A5-14 in urban/city site


#1

I would like to know if any of you are running micro pops in the city with the A5-14? If so are they all on the same channel? Do they work with all the noise from the indoor APs or outdoor ptmp?
I have built a large ring throughout the city using 24, 60 and 80GHz equipment, I would like to start setting up micro pops at each hop. However I don’t want to kill my ptmp 5GHz ubnt and cambium system. The plan is to move as many customers as possible to the micro pops so we can increase their speeds to 500Mbps.
I have just received some ptmp test units from Ignitenet that will do the speeds I need but, I would need to install a lot of the panels to get 360 coverage. The A5-14 is one small easy to install unit.
Any helpful thoughts are welcome.


#2

I have been reading up a lot on the Mimosa’s including blogs and this forum but although some claim they have selling plans going up to 250Mbps I have never seen or heard any trying to deliver 500Mbps to its wireless clients.

I have been looking to the Ignitenet 60Ghz PtMP solution as well but to get a full 360 degrees sector you need 12 sectors and therefore overlap in channels plus the 60Ghz CPE’s.
I hardly can imagine a business plan that can find that many customers for such an expensive tower to give a reasonable ROI.
Secondly, who wants 500Mbps?

But imho the swap from ubnt/Cambium to Mimosa without changing all CPE’s means probably you typed one “0” too much and you are going to sell 50Mbps?
In that case you’ll have no issue. I’ll do that with Mikrotik clients, 60 per A5-14 in a very congested spectrum and 50Mbps is not a problem per client. Total throughput over the AP I can push to some 100+Mbps in a 40Mhz channel.
Tried 80Mhz wide once (which means an overlap on a neighbouring AP) and could go to 160-180Mbps.


#3

Oh, by the way: Don’t expect the channel reuse option to work in the 3rd party modus. This needs full GPS sync of all towers and all CPE’s which is only possible if all network is Mimosa.


#4

So basicly you are telling me there is no reason to use the A514 over any other brand, like Ruckus with the beam forming antennas, if we are not using all Mimosa clients? And if we do use all Mimosa clients I can’t get more than 50 to 100Mbps to the client per AP?


#5

I suggest you study a bit what is written by both Mimosa, other vendors and users in this forum.

  1. Ruckus is nice, but only for very short range. Ideal for conference halls and stadiums but useless if you’d go beyond 300/400 meters. I tried them believe me, for bigger ranges they are useless.

  2. Presently there is no hardware supplier on the market that can deliver 500Mbps to 10 or more wireless clients. Ignitenet comes a long way in their 60Ghz P2MP setup but against a though price ticket and also limited range. (1.5km’s max). LTE or other solutions are even more costly hardware plus licenses on top and still don’t expect too much of it…

  3. Any brand that works with 3rd party CPE’s has to work in 802.11 standard protocol or replace the firmware like eCambium is doing. (Reading a lot of horror stories about their solution plus it comes yet again with a price ticket. (license) and only available for ‘n’ type of ubiquity client plus you’d loose all support from ubnt and upgrades.

Because every brand uses its own ‘proprietary TDMA’ protocol it means you can’t mix with other makes unless you go back to legacy 802.11 CSMA protocol.

  1. Mimosa is the first make that has ‘in the field’ showcases where a reasonable amount of clients can be wireless connected with 200Mbps+ plans. Believe me, NOBODY else can do this up to this moment… But you have to replace ALL units.

  2. Mimosa is very good in working with 3rd party CPE’s. I run 60 Mikrotik CPE’s on one A5-14 and everybody can have 50Mbps plan. Don’t expect that if many use it the plan holds… but this is already 200% or more then what any Ubiquity or Mikrotik AP can achieve…

I suggest you first read in before you make a decision and learn about all the differences.


#7

Hello @Rudy, I’m sorry we miss at MUM in Milan.

