RECEIVERS
INFORMATION
LINKS
Message from Calico at Friday, 14-Aug-09 17:29:23 GMT
Radios..
Are they as good as they look? Or as they have us believe?
Some scrutiny can help separate the serious radios from the commerical hypes. When the League's (ARRL) measurements couldnt explain why "good" radios performance would collapse in contests, Sherwood found out, by studying closely spaced signals.
Don't just buy any radio, take a look at the dynamic range, phase noise, etc figures.Hope this can get YOU a radio that can hear what other big-bucks radios simply can't!
If Icom R8500 comes 9th from the bottom of the top list..you get the idea..more at:
http://www.sherweng.com/table.html
Ergonomics, good looks, bells and whistles, not included in this list, wink ! Value for money though, IS !
73
Tim
Last edited by Calico at Friday, 14-Aug-09 19:00:45 UTC
Are they as good as they look? Or as they have us believe?
Some scrutiny can help separate the serious radios from the commerical hypes. When the League's (ARRL) measurements couldnt explain why "good" radios performance would collapse in contests, Sherwood found out, by studying closely spaced signals.
Don't just buy any radio, take a look at the dynamic range, phase noise, etc figures.Hope this can get YOU a radio that can hear what other big-bucks radios simply can't!
If Icom R8500 comes 9th from the bottom of the top list..you get the idea..more at:
http://www.sherweng.com/table.html
Ergonomics, good looks, bells and whistles, not included in this list, wink ! Value for money though, IS !
73
Tim
Last edited by Calico at Friday, 14-Aug-09 19:00:45 UTC
Message from DOCTORWHO at Thursday, 27-Aug-09 03:16:25 GMT
HELP...
I ran thru instructions, and still cannot change frequencies..I put the frequency in on the right side..what do we press to get it on the receiver itsself
DOCTORWHO
I ran thru instructions, and still cannot change frequencies..I put the frequency in on the right side..what do we press to get it on the receiver itsself
DOCTORWHO
Message from sv1btl at Thursday, 27-Aug-09 17:33:48 GMT
Dear Tim,
You cannot imagine how right you are!!!... Yes, the choice of a receiver is a big trouble. Unfortunately, most of the users are convinced that "the more you pay, the better you bye", but this is not true. Sherwood Engineers are maybe the only laboratories that respect the simple truths and they don't play games with the consumers. Of course the best receiver is the antenna, and all of us must have this in mind.
73 de Emmanuel SV1BTL
You cannot imagine how right you are!!!... Yes, the choice of a receiver is a big trouble. Unfortunately, most of the users are convinced that "the more you pay, the better you bye", but this is not true. Sherwood Engineers are maybe the only laboratories that respect the simple truths and they don't play games with the consumers. Of course the best receiver is the antenna, and all of us must have this in mind.
73 de Emmanuel SV1BTL
Message from Token at Sunday, 30-Aug-09 23:02:13 GMT
Absolutely you must pay much attention to multiple factors in selecting a radio, and cost should not be a large factor (unless you can’t afford what you really want). However, how the information is presented is also very important.
For example, that list is sorted by dynamic range narrow spaced, and so, as you rightly pointed out, the R8500 comes in 9th from the bottom. But, I will tell you that my R8500 out performs my FRG-7700 under almost every condition, normally by a lot, despite the FRG-7700 being ranked 16 spaces above the R8500. Also, the FRG-7700 is ranked 6 above the FT-2000, but there simply is no comparison in real world operation, the FT-2000 smokes the FRG-7700 and the R8500. After using both for many years I would take the R8500 over the FRG-7700 anytime, despite the narrow spaced dynamic range performance advantage of the 7700.
Certainly my Heatkit SB-104 does not perform better under any condition I have ever encountered than my FT-2000, although it might in the lab under specific conditions and so it is ranked higher on this list.
And although I think my Flex 5000 is slightly better than my FT-2000, there is certainly nothing in normal use that would match the fact that the Flex is rated more than 70 places above the FT-2000. Having used them both side by side in tough conditions I can say there simply is not that wide a gap between them, in fact the gap is very, very small.
My point is not to say "they are wrong" or "they are right". They have taken solid data. However how the data is presented can bias the perception of the reader. For example, dynamic range on narrow spaced signals is very desirable, a key feature even, but not if the radio is 20 dB less sensitive than the competition. Sort that list on noise floor and it changes totally, again, sort it on sensitivity and it changes. And if you need one rig to listen to 11175 kHz and 167.475 MHz, the Flex 5000 is not going to beat the R8500.
Spending big money does not insure you get big quality. But, quality is not cheap, so you are not going to get big quality for little money. So, the take-away for me has to be research what you want, balance that against what you need, and buy the best performance you can afford. Don’t do it in a vacuum, and don’t do it because someone told you that $2500 radio was better than that $1200 one, but don’t buy the $125 portable and expect it to compete with either of the other radios.
T!
For example, that list is sorted by dynamic range narrow spaced, and so, as you rightly pointed out, the R8500 comes in 9th from the bottom. But, I will tell you that my R8500 out performs my FRG-7700 under almost every condition, normally by a lot, despite the FRG-7700 being ranked 16 spaces above the R8500. Also, the FRG-7700 is ranked 6 above the FT-2000, but there simply is no comparison in real world operation, the FT-2000 smokes the FRG-7700 and the R8500. After using both for many years I would take the R8500 over the FRG-7700 anytime, despite the narrow spaced dynamic range performance advantage of the 7700.
Certainly my Heatkit SB-104 does not perform better under any condition I have ever encountered than my FT-2000, although it might in the lab under specific conditions and so it is ranked higher on this list.
And although I think my Flex 5000 is slightly better than my FT-2000, there is certainly nothing in normal use that would match the fact that the Flex is rated more than 70 places above the FT-2000. Having used them both side by side in tough conditions I can say there simply is not that wide a gap between them, in fact the gap is very, very small.
My point is not to say "they are wrong" or "they are right". They have taken solid data. However how the data is presented can bias the perception of the reader. For example, dynamic range on narrow spaced signals is very desirable, a key feature even, but not if the radio is 20 dB less sensitive than the competition. Sort that list on noise floor and it changes totally, again, sort it on sensitivity and it changes. And if you need one rig to listen to 11175 kHz and 167.475 MHz, the Flex 5000 is not going to beat the R8500.
