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 [7]   Go Down
  Print  
Author Topic: The Longest Journey Starts With A Single Step  (Read 52878 times)
W3RSW
Member

Posts: 606




Ignore
« Reply #90 on: January 24, 2015, 06:22:37 AM »

It'll be interesting to see an Icom take on this thread, if ever.

We could start bugging car and boat makers to go to a 54 volt DC standard.  Grin
Actually higher voltages are already here in the electrics.
Logged

Rick, W3RSW
N4OI
Member

Posts: 401




Ignore
« Reply #91 on: January 24, 2015, 08:32:50 AM »

[...] Actually higher voltages are already here in the electrics.

My Camry hybrid has about 245 VDC available at the traction battery, and that is boosted to about 650V for driving the motor-generators. 

73
Logged
DL8OV
Member

Posts: 1057




Ignore
« Reply #92 on: January 24, 2015, 11:10:05 AM »

I want Icom in on this discussion as well (along with Yaesu). This is why I wrote to them back in December 2014 but so far all I have is a boilerplate email from Icom Europe. Do they care?

Peter DL8OV















i
Logged
VA2PBJ
Member

Posts: 319




Ignore
« Reply #93 on: January 24, 2015, 05:41:16 PM »

It's not much of an invitation when a public flogging and possible loss of employment is the end result. Yes, execs have bosses too. I can't imagine a positive result lending a public misinterpretation to anything they say, and the internet is so permanent. Re-posting email or verbal conversations meant to be private does not instill trust.

Changes cost money and these companies are "for profit". Do you honestly think they will say "sure" and loose money....just because? That's dreamworld. If you want improvement, change the laws. It saves a lot of face all the way around and will actually get the deed done.
Logged

Peter
DL8OV
Member

Posts: 1057




Ignore
« Reply #94 on: January 25, 2015, 02:27:32 AM »

Well, to be fair I would not regard this as a "public flogging" of Icom and Kenwood, until the work done by Wayne at Elecraft we had all assumed that marine units were cleaner than their ham radio cousins and all it would have taken was a transplant of existing technology. I issued them an open invitation, join in the discussion and help make their company's rigs the cleanest on the market, something that would be a positive selling point leading to an increase in those profits.

To his credit Wayne N6KR DID join in after the discussion widened to include the K3.

Should an engineer or an engineering manager from Icom or Kenwood join in the discussion they will be welcomed and then, hopefully, we can get down to the engineering details. For example, one or two 12V powered ham transceivers manufactured in the past had much cleaner transmitter output, why?

Peter DL8OV
Logged
W3RSW
Member

Posts: 606




Ignore
« Reply #95 on: January 25, 2015, 05:25:32 AM »

Are you really sure with what you now know? Given the data sheet verified characteristics of 12 volt devices available or higher voltage devices derated to 12 volt rails, and assuming the rigs your talking about predate adaptive predistortion techniques, I wonder.
Logged

Rick, W3RSW
W8JI
Member

Posts: 9748


WWW

Ignore
« Reply #96 on: January 25, 2015, 05:27:51 AM »

Since IMD can vary considerably with impedance presented at the output devices (more so than with quiescent current bias changes), and since radios do not always run into 50 j0 loads, it seems to me load impedance ranges should be included in any test or specification.

Ten Tec, for example, has no power fold back with load mismatch. While the transistors might be heavy enough to not blow up, the bandwidth blows up instead.

If the output devices are at the limits of linearity from current, reducing impedance at the device kills IMD performance. If at the limits of voltage, increasing impedance makes it worse.

The device load impedance, of course, does not necessarily move the same direction as the antenna port impedance change.

When someone lives in a city, has poor antenna directivity, and operates wide modes on noisy crowded bands, IMD is not an issue.  The same equipment, heard in the country on narrow beamwidth antennas, can be a real issue.  The worse case is working narrow modes with weak signals next to wide modes.

This is why we never should let CW and narrow modes co-mingle with wider modes when band planning.  This is also why some operators think some things are never issues, while other operators have a problem.