Anyway I always would like to ask something:

  1. You are running A5-14 into 802.11 mode. What are the distances on all customers ?
  2. Can you share your spectrum if you haven’t just done elsewhere in this forum ?
  3. What are the settings on Mikrotik CPE to best performance ?
  4. You hint to run A5-14 into CSMA mode instead of GPS considering the high Mimosa’s CPE prices ?

#8
  1. From 50 meters to 400 meters max.
    2). 2 Scans attached. One with 5550 as centre frequency and one with 5710 as centre frequency. I have done this to show that where the 5550 channel use also show high noise level, when 5710 selected this high level is not there. This is because when Mimosa AP works with only Mimosa C5 clients, their signals are suppressed to get a ‘clean’ noise reading of the scan. Whereas with all 3rd party CPE’s this suppression doesn’t work and thus all these CPE’s wanting to ‘talk’ to the AP are showing a high noise at its working frequency in the AP.

3). RTS/CTS always ‘on’ (Very important!), long preamble.

Code:
/interface wireless
set [ find default-name=wlan2 ] adaptive-noise-immunity=client-mode band=5ghz-a/n/ac channel-width=20/40/80mhz-Ceee country=spain
default-authentication=no disabled=no frequency-mode=superchannel hw-protection-mode=rts-cts hw-retries=15 name=wlan1
nv2-preshared-key=***************** nv2-security=enabled preamble-mode=long radio-name=B-013 scan-list=5400-5850 ssid="" wmm-support=e
/interface wireless connect-list
add comment=“Connect to MC_AP-Ba (Mimosa)” interface=wlan1 security-profile=Mimosa ssid=MC_AP-Ba wireless-protocol=802.11
add comment=“Connect to MC_AP-Bb (Omnitik)” interface=wlan1 security-profile=default ssid=MC_AP-Bb wireless-protocol=tdma

4). First my network was like many full tdma running (Mikrotik = NV2) with OmniTiks or Netmetals or SXT-ac’s as APs.
In using this A5-14 of Mimosa we had to abandon tdma since the Mimosa ‘interop’ mode means ‘802.11’ default protocol. I have one other Mikrotik AP (NetMetal) that I prepared for Mimosa and all clients can connect to a ‘802.11’ AP too. I can swap between Mikrotik NV2 and 802.11an. This AP has 30-34 associated clients but in 802.11 mode it becomes very unstable with many disconnects. This has always been the experience with Mikrotik.

On our A5-14 the 802.11 protocol works much, much better. This AP serves now 60-64 clients that before were connected to two separate OmniTiks in tdma mode. The OmniTiks worked in 20Mhz wide channel where the A5-14 now is set to work with 40Mhz channel. We tried also 80Mhz and that worked fine too but we have no real need for that…

What we saw is on the A5-14 we can now serve double as much clients on one AP as before with Mikrotik AP and still per client we can have higher throughput and the total AP throughput easily reaches 100Mb or more where an Omnitik (or even the Netmetal in the other location) will never be able to reach that. Omnitiks max at around 50-60Mb with 30 clients where the Netmetal reaches about 80-90 with 30 clients.
The A5 I pushed once (in 80Mhz mode) with bandwidth tests running from several SXT’s to one central Mikrotik CCR to 190Mbps download and 30Mbps upload. So over 200Mbps aggregated! In a very ‘dirty’ spectrum.

This is more then a 200% improvement compared to a full Mikrotik environment. And only by investing in one Mimosa AP!

Now off course we can go even further. Replace all 64 clients for C5’s and wait for the GPS firmware that support more then the present 30 associated clients and we will probably have the same results as some on this forum/blog are claiming; 200-400Mbp throughput to the client with almost a gig aggregated over the AP!

This has to come to the expense of 64 times € 115,- meaning over 7K, a heavy investment if you’d ask me. Especial when it will give me speeds nobody wants to pay for. So this extra investment cannot be justified economically for my business. The more since we already saw an 200-250% improvement anyway in only replacing the AP.

