Suggestions for a 100 Watt Class amplifier for a FLEX-1500?

[K6AER]

I measured the IMD on my HLA-300V and with ten watts drive through a 90 degree hybrid combiner, 6 dB pads in the input to the hybrid for 30 dB port to port isolation.  The HLA-300V second to third IMD was 32 dB. That is about 6 dB better than the exciter with a two tone audio measurement. Power out was 261 watts with 14.4 VDC. At three hundred watts out the two tone the IMD dropped to 28 dB second to third.  Measurements were made on 20 and 40 meters. An Agilent ESM-4402B spectrum analyzer was used. Exciters were a pair of IC-706 MK-IIG's in CW.

Mind you when the ARRL measures an amplifier the dB dynamic range is measured from a single tone which inflates the third order IMD numbers 6 dB.

The amplifier is plenty clean enough if not overdriven. More splatter is created in HF transceiver with compression and mic gain than in most amplifiers.

On another note: If your watt meter shows 250 watts PEP out of a HLA-150 that has two 80 watt devices the power meter needs calibration.

[KH2BR]

I just built a Juma pa100 for a ham that wanted to use it with his flex 1500 and he is very happy with its performance. 100 watts output, auto band selection. Just a very nice amp.

[ZENKI]

Reporting IMD figures and proclaiming that a amp is clean is misleading.  You really need  to include all the data and that includes all the higher order products. The higher order products are more of a worry
than the 3rd order products which are  very close to the nominal occupied bandwidth. As everyone knows 2 tone figures have no relation to how any amplifier is going to work in the real world with a real human voice. This is
especially so when the amplifier has marginal amplifier like this one.

The real problem is that when amplifiers like these  are used with ham transceivers with known ALC or power overshoot design issues.  This combination of marginal CB 12 volt amplifiers and poor transceiver design makes
a mess of the band. This is especially so under voice conditions, we dont talk or modulate in 2 tones, so a dynamic IMD test under real voice conditions would be preferable before calling any amplifier and transceiver
combination clean. These RM amplifiers are CB junk and everyone that I have heard on the air causes splatter. Hams need to remember most high performance receivers today have better dynamic ranges than most spectrum analyzers and when combined withe SDR receiver gives you a clear picture of the REAL IMD performance of any station. Old clunky sweeping spectrum analyzers  are not fast enough nor have enough dynamic range to give you
a true picture of how any amplifier is working in the real world.

The ARRL needs to change its IMD testing regime from 2 tones towards real dynamic IMD testing. Even a white noise loaded notch test would be a better test than the almost useless 2 tone test.

These RM Italy amplifiers are still CB garbage amplifiers and when used with such a poor transmitter like the Flex1500 that has poor IMD performance its a disaster for other users of the band.

I measured the IMD on my HLA-300V and with ten watts drive through a 90 degree hybrid combiner, 6 dB pads in the input to the hybrid for 30 dB port to port isolation.  The HLA-300V second to third IMD was 32 dB. That is about 6 dB better than the exciter with a two tone audio measurement. Power out was 261 watts with 14.4 VDC. At three hundred watts out the two tone the IMD dropped to 28 dB second to third.  Measurements were made on 20 and 40 meters. An Agilent ESM-4402B spectrum analyzer was used. Exciters were a pair of IC-706 MK-IIG's in CW.

Mind you when the ARRL measures an amplifier the dB dynamic range is measured from a single tone which inflates the third order IMD numbers 6 dB.

The amplifier is plenty clean enough if not overdriven. More splatter is created in HF transceiver with compression and mic gain than in most amplifiers.

On another note: If your watt meter shows 250 watts PEP out of a HLA-150 that has two 80 watt devices the power meter needs calibration.

[ZENKI]

It might be a good well built amplifier but what is its IMD performance like. Hams are quick to say this or that KIT amplifier works well, how many hams actually ask the question  is the  amplifier that I am buying clean.

All the kit amps  on the ham market are just copies of databook designs  whose sole objective is to make the sellers money. Not one ham kit HF amplifier kit on the market is designed to be a clean band friendly amp.
Hams are starting to behave like CB'ers with no regard for other users of the band when buying or selecting equipment which is very disappointing.

We will be making progress when some amplifier manufacturer actually  displays IMD data and boasts how clean their amplifiers are and not how cheap they are.

I just built a Juma pa100 for a ham that wanted to use it with his flex 1500 and he is very happy with its performance. 100 watts output, auto band selection. Just a very nice amp.

[AB2YC]

I just built a Juma pa100 for a ham that wanted to use it with his flex 1500 and he is very happy with its performance. 100 watts output, auto band selection. Just a very nice amp.

I was quite tempted to build their HF Radio

[K6AER]

Zenki,

My three main spactrum analyzers are:

HP 8593C
Agilent ESM-4402B
Agilent 89441B

What are you using?