Spending big money does not insure you get big quality. But, quality is not cheap, so you are not going to get big quality for little money. So, the take-away for me has to be research what you want, balance that against what you need, and buy the best performance you can afford. Don’t do it in a vacuum, and don’t do it because someone told you that $2500 radio was better than that $1200 one, but don’t buy the $125 portable and expect it to compete with either of the other radios.
T!
Message from Calico at Tuesday, 01-Sep-09 01:49:00 GMT
Hi again, nice inputs Emanuel and Token, cheers guys !
Nice comments about being cautious re. how things are presented and antennas.
It was mentioned in the previous posting that if the results were sorted for example by noise floor or sensitivity then the order would be totally different ? Fortunately this is not so:
The table includes both the Noise Floor (second column) and the Phase Noise (7th Column) as well as the Sensitivity (6th Column)
(http://www.sherweng.com/table.html)
Still, it is very important that Sherwood classify receivers by closely spaced signals dynamic range for two reasons:
1) This is what most of us want to know in competitive or serious HF radio: to be able to hear the underdog, low power DX signal in the presence of close-by strong signals, "big-guns", pile ups, etc.
2) To help us consumers make educated choices about which radio makers can excel in one of the most challenging areas, Narrow Spaced Dynamic Range, when most makers models are still only hopeful.
Not all radios are the same, or can hear the same, and we know it.
Take for example the man behind HamRadio Deluxe, HB9DRV, talking about Icom IC-7800 and Elecraft's K3:
"The receiver is quiet - this is *very* pronounced when using CW and the data modes. I am so glad I sold my IC-7800 - this K3 is much better value for money and much easier to use" (source: http://www.eham.net/reviews/detail/6673?page=12)
Icom's TX/RX flagship (IC-7800) is a hugely expensive radio, however, for the time being it doesn't come close to what tops Sherwood Engineers List.
Note :
In Sherwoods List, regarding Minimum Discernible Signal and Phase Noise, a larger negative number is better, -138 is hence better than -123 by 15 dB. And every 3dB = double the magnitude.
However, too high sensitivity can cause trouble with 3rd Order Intercept Point as well as Strong Signal Dynamic Range.
IP2, IP3, BDR, IMDDR3, MDS, PN, etc, IMHO, would be very useful to be present in one's search for a performance radio.
Likewise as my friend Emanuel (SV1BTL) would most probably add, when we go shopping for antennas, we need to know whether the gain in dB mentioned, is in dBi or dBd, and how the measuruments were made, at what height, distances, reference antennas, etc
I am not sure what was tested with the Yaesu FT-2000 and the FlexRadio, since no test areas or methodology are mentioned. They are also different class radios.
Regarding Icom R-8500, I mentioned it as an "attention-grabber" because it is a well known and recognizeable radio, but it would be unfair to seriously compare it to K3, Perseus or SDR-5000 since it is a much older scanner really and not a thoroughbred HF+6m performance radio.
There is an Austrian saying:
"There is only one perfect child in the world, and every mother has it"
Likewise, for every radio user, maybe there is a different favorite radio, according to what criteria / priorities each of us uses.
In my case it's high performance Short Wave(HF)radios.
A nice summary of important radio indices can be found at: http://www.arrl.org/tis/info/pdf/109435.pdf
73,
Tim/Calico
Nice comments about being cautious re. how things are presented and antennas.
It was mentioned in the previous posting that if the results were sorted for example by noise floor or sensitivity then the order would be totally different ? Fortunately this is not so:
The table includes both the Noise Floor (second column) and the Phase Noise (7th Column) as well as the Sensitivity (6th Column)
(http://www.sherweng.com/table.html)
Still, it is very important that Sherwood classify receivers by closely spaced signals dynamic range for two reasons:
1) This is what most of us want to know in competitive or serious HF radio: to be able to hear the underdog, low power DX signal in the presence of close-by strong signals, "big-guns", pile ups, etc.
2) To help us consumers make educated choices about which radio makers can excel in one of the most challenging areas, Narrow Spaced Dynamic Range, when most makers models are still only hopeful.
Not all radios are the same, or can hear the same, and we know it.
Take for example the man behind HamRadio Deluxe, HB9DRV, talking about Icom IC-7800 and Elecraft's K3:
"The receiver is quiet - this is *very* pronounced when using CW and the data modes. I am so glad I sold my IC-7800 - this K3 is much better value for money and much easier to use" (source: http://www.eham.net/reviews/detail/6673?page=12)
Icom's TX/RX flagship (IC-7800) is a hugely expensive radio, however, for the time being it doesn't come close to what tops Sherwood Engineers List.
Note :
In Sherwoods List, regarding Minimum Discernible Signal and Phase Noise, a larger negative number is better, -138 is hence better than -123 by 15 dB. And every 3dB = double the magnitude.
However, too high sensitivity can cause trouble with 3rd Order Intercept Point as well as Strong Signal Dynamic Range.
IP2, IP3, BDR, IMDDR3, MDS, PN, etc, IMHO, would be very useful to be present in one's search for a performance radio.
Likewise as my friend Emanuel (SV1BTL) would most probably add, when we go shopping for antennas, we need to know whether the gain in dB mentioned, is in dBi or dBd, and how the measuruments were made, at what height, distances, reference antennas, etc
I am not sure what was tested with the Yaesu FT-2000 and the FlexRadio, since no test areas or methodology are mentioned. They are also different class radios.
Regarding Icom R-8500, I mentioned it as an "attention-grabber" because it is a well known and recognizeable radio, but it would be unfair to seriously compare it to K3, Perseus or SDR-5000 since it is a much older scanner really and not a thoroughbred HF+6m performance radio.
There is an Austrian saying:
"There is only one perfect child in the world, and every mother has it"
Likewise, for every radio user, maybe there is a different favorite radio, according to what criteria / priorities each of us uses.
In my case it's high performance Short Wave(HF)radios.