Last week Mimosa gave me a new beta firmware to test since we’ve had occasional ‘network storms’. Since the upgrade we’ve not seen that anymore. So the network is pretty stable now with always more then 55 clients associated.
I must mention though that the average traffic these clients in general is not very high. If I see ‘peaks’ of 50Mbps I would call that “excessively busy”!


#9

Great post. Really useful to me.

I need ask even what you used for Backhaul and what is the PTP link distance. If you think about MicroPOP’s architecture, you will need a PTP for each pop.

When you tell me you use MK CPE, you refer to SXT AC version ?


#10

We still have a lot of plain 5Ghz backhauls (All Mikrotik) but are slowly replacing the most important ones for either Mimosa B5’s or Ubnt Airfibre 5X (they seem to be slightly better then the B5’s, and a bit cheaper.)
And wherever we can (short distance) we make 60Ghz backhauls (Metrolinq).

The particular Mimosa A5 is indeed behind a 60Ghz backhaul that can transport almost a Gb symmetric aggregated traffic. The latency is very low and stable (0-1ms with a very occasional 2ms) and only when we really max it out (in a test scenario) the ping times becoming more erratic and run up.

We are indeed working towards a sort of ‘micro-pop’ scenario but still in ‘interop’ 802.11 mode. Our heart of our network is here in an urbanisation containing some hundreds of houses in only halve a km. radius. We have some 200+ clients here.
We serve(d) them with 6 Mikrotik AP’s and 5 links towards them. We also have 4 backhauls leaving from this estate to serve surrounding towns and one AP that serves distant clients.
7x20Mhz (AP’s) + 4x40Mhz (links to serve 4 AP’s) + 4 x 40Mhz (backhauls leaving) means 1000Mhz of bandwidth needed… We all know we only have 300-350Mhz available…
(That’s without the competition pointing towards us, or fitting their AP’s in the same zone. We write INTERFERENCE in capitals here! Finding relative good frequencies is always a big issue for us. I am always looking for the best, but economic, radio’s and antennas for this reason.)

We have now 3 x 60Ghz links running to free 3x40Mhz in the 5Ghz band. The back hauls leaving this urbanisation are placed as much as possible on the edge of the estate (we are in the middle of the countryside). We consistently work with high gain 45º slant setup disk antennas with noise shields for these links, even to bridge only some km’s. Just to stay out of interference’s and to avoid the signals towards these are not molesting too much my AP’s in this area.
Still its a big channel puzzle to get it done at times. We definitely have overlaps and definitely double use of some frequencies…
[Somebody should develop a sort of matrix calculator that can calculate many radio’s reach and frequency use at once to find out what the best channel scenario would be in a certain point. We use spreadsheets to get the overview and our brains to find the solutions. A tool (radio mobile or any other all are designed for one radio, or maybe two, at best.) that could handle 40 radios that can all ‘see’ each other.+]

In regard of the Mikrotik CPE’s we came from a position where they were all ‘n’ type of the SXT-Lite units but now 80% of these connected to the Mimosa AP are ‘ac’ units. Although they cost a bit more they come with the benefit of the faster CPU (more cycles available to faster handle radio traffic and the decoding of the now in use encryption. We never used encryption in tdma, but now in 802.11ac we need off course) and the fact they can work in the 80Mhz band too if needed. Off course they are still only some 30-40% of the cost of a Mimosa C5.

Within a month we are planning to change two more of the NetMetal AP’s in our area for Mimosa AP’s.
Still all working in ‘interop’ mode… we simply cannot afford to use all our budget for the next half year to replace all 150+ SXT’s at once into C5’s. We would have no money left for the rest of my network that also needs investments in upgrade and up-scaling.
Plus the fact Mimosa AP can up to today only serve maximal 30 C5’s in GPS sync mode…not good enough.

