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

   Home   Help Search  
Pages: [1] 2 Next   Go Down
  Print  
Author Topic: A filtering question  (Read 2970 times)
VK4CH
Member

Posts: 13




Ignore
« on: April 01, 2014, 10:02:50 PM »

Gooday All

I have what might be considered a "dumb" question with regard to filtering.
I am currently operation an ANAN-10 SDR pushing 10w into an Elecraft KPA500 amp. This gives me about 170-200w output depending on band.
If I were to drive the amp a bit harder i.e. 30w it would easily produce the full 400w which is the maximum limit for VK in fact it would be quite capable of exceeding this but that would be naughty.

I have looked at various methods to achieve this goal.
The first was to sell the ANAN-10 and replace it with an ANAN-100 but that will be the best part of $1000 even with all the new models deflating the value of the originals.
The second was to acquire a commercial amp such as the Tokyo HP HL45 or HL50 etc. but these seem quite expensive and will be a pain as their auto band change really is focussed on the FT817.

My third plan, which is the primary focus of this post, is to use one of the broad band PA stages that seem to be everywhere at the moment that have no real band pass filtering as an intermediate stage between the ANAN-10 and KPA500. There seem to be a lot of these for $100-200 that just require a power supply and a bit of switching. Of course the output would be rathey dirty with all sorts for harmonics and junk so connecting it directly to an antenna would be rather silly.

The question is:

Will the KPA500's filtering clean up any rubbish generated in the intermediate stage?

Thanks for reading this.

Tony McRae VK4CH
Logged
K2GWK
Member

Posts: 411


WWW

Ignore
« Reply #1 on: April 03, 2014, 07:26:35 AM »

The low pass filter in the final amplifier will clean up the harmonics but not the intermods that could be generated by the dirty amplifier. The old phrase garbage in garbage out applies.
Logged

K2GWK Website

Stupidity for Dummies (http://stupidityfordummies.com/)

…because sometimes, you just can’t dumb it down enough…
KD0REQ
Member

Posts: 886




Ignore
« Reply #2 on: April 03, 2014, 09:06:14 AM »

yep, looks like a fine setup for a broadband splatterer.  there is the added adventure of possibly overdriving the Elecraft and blowing it up.  looks like minutes of fun.
Logged
K2UE
Member

Posts: 52




Ignore
« Reply #3 on: April 03, 2014, 04:55:37 PM »

Some of the posters seem to assume that just because an amplifier lacks harmonic filters it is "dirty".  There is no reason why you can't use a low cost amplifier without harmonic filtering, provided you do not directly connect it to the antenna, i.e. 10W from the Anan-10 or full-bore.  The biggest issue with FET amps is inadequate bias, leading to poor IM at full power, and even worse IM at lower power.  But most amps are capable of -30 to -40dB IM if heavily biased -- look for amps that say they are OK all the way to Class A, where distortion will be quite low.  Class AB is easy to understand:  if we draw the best straight line on the high current transfer characteristic of a device and note its slope, then find the point where the slope is exactly half that -- that is the optimum AB bias point:  for small signals both devices conduct with half Gm each, and for large signals only one conduct with twice the Gm.  When you do this you will see that most amps ate underbiased.  The popular RD100 devices used in many 100W PA's are typically biased at 500ma or 1A, but in fact IM continues to decrease until an optimum at about 1.5A.

The fact is nobody inserts harmonic filters before the output, because they are unneeded earlier.  You would just be adding another stage to your amp, that just happens to be outside the box.
Logged
K6AER
Member

Posts: 3498




Ignore
« Reply #4 on: April 03, 2014, 05:04:32 PM »

The KPA-500 PA schematic appears to have an input attenuator. Some of this might be for impedance matching but my guess  it is to limit the gain which is required by the FCC to 15 dB for any amplifier sold as a commercial product. I would give a call to Elecraft and see what they say about modifying the input attenuator for 3 dB more gain. You will have to reset the fault settings so as not to over drive the amplifier.

Most push/pull HF finals have more then 20 dB of gain.
Logged
K6AER
Member

Posts: 3498




Ignore
« Reply #5 on: April 03, 2014, 05:26:02 PM »

Took a few minutes to do more research on the  VRF-2933 MOSFET’s used in push/pull configuration. The devices are rated for 300 watts each CW at 50 VDC. Gain is listed at 20 dB from 2-175 MHz.  More gain at 2 MHz.

