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 ... 8 9 10 11 12 [13] 14 15 16 Next   Go Down
  Print  
Author Topic: Better IMD please  (Read 121758 times)
TANAKASAN
Member

Posts: 933




Ignore
« Reply #180 on: December 22, 2011, 09:49:48 AM »

"The real way to get the xcvr clean is to run a 300w pa..at 100w...AND design it around that 100w level."

How? Is it just a question of adjusting the bias levels and maybe transformers of a 300W PA for cleanest output at 100W or does some other work need to be done?

Tanakasan
Logged
ZENKI
Member

Posts: 1648




Ignore
« Reply #181 on: December 23, 2011, 04:05:11 AM »

We will get change when magazines such as QST start naming and shaming the poor transmitters. I dont hold much hope that the ARRL
will actually use the words "poor IMD performance" in any review.

A design code of conduct which is  developed  by consensus and delivered to all the manufacturers as the suggested minimum standards would  be a good next step.

The first transceiver manufacturer that has bragging rights for producing  the best receiver and transmitter  and markets such a radio will also
bring about market change. The Japanese companies are a big and slow, they are never leaders  but always followers. A company like Elecraft is probably small and proactive enough maybe to set the future standards of performance. Maybe the K4 will be a real killer radio with a properly designed clean transmitter.

All hope is not lost while there are viable homebrew solutions like the open source High Performance Software Defined Radio group. The Pennylane is already a bench mark pre-driver in terms of its IMD performance.  The design will accommodate adaptive pre-distortion in the near future. 

Hams just need to be proactive and forthright by making technical demands on manufacturers and we will get what we want. Besides once people know that they are buying crap at a  very high price they wont put up with that for much longer. All is not lost, the question that needs to be asked is simply this, is your transmitter clean yes or no? If the answer is no tell them  that you wont be buying it.
Logged
VE7RF
Member

Posts: 212




Ignore
« Reply #182 on: December 23, 2011, 07:21:46 AM »

"The real way to get the xcvr clean is to run a 300w pa..at 100w...AND design it around that 100w level."

How? Is it just a question of adjusting the bias levels and maybe transformers of a 300W PA for cleanest output at 100W or does some other work need to be done?

Tanakasan

##  yes, that's exactly how to do it.   Use the highest voltage the device will take..say 50 vdc.  Then let's assume the push pull pair will be designed to run no more than 100 w pep..with bias tweaked for lowest imd... in [Class AB]   Knowing that, it's then simple maths to calculate the load Z, etc, etc, etc.  Then  design the combiner on the basis that it has to match a XXX  load Z  to  50 ohms.   The LP filters and power supply is a given. Then use  a big heat sink, and don't go cheap on either the combiner or the toroid's used for the LP filter. 

##  By using max V, and xfmr's tweaked for ONE given power level, then the end result is MIN imd.


### I just looked at the SPE 2 kw output SS amp..and to get around the lousy eff, when run at low power levels, they will drop the 48 vdc [2kw out]...down to 35vdc [1kw out]...then down to 30 vdc [500w out]. That concept will maintain better eff, but IMD will be sure to suffer.

Later... Jim  VE7RF
Logged
VE7RF
Member

Posts: 212




Ignore
« Reply #183 on: December 23, 2011, 07:39:36 AM »

We will get change when magazines such as QST start naming and shaming the poor transmitters. I dont hold much hope that the ARRL
will actually use the words "poor IMD performance" in any review.

A design code of conduct which is  developed  by consensus and delivered to all the manufacturers as the suggested minimum standards would  be a good next step.

The first transceiver manufacturer that has bragging rights for producing  the best receiver and transmitter  and markets such a radio will also
bring about market change. The Japanese companies are a big and slow, they are never leaders  but always followers. A company like Elecraft is probably small and proactive enough maybe to set the future standards of performance. Maybe the K4 will be a real killer radio with a properly designed clean transmitter.