[ZENKI]

I use a Rohde  FSU. But I did not question the accuracy of your test equipment. I questioned your credibility  by proclaiming that this CB amplifier is clean by only reporting 3rd order IMD figures.
If you do want to make a claim that any amplifier is clean based on 2 tone measurements, you could do your credibility some good by publishing all the levels of all the measured IMD products on different frequencies.
Anyone can cherry picky the best IMD results and make wild claims about the performance.

 There seems to be a growing trend from CB hams wanting too promote these rubbish amplifiers for ham radio use and I see that you are very active in promoting this brand on various forums and reviews.
When I see data that  is lacking is details and claims made that the amp is clean with no data to support this claim I have a right to question your assertion that this amp is clean.
 
I know that you have a RF Engineering  background Mike, so the last thing that I want to do is give you lectures on your  profession. However I dont agree that this amplifier is clean because even
at the low power  that you measured  the amp at, the higher products above the 9th order are from  acceptable. These are the products that cause  other operators on the band more grief than the 3rd
order products that are close to the nominal occupied bandwidth.  Look at the FCC  Part 80 Rules for the maritime HF service. The IMD mask  places more emphasis on the higher order product suppression than
the  the 3rd order products. The Part 80 rules for maritime HF service would be a good test of amplifier linearity  for the ham radio service. Most transceivers would fail purely because their high order suppression products are so poor.

I wont change my views because I have measured the HLA300 using  a R&S real time spectrum analyzer when used with a  Icom 706 MK2g, anybody using this combination will be splattering  because of poor IMD  suppression
and ALC overshoot. Everyone of these amplifiers that I have heard on the ham bands causes excessive amount of splatter that makes operating  anywhere less that 10khz away very difficult. If you have to operate this far away from any SSB station that  stations signal is filthy, and in the case of RM Italy amplifiers on the ham bands this is norm.

Zenki,

My three main spactrum analyzers are:

HP 8593C
Agilent ESM-4402B
Agilent 89441B

What are you using?

[WD5GWY]

Zenki,
  Interesting information. Since you did test the HLA-300, what were the
readings that you got. And how hard were you driving it with the 706?
Not defending the HLA-300 but, if you were driving it at full output power,
I'm certain that contributed to the poor IMD performance you noted.
W8JI tested the HLA-150 from RMItaly and with less than 5 watts drive
stated that the IMD products were "acceptable" as long as output was
kept to 90 watts or a bit less.
  One question that I have wanted to ask you, (no not your call sign I know
you won't reveal that) is why do you consider 12 volt devices to be bad as
far as being used in power amplifiers? Can they not be used in Class AB amplifiers
at all? I also note that you feel that higher voltage parts (similar to the finals in my
FT-1000MP MKV) are better. So, is it a voltage problem or simply a design problem
with the devices? I have been trying to read up more on IMD and solid state devices
and have yet to find a clear answer to this question. And since you seem to really
feel very strongly about it, I figured to ask you and ask here. That way, all of us
can learn.

james
WD5GWY
 

[WD5GWY]

Guess I asked too many questions!! 
james
WD5GWY

[WD5GWY]

The silence is deafening. No wonder some people question
ZENKI and his motives. At the very least, he could have
posted his data for HIS HLA-300 test.
Just using the "CB Junk" line is a cop-out. If you're going to
trash something and claim to have actually tested it, then
post the results as others have done. OR quite complaining.
james
WD5GWY

 

[TANAKASAN]

I've done a lot of experimentation with homebrew HF amplifiers and, across all of the designs, one thing stands out. The higher the source drain voltage (or emitter collector) the cleaner the amplifier. I don't know the theory behind this but empirical results have shown that a 48V amplifier is a much better beast than a 12V one.

One other thing helps. About half of the amplifiers have a center tap on the output transformer and use this to feed the DC into the amp, this increases IMD by a small amount but reduces costs and PCB board space. A separate DC feed through a bifilar transformer is the only way to do it.

Tanakasan

[WD5GWY]

I've done a lot of experimentation with homebrew HF amplifiers and, across all of the designs, one thing stands out. The higher the source drain voltage (or emitter collector) the cleaner the amplifier. I don't know the theory behind this but empirical results have shown that a 48V amplifier is a much better beast than a 12V one.

One other thing helps. About half of the amplifiers have a center tap on the output transformer and use this to feed the DC into the amp, this increases IMD by a small amount but reduces costs and PCB board space. A separate DC feed through a bifilar transformer is the only way to do it.

Tanakasan

Tanakasan, thank you!! That is the kind of response I had been hoping for. It would be
interesting to know why the voltage makes such a difference. It would seem that it is a
design flaw in the device itself. But, I do not know that for sure. That is why I was hoping
ZINKI would chime in with some details about his tests with the RM Italy HLA-300 and
what the issue is and why that is a problem.
Again, I appreciate your response. I hope others here with more knowledge than myself
will add to the discussion.
james
WD5GWY

[K6AER]

Wither using a bifilar wound transformer to DC feed a push pull amplifier or using a center tap transformer will neither add nor reduce the IMD of the amplifier. The linearity of the device and how it is being bias has a lot more addition to the IMD being produced.  Beyond that, the magnetics of the frequency being passed through the matching network can affect the efficiency.  Because of the broad band design goal, harmonics are reduced with the addition of low pass or diplexer filters for out of band energy.