The input circuitry has two relays to change the input network. Relay K1 is for HF/6 meters. Relay K2 is for lower HF or higher HF. The RF input detector is buffered through U5. This probably sets up the default maximum input power. You did not mention what bands of operation you are interested but I would give Gary at Elecraft a call.

Also of note the charts show 450 watts out of a single device with 8 watts of drive at 30 MHz. So it appears the devices have power to spare minus dissipation factors and network matching corrections.
Logged
VK4CH
Member

Posts: 13




Ignore
« Reply #6 on: April 04, 2014, 12:53:19 AM »

Thank you all very much for your replies.

My main bands of operation are 20m to 6mwith the main focus on 12, 10 and 6.

VK has a 400w limit so I have no need to turn it up to "11", the KPA-500, I suspect will be quite happy at that level.
I will contact Elecraft, I have done so before when I built, well assembled anyway, one of their transverters and they seem exceptionally professional people to deal with.
Possibly it may be just a firmware mod.

Logged
G3RZP
Member

Posts: 4451




Ignore
« Reply #7 on: April 04, 2014, 01:01:47 AM »

Surely it is very dependent on the harmonic level presented to the Elecraft amplifier? If there's a large harmonic component in the drive, then the reflected harmonic component from the Elecraft filters will be bigger because the harmonics have been amplified as well as the wanted signal and so may cause problems - unless they use absorptive filters, which need a lot more components for each band.
Logged
ZENKI
Member

Posts: 916




Ignore
« Reply #8 on: April 04, 2014, 01:46:44 AM »

Despite the excellent performance of the diplexer filters many manufacturers dont even bother using them. The performance advantage of the diplexer filter makes the  extra component count increase look trivial in comparison to the performance advantage gained in terms of further suppressing harmonics. Even a slight reactance can cause filter performance degradation.

The Russian Made Hermes Amplifier  uses 50 volt mosfets have superb IMD and harmonic suppression. This amplifier uses diplexer filters in its design. One wonders why Apache labs could not have designed a proper low IMD PA much like this Russian PA which excels in all measured performance parameters. One  can only imagine how clean the  signal from this Russian Amplifier would be once the adaptive predistortion is applied to the PA.

Its a technical embarrassment building an advance radio like the Anan series of radios and feeding the Hermes card into a filthy 12 volt PA either of internal or external design. This design incompetence is  normal for amateur radio manufacturers.

The KPA500 when used with the K3 has one of the cleanest on air dynamic performance that I have measured from a solid state amplifier. This is despite the medicocre IMD performance of the devices used in the KPA500. Other leading name solid state amplifiers and transceivers are woeful  by comparison. There is more to achieving excellent IMD performance than just the raw IMD data of the PA devices.

Surely it is very dependent on the harmonic level presented to the Elecraft amplifier? If there's a large harmonic component in the drive, then the reflected harmonic component from the Elecraft filters will be bigger because the harmonics have been amplified as well as the wanted signal and so may cause problems - unless they use absorptive filters, which need a lot more components for each band.
Logged
ZENKI
Member

Posts: 916




Ignore
« Reply #9 on: April 04, 2014, 01:56:13 AM »

Consider buying the 50 volt Russian PA  for the Anan and drive it directly from the Hermes. The Pure Signal pre-distortion will be more effective with high voltage mosfets.
The Russian Amplifier can be driven directly from the Hermes to about 300 watts output. They also have a version that can be driven by the Anan 10. Although the PA's in the Anan are no better than the average CB amplifier in terms of IMD performance. The IMD figures do increase when pre-distortion is applied.
I shudder  every time I read about users feeding   their SDR radios into what is basically CB grade amplifiers. The horrendous splatter from the RM Italy 300 amplifiers on the higher bands  are real curse. From a technical standards point of view no fixed station should be using a 12 volt PA at high power.

Driving the KPA500 directly and applying pre-distortion would be the way to go. I would keep the Anan 10  and junk its sub standard PA. Buy the Russian PA and drive it directly from the Hermes. This setup will drive the KPA500 directly. With pre-distortion and 50 volt mosfets you should have the cleanest signal on the planet.

