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: Prev 1 [2] 3 4 Next   Go Down
  Print  
Author Topic: KX3 / Sherwood Engineering Inc.  (Read 48217 times)
KB1GMX
Member

Posts: 810




Ignore
« Reply #15 on: August 15, 2012, 06:07:22 PM »

>>Theres nothing basically wrong with the radio, its performance has been measured and acknowledged.
Its just that people sit here and imagine  all sorts of imaginary scenarios that are totally unrealistic for a portable radio operating on batteries.<<

Maybe!  Lots of claims and opine.

I've spent the last few weeks running 4W SSB on 20M mobile with good results when the band is not totally dead.  That and A PRC1099 with its 3M whip portable at 5W on 80, 40 and other bands.

Yes, low power and poor antenna are a not a KW and a monoband beam for each band.

On the other hand I put up a PAR EF40/20/10 and ran the same 4W into it as a vertical
and it was very very good.  With a trivial 6oz wire I erased the "inefficient" antenna.


>>Its gets worst when you using a wire up the tree or loaded short vertical. Do people ever estimate what their effective ERP is, or what gain  they have at low takeoff angles etc etc.
They then go onto imagine that using  such inefficient small antennas requires that your receiver is going to need chart topping receiving numbers.
The problem is not the received signal strength  its going to be  the transmitter signal level from inefficient antennas and low power.
Some might prefer to have 20 db of receiver numbers you cant use rather  than 3db more power,  that makes more sense NOT?
I would rather have 4db of transmitter gain than 30db of receiver performance that cannot be used on todays bands.

You cant hear them you will never work them.  Also a small inefficient antenna needs a better RX but the senstivity of just about anything useful these days is adequate it's AGC, selectivity and VFO(digital or otherwise) stability that are important when your out int he cold or hot.

The receiver numbers come at no cost to TX power.  However the TX power is limited to the battery one is willing to carry.  That weight grows fast if you want 20W SSB or more.  Why?  The average SSB radio is
only about 30 to 40 percent efficient in converting raw DC to output power and its the linear amps doing
that.  That PRC1099 weighs in at a heavy 27pounds and much of that is the 6AH 12V gell cell so it can do 20W (high power) but in the 5W setting a far lighter 2.3AH gell cell runs it just as long.  Ever run a FT857
on internal battery at 20W it's fairly short life on TX.

Power is weight, Its battery to power it and heat sinks to dissipate the resulting heat.  It's a pack friendly radio, not a base station. 

With that said compare it to the peers like the K2, IC703 (or the plus version) and the FT817n.
There is not much else that does more than one band and weighs less than 10 pounds.
That eliminates the K3, 590 and the Flex (unless you cripple it with a really light laptop).

Oh and it's not a contest radio unless your contest if Field Day, FYBO or other get out in
the field event.

Also if you have never used an IQ radio even 50db down on the opposing sideband is way down
(thats 100,000 times less loud than the loud guy giving you an ear ache) and a number that an
amazing amount of radios cant beat either.  You have to understand what the specs mean
when your ears are plugged into a Proset and listening.  This number is bigger than that often
means nothing.

As to the substandard transmitter thing.  I agree create a new topic on this as this radio
is only in the pack and not the worst by far.  But keep in mind, build a better amp, but
your restrictions are, small, light, no more than 12V, manufactureble for a price, and
keep the efficiency up while your at it.  See how you do, I've done it but it's never
adequately cheap.


Allison
Logged
NI0Z
Member

Posts: 583


WWW

Ignore
« Reply #16 on: August 15, 2012, 08:18:24 PM »

I think that this is all interesting fodder.  It's funny to think that it comes after a new receiver tops the charts.

I think we will finally see some more new radios earn spots up top as well.  We have the new Yaesu coming, should score as well as the FT5000.  We have the new Flex which if it lives up to hype will be number 1.  We have a new Kenwood coming too and if it's better than the 590 it too should earn a spot up there as well.

Two points I will make that being said.  It is invitable that things will change as new radios get released.

Probably was most remarkable about the KX3 score is the size of the package and radio price, not that a new radio has topped the charts.

Ok, I am an owner, ironically I haven't even opened the box yet and won't till next week.  So let's get that out of the way because I will likely be labeled as one of them!  Oh man, another lable, lol, it's just nonsense!

Come on folks, it doesn't matter if you own or never own this radio.. It's good news for everyone and everyone will win at the end of the day!