Logged
DL8OV
Member

Posts: 1057




Ignore
« Reply #97 on: January 25, 2015, 06:46:49 AM »

Charles, welcome to the discussion. The effects of load mismatch on the IMD figure were mentioned some time ago and at that point I suggested a minor change in transceiver architecture, fit the inbuilt SWR meter between the PA output and the low-pass filters. I have a nagging doubt that the complex load of a LPF could be causing problems and the actions of the filter (similar to the old tank circuits) could be masking the error.

One possible solution was put forward by Bill Sabin when he designed a power amplifier and set of filters that were diplexers, a low pass filter dealt with the wanted signal and a high pass filter plus dummy load disposed of the harmonics. I have yet to try this solution as an order of ferrites seems stuck somewhere over the Atlantic.

Any comments from you regarding the discrepancy in M802 output figures will be most welcome. The initial graph gave us a very high mark to aim for but the check on the second unit revealed a transceiver that was not a high performer in the IMD stakes. Something is amiss.

Peter DL8OV
Logged
W8JI
Member

Posts: 9748


WWW

Ignore
« Reply #98 on: January 25, 2015, 09:38:51 AM »

Passing FCC in a filing has little to do with how the radio actually performs in a two-tone test, just as a two-tone doesn't mean much for voice. I can build a pretty crappy SSB voice amplifier that looks excellent on a two tone, or do the opposite.

For example, an unregulated screen or bias supply in a tetrode can look great in a two tone test but totally fall apart with voice.

The IMD vs. load issue is not simple SWR. The load issue is how the PA behaves with different load impedances. It can get better, worse, or not change much as impedance changes.

The whole measurement method and system is such a mess, it might be premature to worry about getting a number. If you recall, receivers were pretty much designed to look OK using the ARRL testing method. Keyclicks were the same. For many years the Handbook showed a horrible waveform as "ideal" for CW.

Unless the test method means something, nothing really matters. Why not work through the test method first, before comparing things?
Logged
W1QJ
Member

Posts: 2983




Ignore
« Reply #99 on: January 25, 2015, 05:39:28 PM »

.

Ten Tec, for example, has no power fold back with load mismatch. While the transistors might be heavy enough to not blow up, the bandwidth blows up instead.


Correct, but, Ten Tec instead has a high swr  cut out circuit.  If the swr to the radio final is over a certain amount, the power is cut off to the radio and the radio cuts out and goes off line and shuts down.
Logged
N6KR
Member

Posts: 21




Ignore
« Reply #100 on: January 26, 2015, 12:21:05 AM »

... For example, an unregulated screen or bias supply in a tetrode can look great in a two tone test but totally fall apart with voice.

The IMD vs. load issue is not simple SWR. The load issue is how the PA behaves with different load impedances. It can get better, worse, or not change much as impedance changes.

The whole measurement method and system is such a mess, it might be premature to worry about getting a number. If you recall, receivers were pretty much designed to look OK using the ARRL testing method. Keyclicks were the same. For many years the Handbook showed a horrible waveform as "ideal" for CW.

Unless the test method means something, nothing really matters. Why not work through the test method first, before comparing things?

Hi Tom--

You're absolutely right. This is why K3 (and KX3) firmware monitors so many parameters -- we try to preserve decent IMD performance under less-than-ideal conditions. Power is rolled back an appropriate amount when:

  - current drain of either the PA stage or the non-PA circuitry is too high (tracked separately)
  - supply voltage is too low (or too high)
  - drive to the PA stage is too high relative to the drive required at a per-band calibration point
  - reflected power is at an unsafe level
  - operating temperature is out of range

Of course this can't catch every case of marginal load impedance, but it's pretty close. We try to reduce power enough to ensure safe and clean operation, but keep the rig on the air. (The KPA500 amp monitors a similar list of parameters.)

Regarding that clicky, "old standard" keying waveform based on a simple RC time constant: I think all radio designers have used that at one time or another and learned the hard way (pun intended). Instead, we use an optimized sigmoidal waveform for both the rising and falling edges, and apply absolutely no ALC during envelope generation, so that the waveform is not distorted. In CW mode, ALC is pre-calculated to approximately the right level for the power requested, then fine-tuned slowly thereafter, pre-crystal-filter, so any adjustment no matter how small is inherently bandwidth-limited.

We won't be using any tetrodes that I can foresee.

Wayne
N6KR


Logged
Pages: Prev 1 2 3 4 5 6 [7]   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!