A nice summary of important radio indices can be found at: http://www.arrl.org/tis/info/pdf/109435.pdf
73,
Tim/Calico
Message from Token at Friday, 04-Sep-09 15:29:52 GMT
That table does indeed include the data I quoted, noise floor and sensitivity, and that is exactly why I chose to specify those parameters. To highlight the fact that how the data is presented can influence the reader and so the reader must look closely at the data, not just accept it at first glance.
For example, because that list is sorted on dynamic range narrow spaced the Perseus is 2nd from the top. But, if the list were sorted on sensitivity the Perseus would be around 2nd from the bottom. If the list were sorted on noise floor the Perseus would be well down the list, about mid pack.
I do not mean to say that noise floor or sensitivity are more important than narrow spaced dynamic range, or to imply the list should have been sorted that way. I am only saying if it had (for what ever reason) been sorted primarily on another parameter the order of the listing would be very different. And so the person referencing the list must understand the reasoning behind presenting the list as it is.
The overall performance of a receiver does not generally come down to one specific parameter, but rather it is a combination of attributes. To be sure, there are several "most important" parameters for performance (and close dynamic range is very important), but it is still a combination of things that results in a great radio.
I was not defending the R8500 (or any of the other radios I mentioned in my post) in any way, I was just using them as examples of overall performance that is inconsistent with the sorting of that list. Certainly the R8500 is not a high performance HF radio, but it is a decent performance HF radio that also includes the ability to go to over 1 GHz. So, if a user wants good short wave performance and good VHF/UHF performance in one package it moves up near the top of the list very quickly. But if you want high HF performance then obviously it is no place near the top, getting all that bandwidth in one package compromises performance unless you are willing to spend large sums of money. ( ie, R9500, rightfully much further up the list, but at many times the cost).
Your antenna example is perfect. Two antennas can be marketed, one rated at 9.7 and the other at 9.9 dB of forward gain. But, without knowing the rest of the story you really can not tell if those antennas are near equal or not. Is one rated in dBd possibly actually measured on an antenna range? Is the other a calculated theoretical or modeled gain and referenced to an isotropic antenna? There are too many variables undefined to use that number alone to pick the "better" antenna.
Indeed, without a technical background list like the Sherwood list can be confusing to a user. Some numbers are better when large (example dynamic range, wide or close spaced), some are better when small. Numbers are expressed in different units. Etc, etc. It is easy for a user to be misled, to misunderstand that even if their NRD-545 is way down the list from a Collins R-390A the 545 is going to be more useful to many users than the big, heavy, clunky 55 year old design of the Collins.
Right now my most used radios are all software defined. The Flex 5000, Perseus (not mine, I have borrowed a friends, but as a result of using his I will probably get one in the next few weeks), SDR-14, and even the inexpensive SDR-IQ. In fact I think the SDR-IQ is the best "value", I currently have two in the shack and am considering a third. Despite it's image issues above 15 MHz and not so great noise floor it is possibly the best performing radio I have ever used for a new cost of under $500 USD. But, if I lived anyplace near a high power MW BC station I would probably not think that. I am lucky enough to live in a quiet RF location, so some issues that plague low cost radios are not much of a problem.
But, my favorite radios glow in the dark.
T!
Last edited by Token at Friday, 04-Sep-09 15:32:36 UTC
For example, because that list is sorted on dynamic range narrow spaced the Perseus is 2nd from the top. But, if the list were sorted on sensitivity the Perseus would be around 2nd from the bottom. If the list were sorted on noise floor the Perseus would be well down the list, about mid pack.
I do not mean to say that noise floor or sensitivity are more important than narrow spaced dynamic range, or to imply the list should have been sorted that way. I am only saying if it had (for what ever reason) been sorted primarily on another parameter the order of the listing would be very different. And so the person referencing the list must understand the reasoning behind presenting the list as it is.
The overall performance of a receiver does not generally come down to one specific parameter, but rather it is a combination of attributes. To be sure, there are several "most important" parameters for performance (and close dynamic range is very important), but it is still a combination of things that results in a great radio.
I was not defending the R8500 (or any of the other radios I mentioned in my post) in any way, I was just using them as examples of overall performance that is inconsistent with the sorting of that list. Certainly the R8500 is not a high performance HF radio, but it is a decent performance HF radio that also includes the ability to go to over 1 GHz. So, if a user wants good short wave performance and good VHF/UHF performance in one package it moves up near the top of the list very quickly. But if you want high HF performance then obviously it is no place near the top, getting all that bandwidth in one package compromises performance unless you are willing to spend large sums of money. ( ie, R9500, rightfully much further up the list, but at many times the cost).
Your antenna example is perfect. Two antennas can be marketed, one rated at 9.7 and the other at 9.9 dB of forward gain. But, without knowing the rest of the story you really can not tell if those antennas are near equal or not. Is one rated in dBd possibly actually measured on an antenna range? Is the other a calculated theoretical or modeled gain and referenced to an isotropic antenna? There are too many variables undefined to use that number alone to pick the "better" antenna.
Indeed, without a technical background list like the Sherwood list can be confusing to a user. Some numbers are better when large (example dynamic range, wide or close spaced), some are better when small. Numbers are expressed in different units. Etc, etc. It is easy for a user to be misled, to misunderstand that even if their NRD-545 is way down the list from a Collins R-390A the 545 is going to be more useful to many users than the big, heavy, clunky 55 year old design of the Collins.
Right now my most used radios are all software defined. The Flex 5000, Perseus (not mine, I have borrowed a friends, but as a result of using his I will probably get one in the next few weeks), SDR-14, and even the inexpensive SDR-IQ. In fact I think the SDR-IQ is the best "value", I currently have two in the shack and am considering a third. Despite it's image issues above 15 MHz and not so great noise floor it is possibly the best performing radio I have ever used for a new cost of under $500 USD. But, if I lived anyplace near a high power MW BC station I would probably not think that. I am lucky enough to live in a quiet RF location, so some issues that plague low cost radios are not much of a problem.
But, my favorite radios glow in the dark.
T!
Last edited by Token at Friday, 04-Sep-09 15:32:36 UTC
Message from Calico at Saturday, 05-Sep-09 23:31:08 GMT
Hi,
In the original and first posting in this thread it was mentioned that Sherwood helped us understand why otherwise considered “good radios” until then were not up-to-the-task when it came to doing competitive radio (e.g. H.F. DX-ing or Contesting).