I bought this radio for what I believe it is, a Swiss army knife ham radio, very lite, very functional, very capable.  I wanted something for when I go camping.  I wanted to play real QRP where there was no temptation to cheat.  I wanted to compare its Reciever to the flexs.  I like that one can setup a Panadaptor with no additional hardware, and also it will work with iSDR on the iPad.  I think it's cool you can work some digital with the paddles.  I like that in significant part it's software upgradeable.  Also liked that it will support 2 meters eventually.  Best of all I liked that base price was $1000.

I will give it to Elecraft, and you can look this up over in the yahoo group it you want to confirm it.  But after I cancelled my first KX3 order and before I finalized ordering it the second time, I called them and asked them how it would perform.  They told me in most case the same or better than the KX3 .... Except.... Where the K3 had roofing filters and where the KX3 didn't ( not as many filters available for KX3) then the K3 would perform better.  That was in April... Guess what?  They didn't lie!  Oh, and they didn't charge my card till my KX3 shipped.

I am sure that any hams knowing me well enough would agree when I say that it will be more radio than I'll know what to do with.  I don't think I am the only ham this could be said for.

Specs shouldn't make people so grumpy!  Lol

It's real world preformance that's going to really count.

I would say given how light, portable and capable this radio is that the rule of its better to have more than you need than not enough applies.  Hams have options coming if they want less and hopefully want to spend less.  There is the 817 as well, definitely costs less.

Probably all I will say for a good while, because I won't know much for a good long while after I have actually used it in a variety of situations.

And that's my final point, I'd rather hear from real users doing real work on these radios and honest feedback on what they actually do well and don't do well.  Conversations on specs are limited in their value.  Are there any well respected non brand haters that have reported on this radio, let's hear from them.  And being a new ham and not knowing whoses who, please forgive my ignorance of who that might be if they have already commented on this thread.  Again, a user that has owned one and ran it through all it's paces and who can provide objective feedback.

73 folks!
Logged

KE5JPP
Member

Posts: 0




Ignore
« Reply #17 on: August 16, 2012, 03:06:51 AM »

Very small changes in the precision analog components following the QSD will cause the image rejection to degrade.  It will be interesting to see if the KX3 image rejection degrades over time/aging. 

Gene
Logged
NI0Z
Member

Posts: 583


WWW

Ignore
« Reply #18 on: August 16, 2012, 09:46:37 AM »

From Rob Sherwood:

"With the default zero kHz IF setup, a strong signal on CW on the opposite sideband (other side of zerobeat) will only be down 60 to 70 dB due to the limitations of the I&Q cancellation. Neither the K3 nor the FT-590S have this problem with their CW roofing filters and a more standard design. The KX3 can be set to an 8-kHz IF, much like the Flex radios, but then the image would be 16 kHz away and still a potential problem in something like CQ WW with the whole CW band segment full of very strong signals. "

More proof that making a purchasing decision based solely on number in a table is not a good idea.

Gene


Seems more light was needing to be cast upon this quote.
http://groups.yahoo.com/group/KX3/message/13127
Logged

KE5JPP
Member

Posts: 0




Ignore
« Reply #19 on: August 16, 2012, 01:11:28 PM »

From Rob Sherwood:

"With the default zero kHz IF setup, a strong signal on CW on the opposite sideband (other side of zerobeat) will only be down 60 to 70 dB due to the limitations of the I&Q cancellation. Neither the K3 nor the FT-590S have this problem with their CW roofing filters and a more standard design. The KX3 can be set to an 8-kHz IF, much like the Flex radios, but then the image would be 16 kHz away and still a potential problem in something like CQ WW with the whole CW band segment full of very strong signals. "

More proof that making a purchasing decision based solely on number in a table is not a good idea.

Gene


Seems more light was needing to be cast upon this quote.
http://groups.yahoo.com/group/KX3/message/13127

Yes, it further proves what I stated earlier: "proof that making a purchasing decision based solely on number in a table is not a good idea".

Gene
Logged
KE5JPP
Member

Posts: 0




Ignore
« Reply #20 on: August 16, 2012, 03:06:37 PM »

From Rob Sherwood:

"With the default zero kHz IF setup, a strong signal on CW on the opposite sideband (other side of zerobeat) will only be down 60 to 70 dB due to the limitations of the I&Q cancellation. Neither the K3 nor the FT-590S have this problem with their CW roofing filters and a more standard design. The KX3 can be set to an 8-kHz IF, much like the Flex radios, but then the image would be 16 kHz away and still a potential problem in something like CQ WW with the whole CW band segment full of very strong signals. "

More proof that making a purchasing decision based solely on number in a table is not a good idea.

