EMRFD Message Archive 3109

Message Date From Subject
3109 2009-05-24 07:34:39 Chris Trask Re: Driving Mixers
>
> Any switching mixer where the main elements are driven by the
> sum of the RF and LO sinewave sources across two terminals
> have the same problem.
>
> Ideally, the switching elements should have their impedance
> modulated only by the LO, independently of the level of RF.
>
> For the sum of LO and RF hard-switching a diode however, the
> zero-crossings of the LO are phase-modulated by the RF,
> resulting in non-ideal mixing products. The larger the ratio
> LO/RF current through the diodes, the better.
>

Yes, very much so. The early investigations bu Tucker and others as to the overall effectiveness in optimizing the LO waveform showed that the finite rise and fall times were the primary source of IMD products in the diode ring mixer.

The same is true for commutative (aka Gilbert Cell) mixers. The AD831 makes a good attempt at achieving fast rise and fall times by having a limiting amplifier between the LO terminals and the mixer core. Still, the times remain finite but it's a more practical realization than using a Linville regenerative switch.

>
> The ideal active elements to use instead of diodes (in a ring
> mixer) would be two-port (4 terminal) devices, where the
> impedance between the RF terminals is modulated by the LO
> applied to the other two terminals. This means the impedance
> across the RF terminals is a *linear* time-varying characteristic.
>

Which is exactly what I was pointing towards earlier in stating that a linearized analogue multiplier would make a far better mixer for both NF and IMD performance. This and the earlier thread on the subject of mixers encouraged me to drag out my earlier work on a linearized RF analogue multiplying mixer, and I was able to devise a very simple method of linearizing both the RF and LO signals with lossless (ie - noiseless) feedback. It adds two transistors and two or three transformers to the earlier design. The earlier problem had to do with effectively dealing with the inherent errors in the RF logarithmic converters in the overall translinear scheme.





>A LO modulated light source pointed at a really fast light-dependent resistor
>would work (if one existed). Parametric mixing.
>
>Another way to minimize zero-crossing LO modulation is to simulate a really
>large sinewave by using a square-wave LO. Then the dominant distortion
>products are caused purely by the dependence of diode on-resistance on
>the sum of LO and RF current. That is minimized by increasing the amplitude
>of the squarewave LO.
>
>> And that means, we can close the circle by using a clock generator out
>> of the many intended for the PC industry mainboards.
>>
>> I already mentioned the CY22393. Unfortunately it is not very well
>> specified as a rf signal source and Cypress was not be willing to give
>> me details. But I can say that it works! Again, I have no values because
>> I don't own sophisticated measurement equipment.
>> So I'm interested to hear from other what they think about integrated
>> clock generators beside the famous Si57x (single-sourced, expensive,
>> power-hungry). Such a device is cheap, compact, simple to use (program).
>
>The main thing to look out for in any digital device is that there is
>no other asynchronous processes such as state-machines running in them.
>If there are, noise is induced into the PLL elements via ground-bounce
>spikes in the bond wires of the IC, and could also be via internal
>stray couplings within the IC. With the proper design in an IC,
>multiple PLLs can be isolated better by adding extra bond wires
>and ground/supply pins.
>
>This can preclude running more than one PLL in a three-PLL device at
>a time. Of course, it depends on how much noise your application can
>tolerate.
>
>Apart from that, the only noise that remains (apart from the power supply)
>is phase jitter which just comes from the transistors used in the dividers.
>
>
>------------------------------------
>
>Yahoo! Groups Links
>
>
>


Chris

,----------------------. High Performance Mixers and
/ What's all this \ Amplifiers for RF Communications
/ extinct stuff, anyhow? /
\ _______,--------------' Chris Trask / N7ZWY
_
3111 2009-05-24 08:58:58 Lasse Re: Driving Mixers
I remember seeing the inside a Steinbrecher "Paramixer", which
apparently was patented by Donald (?) Steinbrecher, think he were at MIT
until starting his own company. This mixer offers very high IP3... but
if I remember correctly, looked like a high-level diode ring mixer plus
a pcb with LO driving stuff, LO input was 0 dBm. Have not seen the
patent, I guess it was the LO driver which allowed for these good
numbers, think he used FET-switches to feed the mixer diodes.

Think they offered something like +45 to +50 dBm, and had two versions,
optimized either for HF or VHF. He once told us they stopped selling
the mixers as no-one could ever utilize the performance, and thus only
sold ready made receivers using this key-component.