In the same posting it was suggested that we should take a look "at the dynamic range figures, phase noise, etc", (NOT only the dynamic range).
It was also raised that it helps to read between the lines when examining whether a radio is high performance or not really. This was raised since we often see adverts referring to large dynamic range figures, which at closer inspection apply to signals spaced too far away (20 kHz or even more) from our target signal.
As far as competitive-high performing (H.F.) radios are concerned, I still maintain my initial position: It is critial that radios are ranked as per their narrow space dynamic range and not by sensitivity or noise floor as suggested for consideration in a previous postings
In fact ranking HF radios (e.g. for SSB, C.W. DXing) by sensitivity or noise floor instead may be misleading and here is why:
1/ Sensitivity
Put simply, it’s the level of input signal required to produce a given signal output, and is expressed in uV, dBm, or dBuV.
Most H.F. radios on the list are “sensitive” enough already. In fact they are already so sensitive, that even the Perseus, (0.6 μV and “worst” case scenario in the top 3 of the list) is more interested in reducing its sensitivity instead of increasing it. This is eloquently shown by its preamplifier being only +2 dB but its attenuator providing attenuation of -10, -20, -30 dB. Other famous radio designers (e.g. John Thorpe) in the past had chosen a similar path too with plenty of attenuation levels available.
Actually it is not difficult to design a sensitive radio, as it is to design a radio that can handle both the weak and the huge signals without losing linearity. For this reason, we are very interested in what comes next, Minimum Discernible Signal and Noise Floor (in dBm or μV).
2/ Minimum Discernible Signal (also known as. Noise Floor)
Put simply, it is the level of input signal that gives a desired audio output level that is equal to the noise output level. This is sometimes called "minimum discernible signal " (MDS), although a skilled operator can copy a signal at considerably less than the noise floor. Most modern receivers have a noise floor within a few dB of "perfect." A perfect receiver would hear only the noise of a resistor at room temperature. However, especially for HF receiving systems, the system noise is rarely determined by the receiver.
In most cases, external noise is many dB higher than the receiver's internal noise. In this case, it is the external factors that determine the system noise performance.
Making the receiver more sensitive will only allow it to hear more noise. It will also be more prone to overload.
In many cases, especially in the lower HF bands, receiver performance can be improved by sacrificing unneeded sensitivity by placing an attenuator in front of the receiver (as we saw in the Perseus example).
3/ Dynamic Range
Put simply, it is the difference between the weakest signal a receiver can hear and the strongest signal a receiver can simultaneously accommodate without noticeable degradation in performance. Although there are four types of dynamic range (including phase noise limited DR, or reciprocal mixing), let’s focus on the ones in the Sherwood list, 5th column Blocking Dynamic. Range (1dB compression) and IMD DR (Two-tone-third order dynamic range) in the11th and 13th column.
Procedurally, two accurate signal generators, hybrid combiners, step attenuators and audio voltmeters are involved in this measurement. It is the difference between the noise-floor level and the level of signal that causes blocking that we read in the 5th column, the blocking dynamic range (BDR) and which is usually expressed in dB.
Most of us have seen what happens if we connect a low-quality front-end radio (naming no names, pointing no fingers) to a large antenna system: the radio “loses it”, pops, pumping, false signals etc start appearing all over the place and more often than not ,the only signal we don’t hear is the desired signal !
The all lovely up to now “Low-noise floor”, starts rising and things go ugly.
It's been a long time challenge; in real life we have multiple offending signals, usually clustered very close together (as easily seen on CW / SSB allocations in bandplans) which takes us to Two-Tone, Third-Order Intermodulation Distortion Dynamic Range, 11th and 13th columns.
When multiple signals are present at the input of a receiver (a normal situation), they can mix together and create additional signals that were not originally present. Some of these signals can cause interference to desired reception. All amplifiers and mixers are nonlinear to a degree. When two or more signals are present in the amplifier at the same time, additional signals will be created as the input signals mix together. These can range from simple harmonic generation or sum and difference products to more complex mixing involving the fundamental input signals and their harmonics. Frusrating!
The radios which have superior Dynamic Range performance can make good use of a well designed receiver, with serious roofing filters, clean local oscillators providing good AGC and low phase noise to name in passing the rest of the columns in Sherwood Engineers List.
I hope it becomes now more clear why ranking by Noise Floor or Sensitivity instead of Dynamic Range(s) may be misleading and bias people to spend small fortunes unnecessarily.
All one has to do is see how much they would have to spend to get e.g. the top of Icom’s HF radios IC-7800 whose Noise Floor and Sensitivity are better than a K3 or Perseus, or Flex 5000A or even the fantastic and much older AOR AR-7030.
Still both lab and hands-on findings have shown that there the IC-7800 was definitely not the optimum choice (for example see previous reference to HamRadio Deluxe) because of its average Dynamic range performance (only 80dB at 2kHz).
In a previous posting a comparison between Flex 5000A and Yaesu FT2000 was made and it was suggested that the Flex is only slightly better than the FT2000.I do not know how the comparisons were made, and the methodology followed.
However, it would be difficult to follow how in a crowded cluster of strong stations any weak station could be copied only “slightly better” with the Flex than the FT2000 in the light that:
Measured at 500Hz space and 14 MHz, the Flex gives immunity of 96 to 99 dB and the Yaesu FT2000 gives only 64 dB. (!)
That’s a hopping 32 to 35 dB difference, not what would one call “slight” difference. Mat Youngblood of Flex would be a bit sad to read this 35dB superiority reduced to such an analogy.
I have radios at similar or better specification (including a PC deependent SDR) and when it comes to pile-ups for a DX station, the comparison is clear comfortable copy versus frustrating / fatigued copy for SSB/CW with flat settings , on the same antenna with descent quality of isolation and distribution. It’s not necessary to mention/ promote any makes and models.What is at the top of the list today, may not be at the top of the list two months down the road and none of us own stock of their shares. ;)
I fully agree that the Icom R-8500 is a descent radio for what it was designed to be, a general coverage scanner. I have it too, like it and I respect it for its reliability, and general performance.