Anan have announced the Anan 200D, they have made the same mistake by putting in a crap PA design in a world class radio. You cant take the cheap and nasty out of ham manufacturers.


Gooday All

I have what might be considered a "dumb" question with regard to filtering.
I am currently operation an ANAN-10 SDR pushing 10w into an Elecraft KPA500 amp. This gives me about 170-200w output depending on band.
If I were to drive the amp a bit harder i.e. 30w it would easily produce the full 400w which is the maximum limit for VK in fact it would be quite capable of exceeding this but that would be naughty.

I have looked at various methods to achieve this goal.
The first was to sell the ANAN-10 and replace it with an ANAN-100 but that will be the best part of $1000 even with all the new models deflating the value of the originals.
The second was to acquire a commercial amp such as the Tokyo HP HL45 or HL50 etc. but these seem quite expensive and will be a pain as their auto band change really is focussed on the FT817.

My third plan, which is the primary focus of this post, is to use one of the broad band PA stages that seem to be everywhere at the moment that have no real band pass filtering as an intermediate stage between the ANAN-10 and KPA500. There seem to be a lot of these for $100-200 that just require a power supply and a bit of switching. Of course the output would be rathey dirty with all sorts for harmonics and junk so connecting it directly to an antenna would be rather silly.

The question is:

Will the KPA500's filtering clean up any rubbish generated in the intermediate stage?

Thanks for reading this.

Tony McRae VK4CH
Logged
K6AER
Member

Posts: 3498




Ignore
« Reply #10 on: April 04, 2014, 09:53:16 AM »

Zenki,

So what is the two tone IMD out of an Anan 10?

30 dB, 35dB.  What is necessary for a clean signal on the band. 30 dB IMD will be fine onto a dipole at 30 feet. With an amplifier and a 4 element SteppIR at 125 feet you will need a lot more. I have had on occasions a +50 dB over S9 signal 2000 miles away.  My spurious would still be S9 with the best amplifier ever built.

On a S9 noise level during a contest it doesn’t matter. Also most receiver designs on the under $2000 category radios will not withstand high levels of adjacent channel signal no matter how clean the transmissions.

The best radio on the market, the IC-7800 has a transmit two tone, third order IMD, of only 32 dB.  Anything over that in an amplifier is wasted money. Your amplifier IMD is set by the driver radio. Most tube amplifiers and even tetrodes have an IMD of better than 42 dB. My 8410 measured 43 minimum on all bands.

Now put on your marketing hat and try to market an amplifier with the best IMD possible. Hams want cheap reliable power, and good resale ownership.   Everything else is wasted money.  You are promoting the attributes of a fine Merlot to an audience with a beer budget.
Logged
W3RSW
Member

Posts: 85




Ignore
« Reply #11 on: April 04, 2014, 01:46:46 PM »

13.8 volt PA's are designed for easy compatibility with mobile applications as well as meeting current US FCC requirements.

Careful what we wish for in the way of purity standards.  I think we may want to keep emission  standards consistent with what can be homemade by a Ham, be it a single 6CL6 CW transmitter or a QS1R DDC SDR clone.  Frequency and power dependent standards already exist. They exist to protect services outside the ham bands.  I'd vote to keep emissions as unregulated as possible.

Large manufacturers might deign to sell a rig with either a 50 volt set of finals or a 12 volt set.
 Yes, "might."
Logged
ZL3OF
Member

Posts: 20




Ignore
« Reply #12 on: April 04, 2014, 11:00:54 PM »

How about using a Kenwood TL 120 linear amplifier, made for boosting a 10 watt TS120V up to 100 watts. It also has filtering. There is one for sale here. And no, I'm not the seller.
http://www.trademe.co.nz/electronics-photography/radio-equipment/amateur-radio/auction-713126776.htm
« Last Edit: April 04, 2014, 11:09:04 PM by ZL3OF » Logged
ZENKI
Member

Posts: 916




Ignore
« Reply #13 on: April 04, 2014, 11:22:41 PM »

Mike putting aside the 2 tone numbers for the moment. Lets look at this issue from a receivers perspective.