We will have to be creative in setting the 40Mhz bandwidth channels for these new Mimosa units in relation to all the others. The idea is that although the secondary channel can have some overlap with another radio, the main 20Mhz channel will have to be only used once in this area. The ‘ac’ protocol is supposed to handle this much better than the ‘n’ protocol can. We could even extend this idea to 80Mhz wide channel to get higher speeds to those clients that can work in 80Mhz. (It’frees’ the channel faster, and thus benefits other radio’s too.)

And we still have two small fine adjustment tools at hand:
RF elements has brackets for the SXT’s so we can give them a 45º slant fitting. Like the C5’s. This will eliminate a bit interference from other H+V pol radio’s. It won’t be a lot but all bits help…
We can buy these noise shields for the SXTs. That will help another little bit.

The full Mimosa GPS sync setup will be a dream for us not soon to be materialised. It just costs too much.

Last remark in respect of the back-hauls we’d use:
We’ve had an early look into some very interesting new products of IgniteNet’s Metrolinq’s. They are improving their PTP units but also are working hard on a triple radio PtMP AP with 2,5Gb combined throughput!. They work with a new chipset to (like Mimosa, but another one) and I hope to become one of their first European beta testers for these Omni’s. I am specially interested in how the 802.11ac radio would perform.

If that would be in the same range as Mimosa it will open interesting ‘micro pop’ developments.
You could for instance build a 60Ghz backbone network where the radio could even support some high capacity demanding clients (they need a 60Ghz CPE) and where the main Omni as the centre of a star network can serve several 60Ghz bachauls to end points where other 60/5/2,4Ghz AP’s can serve the surroundings and yet again another backhaul etc.
These Omnis with their 2+ Giga throughput (and 1+ giga wireless throughput) can then easily deal with my 500 to 1Gb aggregated client traffic for the next years to come… and we free a lot of 5Ghz frequencies with an ample need to buy new CPE’s (only 60Ghz units for some clients and the back-hauls).

I am also waiting for these 60Ghz Mikrotik devices to hit the market. If what I’d expect they will be very economic it would become easy to make many high capacity short backhauls to reach every corner in a urban region without any interference! How nice will that be? Make a ‘fibre alike’ wireless backbone structure and many Mimosa AP’s on top of it for the micro cells. The future is bright! :slight_smile:


#11

@Rudy, I would even ask you if you never tried Cambium ePMP.

Actually I use ePMP 1000 and 2000. I’m curious about your think about that and experience.


#12

@Rudy do you know some roadmap and availability about that new 60 Ghz PtMP Omni’s from Igninet ?


#13

Well, two years ago when I started to look around for some alternative Wifi products that were affordable and could bring some improvement over my existing full Mikrotik (MT) networks. I picked both Mimosa and eCambium (eC) as interested alternatives to MT
Initially I was a bit in favour over eC since after all, coming from Motorola backgrounds they should have lots more radio technology in their blood that e newcomer like Mimosa.

Soon though I found out that in these days going for eC meant replacing your whole network (AP and CPE’s) which would make it too expensive for me and at the same time still only based upon ‘n’ protocol the performance gain over an existing fine tuned MT’s tdma network could not be phenomenal and thus hard to justify. The usual very enthusiast presentations of any product line taken with a bit of scepticism (this counts for ANY new product line of ANY brand) wouldn’t put it very much ahead over my present networks.

So, if I then had to go for something new and put my money to the test it better would be a product line that from day one had more to offer then eC could.
What did Mimosa promise/have that eC didn’t/couldn’t?

  1. ‘ac’ protocol
  2. complete new chip-set (Quantenna) that according some online reviews promised some real improvement over existing wifi chip-sets.
  3. ‘interop’ protocol. I could test the first AP without the need to change my whole PtMP network overnight and thus safe considerably on investment.