Likewise I respect the JRC and Collins legends, have worked the JRC, SAIT, etc professional consoles in the past, but times have moved on and what they used to teach us in College once upon a time about the 3Ss (Sensitivity, Selectivity and Stability) are only a small part of a much more complex understanding of signal handling nowadays.
Last but not least, I too am excited about SDRs , and can’t wait to see more competition entering the game especially in the transceiver market. Exciting and fun to use radios, I carry my SDR RX to work sometimes. I too think the SDR-IQ is a good value for money solution.
A small reservation though about SDRs so far.. Since most of them (but not all) can’t work without a computer, for the time being we are limited to how RF quiet/clean our computer, its peripherals and power supplied are.
I hope that the exchange of these postings for the benefit of our Forum readers, have contributed to the understanding of what counts and why when considering and testing a new HF radio so as to arm us against hypes and protect our investment in radios.
Resources used to supplement and support the above posting chosen for easy access by anyone
The 1989 (and 2007) ARRL Handbook for radio communications
NCJ March/ April2008
QEX Setp/Oct 2002
QEX June (year info not surviving, Article: Swept Receiver Dynamic Range Testing)
QST October 1994
QST Aug 2002
QST July 2008
QST February 2007
JRC JST-535D Serial: 68926 test
QST June 2009
Sherwood Engineers (updated datat September 3, 2009)
http://www.icomuk.co.uk/files/icom/fieldsetField/254/IC7800leaflet.pdf
73
Calico (Dorset Radio, UK)
Last edited by Calico at Sunday, 06-Sep-09 11:00:20 UTC
In the original and first posting in this thread it was mentioned that Sherwood helped us understand why otherwise considered “good radios” until then were not up-to-the-task when it came to doing competitive radio (e.g. H.F. DX-ing or Contesting).
In the same posting it was suggested that we should take a look "at the dynamic range figures, phase noise, etc", (NOT only the dynamic range).
It was also raised that it helps to read between the lines when examining whether a radio is high performance or not really. This was raised since we often see adverts referring to large dynamic range figures, which at closer inspection apply to signals spaced too far away (20 kHz or even more) from our target signal.
As far as competitive-high performing (H.F.) radios are concerned, I still maintain my initial position: It is critial that radios are ranked as per their narrow space dynamic range and not by sensitivity or noise floor as suggested for consideration in a previous postings
In fact ranking HF radios (e.g. for SSB, C.W. DXing) by sensitivity or noise floor instead may be misleading and here is why:
1/ Sensitivity
Put simply, it’s the level of input signal required to produce a given signal output, and is expressed in uV, dBm, or dBuV.
Most H.F. radios on the list are “sensitive” enough already. In fact they are already so sensitive, that even the Perseus, (0.6 μV and “worst” case scenario in the top 3 of the list) is more interested in reducing its sensitivity instead of increasing it. This is eloquently shown by its preamplifier being only +2 dB but its attenuator providing attenuation of -10, -20, -30 dB. Other famous radio designers (e.g. John Thorpe) in the past had chosen a similar path too with plenty of attenuation levels available.
Actually it is not difficult to design a sensitive radio, as it is to design a radio that can handle both the weak and the huge signals without losing linearity. For this reason, we are very interested in what comes next, Minimum Discernible Signal and Noise Floor (in dBm or μV).
2/ Minimum Discernible Signal (also known as. Noise Floor)
Put simply, it is the level of input signal that gives a desired audio output level that is equal to the noise output level. This is sometimes called "minimum discernible signal " (MDS), although a skilled operator can copy a signal at considerably less than the noise floor. Most modern receivers have a noise floor within a few dB of "perfect." A perfect receiver would hear only the noise of a resistor at room temperature. However, especially for HF receiving systems, the system noise is rarely determined by the receiver.
In most cases, external noise is many dB higher than the receiver's internal noise. In this case, it is the external factors that determine the system noise performance.
Making the receiver more sensitive will only allow it to hear more noise. It will also be more prone to overload.
In many cases, especially in the lower HF bands, receiver performance can be improved by sacrificing unneeded sensitivity by placing an attenuator in front of the receiver (as we saw in the Perseus example).
3/ Dynamic Range
Put simply, it is the difference between the weakest signal a receiver can hear and the strongest signal a receiver can simultaneously accommodate without noticeable degradation in performance. Although there are four types of dynamic range (including phase noise limited DR, or reciprocal mixing), let’s focus on the ones in the Sherwood list, 5th column Blocking Dynamic. Range (1dB compression) and IMD DR (Two-tone-third order dynamic range) in the11th and 13th column.
Procedurally, two accurate signal generators, hybrid combiners, step attenuators and audio voltmeters are involved in this measurement. It is the difference between the noise-floor level and the level of signal that causes blocking that we read in the 5th column, the blocking dynamic range (BDR) and which is usually expressed in dB.
Most of us have seen what happens if we connect a low-quality front-end radio (naming no names, pointing no fingers) to a large antenna system: the radio “loses it”, pops, pumping, false signals etc start appearing all over the place and more often than not ,the only signal we don’t hear is the desired signal !
The all lovely up to now “Low-noise floor”, starts rising and things go ugly.
It's been a long time challenge; in real life we have multiple offending signals, usually clustered very close together (as easily seen on CW / SSB allocations in bandplans) which takes us to Two-Tone, Third-Order Intermodulation Distortion Dynamic Range, 11th and 13th columns.
When multiple signals are present at the input of a receiver (a normal situation), they can mix together and create additional signals that were not originally present. Some of these signals can cause interference to desired reception. All amplifiers and mixers are nonlinear to a degree. When two or more signals are present in the amplifier at the same time, additional signals will be created as the input signals mix together. These can range from simple harmonic generation or sum and difference products to more complex mixing involving the fundamental input signals and their harmonics. Frusrating!
The radios which have superior Dynamic Range performance can make good use of a well designed receiver, with serious roofing filters, clean local oscillators providing good AGC and low phase noise to name in passing the rest of the columns in Sherwood Engineers List.
I hope it becomes now more clear why ranking by Noise Floor or Sensitivity instead of Dynamic Range(s) may be misleading and bias people to spend small fortunes unnecessarily.
