New-ish patent from Flexradio

[applewiz2000]

Hi group,

I am currently studying the Fleradio patent here:
http://www.faqs.org/patents/app/20110043286

He's basically put the analog switches inside the amplifier loop. Any opinions from the group or are you all happy with the mediocre performance of the SR V6.3 ??

~Rob

[W8AEF]

What do you mean about mediocre performance of the SR V6.3?

 

I am very impressed with my V6.3’s performance.

 

de Paul, W8AEF

ZF2JI/ZF2TA FO8DX/FO8PLA 8Q7AA ZX0A VU7RG TX5C A52PP

[applewiz2000]

Paul,

I think you're easily impressed then!! The system is a hedgehog with low IP3. It is also deaf above 21MHz.

The basic design of QSD and implementation of the rest of the hardware is frankly rubbish compared to what Flex sell, admittedly as a commercial product.

~Rob

[John Greusel]

Rob,

Are you revving up to introduce something of your design?

John

KC9OJV

 

---

From: applewiz2000 <robb...@nildram.co.uk>
To: MOBO...@googlegroups.com
Sent: Saturday, May 19, 2012 11:25 AM
Subject: Re: [MOBOKITS] New-ish patent from Flexradio

[Andrea Montefusco]

In any case, it seems they are abandoning the QSD.
Did you see what are showing at Dayton ?

*am*

On 05/19/2012 06:42 PM, John Greusel wrote:
> Rob,
>
> Are you revving up to introduce something of your design?
>
> John
> KC9OJV
>
>

> ----------------------------------------------------------------------------------------------------
> *From:* applewiz2000 <robb...@nildram.co.uk>
> *To:* MOBO...@googlegroups.com
> *Sent:* Saturday, May 19, 2012 11:25 AM
> *Subject:* Re: [MOBOKITS] New-ish patent from Flexradio

>
> Paul,
>
> I think you're easily impressed then!! The system is a hedgehog with low IP3. It is also deaf above
> 21MHz.
>
> The basic design of QSD and implementation of the rest of the hardware is frankly rubbish compared
> to what Flex sell, admittedly as a commercial product.
>
> ~Rob
>
> On Saturday, 19 May 2012 14:52:40 UTC+1, W8AEF wrote:
>
> What do you mean about mediocre performance of the SR V6.3?

> I am very impressed with my V6.3’s performance.
>
>
>


--
*am*

---------------------------------------------------------
Andrea Montefusco iw0hdv http://www.montefusco.com
tel: +393356992791 fax: +390623318709
---------------------------------------------------------

[Sid Boyce]

Very impressive indeed.
I guess the price tag will be quite high but buyers would be getting
money's worth.
73 ... Sid.

Sid Boyce ... Hamradio License G3VBV, Licensed Private Pilot
Emeritus IBM/Amdahl Mainframes and Sun/Fujitsu Servers Tech Support
Senior Staff Specialist, Cricket Coach
Microsoft Windows Free Zone - Linux used for all Computing Tasks

[Edward Currie]

Where can I get a reference to Flex's latest patent?

[Sid Boyce]

I am very impressed with my V6.3’s performance.

 

High Performance == High cost.

Softrock does what it does at a very low price and that's no crime or shame.

The numbers of Flex radios out there, counting in just numbers, if compared to Softrock would make Flex a very rich company.

I'd expect them to dent the sales out of Japan until the Japanese manufacturers wake up to the fact the SDR is the future and no doubt can be made much cheaper than heir current crop with about a thousand knobs - I haven't counted them, I went bleary eyed before I could complete the count.
73 ... Sid. 

[Roger Critchlow]

The first email in the thread linked to a copy.

-- rec --

[Rainer Feldkamp]

Hi all,

There seems to be some misunderstanding in this matter.

1. Both the US and the European Patent Application are in the name of
Gerald Youngblood himself and as far as I remember, Youngblood is not
necessarily identical to FLEX.

2. No patent has been granted so far neither on the US nor on the
European Patent Application, but it will be quite interesting with
what state of the art the respective examiners will come up. Perhaps
there are other interesting news.

