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

donate to eham
   Home   Help Search  
Pages: Prev 1 [2] 3 4 5 6 Next   Go Down
  Print  
Author Topic: Flex power genious xl  (Read 5348 times)
VK3BL
Member

Posts: 1790


WWW

Ignore
« Reply #15 on: April 08, 2019, 06:06:01 AM »

Most very high power solid state amplifiers have a very strong third harmonic component. The Genius amplifier is using two 1500 watt pallets and even at legal output could see a third harmonic of 150 watts.

Second harmonic component.

If it was third, then by definition IMD3 would be -10dB in the above scenario, and not filterable by any practical means.

Even Harmonics = Filterable

Odd Harmonics = Passband

I can't believe all the pompous talk and postulation below when the first comment made should have been immediately picked up on. We should be Elmering here not telling people they've got no chance unless they have a college degree.

Logged

J.D. Mitchell BA  - VK3BL / XU7AGA - https://www.youtube.com/ratemyradio - Honesty & Integrity
W9IQ
Member

Posts: 3563




Ignore
« Reply #16 on: April 08, 2019, 06:20:59 AM »

Jarrad,

I believe you will find that the diplexer filter design is primarily there to absorb, rather than reflect, the third harmonic of the fundamental RF frequency. That is the minus -10 dB point. It has nothing practical to do with IMD.

 - Glenn W9IQ
Logged

- Glenn W9IQ

I never make a mistake. I thought I did once but I was wrong.
VK3BL
Member

Posts: 1790


WWW

Ignore
« Reply #17 on: April 08, 2019, 06:39:27 AM »

Jarrad,

I believe you will find that the diplexer filter design is primarily there to absorb, rather than reflect, the third harmonic of the fundamental RF frequency. That is the minus -10 dB point. It has nothing practical to do with IMD.

 - Glenn W9IQ

I understand the principle purpose of a diplexer filter, nor dispute that the claimed levels are not right; what I'm not certain of is whether it makes any sense to speak of it as absorbing the 'third harmonic', when we could talk about it absorbing the second, fourth, 6th or whatever, and retain compatibility with common Ham convention.

I'm sure there is a good enough reason to refer to it as such as you suggest.

Certainly, it is worth exploring the different conventions when half the thread is about the 'third order' IMD performance of Diplexer based vs Traditional Filter based dual LDMOS designs. 

As you suggest, if the primary purpose of the filter is to absorb the third (6th?) harmonic of the fundamental, then one cannot directly draw any correlation with 3rd order IMD performance improvements.
« Last Edit: April 08, 2019, 06:42:22 AM by VK3BL » Logged

J.D. Mitchell BA  - VK3BL / XU7AGA - https://www.youtube.com/ratemyradio - Honesty & Integrity
W9IQ
Member

Posts: 3563




Ignore
« Reply #18 on: April 08, 2019, 07:37:19 AM »

Jarrad,

You are quite right about the higher order harmonics. These will also go into the dump resistor since that part of a diplexer is simply a high pass filter. My comment about the 3rd harmonic was simply contextual to the thread.

There is probably a case to be made that the diplexer helps to stabilize the finals and thus improves linearity. I don't know if that is the causal to improved IMD performance. Perhaps someone can enlighten us.

- Glenn W9IQ
Logged

- Glenn W9IQ

I never make a mistake. I thought I did once but I was wrong.
K6AER
Member

Posts: 5745




Ignore
« Reply #19 on: April 08, 2019, 07:41:51 AM »

A diplexer is not just to channel harmonics into a termination load but keeps the power from heating the device when power would be reflected under a normal low pass filter design. The third harmonic has most of the harmonic power in a solid state design application.

IMD, that is the  reciprocal mixing of a two tone signal has nothing to do with primary signal harmonics.
Logged
W9IQ
Member

Posts: 3563




Ignore
« Reply #20 on: April 08, 2019, 07:51:39 AM »

A diplexer is not just to channel harmonics into a termination load but keeps the power from heating the device when power would be reflected under a normal low pass filter design. The third harmonic has most of the harmonic power in a solid state design application.

IMD, that is the  reciprocal mixing of a two tone signal has nothing to do with primary signal harmonics.

Isn't that what I said in my two posts above? I am trying to determine if there is a subtle difference that I am missing...