All one has to do is see how much they would have to spend to get e.g. the top of Icom’s HF radios IC-7800 whose Noise Floor and Sensitivity are better than a K3 or Perseus, or Flex 5000A or even the fantastic and much older AOR AR-7030.
Still both lab and hands-on findings have shown that there the IC-7800 was definitely not the optimum choice (for example see previous reference to HamRadio Deluxe) because of its average Dynamic range performance (only 80dB at 2kHz).
In a previous posting a comparison between Flex 5000A and Yaesu FT2000 was made and it was suggested that the Flex is only slightly better than the FT2000.I do not know how the comparisons were made, and the methodology followed.
However, it would be difficult to follow how in a crowded cluster of strong stations any weak station could be copied only “slightly better” with the Flex than the FT2000 in the light that:
Measured at 500Hz space and 14 MHz, the Flex gives immunity of 96 to 99 dB and the Yaesu FT2000 gives only 64 dB. (!)
That’s a hopping 32 to 35 dB difference, not what would one call “slight” difference. Mat Youngblood of Flex would be a bit sad to read this 35dB superiority reduced to such an analogy.
I have radios at similar or better specification (including a PC deependent SDR) and when it comes to pile-ups for a DX station, the comparison is clear comfortable copy versus frustrating / fatigued copy for SSB/CW with flat settings , on the same antenna with descent quality of isolation and distribution. It’s not necessary to mention/ promote any makes and models.What is at the top of the list today, may not be at the top of the list two months down the road and none of us own stock of their shares. ;)
I fully agree that the Icom R-8500 is a descent radio for what it was designed to be, a general coverage scanner. I have it too, like it and I respect it for its reliability, and general performance.
Likewise I respect the JRC and Collins legends, have worked the JRC, SAIT, etc professional consoles in the past, but times have moved on and what they used to teach us in College once upon a time about the 3Ss (Sensitivity, Selectivity and Stability) are only a small part of a much more complex understanding of signal handling nowadays.
Last but not least, I too am excited about SDRs , and can’t wait to see more competition entering the game especially in the transceiver market. Exciting and fun to use radios, I carry my SDR RX to work sometimes. I too think the SDR-IQ is a good value for money solution.
A small reservation though about SDRs so far.. Since most of them (but not all) can’t work without a computer, for the time being we are limited to how RF quiet/clean our computer, its peripherals and power supplied are.
I hope that the exchange of these postings for the benefit of our Forum readers, have contributed to the understanding of what counts and why when considering and testing a new HF radio so as to arm us against hypes and protect our investment in radios.
Resources used to supplement and support the above posting chosen for easy access by anyone
The 1989 (and 2007) ARRL Handbook for radio communications
NCJ March/ April2008
QEX Setp/Oct 2002
QEX June (year info not surviving, Article: Swept Receiver Dynamic Range Testing)
QST October 1994
QST Aug 2002
QST July 2008
QST February 2007
JRC JST-535D Serial: 68926 test
QST June 2009
Sherwood Engineers (updated datat September 3, 2009)
http://www.icomuk.co.uk/files/icom/fieldsetField/254/IC7800leaflet.pdf
73
Calico (Dorset Radio, UK)
Last edited by Calico at Sunday, 06-Sep-09 11:00:20 UTC
Message from Token at Sunday, 13-Sep-09 15:11:02 GMT
"In a previous posting a comparison between Flex 5000A and Yaesu FT2000 was made and it was suggested that the Flex is only slightly better than the FT2000.I do not know how the comparisons were made, and the methodology followed.
However, it would be difficult to follow how in a crowded cluster of strong stations any weak station could be copied only “slightly better” with the Flex than the FT2000 in the light that:
Measured at 500Hz space and 14 MHz, the Flex gives immunity of 96 to 99 dB and the Yaesu FT2000 gives only 64 dB. (!)
That’s a hopping 32 to 35 dB difference, not what would one call “slight” difference. Mat Youngblood of Flex would be a bit sad to read this 35dB superiority reduced to such an analogy."
I am the one who made this comparison. I have no doubt the performance figures measured in the lab indicate the Flex 5000 is "much" better in the area of close spaced dynamic range, 32 dB is very significant. Anytime you are talking about one item being over 1000 times more resistant to a problem that is significant.
I made that comparison, as I said in the original post, based on using the radios side-by-side. I currently own both radios (and have for some time now), and simply put in actual use, in pile-ups, I find the Flex 5000A to be slightly better than the FT-2000 at pulling out the deep signals on voice transmissions. In the CW realm I find the Flex more noticeably better. But, neither of these in actual use are as striking as the more than 1000 times better that the lab test indicates. And that was my point, raw numbers are great to have, but unless the end user understands what the numbers mean they can be a bit misleading. I did not mean to impugn the testing, or to imply it was wrong, I only meant to say that lab test are a great starting spot, they give a quantitative scale, but in actual application the user might find that they like the general "feel" (all aspects combined) of a radio ranked 6th over the performance of a radio ranked 1st (that was not a jab at the K3, I have never used one). Things like sound quality, build quality, etc can combine to minimize the impact of (but never remove) the performance differences once the radio is on the shack desk.
Yes, this is completely non-scientific. But, it is what the end users has to deal with in reality and in use, the total package, not just the numbers from lab results. We are in agreement on what you are saying, I am only adding that data must be taken as part of the whole.
If you think I am disagreeing with you there has been a misunderstanding, I agree, I just want to add that the user needs to be careful in consulting such list, there is a bit to take in to have the full picture.
I also regularly take my SDR to work with me. Small, compact, easy to use, and very entertaining if are fortunate enough to have access to a good antenna at the work location. And, the waterfall display running on a monitor on my desk has started more than one conversation at work.
T!
Mohave Desert, CA, USA
However, it would be difficult to follow how in a crowded cluster of strong stations any weak station could be copied only “slightly better” with the Flex than the FT2000 in the light that:
Measured at 500Hz space and 14 MHz, the Flex gives immunity of 96 to 99 dB and the Yaesu FT2000 gives only 64 dB. (!)
That’s a hopping 32 to 35 dB difference, not what would one call “slight” difference. Mat Youngblood of Flex would be a bit sad to read this 35dB superiority reduced to such an analogy."