So even although I have been given 2 ePMP1000’s for a test we never went for it. I decided to go for Mimosa. And I hoped to have the first Mimosa AP running in Europe or at least Spain. (According supply chain I have the first Spanish A5-14 and since nobody else in Europe could supply me with one in these days I tend to believe it is indeed the first European unit…)

In the mean time we ran into some ‘child diseases’ of the Mimosa fws (that now seems to be solved in the latest fw.) and several delays in the availability since 2014.

eC in the mean time came up with their ePMP Elevate solution. This could be a solution compatible to what Mimosa is doing with ‘interop’ but with the difference they use a firmware import on 3rd part devices so that device will work with eCambium AP and its tdma and GPS sync.

There are still some caveats though:

  1. No firmware for Mikrotik devices yet available. Promised ‘soon’ since the solution has been offered, it is still not there. And yet again last week it was promised ‘next month’. Well, lets see… Mimosa in ‘interop mode’ works with any 802.11a/n/c system and thus in my MT enviroment…
  2. License needed. Meaning even if you are not replacing CPE’s you still have to pay something for each… Mimosa doesn’t need that so only the purchase of an A5 will give you a start. (And so far works outstanding! In ‘interop’ mode!)
  3. eCambium is still based on the ‘n’ protocol. It’s been said again (like a year ago) that the ‘ac’ product line would come 'later this year… Mimosa is fully ‘ac’ compliant. And although MT is also full ‘ac’ compliant the Mimosa makes definitely better use of it then I’ve seen with MT.

I still think eCambium with the big, experienced and resources rich, company behind them has a pro in just that. But imho they are not doing enough yet to ‘win’ lost market share they lost in the first decade of this century to Ubiquity and the lot.
And Mimosa on the other hand should not delay too much more their improvements (In sync mode only can have a very limited CPE’s attached yet. That HAS to be improved! The GUI could definitely use some improvements) and not only focus of the niche market of ‘new WISPS in urban developments’ or they will miss the advance.
eCambium or Ubiquity will follow or take over the lead again soon…
If Mimosa wants to make big scale market penetrations they have to do something to their C5’ price levels too. 130Usd of a relative simple CPE product it not cheap…

Hence eCambium product haven’t been tried… ‘yet’. “Yet” because in the last workshop we won a raffle consisting out of a kit ePMP2000 AP + Smart Antenna for the beamforming and the ePMP sector antenna kit and two CPE units.
Together with the 2 PMP1000’s we have we plan now to replace a small AP that we presently run somewhere with only 14 or so clients.
We could start testing their stuff on that network. The clients are presently Mikrotik but I do have some ubiquity units on the shelf and could buy a couple just for the test purpose if the MT firmware doesn’t come soon.

At the same time next week I expect 4 new Mimosa AP’s and we ordered some MTI 4x4 Mimo duo slant beam forming and a Wireless Instruments similar product antenna to do some tests with an A5c and beamforming.
So we should get some interesting results and experiences over the summer. :slight_smile:


#14

No, I have no roadmap. I’ve seen some brochure and got some details but when I asked about availability the reply was that they are still under development/test.


#15

Need to re think this too long post…


#16

Permit me to say that I follow Mimosa on web from first development and announcement, and I have waited about 1 Y and 6 months to see first A5. Don’t speak about GPS, I don’t know how many times I waited for.

I can understood difficult to develop, test, rewrite code, re-test, on field test, recode, etc, but even Mimosa on that side, is really slow to release products. The demonstration is we are all waiting for C5c yet, and we don’t know how much time is needed more.

I’m using ePMP 1000, ePMP 2000 now, and I can tell you is really good product, but surely is improvable. Firstly the spectrum scanner features is a nightmare. It’s a pity that Spectrum Scanner is one of most useful feature we have to troubleshoot. In a really infested scenario, ePMP 2000 seems to work really good, what I guess I can’t say of Mimosa. To do same thing with Mimosa you need drastically reduce the distance of customer’s CPE. With Cambium with -74/-77 signal at about 10Km I could see 20 Mbps in download and 8 Mbps in upload. The choose of 802.11n is not the best, but the GPS is really working and it really help.