- Glenn W9IQ
Logged

- Glenn W9IQ

I never make a mistake. I thought I did once but I was wrong.
M0HCN
Member

Posts: 566




Ignore
« Reply #21 on: April 08, 2019, 07:57:28 AM »

Not sure I have ever seen the 3F component referred to as the second harmonic before, certainly it is not common RF usage.

In a push-pull stage (which almost all the high power HF solid state designs are), the harmonic at three times the carrier is usually both the biggest and the most difficult to filter sufficiently (Particularly if, as is common practise, you are trying to cover multiple bands with a single LPF).

While it is possible to view the harmonics as a degenerate case of IMD, it is not usually all that useful because the IMD components that we care about are those that cannot be filtered out, which are typically the difference terms where the difference falls in band, so 2F1-F2, 3F1-2F2 and such. The harmonic terms 2F1, 3F1, 4F1, 5F1 and such are easily filtered as they appear well above the band we are working...

Normal ham practise seems to be to consider the multiples of a single tone to be harmonics while the difference components that fall in band are considered to be IMD, which makes sense because the treatment of the two is different.

The reflected harmonics from a filter do NOT have to wind up as heat in the finals, and can actually be used to reduce heating by making the voltage (or current) waveform more closely approximate a square wave, which is arguably a better use for them then heating up a dump resistor.

73, Dan.
Logged
W9IQ
Member

Posts: 3563




Ignore
« Reply #22 on: April 08, 2019, 08:10:55 AM »

Dan,

Can you elaborate on this last point?

Quote
The reflected harmonics from a filter do NOT have to wind up as heat in the finals, and can actually be used to reduce heating by making the voltage (or current) waveform more closely approximate a square wave, which is arguably a better use for them then heating up a dump resistor.

Thanks,

- Glenn W9IQ
Logged

- Glenn W9IQ

I never make a mistake. I thought I did once but I was wrong.
K6BRN
Member

Posts: 1355




Ignore
« Reply #23 on: April 08, 2019, 08:23:55 AM »

Hi Dan:

Hacking is nice, but generally does not pay well.  Businesses, like people, cast their real votes with their wallet.

Engineers emerging with a 4 or 5 year (not three, BTW).undergraduate degree in EE, plus FPGA/ASIC and/or Systems Engineering skills are VERY well paid, right out of the box and generally have no problem finding work.

There is no real substitute for training, in the professional world.  Only a very, very few can "make it" as a professional on their own.

BTW... the reason you'd want to "code up" your own FFT in Matlab is to perform clockwise/bitwise co-verification with a power optimized VHDL/Verlog simulation.  There are about a million ways to reduce complexity, power and cost in an FFT implementation, and the built in Matlab FFT function knows nothing about optimizations that yield simpler designs, perhaps at the expense of some small error terms.

Its all about training.  Which is why trained professionals are so valuable to a business that needs to develop a competitive product and be credible to investors and clients.

No such requirements in amateur radio... its all about fun... and learning.  Because its a hobby, not a paying profession.  The two are not equivalent in any way.

Brian - K6BRN
Logged
K6BRN
Member

Posts: 1355




Ignore
« Reply #24 on: April 08, 2019, 08:46:40 AM »

Glenn:

Quote
Dan,

Can you elaborate on this last point?

Quote
The reflected harmonics from a filter do NOT have to wind up as heat in the finals, and can actually be used to reduce heating by making the voltage (or current) waveform more closely approximate a square wave, which is arguably a better use for them then heating up a dump resistor.

I presume that's a rhetorical question?  Smiley

Brian - K6BRN
Logged
W9IQ
Member

Posts: 3563




Ignore
« Reply #25 on: April 08, 2019, 08:49:31 AM »

Not at all. He seemed to have an idea that I have not heard of before and I am interested in learning about it.

Diplex filters for linear amplifiers, on the other hand, were well described in the prior century.

- Glenn W9IQ
« Last Edit: April 08, 2019, 08:55:50 AM by W9IQ » Logged

- Glenn W9IQ

I never make a mistake. I thought I did once but I was wrong.
K6BRN
Member

Posts: 1355




Ignore
« Reply #26 on: April 08, 2019, 08:59:17 AM »

Glenn:

I see.  Very appropriate.

I'd like to learn,  too.  Other than step function testing, when is driving current or voltage to look like a square wave a good idea in a communications system?

Brian - K6BRN
Logged
M0HCN
Member

Posts: 566




Ignore
« Reply #27 on: April 08, 2019, 09:02:50 AM »

W9IQ Consider a saturated power stage (If the stage is not in saturation then there are likely other things you can do to improve the efficiency), then with a broadband termination, the voltage across the device and current thru it overlap, producing heat.