Anyone have more information on this "Paramixer" ?

We do have two mixers in a receiver system, at the "office" but they are
in-use right now, and it would not be a good move for me to just have a
quick look inside, if something would happen, I'd be toast..
Steinbrecher (the company) was bought by Ericsson and then vanished from
the earth after the first big cell phone shake down :(

/Lasse
Chris Trask skrev:
>
> Yes, very much so. The early investigations bu Tucker and others as to the overall effectiveness in optimizing the LO waveform showed that the finite rise and fall times were the primary source of IMD products in the diode ring mixer.
>
> The same is true for commutative (aka Gilbert Cell) mixers. The AD831 makes a good attempt at achieving fast rise and fall times by having a limiting amplifier between the LO terminals and the mixer core. Still, the times remain finite but it's a more practical realization than using a Linville regenerative switch.
>
>
3114 2009-05-24 09:16:29 ehydra Re: Driving Mixers
Lasse schrieb:
> I remember seeing the inside a Steinbrecher "Paramixer", which
> apparently was patented by Donald (?) Steinbrecher, think he were at MIT
> until starting his own company. This mixer offers very high IP3... but
> if I remember correctly, looked like a high-level diode ring mixer plus
> a pcb with LO driving stuff, LO input was 0 dBm. Have not seen the
> patent, I guess it was the LO driver which allowed for these good
> numbers, think he used FET-switches to feed the mixer diodes.
>
> Think they offered something like +45 to +50 dBm, and had two versions,
> optimized either for HF or VHF. He once told us they stopped selling
> the mixers as no-one could ever utilize the performance, and thus only
> sold ready made receivers using this key-component.
>
> Anyone have more information on this "Paramixer" ?
>
> We do have two mixers in a receiver system, at the "office" but they are
> in-use right now, and it would not be a good move for me to just have a
> quick look inside, if something would happen, I'd be toast..
> Steinbrecher (the company) was bought by Ericsson and then vanished from
> the earth after the first big cell phone shake down :(
>

Hm. In our time now maybe a LVDS driver should to a good job as mixer
driver?
All this very high bitrate stuff is optimized for minimum jitter. And if
I'm right low-jitter characteristic translates to low phase-noise. Just
another view of the same problem: uncertainly or indetermination.

As far as I understood it, high LO drive will minimize the asymmetrics
in the mixer but will drive the noise high just because the current
induced noise.

And amplitude noise is translated to phase noise by the clipping action!


- Henry

--
www.ehydra.dyndns.info
3115 2009-05-24 09:18:57 Chris Trask Re: Driving Mixers
>
>I remember seeing the inside a Steinbrecher "Paramixer", which
>apparently was patented by Donald (?) Steinbrecher, think he were at MIT
>until starting his own company. This mixer offers very high IP3... but
>if I remember correctly, looked like a high-level diode ring mixer plus
>a pcb with LO driving stuff, LO input was 0 dBm. Have not seen the
>patent, I guess it was the LO driver which allowed for these good
>numbers, think he used FET-switches to feed the mixer diodes.
>
>Think they offered something like +45 to +50 dBm, and had two versions,
>optimized either for HF or VHF. He once told us they stopped selling
>the mixers as no-one could ever utilize the performance, and thus only
>sold ready made receivers using this key-component.
>
>Anyone have more information on this "Paramixer" ?
>

That might be US Patent #5493720, which has the Steinbrecher Corporation as the assignee.

An interesting one that I ran across the other day while sifting through some yet-to-be-filed patents was one in which a diode ring mixer was the load for a bridge amplifier. The bridge amplifier was driven by the LO signal, and each side of the bridge acted much like an ECL inverter gate. I'll dig that one out when I return home later this morning as it has some merit in improving the rise/fall times of the LO.


