it represents a load which may be encountered with a real antenna.
The Knight Kit was designed to drive the 3 meter (or 10 foot) radiator and provides no loading coil or other tuning for the antenna system yet it did deliver power to the simulated antenna. A comparison of this with another transmitter assumes both can drive a real antenna with an impedance similar to that of the bench load. Granted the bench load may not be the optimum for each transmitter but it does represent a realistic value to be expected for an antenna system.
Now, let's compare the Knight Kit bench measurement with my home brew transmitter measurement. The KK produced 5 mW into the load compared with mine which produces 83 mW. At first glance it appears that mine will produce a stronger radiated signal but there is a big "gotcha" with this. My transmitter is designed to use an antenna system with a loading coil but for the bench test I chose to use just a 29 ohm resistor without the series capacitor assuming that the loading coil resonates with the antenna capacitance and the transmitter sees no reactive component in the load. The 83 mW compared to 5 mW may not be reasonable in predicting relative radiated field strength since the 83 mW does not include losses in the loading coil and the tests were done under different load conditions.
Including the loading coil in the bench test introduces another non standard variable which would be difficult to reproduce elsewhere and would significantly reduce the power delivered to the load on a bench test. By how much? Using the measured RF resistance of my coil I calculate that my transmitter bench tested with Phil's standard load and including my coil would deliver 57 mW to the 30 ohm resistor. ( I choose not to make the actual measurement since I would have to disassemble my antenna system and move the transmitter).
If my attempt to account for the loading coil is accurate then we can compare the 5 mW from the KK to 57 mW from my transmitter. According to RFB this may not be an accurate predictor of the comparative range of the transmitters but the 11 to 1 ratio of the powers suggests that my transmitter will achieve greater range than the KK. Time permitting I will do a comparative range test.
This rather involved example illustrates the difficulty which will be encountered in getting useful bench test results for various transmitter designs. However, if comparing transmitters with internal loading coils (Rangemaster, AMT-5000, Talking House, Procaster) this task is greatly simplified.
Neil
After reading all the many points of view, I came up with a possible name for this reviewing standards project:
OPERATION JELLO
Jello Carl, Jow are you?
Yes, it is getting complicated but that is the reality of this situation. My hope is that it will converge such that something meaningful will result. What we are trying to do is eliminate as many cofactors as possible in defining a standard test. I think we are on the right track to compare internal loading coil transmitters. Comparing other designs may require full disclosure of the test details so the reader can assess for themselves the data.
I am not as skeptical as RFB regarding the utility of bench test results as proposed. In an attempted correlation between bench test power and actual performance with an antenna there will be a margin of discrimination. I doubt this margin will differentiate 70 mW performance from that of 80 mW but this remains to be determined.
Neil
FIM readings maybe? What do you propose, RFB?
I think bench testing is worth doing,
and the more you learn there, the better.
But I do think the outside testing means
a lot, too. Many years ago Radio World
ran an article about testing the LPB
AM-2000 - was that the name of it??
It was their 100 mW AM transmitter.
What a nice package. Too bad it cost
so much money. Anyway - I seem to
remember that they had to move it
around to a couple of different outside
locations so get it to work well.
(This statement will sound a little
dramatic - - but I sort of sit here
in awe at what you guys know.)
Meanwhile, I'm living over at the Carrier
Current thread. The more I learn - the
more I see that I really know less and
less and less. Then I see there is
more and more to learn. (Huh?)
Keep doing what you are doing, gentlemen.
Bruce, DOGGRADIO
Radio That's Going To The Dogs