...Some examples of base loaded antennas ground the bottom of the loading coil. The feed point is tapped at a point above the bottom of the coil where ideally the load can be matched to the transmitter. ...
This could be worth a try, but probably there is no point along the length of a coil with one end connected to r-f ground where the reactance is zero, so the SWR at those points will be high.
Another consideration is that the loading coils are internal in many modern Part 15 AM transmitters, making it difficult to find a tap point by experiement. Also I think neither end of those coils connect to r-f ground.
On other forums there have been posts claiming the AMT5000 does not operate in class E mode and that tuning according to the procedure described in the manual will not result in optimum output. Additionally, it has been claimed that variations in the ground loss resistance of various installations requires adjustment of the drain capacitor.
First of all, quoting Nathan Sokal, the original patent holder for the class E amplifier, he defines Class E as follows: "In Class-E, the transistor operates as an on/off switch and the load network shapes the voltage and current waveforms to prevent simultaneous high voltage and high current in the transistor; that minimizes power dissipation, especially during the switching transitions". The AMT5000 most certainly conforms to this definition when tuned anywhere near resonance. Efficiency varies with tuning, but in all cases it's still operating in Class E mode. An oscilloscope connected to the output transistor drain (test point T4 on the AMT5000 board) would certainly be helpful when tuning, but a scope is not at all necessary, and a scope should not be a requirement as most users probably don't have one available. The instruction manual describes how to set the toroid coil tap jumpers and how to adjust the trimmer capacitor (C1) to the peak input current value. The instructions then say to rotate the trimmer one full turn clockwise for best efficiency. One full turn of C1 increases the capacitance by about 0.7pF.
The following graph plots the input current against the C1 capacitance. The best output transistor drain efficiency region is in red. The efficiency varies from a peak of 98% by only two percent over this region. The curve is substantially the same over the frequency range of 1350kHz - 1700kHz. We don't need an oscilloscope when we can get the same result without one. 
Here is a curious thing. With the power control at a FIXED setting, the input power will vary with the same curve. So, the input power will peak at the peak input current measurement point. But, that's not the optimum tuning point! The instructions say to set the power control for 100mW AFTER tuning to the optimum setting. If you do that and then reduce C1 back to the maximum input current point, the input power will increase dramatically, well over the 100mW limit and the efficiency will reduce to about 80%. Even at the reduced efficiency, the power increase above 100mW will likely produce a higher field strength than at optimum tuning, but you will be violalating the 100mW rule. If you then decrease the power control to 100mW, the field strength will be reduced below what you had at the the optimum efficiency tuning point.
The AMT5000 is optimized for an antenna system with a ground loss resistance of 30 ohms. This optimization is achieved with the default S8 jumper ON setting. This connects a 220pF capacitor to the output transistor drain. With this setting, the efficiency varies by 2 percent over the ground resistance range of 10 to 70 ohms with a peak at 30 ohms. The efficiency drops almost linearly in the 70 ohms to 100 ohms region to 88%. Removing jumper S8 will bring the efficiency up at 100 ohms, but that isn't mentioned in the manual because the field strength difference would be negligible. Besides, users who buy a more expensive transmitter are likely concerned with getting good range. Even a basic radial wire ground system should easily bring the ground resistance below 70 ohms, hopefully more like 30 ohms or less.
I had never read this 219 rule: " This regulation does not dictate a field strength limit but specifies that the combined length of the antenna and ground lead cannot exceed 3 meters "
I mean it looks like a strange rule, the FCC 15 does not tell anything about the max height the transmitter can be placed nut this 219 is a tricky one, someone lives on the 150th floor of a skyscratcher, at his/her balcony ( one that is surrounding the complete floor ) he/she places a legal FCC 15 part AM transmitter, a 3 metre antenna and the grounding on the metal of the balcony ballustrade metal. The grounding in this case is much more as the regulated 3 metres nut it is not possible to ground it at another manner.
Same is if someone uses the LPAM transitter indoors, than is the most used grounding, the central heating system, that has also a lot of more metres of metal....
If used outside, than a grounding of at least 10 metres is needed to protect the aquipment also against lightning. In all 3 the above examples the transmitter is suddenly illegal..
Call me a donkey but is the FCC really going to control the depth of the grounding or giving a fine if the grounding is connected to the central heating or metal balcony ballustrade?
I guess the short answer to your last question is - they might. Because they consider the balcony to be part of the ground.
But in that case, you would be better off to use something like a Talking House or Talking Sign transmitter. Both are grounded to the power outlet (via the neutral, I believe) and they are both FCC certified for use in that fashion. String the wire antenna up vertically on that balcony, and go for it.
Now, if you use the ATU with the Talking House, that might be another matter. I believe that the Talking House with ATU is certified for use with the ATU being a meter off the ground, which you obviously can't do high up in an apartment. There has been lots of discussion here whether or not an elevated ATU would be legal, even if you didn't ground it (if you do ground it, you run into the same problem with other elevated installs, in that the length of the ground + antenna would be greater than 3 meters or 10 feet).
It's really up to the FCC inspector to determine whether your installation is legal or not if you are so unfortunate as to get the visit. As you can see, even a certified transmitter can be installed illegally.
Yes, they were all tested on the same day. However, as I've mentioned previously, I've noted wide variations in range (which is directly related to field strength) and sound during different parts of the day on my transmitters, and that is something that wasn't accounted for in the testing.
And yes, all transmitters were tested with identical grounds (even the Talking House's power plug neutral pin was grounded to the copper stake) and their own respective antennas, however efficient or inefficient those may be (i.e., the ProCaster with its tube antenna, the Talking House with its wire antenna, etc.) It looks like for those aspects of the testing there was consistency. And certainly, with broadcast engineers with a claimed 1,767 (or so) combined years of experience, I don't think that you can cast aspersions on the measurements that were obtained, as far as they went.