In the ideal case, by reflecting the harmonics with the correct phase relationships you can square one of these waveforms up producing far less overlap and hence less heat, this is class F, and is in the literature. In practise in a HF amplifier, I have not managed better then about a 10% improvement in DC->RF efficiency, but that is worth having as it makes the cooling easier. It is instructive to take a power amp and measure the efficiency with various lengths of line between the amp and the LPF. You can do much better in a microwave or UHF stage because stub lines are just more practical up there.

For an example in a current fed (Choke in the DC feed) 6M amplifier you could use a 1/4 wave coax line at 156MHz to provide a short circuit termination to the drains at third harmonic, as well as providing some of the capacitance needed to tune out the parasitic inductance at 6M, not tried that, but it is a simple example, and might work quite well.

K6BRN, The four year undergrad thing is a US perversion, ours get it done in three by not doing the Major/Minor thing.

I work designing the doings of equipment for the TV broadcast industry so yea, it can be done without the paper, I have product out at NAB this week in contention for an award, and while the training is nice, If you can design digital logic, VHDL and constraints files are not a big leap even for an amateur.

Personally, if writing a bit exact simulation model for a VHDL core, I would be reaching for C or C++ in all probability, Matlab is a wonderful tool but not in my view for this.

73 Dan.
Logged
W9IQ
Member

Posts: 3563




Ignore
« Reply #28 on: April 08, 2019, 09:27:29 AM »

Thanks, Dan for the explanation. I was thinking class AB. Now I understand your intent.

- Glenn W9IQ
Logged

- Glenn W9IQ

I never make a mistake. I thought I did once but I was wrong.
K6BRN
Member

Posts: 1355




Ignore
« Reply #29 on: April 08, 2019, 12:59:36 PM »

Hi Dan:

Quote
K6BRN, The four year undergrad thing is a US perversion, ours get it done in three by not doing the Major/Minor thing.

I work designing the doings of equipment for the TV broadcast industry so yea, it can be done without the paper, I have product out at NAB this week in contention for an award, and while the training is nice, If you can design digital logic, VHDL and constraints files are not a big leap even for an amateur.

Nothing perverted at all about the U.S. higher educational system,  Seems to work pretty good.  As does the U.K.  system.  I was at the University of Edinburgh not to long ago and spent a week with students there:  Scottish, English, German, American....  I was impressed.

Once of my critical mentors was a Brit by the name of Jack Bryan.  Ex RAF.  Very smart, very capable, very trained.

Also did plenty of co-development work with GEC Plessy, back in the day and got to know their engineering staff very well over more than a few pints.  (pre-ITAR)  Also well trained.  Kind of like in the U.S.

So I know it works about the same over here as in your neck of the woods.  There are some differences, but training counts.  Always. 

Glad to hear you skipped all that and were able to make it on your own.  Very few do.  It's a real achivement.  Really.

Also glad to hear that you think engineering is simple and are not intimidated by it.  Of course its simple.  All of it.  Kids do it in High school with robotics.  Easy to blast through 4-5 years of university and come out with a degree, right?

But qualification in ham radio is not what's required to be professionally successful.  Some incredibly hard work in school and in engineering is the usual path, for those that can do it.  Again, the minority.  You see, for me, it took effort.  And still does.

But that is NOT what my replies above were about.  They were about the (non-existant) requirement that amateur radio operators have in-depth technical knowledge of radios, antennas, etc.  And the fact that a lot of pseudo-experts on the forums, who could not make a living at what they profess to be experts in if their lives depended on it, hammer on what they call "appliance users".

As I KEEP ON SAYING - ITS A HOBBY OPEN TO EVERYONE.  AND IF THE PERSON WHO JOINS THE HOBBY AND ENJOYS IT AND EVEN LEARNS, THEY ARE SUCCESSFUL.  NO NEED FOR IN DEPTH TECHNICAL EDUCATION OR WEALTH OR FAME.  JUST ENJOY IT.

Regarding criticism of education - my obsevation is that the detractors are invariably the ones who've not tried it.  Sounds a lot like some antenna arguments over wire antennas that are popular on the forums.  And just as silly,

Best Regards,

Brian - K6BRN



Logged
Pages: Prev 1 [2] 3 4 5 6 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!