Coming to price:
GIGAPOPs

  • ePMP 1000 GPS = 360 Euro + antenna 90° 200 Euro = 560 Euro x 4 sectors => 2240 Euro
  • ePMP 2000 GPS = 460 Euro + antenna 90/120° 200 Euro + Smart Antenna 350 Euro (too much) = 1010 Euro x 4 sectors => 4000 Euro
  • eForce 180 CPE 75 Euro (18dB useless), eForce 200 CPE (25 dB dish) 105 Euro
    MICROPOP
  • ePMP 1000 GPS = 360 Euro + KP 360° 250 Euro = 610 Euro

GIGAPOP

  • Mimosa A5c GPS = 730 Euro + antenna 65°(!!!)/90° from 250 to 500 Euro => from 900 to 1200 Euro x 4 sectors or more. Minimum 5000 Euro for single tower
    Here you can’t build GIGApop cause CPE connectorized missing so now it’s impracticable.
    MICROPOP
  • Mimosa A5 GPS = 850 Euro => only 44 cpe connected actually with SRS firmware
  • Mimosa C5 20 dB (too poor) = 100 Euro (!!!)
    If you are Rockfeller is not a problem, but if you are a small WISP, it’s a big problem considering some users once disconnected never give back antenna.

The only possible choice seems to use A5 in microPOP avoiding GPS SRS firmware and use Mikrotik or Ubnt CPE. In this case SXT 5 ac cost 49 Euro, and Ubnt is near to that price.

Or other choice is to use ePMP 1000 + Omni antenna and give a pinch to the belly spending 75 Euro each customers. Considering to give 20/5 average profile, Cambium is a great choice too, considering GPS, about 200 CPE maximum (80-90 really), and no other limitation or problem I can see.


#17

Like said I agree on the slow availability of promised products by Mimosa. Two years ago when de ePMP1000 was presented to us in Madrid they already nodded ‘ac’ was ‘on route’ but now 2 years later still not there….

And yes, they surprised us with the ‘elevate’ solution in the meantime but yet again, promised firmware for Mikrotik is still nothing more than just that, a promise….

Mikrotik has shown in the past (and yet again in Milan!) that hardware solutions presented can sometimes take almost a year before they are really available. And If heard on the forum Ubiquity is not so much different. So I’ll guess it is an industry wide theme…. Presenting to be the first with something new but delaying the real delivery to overcome all the setbacks in the production lines and real implementations…

And I do indeed believe that eCambium delivers a good product. The ePMP2000 has some features that are really interesting (like the channel filter).

And comparing different networks is always a bit of a hassle since one has this, the other has that, etc. etc.

One major factor though in migrating smaller WISP networks is the always imminent lack of resources while at the same time trying to keep the resources rich raiders out of your backgarden. So deliver the best quality but try to be clever in the economics….

Hence in regard to the prices some remarks:

You have to do some shopping. Where Cambium has a strict sales policies with protected dealer network, so price are pretty fixed, Mimosa is a bit more easy to bargain for. I found suppliers in Europe that give me some 15-20% better prices than the national supplier. And since I am a profound European :wink: I see no reason why I should not shop somewhere else if I can save me some bucks….

Based upon what I pay:

GIGAPOP

  • Mimosa A5c GPS = 570 Euro + antenna 70°/100° = 100€ (carrier class RF elements) => 2x570 + 4x100= TTL 1540€ Or 3xA5c + 6x100= 2.310€ That’s 2.500€ for a full blown 6 sector 150+ clients serving tower. That’s ‘interop’ mode still……

    With 3 A5c’s and 6 sectors and some 150 clients costing me 75€ each (mix of LHG’s, LDF5’s and Dynadish) I have to spend 13.700,- say 14.000€ for a full 150 clients serving tower and everybody should be able to get peak downloads up to 30-40Mbps. This option uses 3 x 40Mhz + 5Mhz band protection = 135Mhz of spectrum.