I still believe, however, that the testing was skewed positively towards the factory assembled units. If test units could not have been obtained from the manufacturer that were guaranteed to be the most effective they could be (and I understand that that was the case in several instances), then more than one example of each transmitter should have been tested. And from my first point, they should have each been tested at similar points in the day, if possible (or at least close). Since the results were all relative, it really wouldn't have mattered if the tests were spread over multiple days, although I can certainly understand that that would have meant a lot more work.
For whatever reason, I can now access the Challenge document on the other site, as well as their Forums (I guess they realized how stupid it was to block access, or perhaps it was always meant to be temporary). I hope that anyone from 'over there' reading this post recognizes that these criticisms are meant in a positive way, as I think that with a bit more testing, the issues could be laid to rest. And I certainly have an interest as to which transmitter is the most efficient at getting a signal out, although for me, that isn't the be all and end all. Some transmitters, for example, allow asymetric overmodulation (such as the Rangemaster), which gives them even more of an advantage over those that don't (such as the Talking House). Some simply sound better. However, I don't hold out much hope for that, as they're still defending themselves against the 'know nothings', and continuing to harp on the 2,957 (or so) combined years of broadcast engineering experience the testers and reviewers have.
<Sigh>
Wow, I haven't been on this forum for several years, and it's still the same hotbed that it was the last time I logged in. Edwin Armstrong would be right at home. <g>
Personally, I was very intrigued by the test and appreciated all the time and trouble that went into it. We need more of these kinds of experiments. For those that weren't happy with the way they came out, feel free to organize your own shootout. The more we test, the more we will learn.
I have no problem with the test (and the results) - as far as it went. But it was hardly complete and to publish the results as the (implied) final word, with winners and losers, is premature.
Because some of the results were surprising, more testing should have been done to eliminate any chance of error. That's just plain common sense, whether you're a scientist or an engineer or whatever.
And for whatever reason, the testers have been unwilling to listen to ANY comments or suggestions (or criticisms), taking them VERY personally. That was not the intent, at least from this quarter. And their attacks back at those who were attempting to help just make me wonder what they have to hide.
I don't particularly care what the testers backgrounds are, and how many umpteen years of broadcast engineering experience they claim to have and how many others support them for their own personal reasons. The testing could have been more rigorous, as per my previous posts.
Maybe next time it will be.
I think the object was to keep things simple. Use a common ground, fixed measuring distance, tune for max legal smoke whether manual or automatic and use a standard measuring device. Now if you guys want rigorous, then they should have measured transmit bandwidth, audio frequency response, second and third harmonics, frequency stability in ppm, IMD etc.
There's simplicity and then there's simplicity. If you simplify too much you risk compromising the results, as I believe may have been the case here.
Really, one set of results doesn't mean all that much. Particularly given the various sources that the transmitters came from (it could have affected the results either positively or negatively). Now THAT'S pretty simple to understand, and seems to have been overlooked.
And regarding the other measurements, they stated quite clearly up front they were just measuring the field strength, not the sound quality and other aspects of the signal. The field strength is the primary driver of range, but it's not all of it (compression, asymetric overmodulation, etc. can all positively affect range, even the type of programming). I guess that's my other issue with the entire exercise - a casual reader could look at the results and interpret that the 'winner' was the 'best' transmitter, when there ARE other factors that would determine that (although I'm sure that there would be a lot of discussion as to how to rank those other factors).
Being unlicensed is a proud tradition that is helped by events like the "Transmitter Challenge," because it brings attention and discussion about the world's most intelligent hobby (the one possible exception being hobby doctor).
Pity that rational discussion was viewed by some as personal attack, but even hurt feelings haven't muddied the overall beneficial result of so much part 15 publicity.
If one comparison was overlooked in trying to test the transmitters on equal footing, it is this:
Whereas the certified transmitters all have particular antennas under which certification was granted, and deliberately were employed in making the measurements, the same cannot be said for the AMT5000.
In the case of the AMT5000 a minimal starter antenna is supplied with the kit in the form of a 10-foot white wire so the device can be tested right away after construction.
It is common knowledge that better antennas can be attached to the AMT5000 in any form desired by the user, and performing an antenna upgrade may reasonably be expected to enhance the performance.
As I've said before, but was ignored at the time, the fiberglass pole used to suspend the white-wire antenna has unknown properties that, if a helical-wire were embedded as is sometimes the case, would place a capacitive drain against the antenna and lower the result.
There is yet another variable that has not been disclosed or questioned, but was mentioned to me in a private e-mail, which I will leave alone at this time until we see if someone else mentions it.
It would have been fairer for all the transmitters to have been tested with the same antenna, just as they were all tested with the same ground system. The results from the I.A.M. Talking House show that there are significant differences with the SAME transmitter and 2 different antennas.
Good point.
Maybe they should have never tested anything at all.
Then WDCX, we wouldn't have this splendid thread.
Or, rather than testing nothing, they could have a Challenge II and put to use all of the excellent ideas put forward by part 15 thinkers?
Somehow I doubt that those guys will do a Challenge II. After all, their first Challenge was perfect in every way, according to them, and they have 5,327 combined years (or so) of claimed broadcast engineering experience.
No, I suspect that it will have to be some truly unbiased and freer thinking individuals, perhaps from here. If I had a FIM, I'd do it myself (with a Rangemaster, AMT5000, Talking House, ProCaster and Talking Sign). I even have several ProCasters left, as well as multiple Talking Signs (the other transmitter measurements can be compared to the Challenge I results). Who knows, perhaps I will eventually find that elusive FIM.
But in the meantime, I'm taking those first set of singular results with a healthy grain of salt.