I'm trying to figure out a way to use 802.11a 5GHz radios in the 10 GHz US HAM band. We're looking to put up a link in an area that is saturated in all of the lower bands and we'd like to stay away from licensed microwave gear and the related licenses, frequency allocations, etc.

My first question is under Dual NStreme, is one radio EXCLUSIVELY TX and the other RX? I mean 1000000% exclusive. If that is the case, I can use a combination of band pass filters, preamps and frequency dividers on the RX side to bring the 10 GHz signal down to 5 GHz. If anyone has a better suggestion how to go about this, I'm all for other ideas since most (affordable) frequency dividers need ~ -10db or higher input power to work correctly.

Second, what would everyone suggest for a method to get a 10 GHz TX signal? I was either going to go with a passive frequency doubler, where I'd end up losing ~15db but when driven with a 27 db radio I should still have end up with the 10mW of power needed to drive a 10 GHz low noise amplifier or an X Band amplifier of some type. My other option was just amplify the second harmonic but that would be a last resort in my opinion...

The other options I was considering were:
http://www.kuhne-electronic.de/en/shop/ ... U_UP_107_A
or
http://www.kuhne-electronic.de/en/shop/ ... U_XN_10_TX

Combined with:
http://www.kuhne-electronic.de/en/shop/ ... _LNC_107_A

I'd have to see if they would change the local oscillator frequency in the KU LNC to get a 2.4 GHz output but the combination of the two converters should provide a 10 GHz link if Dual Nstreme is exclusive.

Let me know what you think.
Thanks

Wow this is extremely interesting, but is probably beyond the scope of 99% of the users here. Just curious as to what you'll use an antenna? A satellite dish with a 10GHz feedhorn? How much does the up and down converter and frequency multiplier cost?

I will be using 2 meter C band dishes converted to an Axial Displaced Ellipse style feed on one end of the system and on the other end I intend to use KS 15676 Horn Reflectors. The horn reflectors should provide somewhere around 45-46 db of gain and I haven't done any real studying of how the 2 meter C band dishes will perform but I would assume on the order of 39-40 db, possibly even higher. The C bands actually might be replaced as well with smaller dishes as there may be no need for such extreme gain, time will tell.

Do you have any insight into if the Nstreme is actually 100% exclusive? Any input would be great. Thanks.

You might get better luck emailing mikrotik directly, support (at) mikrotik.com

Please let us know how your project turns out.

Please tell how much are you available to spend on a pair of 10 ghz radios, chinese manufacture, producer specification say can carry gigabit speeds, for up to 6 km.
I've seen them on a czech distributor, i don't know exactly if it's this or some other:
http://www.i4wifi.cz/?cls=stoitem&stiid=739
You might get better this way, without the complexity, but with the possible higher price tag.

Just for all of those who might be wondering, I emailed support directly and I got a reply stating that the RX radio will never try to TX in a Nstreme 2 system. This opens up a whole world of possibilities for modification of the signal as there is no need to worry about TX power damaging components up stream. If anybody knows of a good place to find microwave mixers I'd be really interested as I think the best scenario will be to use a mixer combined with a fixed local oscillator to adjust the frequencies of both the TX and RX radios on each end. The bigget problem I think that might occur will be drifting of the oscillator but that should be less that 1 khz after warmup so I doubt there will be any issues. I'm actually looking right now into retuning a microwave combining network and getting some waveguide launchers setup to test the link I am looking at first deploying this on.

I haven't got a quote back on the up and down converters yet but I have been in contact with the company and asked about the possibility of modifying the local oscillator frequency in their products so I wouldn't have to come up with my own solution.

I'll keep everyone updated on how this turns out, if it works out well, look for a 36 GHz test after I can make a 10 GHz link.

Cards with 2 ant connectors (main & aux) may be used for Rx and Tx exclusively,

I'm trying to figure out a way to use 802.11a 5GHz radios in the 10 GHz US HAM band. We're looking to put up a link in an area that is saturated in all of the lower bands and we'd like to stay away from licensed microwave gear and the related licenses, frequency allocations, etc.

You'll find that the cost (both in terms of components and time) for licensed microwave gear is far lower than making 802.11a work at 10 GHz (the legality of which is questionable in the first place).