Gene


Seems more light was needing to be cast upon this quote.
http://groups.yahoo.com/group/KX3/message/13127

None of those numbers in the Sherwood list tells you anything other than how the particular radio performs in a non-real world laboratory environment where the only signals it sees are those generated by a piece of laboratory test equipment.  

Gene
Logged
KB1GMX
Member

Posts: 810




Ignore
« Reply #21 on: August 17, 2012, 12:31:11 PM »

>>Very small changes in the precision analog components following the QSD will cause the image rejection to degrade.  It will be interesting to see if the KX3 image rejection degrades over time/aging.<<

The KX3 does all the heavy lifting in the digital realm.  The analog part influence that less.
If you use quality componenets stability is far better than the days of the Central Electronics
or Hallicrafters Phasing rigs with old carbon resistors that drift with heat and wax paper caps.

I use a KK7B designed MiniR2/T2 based radio for 6M and that's a full out IO analog design.
Once I set it up I haven't opened the box, that was 6 years ago for a analog design with a
lot of my own HB stuff around it.   I expect it to age as well as any other radio, maybe better.
Fewer parts to fail.

Allison

Logged
KE5JPP
Member

Posts: 0




Ignore
« Reply #22 on: August 17, 2012, 01:09:20 PM »

>>Very small changes in the precision analog components following the QSD will cause the image rejection to degrade.  It will be interesting to see if the KX3 image rejection degrades over time/aging.<<

The KX3 does all the heavy lifting in the digital realm.  The analog part influence that less.
If you use quality componenets stability is far better than the days of the Central Electronics
or Hallicrafters Phasing rigs with old carbon resistors that drift with heat and wax paper caps.

I use a KK7B designed MiniR2/T2 based radio for 6M and that's a full out IO analog design.
Once I set it up I haven't opened the box, that was 6 years ago for a analog design with a
lot of my own HB stuff around it.   I expect it to age as well as any other radio, maybe better.
Fewer parts to fail.

Allison



The QSD is susceptible to I/Q unbalance in part because of component tolerances in the audio amplifier that follows the QSD switches.   Even a very small variation in component values affects the I/Q balance and thus the image suppression greatly.   That is why Flex finally came up with dynamic I/Q balancing.  

It's OK if you are happy with 30-40 dB of image suppression.  Otherwise it becomes a problem with component aging, temperature, QSD switch frequency and duty cycle, and a lot of other contributors to I/Q imbalance.  If it was not a problem as you indicate, then Flex would have been able to do a factory I/Q balance and then just forget about it.  That does not work for a QSD.   And this is ALL in the ANALOG realm, before the I/Q signals get digitized.  So the ANALOG part influences this GREATLY.

Gene
« Last Edit: August 17, 2012, 01:11:20 PM by KE5JPP » Logged
KB1GMX
Member

Posts: 810




Ignore
« Reply #23 on: August 18, 2012, 05:34:20 PM »

>>It's OK if you are happy with 30-40 dB of image suppression. <<

Actually the QSD down converters are fairly stable and the audio that follows
unless poorly done should be too especially those built with opamps.  Where
are are issues it's the capacitors and more likely inductors where obtaining those
to 1% values is near unlikely.  However they are fairly stable once dialed in.
For manufacturing if you can avoid anything that must be adjusted all the better
as people doing that cost.

The old 2Q4 network still does 34DB nearly 6 decades later in my hallicrafters.
Considering the parts inside that's not all that bad.  Is it enough, there is a lot of
opinion.  However we can easily do 10-20db better now using opamps even
outside the digital realm.  But I have build receivers around one.  Once dialed
in it sounded very good and was without any distortion or digital artifacts.  I
might add that without AGC pumping up signals 34db or 2500 times weaker
really does sound good.  In fact good enough to hear all the other guys
distortion products as distinct signals.

If you want to try an IQ or image rejecting radio build a MicroR2 and listen to 40M
for a while on a decent antenna.  It's not the highest opposing sideband rejection
at about 37DB but enough to get a feel what that really means on the air especially
at less than 80ma total drain and very tiny.  It's pretty good.

Modern parts and designs are better than that and stability is an issue when you are
pushing into the 50DB range or greater.  Who says 40db is not enough for practical use?
Its still 10,000 times weaker than the guy you want to hear.