Chris

,----------------------. High Performance Mixers and
/ What's all this \ Amplifiers for RF Communications
/ extinct stuff, anyhow? /
\ _______,--------------' Chris Trask / N7ZWY
_
3116 2009-05-24 09:19:55 ehydra Re: Driving Mixers
Lasse schrieb:
> I remember seeing the inside a Steinbrecher "Paramixer", which
> apparently was patented by Donald (?) Steinbrecher, think he were at MIT
> until starting his own company. This mixer offers very high IP3... but
> if I remember correctly, looked like a high-level diode ring mixer plus
> a pcb with LO driving stuff, LO input was 0 dBm. Have not seen the
> patent, I guess it was the LO driver which allowed for these good
> numbers, think he used FET-switches to feed the mixer diodes.
>
> Think they offered something like +45 to +50 dBm, and had two versions,
> optimized either for HF or VHF. He once told us they stopped selling
> the mixers as no-one could ever utilize the performance, and thus only
> sold ready made receivers using this key-component.
>
> Anyone have more information on this "Paramixer" ?
>

Have a view:
http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=1&u=%2Fnetahtml%2FPTO%2Fsearch-bool.html&r=0&f=S&l=50&TERM1=Steinbrecher&FIELD1=&co1=AND&TERM2=mixer&FIELD2=&d=PTXT


- Henry

--
www.ehydra.dyndns.info
3117 2009-05-24 11:07:56 Chris Trask Re: Driving Mixers
>
> An interesting one that I ran across the other day while sifting
through some
> yet-to-be-filed patents was one in which a diode ring mixer was the load
for a
> bridge amplifier. The bridge amplifier was driven by the LO signal, and
each side
> of the bridge acted much like an ECL inverter gate. I'll dig that one out
when I
> return home later this morning as it has some merit in improving the
rise/fall
> times of the LO.
>

The patent I was think of here is:

Kobayashi, K.W., "High Linearity Active Balanced Mixer,"
US Patent 6,078,802, 20 June 2000

Chris

,----------------------. High Performance Mixers and
/ What's all this \ Amplifiers for RF Communications
/ extinct stuff, anyhow? /
\ _______,--------------' Chris Trask / N7ZWY
_
3118 2009-05-24 12:12:45 Lasse Re: Driving Mixers
I'll think I can come up with the patent's number, give me a few days,
but is should be filed in the early/mid eighties...

/Lasse

ehydra skrev:
> Have a view:
> http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=1&u=%2Fnetahtml%2FPTO%2Fsearch-bool.html&r=0&f=S&l=50&TERM1=Steinbrecher&FIELD1=&co1=AND&TERM2=mixer&FIELD2=&d=PTXT
>
>
> - Henry
>
>
3119 2009-05-24 12:32:32 Lasse Re: Driving Mixers
No pat. no yet, but Mr Donald Steinbrecher seems to have been deeply
into high-performance mixers...
http://massis.lcs.mit.edu/archives/george.gilder.essays/auctioning.the.airwaves
"............

From this experience, Steinbrecher concluded in 1968 that
receivers could be built with at least a thousand times more dynamic
range than was currently believed possible. He assigned his student
Robert Snyder to investigate the issue mathematically, integrating the
possible performance of each component into the performance of a
mixer. Snyder's results stunningly continued Steinbrecher's
hypothesis. They predicted that in principle -- with unlimited time
and effort -- the linearity and dynamic range of a radio could be
improved to any arbitrary standard. In a key invention, Steinbrecher
figured out how to create a diode circuit that could produce a perfect
square wave, creating a diode with essentially zero switching time.

Steinbrecher then proceeded to put his theory into practice by
developing the crucial diode and field-effect transistor arrays,
mixers, amplifiers and other components necessary to build a working
system of unparalleled dynamic range. Most of their advances required
detailed knowledge of the behavior of P/N junctions. To this day, the
performance of Steinbrecher's equipment depends on adjustment to
unexpected nonlinearities and noise sources that were discovered as
part of Robert Snyder's work but are still not integrated into the
prevailing models of diode behavior.
.........."

http://www.sdrforum.org/sdr08/steinbrecher.html
"......In 1975, Dr. Steinbrecher established a fundamental performance limit
for the dynamic range of frequency converters, which resulted in two
patents defining the /ParamixerTM/ frequency translator. The /Paramixer/
became the core technology for the development of numerous broadband
software-defined signals acquisition systems. He later derived a
fundamental performance limit for the dynamic range of the
RF-to-digital conversion process and teamed with Professor fred harris
of San Diego State University to teach a series of short courses on the
optimum design of software-defined radios. Dr. Steinbrecher has
recently discovered a fundamental performance limit on the ability of
signals-acquisition systems to create wide-band high dynamic-range
digital images of propagating RF signals. This fundamental performance
limit incorporates the method of RF-energy capture and shows that there
is an optimum air interface for software-defined radio systems. The
predicted optimum air interface design is described in three patents.
Dr. Steinbrecher has been allowed 14 patents for his work on high
performance signals intercept concepts......"