Consider the following. A carrier grade full duplex 38 GHz link running at 150 Mbps costs $2000 (including 2 IDU,2 ODU, and two 1' antennas).
Licensing is probably going to be another $2000/year to guarantee that you never get any noise.

You might get better this way, without the complexity, but with the possible higher price tag.

Like it's been said !
Maybe the man wants to learn and practice.

You'll find that the cost (both in terms of components and time) for licensed microwave gear is far lower than making 802.11a work at 10 GHz (the legality of which is questionable in the first place).

I'm fairly certain he is amateur radio operator which means he has unfettered access to 10.0-10.5 GHz.

You'll find that the cost (both in terms of components and time) for licensed microwave gear is far lower than making 802.11a work at 10 GHz (the legality of which is questionable in the first place).

I'm fairly certain he is amateur radio operator which means he has unfettered access to 10.0-10.5 GHz.

Yes, but there are regulations concerning the use of amateur radio for commercial use and data transmission.

Yes, but there are regulations concerning the use of amateur radio for commercial use and data transmission.

All amateur radio operators are aware of this. It's part of the test.

Just because it is in the test doesn't mean someone will abide by it. People do obviously illegal things all the time. There was just a guy on here in the US running 5 watt 900 amps and he knows it is wrong but just laughs about it...

Just because it is in the test doesn't mean someone will abide by it. People do obviously illegal things all the time. There was just a guy on here in the US running 5 watt 900 amps and he knows it is wrong but just laughs about it...

So what would you guys like to do? Call the FCC and make sure he's following regulations? He's a HAM!

Which from my experience makes it that much worse!

Which from my experience makes it that much worse!

Okay I concede.

Wow guys, ok, so why don't we paint all HAM operators as criminals. Let's see, I would guess most people don't get an Extra class license and retain it for extended periods by violating the law. Not to mention a first class PG license and an RP. You're not talking about the local WISP man who got a HAM license because he thought it allowed him to run 1500 watts of PEP, or the guy who runs an amp just because it "makes the signal go farther". Come on guys, I'm doing this to see if it can be done, and if it can, how easily. There has been talk for quite some time about the FCC opening up another ISM band, it would be nice to see if the current gear can be used at a new frequency if and when that happens, considering the slowness of Mikrotik to adopt new technology/drivers. I'm currently working on a PLL'd YIG oscillator mixed with a 2.4 Ghz radio and getting a 10 GHz signal that almost identically matches the 2.4 GHz signal. I think it is close enough that it should be data stable, I just need to come up with 3 more complete setups and I should be able to test a link.

I am well aware that Part 97 Section 113 states that no transmission for hire, as well as no transmission that could reasonably be provided by another radio service. The link will be a test, with no "for hire" communication, and I'd like to see a 100 mile 25 mbit link with any other radio service. As far as the data transmission comment, actually, not so much... Many HAMs are using Part 15 gear in the Part 97 spectrum for remote control links to repeaters and such, anything that qualifies for Part 15 will qualify for Part 97, it just depends on how much power you are allowed. As long as it's not considered spread spectrum (g and a are exempt), you can run up to 1500 watts PEP with unlimited antenna gain.

Furthermore, this could have other applications as well, the 10.7 GHz microwave band in the US allows up to 40 MHz of bandwidth per license and 40-55 db of TX power. Thus, 802.11 gear would qualify under the bandwidth and tolerances, and if it could be retuned to these frequencies, would be a valuable alternative to costly microwave links. Better yet, the 12.2 to 12.7 GHz range allows for up to 500 MHz of bandwidth under MVDDS classification, which I would imagine wireless data transmission could qualify under since satellite data transmission fits. The possibilities are endless since a, g and n are all ruled non spread spectrum technology by the FCC, they basically just have to comply with the bandwidth requirements and fit the tolerances to be allowed in any microwave band.

A carrier grade full duplex 38 GHz link running at 150 Mbps will run how far? Not to mention the cost, $4000 for the initial year. I can put together the Mikrotik boards, radios, YIGs, mixers, GaAs FET amplifier and everything for less than 1/4th of that. Run it on a HAM license for experimentation and if it is viable, with just a little retuning, file for a 10.7 GHz microwave license.