And the analog design I have runs when last tested at 45db, it's enough that the KW 800ft away
is not a factor save for he splatters when he mics up.

>> That is why Flex finally came up with dynamic I/Q balancing.<<

They did that to get into the 60++ DB range to compete with radios using common tech
in that price range.   Where keeping the digitally derived LO was a problem for both
level and phase balance across the the entire tuning range.  Same for the mixer input
splitter.  The audio stages were also a factor but not nearly as much.  But they also
switch sidebands and modulation and demod modes and that requires different
calibration for each mode both receive and transmit.   

The level of difficulty is 30DB, loose and rough, 40db keep it within less than 1degree and 1db,
and 60 db we are to .1degree and .1db.  At 50mhz .1 degree is a mere 1.5mm (.060")
difference in lead or PC trace length.  Even the board physical characteristics are a factor then.
Since there is a computer attached to the end of the wire using it to do adaptive tuning
is expected and appropriate.

Then again like I said there are a lot of filter radios that 50DB of opposite sideband suppression
for the voice bandwidth is a good day for them.  Some of the simple ones have filters that have
a stopband that doesn't make 50DB.  It's not a limiting factor unless you expect IC7800 class
performance. and that is far from portable.  For that fact the flex 1500 is not what
I'd call portable either. And for the price I'd expect a lot of performance because your
buying both a radio and a computer that is integral to it.  Also when Your likely going to
hook that to a linear the transmitted signal is more critical in quality.

But for radios like that even if the opposite sideband is way down is the front end able to do
the levels needed to avoid overload?  There are many ways a radio receiver can fail and
opposite sideband is less an issue than frontend overload.  Being on the east coast 40M at night
with the broadcasters hammering the meter on any radio that's working overload is the prime
performance issue followed by skirt selectivity.

Than again when out in the woods running on penlite cells even 30DB is plenty adequate.
The problem is not hearing but being heard. If you can do that at modest current so the
battery life is available for the transmitter, that is all the better. 

Apples and oranges.  Small and portable radios that are both frequency agile and high
performing are scarce and yet we are comparing them to radios that are typically far
from portable nor in the same price class.

Another phasing radio kit is the JUMA TXC2A, all analog.


Allison



Logged
KE5JPP
Member

Posts: 0




Ignore
« Reply #24 on: August 18, 2012, 07:30:29 PM »

Who says 40db is not enough for practical use?
Its still 10,000 times weaker than the guy you want to hear.

Have you ever really spent some time operating a SDR with a spectrum display?  Because, if so, you would realize the 40 dB of image suppression would make for some very unhappy customers.  A -30 dBm signal in the spectrum display would result in a -70 dBm image (almost S9).  Even 60 dB of image suppression is not enough.  They are shooting for more like 70 to 80 dB of image rejection and at that point even breathing on the analog components that follow the QSD switches disrupts the I/Q balance.  With only 40 db of image suppression, the spectrum display would be almost useless (specially in CW mode) because it would be impossible to sort out images from real weaker signals in the spectrum display.  It would be totally unacceptable.

Although the KX3 does not have a built in spectrum display, it does have an audio I/Q output for use with external software.  The KX3 I/Q balance (which translates to image suppression) needs to be good too with time, aging, temperature, etc..., otherwise it will render any spectrum display pretty useless when used in a crowded band with strong signals for the same reasons given above.

Gene
« Last Edit: August 18, 2012, 07:33:30 PM by KE5JPP » Logged
KB1GMX
Member

Posts: 810




Ignore
« Reply #25 on: August 19, 2012, 06:13:05 AM »

>>Have you ever really spent some time operating a SDR with a spectrum display?<<

Yes. It was a pretty poor display compared to my HP9568B and neither are portable without a hand truck.

>>  Because, if so, you would realize the 40 dB of image suppression would make for some very unhappy customers. <<

If I had a cheap (under 1000$) radio with spectrum display I'd be happy. At 2000$ and up I would not.

>> A -30 dBm signal in the spectrum display would result in a -70 dBm image (almost S9).  Even 60 dB of image suppression is not enough.  They are shooting for more like 70 to 80 dB of image rejection and at that point even breathing on the analog components that follow the QSD switches disrupts the I/Q balance.  With only 40 db of image suppression, the spectrum display would be almost useless (specially in CW mode) because it would be impossible to sort out images from real weaker signals in the spectrum display.  It would be totally unacceptable.<<

Ever use one with less than 50DB ever?  I have it's actual mugh more useful than you believe.
Or is this all speculation or opinion?