So, what exactly is the Paramixer then ??? The plot thickens :)


Lasse skrev:
> I'll think I can come up with the patent's number, give me a few days,
> but is should be filed in the early/mid eighties...
>
> /Lasse
>
>
>
3120 2009-05-24 12:37:10 Chris Trask Re: Driving Mixers
>
> I'll think I can come up with the patent's number, give me a few days,
> but is should be filed in the early/mid eighties...
>

I've come across two mixer patents where Steinbrecher is the named
inventor:

Steinbrecher, D.H., "Doubly Balanced Mixer with Optimized Dynamic
Range,"
US Patent 4,112,374, 5 September 1978

Steinbrecher, D.H., "Doubly Balanced Mixer with Optimized Dynamic
Range,"
US Patent 4,230,956, 28 October 1980

These are both quite a hairy bit of circuitry.

Chris

,----------------------. High Performance Mixers and
/ What's all this \ Amplifiers for RF Communications
/ extinct stuff, anyhow? /
\ _______,--------------' Chris Trask / N7ZWY
_
3121 2009-05-24 12:39:26 Lasse Re: Driving Mixers
It seems the Paramixer is covered in patents
4,112,374 and 4,230,956

/Lasse

Chris Trask skrev:
>> An interesting one that I ran across the other day while sifting
>>
> through some
>
>> yet-to-be-filed patents was one in which a diode ring mixer was the load
>>
> for a
>
>> bridge amplifier. The bridge amplifier was driven by the LO signal, and
>>
> each side
>
>> of the bridge acted much like an ECL inverter gate. I'll dig that one out
>>
> when I
>
>> return home later this morning as it has some merit in improving the
>>
> rise/fall
>
>> times of the LO.
>>
>>
>
> The patent I was think of here is:
>
> Kobayashi, K.W., "High Linearity Active Balanced Mixer,"
> US Patent 6,078,802, 20 June 2000
>
> Chris
>
> ,----------------------. High Performance Mixers and
> / What's all this \ Amplifiers for RF Communications
> / extinct stuff, anyhow? /
> \ _______,--------------' Chris Trask / N7ZWY
> _
3122 2009-05-24 12:43:30 Chris Trask Re: Driving Mixers
>
> "......In 1975, Dr. Steinbrecher established a fundamental performance
limit
> for the dynamic range of frequency converters, which resulted in two
> patents defining the /ParamixerTM/ frequency translator.
>

I believe that the two patents I cited earlier are the ones being
referred to here.

Chris

,----------------------. High Performance Mixers and
/ What's all this \ Amplifiers for RF Communications
/ extinct stuff, anyhow? /
\ _______,--------------' Chris Trask / N7ZWY
_
3123 2009-05-24 13:29:05 Lasse Re: Driving Mixers
Chris,
my bad, sorry!

Wish I could see a diagram over the darn thing... patents are not always
the simplest thing to "decode", especially when the drawings doesn't
show in my browser.
Oh, well I try to have a look inside the Paramixer one of these days to
satisfy my curiosity!
If I remember correctly, when I was at a tour around Steinbrecher in
Woburn, there were a few disassembled uints,
and it looked just as if it were a high level diode ring mixer with some
fancy FET-drivers (Motorola, I think but don't remember the numbers) to
acheive a square wave from 0 dBm LO.

The Paramixer needs good termination on all ports, and I remember we had
some problems with a sloppy assembled diplexer circuit, which we had to
resolder.

/Lasse



Chris Trask skrev:
>
> I believe that the two patents I cited earlier are the ones being
> referred to here.
>
> Chris
3124 2009-05-24 13:55:52 Chris Trask Re: Driving Mixers
>
> my bad, sorry!
>

No problem. We seem to have been working in parallel.