3. Copies of the published patent applications are available for free
from the respective Patent Offices (search for 20110043286 under
uspto.gov or EP2288041 under epo.org or in your national Patent
Offices - ipo.gov.uk for the UK Patent Office is not to be found at
first glance).


If you have problems with 3., let me know.

73, Rainer DJ6OA



Zitat von Edward Currie <dr.c...@gmail.com>:

>>>> I am very impressed with my V6.3âEUR(TM)s performance.

>>>>
>>>>
>>>>
>>>>
>>>
>>>
>>
>> --
>> Sid Boyce ... Hamradio License G3VBV, Licensed Private Pilot
>> Emeritus IBM/Amdahl Mainframes and Sun/Fujitsu Servers Tech Support
>> Senior Staff Specialist, Cricket Coach
>> Microsoft Windows Free Zone - Linux used for all Computing Tasks
>>
>>
>

Fax: +49 89 54014014
e-mail: feld...@wallach.de

[The Silver Fox]

I have always been under the opinion that with a fast enough processor you
would not need QSD. That is, with a fast enough FFT you could do great
things with only digital processing.
73,
Alan - W6ARH

[Andrea Montefusco]

On 05/19/2012 08:42 PM, Sid Boyce wrote:
> Very impressive indeed.
> I guess the price tag will be quite high but buyers would be getting money's worth.

up to 6k USD, BTW they are already getting preorders...

http://cart.flexradio.com/Signature-Series-Radios_c_17.html

[Edward Currie]

The patent link is 

http://www.faqs.org/patents/app/20110043286

(Not sure why it is necessary to tell someone to look elsewhere)

[Sid Boyce]

Ouch!
73 ... Sid.

[Edouard Griffiths]

Hello,

correct me if I am wrong but I can't see any real difference with a QSD.

73s!
Edouard, F4EXB.

[W8AEF]

I don’t consider myself as being easily impressed.

 

My V6.3 installed in a 4.3 MOBO has an MDS of 131dB on 28 MHz and 132dB on 21MHz, 250 Hz bandwidth.

 

I have measured very impressive MDS on both of the softrocks I have built.

 

Receive performance is on a par with my FT-1000D (-134MDS on 20m with 50 Hz apf enabled) and my IC-756PROII with 250 Hz filter.

 

I am a DXer and operate primarily cw with rtty mixed in.  SSB if I really have to, and I am on the top of the Honor Roll.

 

de Paul, W8AEF

ZF2JI/ZF2TA FO8DX/FO8PLA 8Q7AA ZX0A VU7RG TX5C A52PP

[applewiz2000]

 Hi Lads,

Night out so slow reply. The new Flex rig is indeed very impressive but not entirely unexpected. They have been selling a DDC receiver for some time now.

For sure they are not undercutting the Flex5000 because that is still fairly new. But the days for the '5000 are undoubtedly numbered. I agree entirely with the opinion that the Japanese are behind the curve. The new Yaesu FT5000 has great signal handling but is an ergonomic nightmare that looks end-of-line.

There are some other interesting patents out there too. And as said Youngblood is not really Flex. He is probably putting out these ideas for use in low end Flex rigs, but there is a possibility he is putting them out as a hint to the "roots" hams who helped him in the early days of the company. Take the latter point cynically if you like.

Whatever I would be up for designing a better Softrock at some time. Having the switches inside a feedback loop will largely remove the non-linearity and errors. The focus will then fall on the instrumentation amplifier. The SR 6.3 was never designed for high performance but for cheapness.

Whatever I got some activity going on this group!

~Rob

[compucat]

Hello Edouard,

in the patent application, Youngblood claims an enhanced phase sampling detector which in most embodiments is a quadrature phase samplng detector or QSD.

Tbe central idea of Youngblood is to provide 'a current mode quadrature sampling detector having analog switches which operate within the opamp feedback loop'

This should be a substantial improvement since by that step, the voltage drop across the analog switches may be neglected.