All hope is not lost while there are viable homebrew solutions like the open source High Performance Software Defined Radio group. The Pennylane is already a bench mark pre-driver in terms of its IMD performance.  The design will accommodate adaptive pre-distortion in the near future. 

Hams just need to be proactive and forthright by making technical demands on manufacturers and we will get what we want. Besides once people know that they are buying crap at a  very high price they wont put up with that for much longer. All is not lost, the question that needs to be asked is simply this, is your transmitter clean yes or no? If the answer is no tell them  that you wont be buying it.


###  That's a good idea...send em all a nice letter.."design code of conduct". For a laff, I should phone my dealer, and order a new $12K  yaesu 9000MP [400w version with -22 db pep IMD3]....then cancel it  2 x days later...citing the lousy TX imd, sri, no sale, .."too bad, I really needed two of em". 


##  has anybody actually tried contacting yaesu /Icom /Kenwood  directly Huh    Or is it..'no speak english' ?    At least with American co's, you can probably phone up and raise hell.  You may well get nowhere, but at least they will take a hint if they got enough calls and e-mail abt it.

## yaesu will no doubt just tell you to buy their latest 5000..and use Class A.  And if 75w pep out is not enough..then buy their 9000MP..and it runs 100W out in class A.

##  I dunno why QST  doesn't start slamming bad stuff. If they did, where else will the makers of lousy gear gonna go to advertise their junk, CQ rag Huh  Folks are not stupid. If u saw a bad QST review on XXX, and XXX had zero ads in QST...yet advertised in CQ rag, you would put 2 + 2  together real fast. IMO, a real good trashing would give ALL of em a wake up call. "gee, are we next ? "   It would then be  ..'sink or swim time'

Later... Jim  VE7RF
Logged
M0HCN
Member

Posts: 566




Ignore
« Reply #184 on: December 23, 2011, 10:01:11 AM »

### I just looked at the SPE 2 kw output SS amp..and to get around the lousy eff, when run at low power levels, they will drop the 48 vdc [2kw out]...down to 35vdc [1kw out]...then down to 30 vdc [500w out]. That concept will maintain better eff, but IMD will be sure to suffer.
That used to be a very standard trick in low power AM Broadcast back  before everyone went all up EER, and actually you can do it so the effect on IMD is minimal.... Of course the electrically very short aerials with only ~10Khz bandwidth helped!
It was generally known as class H.

AM has no PM component (unlike SSB) so a simple pin diode attenuator at the input, power sensor at the output and most of the AM IMD went away as long as the supply ramped more slowly then the resulting loop bandwidth. I have designed several like that for various people.

Regards, Dan.
Logged
VK4DD
Member

Posts: 79




Ignore
« Reply #185 on: December 23, 2011, 05:18:42 PM »

Here is a man with a vision.

http://ludens.cl/Electron/mosfetamps/amps.html

That is what I like to see.

Ron VK4DD

Logged
TANAKASAN
Member

Posts: 933




Ignore
« Reply #186 on: December 24, 2011, 01:38:30 AM »

I was agreeing with you Ron right up until I got to the point where he wants to run the amplifier straight off the 220v mains supply and water cool it. That's not a project I would like to work on.

Tanakasan
Logged
GM3SEK
Member

Posts: 99




Ignore
« Reply #187 on: December 24, 2011, 08:38:31 AM »

We will get change when magazines such as QST start naming and shaming the poor transmitters. I dont hold much hope that the ARRL will actually use the words "poor IMD performance" in any review.

Although ARRL reviews may not comment on transmitter IMD in the text, IMD3 and IMD9 are prominently included in the the 'Key Performance Features' sidebar on the first page. The positions of the markers on bars shaded green-yellow-red must represent some kind of technical judgement on a scale 'from good to bad' (eg IMD3 of -29dBPEP is in the yellow zone).

Is there any published information about these shaded bars, and what judgements they represent?


