I'm a wisp based in the Ottawa-West Quebec area of Canada. Xagyl comunications is my principal supplier of all things wireless from Mikrotik and antenas to Ubiquiti gear.
The owner is an acomplished electronics engineer that was respossible for the foundation of one the oldest and largest Wisp's in the area but left that to persue being a supplier to the industry. Long story short he has been working on his own line of products to solve problems that he hears from his clients (like me) on a day to day basis.
I have previously tested his high powered 2.4 ghz FLR24G30 1000mW 802.11b/g miniPCI card with great success and now use them where ever I would have formerly used an XR2.
I have been having some very exiting results with tests I'm running on his new 900mhz card the FLR9G30 900MHz 1000mW miniPCI.
As I have had to turn to 900mhz like most Wisps to solve the problems of getting into dificult to serve areas .... I feel your pain !
With 5mhz and 20 mhz being unstable and really only 2 channels that have ever effectivly worked the XR9 has got the job done but it has left a lot to be desired.
This card is the real deal, I have been posting over at dslreports about it and I still have a lot more testing to do.
But more than that for me is that card is really working well in everything I have put it through.
I only have one card but I have seen situations time and again where an XR9-AP<->XR9-STA connection fails is and an XR9-AP<->FLR9G30-STA works.
I have seen stable 10 km links in 5mhz channels ! Something I simply can't do at the same location with a pair of XR9's.
The same thing is true for 20mhz, I have established long and stable links with the FLR9G30 at distances where the XR9 just would not associate.
The noise floor on this card is phenomenal I have been doing tests all over my network where I have MOTO gear deployed by a municipal funded giant.
Where the noise floor in the XR9 show in the -70s -80s when I snap in the FLR9G30 I see -105 to -110 !
The SNR goes from being -14 with the XR9 to -58 with the FLR9G30.
The bottom line here is this card shows enormous promise. I will be doing more testing and reporting but I'm really looking forward to having more cards as its clear that just replacing the AP card as a start is going to clear up a lot of my problems especially with the impending bloom of spring leaves.
I will also be looking at testing 2 cards in a board and a 2.4 card in the board along side of it. (yes the other card will be the FLR24G30)
More details are over at the post in http://www.dslreports.com/forum/r239665 ... real-world

I've heard about these, but I was having a hard time finding info.

I will keep an eye out as the story develops...

i wont believe it until you post some screenshots of your claims. those snapshops over at dslreports look backwards, the 00:15:6D mac is a ubnt mac, and it shows better ? Im confused I guess.

I appreciate you pointing that out to me .
I had named the creen caps so that they where self explantory but the naming isn't visible in the post.
I'm going to name them in the image itself to clarify.

i wasnt looking at the names, I was looking at the MAC addresses in the AP client window.

Yes but it's the Mac address of AP XR9 from the FLR930 based client registration table.
or the Registration of the FLR9G30 in the AP-XR9' Registration Table. It was too difficult to tell.
I have edited the post and corected the screen caps to clarify at dslreports but I'll post them here as well.
I can only do 3 at a time here so more coming.

More Screen caps.

For those interested in more information including the actual specsheet.

http://www.xagyl.com/download/flr9g30_datasheet.pdf

The card on the site http://www.xagyl.com/store/product.php? ... 251&page=1

The Site http://www.xagyl.com/store/home.php

Could you post bandwidth screenshots. I have used other 900 MHz equip that while things may look OK, as soon as you start to pass traffic they become unstable and/or the actual bandwidth isn't that great.

Cheers

I too have some of these cards that I was able to test and I can concur that these cards filter noise much better than the XR9s. After replacing the XR9 in one of my APs that had a noise floor of -85 the new floor is now around -105. SNR increased and disconnects became less frequent. I also saw an improvement in bandwidth. Also, I too was able to use 5MHZ and 20 MHZ channel widths successfully where with an XR9 in the AP, only the strongest clients could ever connect. Also of interest, I was able to use all 4 channels unlike the XR9 which only ever works with the 2 lower ones.

I have a new card that I will be installing in the near future and I will do some pre and post installation screen shots to prove the findings.

I can but it won't be for a couple of days. I do have some speedtest caps from the tests but thats not the same as a point to point test from the AP to the client.
I'll put those up shortly but I hope I don't get trampled for doing it. It was really informally for my own info.
We have bad weather heading in and I have to head out about 30kms to the far end of my network.
I usually try and my testing into other reasons to be there.
Hence the slow production of tests I'm usually at the end of my day when I'm there and almost out of usable light.
I was concerned about the same idea, ok the numbers look good but what can it do.
It was complimentary to the link quality, what you would expect.
When I was seeing pthroughput at 1800kbit I got a test of 2.77 mbs down and 2.17mbs up
But I will post the correct tests when I get them done to prove it.
I'm curious did someone else offer a card that could interoperate with the XR9?

Hi GREG3f.
How many clients on your AP?
I'm eager to put this card into prodution but I can't because I'm still walking through tests.

Only 4 on this one, mainly because it was not working well through the poor noise floor... now that I have the new card in and things seem to be working better, I will be able to expand in this area.

What's this card's FCC id? Or is this not for usage in the US?

I don't know I know that the card I have is prototypical.
I will ask for info on this tomorrow.

Edit the answer from the engineer is
"In process.
FCC ID will be XQBFLR9G
IC will be 7503B-FLR9G30"

Here is the speedtest
This was the FLR9G30 at 10 km and 10 mhz in nstreme.
Pthroughput was reported at 5136 in the client at the time not 1800 as I wrote earlier.

Really, this has won my heart. But Daniel of Xagyl told me that RB433 cannot drive three of this cards (FLR24G30) cos of miniPCI slots space and the power drain. I have found a good alternative to Amplifier.

Regards,

How about 3 R52H and one of those in a RB600 w/48v POE?

thank you oldman. i will check that card and check the board as well...xagyl confirm that rb433ah will take 2 of flr24g30 in terms of space and power requirement

Unless this card turns out to be a complete waste, I am going to replace an XR9 as soon as I possibly can.

It looks like the Xagyl card heat sink is too large to have any card on top of it, but what about the back? Does it have a heatsink on the back? If so, how thick?

I have a card here for testing as well, and can confirm what other testers of this pre production card are finding.

The card is large no doubt, larger and thicker than an XR9, and does take its fair share of power, only 2 will fit on a 333/433. No heat sink on the top, but the RF shielding is thick, there is a heat sink on the bottom, its relatively thin, but Im not sure that this is the final heat sink design, as the one I have is crude, also a white 6 pin connector on the bottom that is a little thicker than the heat sink, not sure what this connector is for, maybe to supply the board with additional power on some platforms that cant supply it via PCI.

Will fill 2 x MT minipci R52nM and one in the middle slot FLR24G30?

will it fit and work?


thx

It would be best to put the FLR24G30 on the bottom, because the heatsink would interfere with the bottom card if it were inserted in the middle slot.

Thank you
Microcom Sales Team
Tel: +1 818 880-8008
sales@microcomtec.com
http://www.microcom.us

Follow Microcom on Twitter: http://twitter.com/MicrocomTweet