Low pass filters have no effect on IMD distortion other than to reduce energy contained in harmonics.

In a classical push/pull design (class AB) the DC being fed through the impedance matching output transformer is very acceptable and in most Cases is the preferred DC feed method. All output stages in 99% of modern transceivers is class AB in a push pull transformer configuration.

I find it interesting that there has been such concern about amplifier IMD when the conventional transceiver in most cases is no better than 30 dB second to third. This in general is 10-20 dB worse than the amplifier they are driving.  Now before the IMD police jump in here I am quoting 2nd to 3rd for it is a fixed and settable number that can be compared and measured in all designs.  Yes there are higher order products but they tend to be less and 2nd to 3rd is very measureable and when the separation is higher (dB)  the higher order products also tend to also be less.

Even order harmonic output (2nd, 4th, 6th etc.) in push pull is generally canceled and there for provide another engineering point to use push pull amplifiers.  

Only in very high output amplifier does the transformer physics become problematic with efficiency’s and heat dissipation.  These amplifiers tend to be single ended in design.

Again going back to the original thread there are many designs that will perform quite well to take a typical 5 watt QRP rig to the 100-200 watt level.

One final note: many expensive amplifiers will produce a high order of IMD and harmonics when improperly driven beyond its design levels.  This includes expensive transceivers as well.

[WD5GWY]

Thank you Mike! I really do appreciate your comments and you posting them here.
That helps everyone learn a bit more. Although I have been a ham for almost 40
years, I have found that there is a lot I still do not know. But, then again, that's a good
thing as it keeps me on my toes! And it give me incentive to learn more.
My dear old dad used to say when the day comes you don't learn anything, you're probably
dead. So, I don't plan to stop learning new things anytime soon. (if I can help it!)
Thanks again, for the explanation.
james
WD5GWY

[ZENKI]

Nah, I dont run away and hide and report deceptive 2 tone figures and tell people that the RM Italy Amps are fine.   Nor do i introduce  the amp has filters and its clean argument. Nobody is concerned about  harmonics because
all radios and amplifiers designed for the ham service have proper levels of harmonic suppression because they do have to pass FCC type acceptance. Amplifiers do not have to pass FCC type acceptance for IMD or splatter in the ham service. My view is that IMD type acceptance for the ham service is a must and should be made the same as that for all commercial HF radios. If these rules were adopted no ham would be using  this CB crap on any ham band.

All I am concerned about is the occupied bandwidth produced when using these amps. The argument that hams drive these amps at half power is a ridiculous  way of suggesting that a piece of equipment that is unacceptable performs in an acceptable manner when a smart CB LID with no test equipment is going to drive it to the highest standard. Many Hams cant even adjust mic gain and ALC properly for a clean signal how the hell are they going to monitor their signals for driving a CB amp cleanly. They cant and there all hams should not use piece of equipment not meant for linear SSB service.

Every ham that I have heard on the air with one of these  RM Italy amps drives them to the maximum. The subsequent mess on the bands is evident on any SDR  receiver which measures dynamic IMD  of a transmission better than largely irrelevant and inaccurate 2 tone IMD measurement data.

When i  measured the HLA300 not  with 2 tones using white noise loading using a IC706 and  HLA300 the IMD was disaster at fully output. The IC706 was putting out between 20 and 25 watts for about 300 to 400 watts across all the bands. Thats  about the manufacturers ratings and where most hams would be running them not at 200 watts. Why would spend money on a 400 watt amplifier and run it at 200 watts?  No normal ham would do this. Its a nice feel good argument for hams who buy this CB splatter garbage  and then run the amp at full power. Everyday on the ham bands we hear these crap boxes spewing out splatter and  occupying 20 khz of bandwidth with splatter  when they drive to the full rated output.

So the end conclusion is that this is CB junk that causes a lot of splatter nothing will change any time soon thats going to make this RM Italy equipment clean.  Reading all the ham forums I wonder if some hams dont have a pecuniary
interest in promoting this CB junk on so many ham forums.

My dynamic IMD measurements of this combination of the IC706 and HL3000 were terrible.

 5 khz 20 db down from PEP
10 khz 27 db down from PEP
15 kHz 35 db down from PEP
20 khz 40 db down from PEP

The results were identical for white noise or voice. On many occasions using  voice these figures were worst because of the crap ALC in the IC706 which cause overshoot. This combination is a disaster for the ham bands
If this amp was to be called clean the 20khz figure would be where the 5 khz  is IE 40db down at 5  khz not at  20KHZ! If you do a 2 tone test on these amps you will have your answer why the IMD performance stinks under voice.
The IMD products after the 7th order plateau   right out past the 17th order. There is no drop in the IMD products  because this amp generates so much distortion. Any hint of over drive generates the worst possible signal.  In anybodies  language this RM Italy amp is a crap box that does not belong on the ham bands because most of the hams that this product appeals to will drive it to the max causing problems for others.