Most of the spectrum displays I've seen have such a compressed scale that trying to see a -120signal
in a maze of -70 grass is pointless.  On 40M here where local noise is usually in the "S5" region on a
quiet day anything less than that would be part of the baseline noise. I'd say that most spectrum displays
are there because they have to be there to compete at that price point. They have little to do with
quality of the receiver.

What you failed to copy in my last was for a portable system IE: back packable you will not have a
lot of things and nor want them for power and weight reasons.  Things that make sense for a portable
transceiver or even mobile make less sense than on the desk.  That doesn't mean that super number performance isn't a great thing or desirable for the desk.   Then again a portable 5 pound with batteries 20W, auto tuner with full display with a bandscope radio would be nice to have for under 1000$.  What
are the odds of that happening?

Allison
Logged
KE5JPP
Member

Posts: 0




Ignore
« Reply #26 on: August 19, 2012, 06:38:22 AM »

>>Have you ever really spent some time operating a SDR with a spectrum display?<<

Yes. It was a pretty poor display compared to my HP9568B and neither are portable without a hand truck.

>>  Because, if so, you would realize the 40 dB of image suppression would make for some very unhappy customers. <<

If I had a cheap (under 1000$) radio with spectrum display I'd be happy. At 2000$ and up I would not.

>> A -30 dBm signal in the spectrum display would result in a -70 dBm image (almost S9).  Even 60 dB of image suppression is not enough.  They are shooting for more like 70 to 80 dB of image rejection and at that point even breathing on the analog components that follow the QSD switches disrupts the I/Q balance.  With only 40 db of image suppression, the spectrum display would be almost useless (specially in CW mode) because it would be impossible to sort out images from real weaker signals in the spectrum display.  It would be totally unacceptable.<<

Ever use one with less than 50DB ever?  I have it's actual mugh more useful than you believe.
Or is this all speculation or opinion?

Most of the spectrum displays I've seen have such a compressed scale that trying to see a -120signal
in a maze of -70 grass is pointless.  On 40M here where local noise is usually in the "S5" region on a
quiet day anything less than that would be part of the baseline noise. I'd say that most spectrum displays
are there because they have to be there to compete at that price point. They have little to do with
quality of the receiver.

What you failed to copy in my last was for a portable system IE: back packable you will not have a
lot of things and nor want them for power and weight reasons.  Things that make sense for a portable
transceiver or even mobile make less sense than on the desk.  That doesn't mean that super number performance isn't a great thing or desirable for the desk.   Then again a portable 5 pound with batteries 20W, auto tuner with full display with a bandscope radio would be nice to have for under 1000$.  What
are the odds of that happening?

Allison


People who actually have used a SDR with a spectrum display will understand what I am talking about.  You seem to be living on another planet altogether.  A SDR with software running on a laptop does not require the HP9568B hand truck and good direct sampling SDRs have much better dynamic range than your HP9568B.

I have used QSD based SDRs like the SoftRock before the I/Q balance adjustment is performed where the image is only 40-50 dB suppressed and it makes CW operation where you have lots of signals within a 1 kHz bandwidth impossible to discern which are real signals and which are images.  I am not taking about using in a laboratory as a spectrum analyzer.  I am talking about using it to hunt down DX on the bands.

Unlike QSD based SDRs, direct sampling SDRs have no problems with I/Q balance or image suppression.  The image is more than 110+ dB down, virtually non-existent.

There is an iPad app called iSDR that will work with the KX3 as a spectrum display.  No hand truck or muscles required for portable operation.  Even with the KX3 and iSDR on an iPad, 50 dB of image suppression would be pretty useless in CW mode.  Don't try to tell me that no one would use something like that for portable operation.  Eric Swartz, WA6HHQ, of Elecraft uses iSDR himself with the KX3.

Come back and argue after you have really has some experience with SDRs.   I get tired of wasting my breath.

Gene
« Last Edit: August 19, 2012, 07:14:52 AM by KE5JPP » Logged
KB1GMX
Member

Posts: 810




Ignore
« Reply #27 on: August 19, 2012, 11:43:49 AM »

>>People who actually have used a SDR with a spectrum display will understand what I am talking about.  You seem to be living on another planet altogether.  A SDR with software running on a laptop does not require the HP9568B hand truck and good direct sampling SDRs have much better dynamic range than your HP9568B.<<

Lessee a decent SDR with dynamic range some some softrock, a computer with a good sound card
and power to run both.  I assume that will be real light compact and easily carried.  You clearly
missed the point. 