I am the one who made this comparison. I have no doubt the performance figures measured in the lab indicate the Flex 5000 is "much" better in the area of close spaced dynamic range, 32 dB is very significant. Anytime you are talking about one item being over 1000 times more resistant to a problem that is significant.
I made that comparison, as I said in the original post, based on using the radios side-by-side. I currently own both radios (and have for some time now), and simply put in actual use, in pile-ups, I find the Flex 5000A to be slightly better than the FT-2000 at pulling out the deep signals on voice transmissions. In the CW realm I find the Flex more noticeably better. But, neither of these in actual use are as striking as the more than 1000 times better that the lab test indicates. And that was my point, raw numbers are great to have, but unless the end user understands what the numbers mean they can be a bit misleading. I did not mean to impugn the testing, or to imply it was wrong, I only meant to say that lab test are a great starting spot, they give a quantitative scale, but in actual application the user might find that they like the general "feel" (all aspects combined) of a radio ranked 6th over the performance of a radio ranked 1st (that was not a jab at the K3, I have never used one). Things like sound quality, build quality, etc can combine to minimize the impact of (but never remove) the performance differences once the radio is on the shack desk.
Yes, this is completely non-scientific. But, it is what the end users has to deal with in reality and in use, the total package, not just the numbers from lab results. We are in agreement on what you are saying, I am only adding that data must be taken as part of the whole.
If you think I am disagreeing with you there has been a misunderstanding, I agree, I just want to add that the user needs to be careful in consulting such list, there is a bit to take in to have the full picture.
I also regularly take my SDR to work with me. Small, compact, easy to use, and very entertaining if are fortunate enough to have access to a good antenna at the work location. And, the waterfall display running on a monitor on my desk has started more than one conversation at work.
T!
Mohave Desert, CA, USA
Message from Herz at Monday, 14-Sep-09 23:56:05 GMT
Hi guys
If you don't mind me asking one question here?
How is the Flex and the Yaesu fed to the same antenna when the tests are made side-by-side?
What cables and/or selection mechanisms connect the two radios presumably on the same antenna ?
Interesting topic indeed
Herz, UK
If you don't mind me asking one question here?
How is the Flex and the Yaesu fed to the same antenna when the tests are made side-by-side?
What cables and/or selection mechanisms connect the two radios presumably on the same antenna ?
Interesting topic indeed
Herz, UK
Message from Calico at Friday, 18-Sep-09 12:20:24 GMT
Choosing a new radio...(continued)
Hi,
Yes! The total package as you mention, is very significant, the feel good factor for a radio someone is going to pay thousands of $ is MAJOR, we are on the same page on that too. And of course I second (support) what you mentioned about a good balance of features (and price, of course!)
No such thing as a perfect radio on the other hand either. But IP2, IP3, BDR etc Sherwood indices, have helped quite a few of us make informed choices and protect against expensive mistakes.
Most radios exhibit about 20-30 dB difference EVEN in Blocking Dynamic Range (that is Desense, when radio loses its sensitivity) when it comes to taking the measurement either at 20 kHz or 5 kHz (let alone any closer).
For Example, let's take the top of the range of the "Big 3" from Japan.
Yaesu FTDX9000 Contest (List Price $ 4 999.00)
128 dB at 20kHz spacing but 97 dB at 5kHz (31 dB worse)
Kenwood TS2000 (List price $ 2 186.00)
121 dB at 20 kHz spacing but 98 dB at 5 kHz (23 dB worse)
Icom IC-7800 (the new version) (List Price: $ 13 956.00)
143 dB at 20 kHz spacing but 114 dB at 5 kHz (29 dB worse)
ICOM IC-R9500 (List Price: $ 15 995.00)
144 dB at 20 kHz spacing but 109 dB at 5 kHz (35 dB worse)
But other designs offer much more consistent performance:
Flex 5000A
123 dB at 20 kHz spacing AND 123 dB at 5 kHz (No difference)
Perseus (Microtelecom)
115 dB at 20 kHz spacing but 98 dB at 5 kHz (17 dB worse)
Elecraft K3
139 dB at 20 kHz spacing AND 139 dB at 5 kHz (No difference)
This new comparison in this posting (in addition to the Third-Order Intermodulation Distortion Dynamic Range in some previous posting) indicates that whereas some designs lose sensitivity as the offending signals come closer, (and that is when unfortunately IMD products massacre the MDS) some other designs allow just as good performance and hence sensitivity both with far away and close by signals. And in real life that counts a lot.
I am embarrassed to admit that one "good" radio (according to ARRL, "Totally Excellent",urgh!) I bought at a hamfest last year, some weeks ago could hear nothing of a signal on a busy 80m band at recent weekend evening which was Q-5 on one of the 3 first radios in Sherwood's list. I hadn't done my homework before buying, serves me right..
It is great that you are pleased with the performance of the Yaesu FT-2000. It's a very popular radio here in the UK too, and if I can remember well, it also interfaces with RF-Space SDR-IQ. One day, I may be tempted to invest in one, when the recession is over ! ;)
However, as I mentioned previously, I cannot comment on the side by side comparison of the FT-2000 and Flex 5000A as:
I don't know the impedance load distribution of the two radios when you tested them and what devices were used?
The existence of buffering of the antenna input so as to ensure stable isolation @ 50 Ω (Ohm) at the RX's antenna terminals.
The spacing AND magnitude in μV (uV) of the "strong" to "weak" signals ratio so as to get an idea of what dB Dynamic Range was needed. If for example and very roughly one radio has 105 dB dynamic range, it will not perform better than a receiver that has 72 dB if only 60dB is required by the offending signal(s).
However as a rule of thumb, if a radio is challenged in a lab environment, things are even harder in real life RF environment especially on the low bands where in addition to the myriads of signals stretching the radio,atmospheric band noise, static crashes, etc are not on our side.
Pity that the Japanese makers don't go by the usual convention for S-Level readings. For example the Flex for a standard S-9 reading, adheres to the 50 μV convention, either with the preamp on or off, it's still the same signal, 50 μV at the antenna terminal, so correctly the Flex won't show more signal on the S-meter when the preamp is on, since the signal remains the same.
