>I have been testing the new EnGenius 600mW ecb-8610s and the range is
>the best I have ever tested.
>
>It is giving me %50 better range than the 400mW ecb-3220 units.
Ok, let's do the math.
dB = 10 log (600/400) = 1.76 dB.
Hmmm.... not much. That's the difference you should see using
Netstumbler or the typical laptop signal strength indication.
As for the range increase, range doubles for every 6dB increase in
system gain. So, a 1.76dB increase will yield:
range_increase = 10^(1.76/20) = 1.22 time
or a 22% increase in range.
To get a 50% increase in range, you would also need to improve the
receiver sensitivity. Since most units are running near the best
sensitivity that can be achieve with current technology, that's not
going to happen this week.
>There is a short video here showing off the product features.
I didn't think You-Tube allowed commercial advertising.
Incidentally, I suggest you ease up on the confusing model numbers.
Unless the viewer already owns one, I doubt if the long model numbers
mean anything to a prospective buyer.
On Nov 15, 7:37 am, Jeff Liebermann <je...@cruzio.com> wrote:
> keenanj <keenanj.303...@no-mx.wirelessforums.org> hath wroth:
>
> >I have been testing the new EnGenius 600mW ecb-8610s and the range is
> >the best I have ever tested.
>
> >It is giving me %50 better range than the 400mW ecb-3220 units.
>
> Ok, let's do the math.
> dB = 10 log (600/400) = 1.76 dB.
> Hmmm.... not much. That's the difference you should see using
> Netstumbler or the typical laptop signal strength indication.
>
> As for the range increase, range doubles for every 6dB increase in
> system gain. So, a 1.76dB increase will yield:
> range_increase = 10^(1.76/20) = 1.22 time
> or a 22% increase in range.
>
> To get a 50% increase in range, you would also need to improve the
> receiver sensitivity. Since most units are running near the best
> sensitivity that can be achieve with current technology, that's not
> going to happen this week.
>
> >There is a short video here showing off the product features.
>
> I didn't think You-Tube allowed commercial advertising.
> Incidentally, I suggest you ease up on the confusing model numbers.
> Unless the viewer already owns one, I doubt if the long model numbers
> mean anything to a prospective buyer.
>
> >'Keenan Systems Wireless Store Featuring Engenius'
> >(http://www.keenansystems.com/store/c....php?products_...)
>
> Good luck, and please lay off the miraculous claims.
>
> --
> Jeff Liebermann je...@cruzio.com
> 150 Felker St #D http://www.LearnByDestroying.com
> Santa Cruz CA 95060http://802.11junk.com
> Skype: JeffLiebermann AE6KS 831-336-2558
I wish the manufacturers would concentrate on lower noise receivers
versus higher power.
On Thu, 15 Nov 2007 11:47:08 -0800 (PST), miso@sushi.com wrote:
>I wish the manufacturers would concentrate on lower noise receivers
>versus higher power.
I'm not sure they can do much more with non-cryogenically cooled front
ends. Most are SiGe or GaAs. The front end noise figures aren't
going to get much better without cooling. With integrated chipsets,
the minimum sensitivity is typically set by the internal noise pickup,
not by the NF.
It would also be nice if they would stop lifting the numbers from the
chipset data sheet and actually test for sensitivity. So far, only
DLink seems to be supplying real numbers. Everyone else lies:
<http://802.11junk.com/jeffl/rx-sens/receiver%20sensitivity.htm>
Personally, I'd like to see the RF section seperate from the digital
stuff, so I can install it high on top of a tower. Alvirion has had
this for a long time, but it's expensive. Note that this implies a
receiver front end that can handle some serious overload and not go
into blocking in high RF environments. Incidentally, that's one
application where a tower mounted power amp and pre-amp can do some
good. However, the rx amps in the boxes that I've seen are actually
worse that what's in the access points.
On Nov 15, 7:01 pm, Jeff Liebermann <je...@comix.santa-cruz.ca.us>
wrote:
> On Thu, 15 Nov 2007 11:47:08 -0800 (PST), m...@sushi.com wrote:
> >I wish the manufacturers would concentrate on lower noise receivers
> >versus higher power.
>
> I'm not sure they can do much more with non-cryogenically cooled front
> ends. Most are SiGe or GaAs. The front end noise figures aren't
> going to get much better without cooling. With integrated chipsets,
> the minimum sensitivity is typically set by the internal noise pickup,
> not by the NF.
>
> It would also be nice if they would stop lifting the numbers from the
> chipset data sheet and actually test for sensitivity. So far, only
> DLink seems to be supplying real numbers. Everyone else lies:
> <http://802.11junk.com/jeffl/rx-sens/receiver%20sensitivity.htm>
>
> Personally, I'd like to see the RF section seperate from the digital
> stuff, so I can install it high on top of a tower. Alvirion has had
> this for a long time, but it's expensive. Note that this implies a
> receiver front end that can handle some serious overload and not go
> into blocking in high RF environments. Incidentally, that's one
> application where a tower mounted power amp and pre-amp can do some
> good. However, the rx amps in the boxes that I've seen are actually
> worse that what's in the access points.