All of the SDR I've used that were really good used a desktop with a high sample rate sound card
not found in any laptop.  Yes they were good.  But the desktop wanted serious power and the 19" display was anything but portable. Oh and it was receive only.  Yes and moving them it did make sense to use a hand truck as I ran out of hands for the display, keyboard and mouse.

Direct sampling have two problems I've seen, dynamic range, and not many reach VHF.  Seems when you need lots of bits for dynamic range you loose that when you try and go up in frequency.  At some point
the only out is down converting.  Though with enough bits at lower frequency they do work well.
Then again my TenTec 6n2 (model 526) is like many direct sampling IF radios where direct sampling
is done at low IF.  Just about any radio these days that feature variable filters are likely to be IF direct sampling.  I generally play at 50mhz and up and my idea of weak signals are way less than -130dbm.

I have ISDR on my iphone and its ok with a good IQ front end.  I have used it portable. It makes a
crappy awkward radio with a tiny display, not because of some number that I cant hear but the whole
setup is a cute toy.  With SDR on my Macbook it's decent but then I need battery adaptors to keep the
Macbook battery going and all.  Again not conveniently portable, not a package.  Even my 600mhz
Blackfin based board is not up to more than a few hundred kilohertz direct sample and that's not at all battery friendly buts its fine for a dual conversion system and IQ direct to wide baseband.

Like I said I have the experience both design and otherwise, I didn't drink the koolaid, yet!  For portable
ops it's got a ways to go but it's a fine experimental thing.  The real stuff for commercial and other TLA
users is a whole different realm.  My feeling is when we start seeing ASICs with the DSP core in hardware
that has enough speed to do the direct sampling at 10 or 20mhz and enough bits so the noise and other artifacts are not ear gratingly present then I'm listening, assuming the audio stages that follow aren't junk.

For the desktop, SDR still has to separate itself from winders (M$Windows) so I don't have to reboot
in a QSO.  A few people (SDR cube and Elecraft) have designs that are not tied to the box and big display with an awkward not meant to be carried keyboard.  I tend to be a person that wants a radio and not something that can play pong part time. 


Allison


Logged
KE5JPP
Member

Posts: 0




Ignore
« Reply #28 on: August 19, 2012, 01:13:30 PM »

>>People who actually have used a SDR with a spectrum display will understand what I am talking about.  You seem to be living on another planet altogether.  A SDR with software running on a laptop does not require the HP9568B hand truck and good direct sampling SDRs have much better dynamic range than your HP9568B.<<

Lessee a decent SDR with dynamic range some some softrock, a computer with a good sound card
and power to run both.  I assume that will be real light compact and easily carried.  You clearly
missed the point.  

All of the SDR I've used that were really good used a desktop with a high sample rate sound card
not found in any laptop.  Yes they were good.  But the desktop wanted serious power and the 19" display was anything but portable. Oh and it was receive only.  Yes and moving them it did make sense to use a hand truck as I ran out of hands for the display, keyboard and mouse.

Direct sampling have two problems I've seen, dynamic range, and not many reach VHF.  Seems when you need lots of bits for dynamic range you loose that when you try and go up in frequency.  At some point
the only out is down converting.  Though with enough bits at lower frequency they do work well.
Then again my TenTec 6n2 (model 526) is like many direct sampling IF radios where direct sampling
is done at low IF.  Just about any radio these days that feature variable filters are likely to be IF direct sampling.  I generally play at 50mhz and up and my idea of weak signals are way less than -130dbm.

I have ISDR on my iphone and its ok with a good IQ front end.  I have used it portable. It makes a
crappy awkward radio with a tiny display, not because of some number that I cant hear but the whole
setup is a cute toy.  With SDR on my Macbook it's decent but then I need battery adaptors to keep the
Macbook battery going and all.  Again not conveniently portable, not a package.  Even my 600mhz
Blackfin based board is not up to more than a few hundred kilohertz direct sample and that's not at all battery friendly buts its fine for a dual conversion system and IQ direct to wide baseband.