Most Japanese radios on the other hand, are not consistent with what strength an S-9 signal is. For example the FT-2000 reads S-9 for a signal of 160 μV with the preamp off, 45 μV with Preamp 1 and 20 μV with Preamp 2 (measurements at 14.200 MHz)
What I am trying to say is that any signal indication comparisons using the two radios would also have to compensate for this inconsistency.
I respect and agree with what you say, and the viewpoint you introduce, also I want to thank you for our constructive and civil exchange of information and experiences.
A little ray of hope regarding our very old indeed mixers, synthesizers, roofing filters, etc etc technology, bypassing component tolerance variations and performane maybe the new generation of completely independent of PCs SDR transceivers, the next wave!
First steps again, expensive too (4600 Swiss Francs), but anyway, here it is:
http://www.adat.ch/p2e_adat.html
http://www.adat.ch/img/adt200a_front.jpg
http://www.adat.ch/p4e_specification.html
For referencing, all measurements mentioned in above posting were made on the 20 m band and were conducted/supplied by the League (ARRL)
That's for now,
73 de Tim and Dorset Radio, UK
Last edited by Calico at Saturday, 19-Sep-09 08:32:03 UTC
Hi,
Yes! The total package as you mention, is very significant, the feel good factor for a radio someone is going to pay thousands of $ is MAJOR, we are on the same page on that too. And of course I second (support) what you mentioned about a good balance of features (and price, of course!)
No such thing as a perfect radio on the other hand either. But IP2, IP3, BDR etc Sherwood indices, have helped quite a few of us make informed choices and protect against expensive mistakes.
Most radios exhibit about 20-30 dB difference EVEN in Blocking Dynamic Range (that is Desense, when radio loses its sensitivity) when it comes to taking the measurement either at 20 kHz or 5 kHz (let alone any closer).
For Example, let's take the top of the range of the "Big 3" from Japan.
Yaesu FTDX9000 Contest (List Price $ 4 999.00)
128 dB at 20kHz spacing but 97 dB at 5kHz (31 dB worse)
Kenwood TS2000 (List price $ 2 186.00)
121 dB at 20 kHz spacing but 98 dB at 5 kHz (23 dB worse)
Icom IC-7800 (the new version) (List Price: $ 13 956.00)
143 dB at 20 kHz spacing but 114 dB at 5 kHz (29 dB worse)
ICOM IC-R9500 (List Price: $ 15 995.00)
144 dB at 20 kHz spacing but 109 dB at 5 kHz (35 dB worse)
But other designs offer much more consistent performance:
Flex 5000A
123 dB at 20 kHz spacing AND 123 dB at 5 kHz (No difference)
Perseus (Microtelecom)
115 dB at 20 kHz spacing but 98 dB at 5 kHz (17 dB worse)
Elecraft K3
139 dB at 20 kHz spacing AND 139 dB at 5 kHz (No difference)
This new comparison in this posting (in addition to the Third-Order Intermodulation Distortion Dynamic Range in some previous posting) indicates that whereas some designs lose sensitivity as the offending signals come closer, (and that is when unfortunately IMD products massacre the MDS) some other designs allow just as good performance and hence sensitivity both with far away and close by signals. And in real life that counts a lot.
I am embarrassed to admit that one "good" radio (according to ARRL, "Totally Excellent",urgh!) I bought at a hamfest last year, some weeks ago could hear nothing of a signal on a busy 80m band at recent weekend evening which was Q-5 on one of the 3 first radios in Sherwood's list. I hadn't done my homework before buying, serves me right..
It is great that you are pleased with the performance of the Yaesu FT-2000. It's a very popular radio here in the UK too, and if I can remember well, it also interfaces with RF-Space SDR-IQ. One day, I may be tempted to invest in one, when the recession is over ! ;)
However, as I mentioned previously, I cannot comment on the side by side comparison of the FT-2000 and Flex 5000A as:
I don't know the impedance load distribution of the two radios when you tested them and what devices were used?
The existence of buffering of the antenna input so as to ensure stable isolation @ 50 Ω (Ohm) at the RX's antenna terminals.
The spacing AND magnitude in μV (uV) of the "strong" to "weak" signals ratio so as to get an idea of what dB Dynamic Range was needed. If for example and very roughly one radio has 105 dB dynamic range, it will not perform better than a receiver that has 72 dB if only 60dB is required by the offending signal(s).
However as a rule of thumb, if a radio is challenged in a lab environment, things are even harder in real life RF environment especially on the low bands where in addition to the myriads of signals stretching the radio,atmospheric band noise, static crashes, etc are not on our side.
Pity that the Japanese makers don't go by the usual convention for S-Level readings. For example the Flex for a standard S-9 reading, adheres to the 50 μV convention, either with the preamp on or off, it's still the same signal, 50 μV at the antenna terminal, so correctly the Flex won't show more signal on the S-meter when the preamp is on, since the signal remains the same.
Most Japanese radios on the other hand, are not consistent with what strength an S-9 signal is. For example the FT-2000 reads S-9 for a signal of 160 μV with the preamp off, 45 μV with Preamp 1 and 20 μV with Preamp 2 (measurements at 14.200 MHz)
What I am trying to say is that any signal indication comparisons using the two radios would also have to compensate for this inconsistency.
I respect and agree with what you say, and the viewpoint you introduce, also I want to thank you for our constructive and civil exchange of information and experiences.
A little ray of hope regarding our very old indeed mixers, synthesizers, roofing filters, etc etc technology, bypassing component tolerance variations and performane maybe the new generation of completely independent of PCs SDR transceivers, the next wave!
First steps again, expensive too (4600 Swiss Francs), but anyway, here it is:
http://www.adat.ch/p2e_adat.html
http://www.adat.ch/img/adt200a_front.jpg
http://www.adat.ch/p4e_specification.html
For referencing, all measurements mentioned in above posting were made on the 20 m band and were conducted/supplied by the League (ARRL)
That's for now,
73 de Tim and Dorset Radio, UK
Last edited by Calico at Saturday, 19-Sep-09 08:32:03 UTC
© 2007-2018 Ivo Smits, UCIS Internet - About GlobalTuners & Contact - Page generated in 2 ms. 0 SQL queries used.