73, Rainer DJ6OA

[applewiz2000]

Paul,

I have little in definitive measurements as my test gear is limited to a network analyser and borrowed time on a 'scope at work. I compared an Icom IC7000 to the SR 6.3/Mobo system on two nearby antennas at a field day site. The IC7000 basically wiped the floor with it. That's even with the improvements at the end of this page:
http://www.m0rzf.co.uk/softrock123/

As for DX honour roll, well I rarely hear anything outside Europe here. What there is gets swamped with spurious from my receiver and people running Chinese SMPSs for 24/7. The latter may be solved by a QTH move.

-----
Rainer,

Yes you are correct. The analog switches can see almost no voltage swing. The problem with switches is the CMOS rail-to-rail type either switch slowly or have high on-resistance (lossy). The N channel switches are faster and wider bandwidth but their voltage swing is limited.

Additionally the gain and phase distortion introduced by the switches is also greatly reduced. This should ease the burden of doing image rejection in software.

For any possible Softrock redesign, analog fundamentals are ... fundamental.

Impressive as the new Flexradio is, their claim to universal remote operation has a weakness. All that fast sampling and processing will be greedy on current. Remotes powerd by alternative energy will need some big solar panels!!

Cheers for responses,
Rob.

[Mike Collins]

On 5/20/2012 4:26 AM, applewiz2000 wrote:
> Paul,
>
> I have little in definitive measurements as my test gear is limited to
> a network analyser and borrowed time on a 'scope at work. I compared
> an Icom IC7000 to the SR 6.3/Mobo system on two nearby antennas at a
> field day site. The IC7000 basically wiped the floor with it. That's
> even with the improvements at the end of this page:
> http://www.m0rzf.co.uk/softrock123/
>

Hi Rob,

What, no measurements?? How can you make all the claims of poor
performance?

Looking at your "improvements" it is likely they are the cause of your
troubles. Reducing the OP amp gain will likely not help your MDS
problems. Also the preamp design will likely kill the IP3 of the
system. Did you measure that??

My measurements of a stock SR6.3 are much like Pauls. Maybe you should
make some actual measurements so you will be able to find your problem
and make real improvements.

Real measurements would also help with a redesign effort....then all the
ideas floating around could be actually verified with a orig design.
Without that: it is all talk. Easy to come up with design
ideas.....hard part is to actually improve the performance.

73, Mike Collins KF4BQ

[applewiz2000]

Hi Mike,

I stand behind every one of my modifications. Many widget users reduced the gain of the QSD amplifiers far more than I did. I only went down by just over 3dB, which raised the clipping level by that amount. The widget soundcard has only a 5V input circuit.

The preamp makes the whole thing usable on 10m, and ADL5535 was chosen as the lowest gain easily available single chip. It also has a high IP3. Of course a discrete component solution would be better, but it would never fit on such a small board.

I did some comparative measurements of a stock SR 6.3 (I have 2 boards) and admittedly the improvement is not much. There is absolutely no "problem" with any of the components anywhere though.

At least I can actually use my system. There is someone I know who relegated his Mobo system to the junk because it is unusable. He is a very experienced builder and would not have made any mistakes in construction.

I agree that it is all talk, but you can't measure or even simulate anything without having a fairly lengthy survey of ideas. Which is where I turned up the Youngblood patent.

~Rob

[W8AEF]

I suspect your problem at Field Day may have been from transmitters on other bands.  Using a W3NQN design bandpass filter between the softrock and antenna would help here.

 

I intend to use my radio for Field Day with W3NQN filters.

 

de Paul, W8AEF

[applewiz2000]

Hi Paul,

Yes I guess we can't really expect any receiver to work well right next door to a transmitter.

I spent time studying the Flex patent. For Figure 6, I will draw the circuit and simulate. They say it has advantages over previous designs:

1. Analog switches inside amplifier loop - reduce distortion and loss
2. Low local oscillator drive - less LO radiation, perhaps less spurious from lower voltage logic
3. Relatively easy to implement, 2:1 quad multiplexers cheap

One possible flaw - needs high perfomance differential amp.

73, Rob.