>
> --
> # Jeff Liebermann 150 Felker St #D Santa Cruz CA 95060
> # 831-336-2558 je...@comix.santa-cruz.ca.us
> #http://802.11junk.com je...@cruzio.com
> #http://www.LearnByDestroying.com AE6KS
The whole intent of SiGe was heavy integration. I don't think there is
enough money in SiGe stand alone amp chips, or perhaps better stated
the money is in the system on a chip. Agreed that the RF on a
different chip than the logic would be better.
Incidentally, I got a 2400Mhz video receiver. There is not a lot of
2400mhz video, but there is enough that I bet some people wonder why
they do their site survey, find a free chanel, then wonder why they
have problems.
>The whole intent of SiGe was heavy integration. I don't think there is
>enough money in SiGe stand alone amp chips, or perhaps better stated
>the money is in the system on a chip.
Probably true for individual transistors, but there's quite a bit of
SiGe in use in the form of simple downconverters and RF front ends.
For example:
<http://www.maxim-ic.com/quick_view2.cfm/qv_pk/1918>
<http://para.maxim-ic.com/results.mvp?q=lna&an_1=Family&av_1=Low-Noise%20Amplifiers>
Most are in the 1.5 to 3.0dB NF range, which is nowhere near the 0.4dB
NF that can be achieved with a properly optimized discrete GaAs FET
design.
>Agreed that the RF on a
>different chip than the logic would be better.
>
>Incidentally, I got a 2400Mhz video receiver. There is not a lot of
>2400mhz video, but there is enough that I bet some people wonder why
>they do their site survey, find a free chanel, then wonder why they
>have problems.
Yep. Lots of security cameras are also wireless. The TV video
"rabbit" links are suppose to only occupy perhaps 12Mhz (double
sideband, not VSB). In theory, they only trash a few channels.
Anyway, that's why I use a spectrum analyzer for dealing with possible
non-802.11 interference.
On Nov 16, 10:38 pm, Jeff Liebermann <je...@cruzio.com> wrote:
> m...@sushi.com hath wroth:
>
> >The whole intent of SiGe was heavy integration. I don't think there is
> >enough money in SiGe stand alone amp chips, or perhaps better stated
> >the money is in the system on a chip.
>
> Probably true for individual transistors, but there's quite a bit of
> SiGe in use in the form of simple downconverters and RF front ends.
> For example:
> <http://www.maxim-ic.com/quick_view2.cfm/qv_pk/1918>
> <http://para.maxim-ic.com/results.mvp?q=lna&an_1=Family&av_1=Low-Noise...>
> Most are in the 1.5 to 3.0dB NF range, which is nowhere near the 0.4dB
> NF that can be achieved with a properly optimized discrete GaAs FET
> design.
>
> >Agreed that the RF on a
> >different chip than the logic would be better.
>
> >Incidentally, I got a 2400Mhz video receiver. There is not a lot of
> >2400mhz video, but there is enough that I bet some people wonder why
> >they do their site survey, find a free chanel, then wonder why they
> >have problems.
>
> Yep. Lots of security cameras are also wireless. The TV video
> "rabbit" links are suppose to only occupy perhaps 12Mhz (double
> sideband, not VSB). In theory, they only trash a few channels.
> Anyway, that's why I use a spectrum analyzer for dealing with possible
> non-802.11 interference.
>
> --
> Jeff Liebermann je...@cruzio.com
> 150 Felker St #D http://www.LearnByDestroying.com
> Santa Cruz CA 95060http://802.11junk.com
> Skype: JeffLiebermann AE6KS 831-336-2558
I worked for Maxim at the time they got their SiGe working. Those
engineers were nearby, so I would visit once in a while to absorb what
I could absorb. [I'm a baseband guy, not RF engineer.] Maxim made
those building blocks because that was all they could do. That doesn't
mean it was a good market. I'm not sure how many of the RF engineers
are even there anymore. My last contact with the group quit for a
start up.
The RF guys had a hell of a time measuing the noise figure of SiGe.
If there was a market, you could make nearly impossible to measure amp
in the VHF band up to say 400MHz. The input stage had to be sized
differently, though I was told you could parallel the building block
amps to do this. The trouble is bipolar amps were pretty good in that
frequency range, so there was no market.
Jeff Liebermann wrote:
> miso@sushi.com hath wroth:
>
>> The whole intent of SiGe was heavy integration. I don't think there is
>> enough money in SiGe stand alone amp chips, or perhaps better stated
>> the money is in the system on a chip.
>
> Probably true for individual transistors, but there's quite a bit of
> SiGe in use in the form of simple downconverters and RF front ends.
> For example:
> <http://www.maxim-ic.com/quick_view2.cfm/qv_pk/1918>
> <http://para.maxim-ic.com/results.mvp?q=lna&an_1=Family&av_1=Low-Noise%20Amplifiers>
> Most are in the 1.5 to 3.0dB NF range, which is nowhere near the 0.4dB
> NF that can be achieved with a properly optimized discrete GaAs FET
> design.