>
> Wish I could see a diagram over the darn thing... patents are not always
> the simplest thing to "decode", especially when the drawings doesn't
> show in my browser.
> Oh, well I try to have a look inside the Paramixer one of these days to
> satisfy my curiosity!
> If I remember correctly, when I was at a tour around Steinbrecher in
> Woburn, there were a few disassembled uints,
> and it looked just as if it were a high level diode ring mixer with some
> fancy FET-drivers (Motorola, I think but don't remember the numbers) to
> acheive a square wave from 0 dBm LO.
>
> The Paramixer needs good termination on all ports, and I remember we had
> some problems with a sloppy assembled diplexer circuit, which we had to
> resolder.
>

The circuitry is a bit ambitious, and I doubt that I would want to try
re-creating one. That other patent I mentioned earlier, 6,078,802, has
given me some inspiration on how to fix the infamous H-Mode mixer. I made
one of those early last year, using diodes instead of the FETs and
reconfiguring the centre transfomer for injecting the LO and RF
simultaneously. I then took it apart and used the exact same parts to make
a traditional ring mixer. The passive H-mode mixer had 6dBm higher IMD
products than did the diode ring mixer, or in other words the OIP3 of the
H-mode mixer was 3dB worse than that of the diode ring mixer. I had some
Siliconix quad FETs in the mail at the time, but due to those results I did
not bother to duplicate the experiment with them.

Russell's earlier comment about switching the diodes with the LO only to
eliminate the phase corruption of the switching voltage with the added RF
signal made me return to the H-mode mixer as it does accomplish this, but
the FETs require a fairly high switching voltage, which aggravates the rise
and fall time problem due to V/uSec limitations. So, I'll be heating up the
soldering iron before long and give this new approach that I have in mind a
try.

Chris

,----------------------. High Performance Mixers and
/ What's all this \ Amplifiers for RF Communications
/ extinct stuff, anyhow? /
\ _______,--------------' Chris Trask / N7ZWY
_
3128 2009-05-26 01:01:52 Johan H. Bodin Re: Driving Mixers
Hej Lasse,

you may wish to try the AlternaTIFF viewer plugin. I am using it to view images
at USPTO with Firefox. Works UFB.

http://www.alternatiff.com/

73
Johan SM6LKM


Lasse wrote:
> Wish I could see a diagram over the darn thing... patents are not always
> the simplest thing to "decode", especially when the drawings doesn't
> show in my browser.
3129 2009-05-26 06:27:53 Chris Trask Re: Driving Mixers
>
> > Wish I could see a diagram over the darn thing... patents
> > are not always the simplest thing to "decode", especially
> > when the drawings doesn't show in my browser.
>
>
> you may wish to try the AlternaTIFF viewer plugin. I am using
> it to view images at USPTO with Firefox. Works UFB.
>
> http://www.alternatiff.com/
>

If you download the patents from Google Patents:

http://www.google.com/advanced_patent_search

they're in PDF format. Same with the European Patent Office (EPO):

http://ep.espacenet.com/advancedSearch?locale=en_EP


Chris
3130 2009-05-26 08:59:07 Leon Heller Re: Driving Mixers
3131 2009-05-26 11:42:15 k5nwa Re: Driving Mixers
On 5/26/2009 9:40 AM, Le
3132 2009-05-26 13:18:08 leon Heller Re: Driving Mixers
----- Original Message -----
3133 2009-05-26 14:18:17 Lasse Re: Driving Mixers
Jag spar tipset om alternatiff till ett senare tillfälle, tack i alla
fall Johan!

Thanks for the tip Chris!
The Google link worked like a charm! Got the two patents you mentioned
down loaded!

Found a picture in an old ad for Steinbrecher... and some data:

135 dB dynamic range
IP3in >45dBm
IP2in >+90dBm
NF < 6,2dB
Compression point 1dB >33dBm
Supression point 1dB >33dBm
LO Power <+20dBm
DC power <7w

They claim it can handle 1 Watt both as a transmitting mixer and in a
rx front-end.
I do not think the mixing diodes are exotic, but the way they shape the
switching voltage.

See if I can get some time to study the patents.

Thanks
Lasse


Chris Trask skrev:
>>> Wish I could see a diagram over the darn thing... patents
>>> are not always the simplest thing to "decode", especially
>>> when the drawings doesn't show in my browser.
>>>
>> you may wish to try the AlternaTIFF viewer plugin. I am using
>> it to view images at USPTO with Firefox. Works UFB.
>>
>> http://www.alternatiff.com/
>>
>>
>
> If you download the patents from Google Patents:
>
> http://www.google.com/advanced_patent_search
>
> they're in PDF format. Same with the European Patent Office (EPO):
>
> http://ep.espacenet.com/advancedSearch?locale=en_EP
>
>
> Chris
>