73 from Ian GM3SEK
Logged
W6RMK
Member

Posts: 680




Ignore
« Reply #188 on: December 24, 2011, 09:30:10 AM »

We will get change when magazines such as QST start naming and shaming the poor transmitters. I dont hold much hope that the ARRL will actually use the words "poor IMD performance" in any review.

Although ARRL reviews may not comment on transmitter IMD in the text, IMD3 and IMD9 are prominently included in the the 'Key Performance Features' sidebar on the first page. The positions of the markers on bars shaded green-yellow-red must represent some kind of technical judgement on a scale 'from good to bad' (eg IMD3 of -29dBPEP is in the yellow zone).

Is there any published information about these shaded bars, and what judgements they represent?

Sure... ARRL described it some years ago.. I'll look for it.

http://www.arrl.org/files/file/Technology/tis/info/pdf/Procedure%20Manual%202010%20with%20page%20breaks.pdf

describes the measurement process, but not the bar structure..

Logged
ZENKI
Member

Posts: 1648




Ignore
« Reply #189 on: December 25, 2011, 12:53:57 AM »

For every day hams a more useful indication would just be using white noise and reporting the IMD levels at 5,10,15 and 20khz. This would include measurements with the ALC set at as per the manual, at  half scale and at full scale. You would use  various power levels,  25 watt steps would be a  good starting point.

Summarizing  a transmitters performance in one line is a tough call, and leaving out the IMD products above the 9th order gives you few  clues of how good the transmitter really is.

Alternatively they could use a standard speech database and use LINRAD in TX test mode too generate real dynamic IMD test data.
Jack Smith at Clifton Laboratories used this approach.

http://www.cliftonlaboratories.com/elecraft_k3_speech_processing.htm

This will all fall on deaf ears, the ARRL rarely responds when prompted for improvements in their test techniques. Look how long it took them to start measuring  receiver data at close signal spacings. If it was not for Peter Hart, Sherwood and W8JI they would still be measuring data at 20khz signal spacings. Maybe Ulrich Rohde  can donate some transmitter  ITU IMD software masks with a Rohde FSU too the ARRL lab. I think the phase noise test jig  was donated by him. I am sure Ulrich understands transmitter TX performance!




We will get change when magazines such as QST start naming and shaming the poor transmitters. I dont hold much hope that the ARRL will actually use the words "poor IMD performance" in any review.

Although ARRL reviews may not comment on transmitter IMD in the text, IMD3 and IMD9 are prominently included in the the 'Key Performance Features' sidebar on the first page. The positions of the markers on bars shaded green-yellow-red must represent some kind of technical judgement on a scale 'from good to bad' (eg IMD3 of -29dBPEP is in the yellow zone).

Is there any published information about these shaded bars, and what judgements they represent?

Sure... ARRL described it some years ago.. I'll look for it.

http://www.arrl.org/files/file/Technology/tis/info/pdf/Procedure%20Manual%202010%20with%20page%20breaks.pdf

describes the measurement process, but not the bar structure..


Logged
GM3SEK
Member

Posts: 99




Ignore
« Reply #190 on: December 25, 2011, 03:40:38 AM »

For every day hams a more useful indication would just be using white noise and reporting the IMD levels at 5,10,15 and 20khz. This would include measurements with the ALC set at as per the manual, at  half scale and at full scale. You would use  various power levels,  25 watt steps would be a  good starting point.

Summarizing  a transmitters performance in one line is a tough call, and leaving out the IMD products above the 9th order gives you few  clues of how good the transmitter really is.

Alternatively they could use a standard speech database and use LINRAD in TX test mode too generate real dynamic IMD test data.
Jack Smith at Clifton Laboratories used this approach.

http://www.cliftonlaboratories.com/elecraft_k3_speech_processing.htm

The problem with the white noise test signal is that it doesn't have the strong syllabic modulation of real speech, so it fails to detect transient IMD caused by poor dynamic regulation of the power supply. I would much prefer a standard file of voice signals; but not necessarily the standard ICAO files that Jack used because pilots and air traffic controllers do not adequately represent the full range of voices found in amateur radio.