Most good contest grade radios have  a reciprocal mixing dynamic range of about 100db at 5khz spacing.
The typical ham transmitter has  dynamic IMD suppression that is anything from 20db to 30db. There are 1 or 2 radios that could be better.
This dynamic transmitter IMD  number is for IMD products  well outside the filter  skirt  and away from nominal occupied bandwidth. This test can be done with white noise or voice.  A good radio with a well designed transmitter and PA could  achieve -50 db or more(with predistortion) dynamic IMD suppression. at 5 khz. This is a huge improvement over the typical ham contest transmitter/amplifier  on the air at the moment. This 20  to 30db improvement is a figure that is not representative of  say moving from -26db 3rd order IMD to say something like -36db third order IMD. The point is that there is no point  focusing in on 2 tone IMD figures because these dont ultimately reflect the final IMD suppression when you account for transmitter faults and other spurious products. In this regard the FCC and ITU standards dont have onerous 2 tone IMD requirements for marine transmitter but their on air IMD products must fall within the mask. The standard reflects by how much the IMD should be suppressed by and at the rate at which it must fall. This should be  the test standard that the ham  industry should adopt.

 We should have a IMD mask and we should have clear figures  by what amount IMD should be suppressed by at 5, 10, 15 and 20khz spacings.  2 tone figures should be set at ITU levels. This would be a lot easier than specifying 2 tone figures that take no account for ALC  induced compression splatter.  While 2 tone numbers are an important engineering test they are not a measure of real on air IMD performance which is more of a concern to hams. When we realize this point and optimize transmitters for low IMD at 5, 10,15 and 20khz spacing  you soon realize that playing with 2 tone numbers is not going to save your butt.

So in this regard I agree with the points that you have made when it comes to the tone numbers. The debate has to be shifted to  what realistic IMD suppression we can achieve by designing a proper transmitter that has IMD minimization as a design goal. Ham transmitters cause IMD and splatter not only because of  good or bad 2 tone numbers. The simplest fix would be to adopt ITU/FCC test procedures and we will ensure  that bad transmitters dont make it to the market. Beyond the minimum would be better of course and hams should strive for  the best possible standard. Now that pre-distortion is readily available for the ANAN this will surely become the new standard for ham transmitters.

This 20 to db of net  dynamic improvement can make a huge difference to what  you can and cannot copy. It also makes  fitting in more stations into a limited bandwidth more practical. On bands like the 75 meter DX window this huge decrease in IMD would  make the band more enjoyable and reduce the amount of  bickering. Currently in the DX window it just takes 1 station with a lousy transmitter to ruin the whole 20khz with bad splatter. If all hams ran transmitters/amplifiers with -50db or greater IMD suppression at  5khz and greater we could easily fit many stations in the DX window with  just about no interference.  Myself and a neighboring ham both own HF8104A transmitters that drive linear 4cx1500B kilowatt amplifiers. We both can operate in the DX window without knowing  that we on the same band. These Collins transmitters achieve  have exceptional IMD performance both from a 2 tone and dynamic IMD suppression. The signal is essentially  a brick wall signal. The Anan radio with pre-distortion could achieve similar levels of IMD performance. Unfortunately its 12 volt PA is not the best  design for use with the predistortion algorithm. You can see the typical IMD performance with the  Anan VS pre-distortion. Who can argue against such improvement?

http://homepage3.nifty.com/ji3gab/english/dpd.html

Again even with the pre-distortion emphasis must be placed on real voice or white noise tests for dynamic IMD performance which would find things like ALC compression and other transmitter design flaws.

On the IC7800. It has average IMD performance. The IC7800 has  the well known ALC induced splatter issue.  The moment you start using moderate levels of ALC the dynamic IMD  figures deteriorate very poorly on the IC7800. I must say that  a IC7800 when adjusted correctly is a very clean transmitter unlike the Icom IC756 and IC7600  radios which are appalling.