>
On Nov 18, 5:33 am, LR <l...@privacy.net> wrote:
> Jeff Liebermann wrote:
> > m...@sushi.com hath wroth:
>
> >> The whole intent of SiGe was heavy integration. I don't think there is
> >> enough money in SiGe stand alone amp chips, or perhaps better stated
> >> the money is in the system on a chip.
>
> > Probably true for individual transistors, but there's quite a bit of
> > SiGe in use in the form of simple downconverters and RF front ends.
> > For example:
> > <http://www.maxim-ic.com/quick_view2.cfm/qv_pk/1918>
> > <http://para.maxim-ic.com/results.mvp?q=lna&an_1=Family&av_1=Low-Noise...>
> > Most are in the 1.5 to 3.0dB NF range, which is nowhere near the 0.4dB
> > NF that can be achieved with a properly optimized discrete GaAs FET
> > design.
>
> Was reading about these last week. BFU725F'shttp://www.spoerle.com/download/en_1184324718115_E584_NXP_BFU725.pdf
>Jeff Liebermann wrote:
>> miso@sushi.com hath wroth:
>>
>>> The whole intent of SiGe was heavy integration. I don't think there is
>>> enough money in SiGe stand alone amp chips, or perhaps better stated
>>> the money is in the system on a chip.
>>
>> Probably true for individual transistors, but there's quite a bit of
>> SiGe in use in the form of simple downconverters and RF front ends.
>> For example:
>> <http://www.maxim-ic.com/quick_view2.cfm/qv_pk/1918>
>> <http://para.maxim-ic.com/results.mvp?q=lna&an_1=Family&av_1=Low-Noise%20Amplifiers>
>> Most are in the 1.5 to 3.0dB NF range, which is nowhere near the 0.4dB
>> NF that can be achieved with a properly optimized discrete GaAs FET
>> design.
Yep. Much easier to bias and protect SiGe xsistors than to deal with
the tangled mess required for GaAs FET's. 0.4dB NF and 26dB gain at
1.8GHz is very close to what a commodity GaAS FET will do. Thanks for
the pointer. That's the first I've heard of SiGeC.
>The RF guys had a hell of a time measuing the noise figure of SiGe.
Same here. I can setup a measurement twice, and get two substantially
different numbers. At one company, I had to setup a temporary RF
shield room to get fairly consistent numbers. See Fig 3-1 on page 12
at:
<http://cp.literature.agilent.com/litweb/pdf/5952-3706E.pdf>
for a clue.
>If there was a market, you could make nearly impossible to measure amp
>in the VHF band up to say 400MHz. The input stage had to be sized
>differently, though I was told you could parallel the building block
>amps to do this. The trouble is bipolar amps were pretty good in that
>frequency range, so there was no market.
For 150 to 400Mhz, I prefer cheap Dual Gate MOSFET amplifiers. They
have almost the same noise figure as GaAs in the VHF frequency range.
The NF of bipolar is far worse. The big different is in handling
intermod and blocking. The Dual Gate MOSFET can handle much more
signal than either bipolar or GaAs as shown by its much higher 3rd
order intercept specification.
However, at 2.4GHz, things change. The DG MOSFET won't work at much
above 500MHz. Si bipolar is rather noisy above 1GHz. That leaves
GaAs FET and SiGe by default.
Jeff Liebermann wrote:
> LR <lrme@privacy.net> hath wroth:
>
>> Jeff Liebermann wrote:
>>> miso@sushi.com hath wroth:
>>>
>>>> The whole intent of SiGe was heavy integration. I don't think there is
>>>> enough money in SiGe stand alone amp chips, or perhaps better stated
>>>> the money is in the system on a chip.
>>> Probably true for individual transistors, but there's quite a bit of
>>> SiGe in use in the form of simple downconverters and RF front ends.
>>> For example:
>>> <http://www.maxim-ic.com/quick_view2.cfm/qv_pk/1918>
>>> <http://para.maxim-ic.com/results.mvp?q=lna&an_1=Family&av_1=Low-Noise%20Amplifiers>
>>> Most are in the 1.5 to 3.0dB NF range, which is nowhere near the 0.4dB
>>> NF that can be achieved with a properly optimized discrete GaAs FET
>>> design.
>
>> Was reading about these last week. BFU725F's
>> http://www.spoerle.com/download/en_1...NXP_BFU725.pdf
>
> Yep. Much easier to bias and protect SiGe xsistors than to deal with
> the tangled mess required for GaAs FET's. 0.4dB NF and 26dB gain at
> 1.8GHz is very close to what a commodity GaAS FET will do. Thanks for
> the pointer. That's the first I've heard of SiGeC.
>
This is from 2003 but worth a read. http://www.gaasmantech.org/Digests/2003/2003PDF/1-3.pdf
Re: 600mW ECB-8610S now shipping 
Linear Mode