I'd suggest that just because you have a bad taste in your mouth about people in your area using amplifiers and other items illegally, don't assume that everyone is out to cheat the system. If it's burning at you so bad, contact the FCC. They would be more than happy to come shutdown the illegal transmitter, confiscate his gear, thus resulting in more customers for you. Let's halt progress though, because if someone is to do something outside the box, with the potential to be operated illegally, they must INTEND to operate it illegally or have no idea what they are doing.

A carrier grade full duplex 38 GHz link running at 150 Mbps will run how far? Not to mention the cost, $4000 for the initial year. I can put together the Mikrotik boards, radios, YIGs, mixers, GaAs FET amplifier and everything for less than 1/4th of that.

Just the 4 radios and 4 RouterBoards will cost you $800. You are telling me that you can make these frequency converters for $50 each?
If you do, I'll be happy to buy some.

Where are you getting a $2000 150mbit microwave link? I'd love to see the website, maybe some data sheets, etc.... You might be right, I might not be able to do it for $50 each, but I have one assembled with 20 db tx power right now for less than $200, mixer, YIG, PLL for the oscillator, filters, etc. So I would imagine in some sort of bulk and with some refinement the pricing could be brought down...

Keep in mind this is done with just off the shelf throw it together components, no custom PCB design with components specifically for this. So, $200 for the MT board and radio, $200 for the converter... We're talking $400 for a functional 10-13 GHz transmitter.... This is pointless to argue cost with you about who can do what cheaper.... Even if I could build it for $50, someone in China could do it for $20. My goal isn't to do it as cheap as possible, or just go online and clicking buy buy buy and having a link. Some things are about proving they can be done, trying something new, etc.... If it works, more power to it... If you want, now you can argue the reliability and how I have no support for my product where your commercial link does. This could go on and on, but rest assured, I'm not going to be pointing a 10 GHz link with 5 gigawatts of power all over the world and running commercial traffic on it to boot, there are no questionable deeds here as your fist post suggested.

Seriously thought, send me a link to that microwave gear you are talking about... $2K for an entire link sounds fascinating... How far can it reach? I'm needing to go at least 100 km.

Seriously thought, send me a link to that microwave gear you are talking about... $2K for an entire link sounds fascinating... How far can it reach? I'm needing to go at least 100 km.

It does about 4 miles at 150 Mbps full duplex. As I mentioned, they are 38 GHz and that means attenuation due to rain is high. Manual says annual uptime is 99.99% at 4 miles, you could probably go a further if you don't mind a little downtime when it pours. I use it in urban areas only.

Send me an email at rk @ rontec.net and I can send you the details.

I'm all for trying new things simply for the sake of showing that it can be done.
My interpretation of your posts is, however, that you are actually trying to build a commercially viable device.

My final goal is to go to 24 GHz actually. I was going to shoot for a 10 GHz link first since everything is so much simpler in terms of design. Off the shelf oscillators, mixers and etc can be used to do the frequency conversion and SMA connectors perform fine at 10 GHz. Going to 24 GHz will require a custom designed subharmonic mixer, a very stable oscillator with many multiplier stages, filters will have to be very carefully built and everything will likely end up being waveguide. That's why I went for the 10 GHz target first. Since I have a license and can actually do things in this band, it seems like a logical place to start. If something is able to be made that can actually link, then moving to 24 GHz is just a matter of rebuilding for the higher frequency.

Once an up and down converter were made that would work at 24 GHz, it opens up possibilities for everyone as 24.0 to 24.25 GHz is an ISM band and thus can be used by anyone. 24 GHz falls into an interesting area as well when it comes to attenuation. Significant losses due to high humidity and fog start to happen above 30 GHz and the water vapor resonant frequency is somewhere around 22.5 GHz. At the standard corrected water vapor, 54% relative humidity at 20C, attenuation is approximately 0.175 dB/km for 24.125 GHz. Much more favorable than 39 GHz equipment. Additionally, corrected oxygen loss is about 0.0235 dB/km as compared to 39 GHz of double that and 60 GHz at nearly 15 db/km.

I'll send you an email later today, I'm guessing you're talking about DS3 or OC3 gear though, and while the cost per megabit for the radio equipment is cheap, the routers on each end to go from ethernet to SONET traffic become expensive. An OC3 card for just about any router is at least 300 dollars used, not including the cost of the router... Even though the gear is cheap and all over the internet, the reason it's so cheap is because nobody wants to fight the issues. At least that's my opinion.