In the Internet Age there would be no problem in compiling our own standard sound file, which can then be distributed around the world as a bit-perfect copy. A neat feature of a peak-hold test is that if someone says "Hey, you forgot about voices like ours!" they can simply be added to the file without affecting the results from voices that are already included.


Quote
This will all fall on deaf ears, the ARRL rarely responds when prompted for improvements in their test techniques. Look how long it took them to start measuring  receiver data at close signal spacings. If it was not for Peter Hart, Sherwood and W8JI they would still be measuring data at 20khz signal spacings. Maybe Ulrich Rohde  can donate some transmitter  ITU IMD software masks with a Rohde FSU too the ARRL lab. I think the phase noise test jig  was donated by him. I am sure Ulrich understands transmitter TX performance!

The way forward is to start testing, and to publish tables in the same way that Rob Sherwood did. A good start is the table of peak-hold IMD results reported by SM5BSZ (Mr Linrad): http://www.sm5bsz.com/dynrange/dubus205/dubus205.htm  The results are in Table 2, but by all means read the whole article (disclosure: I helped to edit this article for DUBUS).

Peak-hold testing no longer requires a high-end spectrum analyzer; today, anyone with an SDR can do it. The main outstanding problem is how to capture the data from different kinds of software, and to standardize the format for display and comparison.



73 and Happy Holidays to all
Ian G/GM3SEK



Logged
GM3SEK
Member

Posts: 99




Ignore
« Reply #191 on: December 25, 2011, 03:17:19 PM »

More good information to digest, from AMPS:

1. Interesting description of IMD: http://lists.contesting.com/archives//html/Amps/2011-12/msg00204.html

2. Direct download link for Vuolevi's PhD thesis on IMD: http://herkules.oulu.fi/isbn9514265149/isbn9514265149.pdf  Too much to digest today (of all days) but any PhD student who had Steve Maas for a referee should be worth reading...


73 from Ian GM3SEK

Logged
ZENKI
Member

Posts: 1648




Ignore
« Reply #192 on: December 25, 2011, 04:14:38 PM »

Very interesting information from a professional in the industry.  At least we know what the starting point is for calling any amplifier and transmitter chain ""clean"" -40db to -50 db is a good starting point. Any decent radio when used with any amplifier on the ham bands should have a transmitter with a minimum of -40db 3rd order IMD.

More good information to digest, from AMPS:

1. Interesting description of IMD: http://lists.contesting.com/archives//html/Amps/2011-12/msg00204.html[/url

2. Direct download link for Vuolevi's PhD thesis on IMD: http://herkules.oulu.fi/isbn9514265149/isbn9514265149.pdf  Too much to digest today (of all days) but any PhD student who had Steve Maas for a referee should be worth reading...


73 from Ian GM3SEK


Logged
KM1H
Member

Posts: 5498




Ignore
« Reply #193 on: December 26, 2011, 05:45:51 PM »

The latest QST has a review of a TE 2M brick and they give it a -17dB 3rd and barely comment its not as good as they would like to see. The other ones arent much better.
Jeez, how can they even accept advertising for that crap?? No conscience required when money crosses hands in ad revenue.

Carl
Logged
ZENKI
Member

Posts: 1648




Ignore
« Reply #194 on: December 26, 2011, 06:30:53 PM »

Yep, they might as well start taking advertising for RM Italy amplifiers, the bar is already so low. 

The latest QST has a review of a TE 2M brick and they give it a -17dB 3rd and barely comment its not as good as they would like to see. The other ones arent much better.
Jeez, how can they even accept advertising for that crap?? No conscience required when money crosses hands in ad revenue.

Carl
Logged
Pages: Prev 1 ... 8 9 10 11 12 [13] 14 15 16 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!