Like I said I have the experience both design and otherwise, I didn't drink the koolaid, yet!  For portable
ops it's got a ways to go but it's a fine experimental thing.  The real stuff for commercial and other TLA
users is a whole different realm.  My feeling is when we start seeing ASICs with the DSP core in hardware
that has enough speed to do the direct sampling at 10 or 20mhz and enough bits so the noise and other artifacts are not ear gratingly present then I'm listening, assuming the audio stages that follow aren't junk.

For the desktop, SDR still has to separate itself from winders (M$Windows) so I don't have to reboot
in a QSO.  A few people (SDR cube and Elecraft) have designs that are not tied to the box and big display with an awkward not meant to be carried keyboard.  I tend to be a person that wants a radio and not something that can play pong part time.  


Allison




Wow, you are way out in left field here.  Where do you come up with the idea that direct sampling SDRs have poor dynamic range?  The Perseus, which is a direct sampling SDR, is in the top 4 receivers of the beloved Sherwood Engineering list for dynamic range.  Way above many other conventional analog receivers.  It was in the top three until the KX3.

As far as frequency coverage, I have my direct sampling QS1R (which is similar in performance to the Perseus) and I can use it up to 500 MHz.  The QS1R's software runs on Windows, OS X, or Linux.  I use it on my Linux box and it has never crashed or required a reboot.  Obviously, you have only ever used a QSD based, sound card based SDR, with Windows based software and not a superior direct sampler.

None of the direct sampling receivers require a sound card and I can plug them into my laptop which does not require a hand truck.  Heck, even the low priced/low performance Rf Space SDR-IQ direct sampling SDR has better dynamic range than your spectrum analyzer, and it is USB powered!

Gene
« Last Edit: August 19, 2012, 01:17:58 PM by KE5JPP » Logged
KB1GMX
Member

Posts: 810




Ignore
« Reply #29 on: August 19, 2012, 03:30:00 PM »

>>>Wow, you are way out in left field here.  Where do you come up with the idea that direct sampling SDRs have poor dynamic range?  The Perseus, which is a direct sampling SDR, is in the top 4 receivers of the beloved Sherwood Engineering list for dynamic range.  Way above many other conventional analog receivers.  It was in the top three until the KX3.<<<

And you don't read.

I will repeat, if you want to sample and very high speeds needed for very high frequencies your have to trade bits in the A->D conversion or limit your frequency span.  If you limit the number of bits you limit the dynamic range unless you add RF gain or conversion so you can add agc to the system to avoid overloading the A/D. 

It takes 20bits to express any number between and about 1million or 100DB. If you sampling 20 bits at
at least 8x the input frequency that's 160Mbits/S for a 1mhz signal  For 100Mhz that is a mere 16Gbits second.  That is far faster than most flash A/Ds that are usually under 10bits (o to 1023 or 30db).
Since systems are not quite that fast you either down covert or use fewer bits to to keep the data rate believable.  The LTC2208 is only 16bits so there must be an agc system to allow a over 100db range as the 16bits is 0 to 65535 or 48db between detection threshold and saturation of the A/D.  Oh, that 100DB is the SFDR not the DR which is lower. 


Now you confirm something I'd figured you didn't understand.  Undersampling is downconversion.
How is that done.. well the sample and hold is run at 130mhz so that amounts to the conversion
osc and it will wrap half that or 62.5 mhz down in the direct sampling frequency.  And what happens at exactly 62.5mhz or 192.5mhz?

Direct quote from their site.
>>QS1R covers 10 kHz through 62.5 MHz in its standard configuration and can be used in undersampling applications to 500 MHz.  <<

Direct to 63.5mhz and undersampling to 500mhz.  What does SSB voice at 432 sound like when you undersample?   How do you handle images above 62.5mhz, higher?

It's a fine RECEIVER, not much of a transmitter.  It makes a fine instrumentation receiver once
you get past all the digital artifacts. and never use it next to the computer with a whip, it's plenty good enough to hear all the spurs from the computer especially thing that are ground conducted over the
USB cable.

 But for $899 and you provide a good computer it better be good.   Is that fair to compare
to a 10W HF transceiver?

Can you get a linux version of power SDR for the FLEX 1500, no.  Did you have to cook your version of software?  I know there is SDR for linux and also for uClinux where I do development, both take a bit
of cooking to make it.. least for some "waveforms".

Your on a bend for specific issue that is not KX3 and not even transceiver.  If your grip is with
Bob Sherwood take it to him.


Allison
Logged
Pages: Prev 1 [2] 3 4 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!