With eCambium I have three options to do sort of the same:

A. Either use the ePMP 2000 GPS + Smart Antenna option for 4K and with 150 eForce 200 clients I end up spending 19.750.- say 20.000€ This is 6.000€ more but ok, fully sync and 2x40 + 15 = 95Mhz of spectrum use.
B. Use same AP setup but convert existing Mikrotik or Ubnt CPE’s with ePMP elevate software. (25€/ud). That would make 4K + 150x85€ (mix) = 16.750€ say 17.000€. Still some 3K more than the Mimosa option.
C. Instead of the ePMP 2000 use the 1000 but the rest stays the same; You’d save some 1.000€ compared to option A or B. Still more than the Mimosa in ‘interop’ mode.

This is all if you’d compare Mimosa with eCambium and you need to setup a new tower with all new clients. Yes, I agree that on the moment due the lack of a C5c device this is not yet even possible with Mimosa in their full GPS syncd tdma setup.

But as I showed, in ‘interop’ mode this is possible and as I have seen myself, ‘interop’ mode doesn’t necessary mean “poor performance”.

When Mimosa would come in the market with C5c’s, it will be a costly system to fully use in its frequency re-use setup but the capacity will probably outdo eCambium 2 or 3 times…. (But who needs a 200-300Mb plan?)

[I am going to write them their C5/C5c price policy really needs to be overlooked.]

Many of us probably are not thinking of setting up new towers to serve new clients. Most of us are looking to upgrade existing towers to serve the same or more clients from one AP and at the same time crank up our consumer’s plans.

Now the differences are a bit more clear:

With a Mimosa tower as prescribed above you spend 2.5 thousand euros and the clients basically can stay the same (given signals are already good enough.)

To upgrade this tower to eCambium usage the APs costs 4K (or 3K) PLUS 150 elevate licenses makes the investment still 7.750€ (or 1K less).

That’s a difference of magnitude of 5K!

Ok, we have a different network now:

Mimosa is 802.11 csma with RTC/CTS and depending on the clients ‘n’ or ‘ac’.

eCambium is 802.11’n’ tdma with GPS sync so you’d save some spectrum and it comes with beamforming.

If you want the beamforming option to work on the Mimosa also then you have to put 6 x A5c’s plus 6 times MTI 65º 4x4 Mimo 2x dual slant antenna which than will cost 5K€ combined. That is still more than 2.5 thousand euros cheaper than the eCambium option but now no doubt you could serve 6x40=240clients on the tower, still all in interop mode!

As you can see, I have strong confidence in the Mimosa range of product but you have to be a bit smart to find the best and most economical solution. And not every network is the same.

At the same time we still need to see big scale real world developments with both product ranges and now Ignitenet is coming with some new micropop solutions things are getting more interesting by the day (….well, month…! :blush: )

Rudy


#18

Where you found RF Carrier Class antenna at 100 euro ??? I found more than 200 euro each


#19

https://www.landashop.com/rf-elements-sec-cc-5-17.html


#20

it’s 2x2 how you connect them ?

360 degrees / 100 = 3,6 rounded to 4 ? And you don’t loose MIMO performance ?


#21

I am not using the A5c yet. But its documented in the forum as well in their docs that the 4x4 can be connected up to two 2x2 sectors and thus server a twice as wide sector. So yes, you should be able to have two A5c’s connected to 2 x 2 100 degree sectors that all have a slight overlap to cover 360 degrees. Off course the 100 degrees are actually only 70 when -3dB margin is taken so if still the full signal is needed in the 360 degrees you can add two more sectors (or some RF elements cone antennas) to reach distant units. Off course with 3 A5c’s your total capacity of the tower also runs up….

But this is all from reading and it makes sense. I haven’t done it myself yet.