I am not promoting anything that cant achieved even with a radio like the IC718.  If the transmitter is designed to be ITU compliant that would be a  huge improvement  over what we have now which is no enforcement of  any transmitter standards. We have harmonic suppression standards do you think adding ITU compliance   for passing the ITU IMD mask would cause a jump in price? I think not. Icom for example has been designing ITU compliant marine transmitters for decade. If you compare the on air dynamic IMD performance of a old Icom  IC700 marine transmitter with the IC7800 it puts the IC7800 to shame. A ITU compliant marine transmitter from Icom costs 1500 dollars. What is interesting is that since the  FCC brought its standards in line with the ITU the marine transmitter standards have fallen. In fact the Icom M802 wont pass  NTIA standards at 150 watts because its IMD performance is so poor. If Icom had to pass this M802 under the old FCC part 80 rules the M802 would pass NTIA rules easily. The point is that the moment you relax the rules the manufactures take advantage and let their  standards drop. This is exactly whats happening in the ham service. We have no standards and the transmitters are poor why is that a surprise?  The ham service because its so crowded should adopt the most stringent standards. This can easily be achieved by both 12 volt and high voltage FET PA's. Icoms old marine transmitters all ran 12 volts and passed the old FCC Part 80 regulations which were exceptional standards that were designed by engineers who understood the issues very well. Look at any USA designed  commercial HF transmitters of that era and you can see what outstanding  transmitters were produced because of these standards.


Zenki,

So what is the two tone IMD out of an Anan 10?

30 dB, 35dB.  What is necessary for a clean signal on the band. 30 dB IMD will be fine onto a dipole at 30 feet. With an amplifier and a 4 element SteppIR at 125 feet you will need a lot more. I have had on occasions a +50 dB over S9 signal 2000 miles away.  My spurious would still be S9 with the best amplifier ever built.

On a S9 noise level during a contest it doesn’t matter. Also most receiver designs on the under $2000 category radios will not withstand high levels of adjacent channel signal no matter how clean the transmissions.

The best radio on the market, the IC-7800 has a transmit two tone, third order IMD, of only 32 dB.  Anything over that in an amplifier is wasted money. Your amplifier IMD is set by the driver radio. Most tube amplifiers and even tetrodes have an IMD of better than 42 dB. My 8410 measured 43 minimum on all bands.

Now put on your marketing hat and try to market an amplifier with the best IMD possible. Hams want cheap reliable power, and good resale ownership.   Everything else is wasted money.  You are promoting the attributes of a fine Merlot to an audience with a beer budget.

Logged
ZENKI
Member

Posts: 916




Ignore
« Reply #14 on: April 04, 2014, 11:38:02 PM »

Anything is possible.   12 volt amplifiers regardless of who makes them make poor home station amplifiers. If you running an advanced radio like the Anan the last thing you want to do  is drive a 12 volt PA and then drive a high power amplifier.

 We have the same messy situation with FT817 users. They buy a FT817 and attach a  lousy CB amplifier and destroy the band in the process. It would have been smarter to buy the FT857 then you would not have to worry about the mess of an external amplifier, their signals would have been a lot cleaner as well.

Hams have to get out of this cheap mentality and think that they smart  and doing the right thing by using any piece of crap equipment because its cheap. This disturbing trend by new hams who  think " i need amplifier" "buy cheap CB amplifier OK" This stupid cheap mentality without regard to how the equipment  performs and affects others on the air is the height of technical stupidity  that does not deserve a place on the ham bands.

 If CB equipment was cheap and met technical standards and did not cause interference I would would jump on that equipment. You pays for what you get and unfortunately some dumb hams want others to suffer because of their poor technical judgement.  What is also disturbing is that stations know that they are buying a marginal CB amplifier yet their first instinct is to drive the living crap out of the CB to make matter much worst. There so many good ham tube and now cheap good solid state amplifiers available for hams why on earth would you consider buying a CB amp in the first place.

 I came across another idiot the other day who was taking out half of 12 meters. He was running a TS480HX which had been screwed up to beyond 300 watts of output. This stupidity was evident on 12 meters. This ham was very arrogant when told of his splatter however this ignorant mentality seems to be creeping into the ham bands day by day. If you want modify radios at least attempt to try and understand what you modifying and measure what the end result will be. Why is that too much to ask for?
Fiddling inside radios without knowing what the end result is going to be is best left on the CB band where its normal stupid practice.

How about using a Kenwood TL 120 linear amplifier, made for boosting a 10 watt TS120V up to 100 watts. It also has filtering. There is one for sale here. And no, I'm not the seller.
http://www.trademe.co.nz/electronics-photography/radio-equipment/amateur-radio/auction-713126776.htm
Logged
Pages: [1] 2 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!