Was wondering if anyone is using ePMP 425 (5 GHz) for nLOS / NLOS PTP?
If so please describe path, environment, mounting heights, configuration of radios and of course performance results.
Stay Safe.
Hi ,
You can go through F425 performance results in this video.
1 Like
Hello Sreejith
Thank you for providing the ePMP Force 425 performance video.
Unfortunately i did not see that it provide any information regarding the link path other than distance and noise floor.
I am assuming that this was NOT a nLOS / NLOS link.
Please confirm.
Hi Lostin,
Yes- this link has LoS.
Please update post when you have nLOS / NLOS results.
5Ghz is really the wrong choice for nLOS/NLOS. There are other products better suited for that use.
1 Like
You would think but according to LinkPlanner the ePMP Force 425 shows significantly higher throughput than my current 450 PTP 900.
It appears that only a portion of the top portion of the 5GHz FZ is unobstructed.
From a previous LinkPlanner experiment I was close to swapping out the 900MHz link with the ePMP Force 200 when some brought the 425 to my attention.
That being said I donāt have money to burn which is why I am looking for results of the 425 IRL.
Well, itās impossible to quantify what you mean by Non or near LOS. If itās mostly clear, with a scattering of obstructions, then youād really have to try it I guess. But, as Jacob and many others have said - if itās an obstructed nLOS/NLOS, then 5ghz isnāt relaly going to be a good choice.
That being said ā from my acreage, I have a PTP shot with Force200 (5Ghz) at about 6.3 km (4 miles) and itās skimming through the treetops, and itās does -65 signals in the summer (with leaves), and does MCS15/MCS15, and I get about 180 Megabit download.
Result
Downlink 180.987 Mbps
Uplink 163.714 Mbps
Again - itās just just skimming the treetops. When I stand at the antenna, I canāt NOT quite physically see the far side. It I move back and forth, I āthinkā I can see the shape of the building on the far side - so I am just skimming though the treetops, and this is a 5Ghz link which works fine.
BUT, it I moved my antenna up 5ā - 10ā Iād be 100% over all the trees. If you need to āpunch throughā, I wouldnāt put much faith in doing that with 5Ghz. However - you might consider trying a couple Force200 @2.4ghz (depending how much bandwidth you need)? Weāve done some short range, rural, remote (less interference) point to point links through a BUNCH of trees with 2.4ghz, which have worked well.
BUT - the long and short of all this, is itāll be impossible for anyone to quantify how many trees you have, and how āobstructedā your site is. I can sent some photos of my location when Iām back home in a week or two, but your millage may vary. 
Your numbers are impressive. I would be doing backflips if I were able to achieve similar results. Look forward to seeing any photos you have and if you wouldnāt mind included a Google Earth/Map view, omitting your actual address or coordinates. Feel free to PM.
I am located in a rural area very very few homes lots of fields (just not were I need them) and trees (just where I donāt need them). The area has limited 900MHz noise (-95 ~ -87) and even less noise in the 2.4 and 5 GHz range.
PTP link distance 5.00 km (3.11 miles)
At sea level, flat, no hills, no valleys, no bodies of water to cross over.
Unobstructed wide open space from SM to trees 1.01 km (.63 miles)
Unobstructed wide open space from AP to trees 1.59 km (.99 miles)
Forested area (pines) covers about 2.39 km (1.49 miles)
Trees vary in height for 30 to 60 feet with a few open areas in the path.
SM height 65 feet
AP height 45 feet
Preparing to perform maintenance/repair/replacement on the SM side. Because of the difficulty in accessing the radios I am open to moving to 2.4 or 5 GHz Force. Something went a miss a while back and has gotten progressively worse.
LinkPlanner indicates that my results, if I were to use the 5 GHz version of the ePMP Force 200 or 425, would be better than my existing 900 MHz link and would perform better than the 2.4 version of the ePMP.
Could be - LinkPlanner(S) always have a tricky time predicting treesā¦ the difference in height plus/minus a few feet can make all the difference in the world. Also trees which are dry, or leafless (winter), or which are filled with snow, or wet leaves when it rainsā¦ all block a dramatically different amount of signal.
BUT, 5Ghz has a smaller (tighter) Fresnel zone too ā so if Linkplanner does think there is a clear enough LOS, then maybe the tight/small Fresnel zone will keep the signal out of the trees. One would have to experiment to really know for sure.
Iāll send picture when Iām back home in a couple weeks. FEEL FREE TO BUMP ME, if I forget to take pictures for you. 
@ninedd Are you still a person of leisure traveling the world?
Looking forward to seeing those pics and any other info regarding your link.
No no - Iāve been back, buried in work. I forgot to take photos. Iāll see when I can have a chance to do itā¦ I got a LOT of balls in the air right now. We had a bunch of solar panels and batteries and generators and chargers stolen, so Iāve got to rebuild a whole solar site here. Iāll take pictures when I get an opportunity too. 
Iāll see about flying my drone up near the antenna, and see if thatās good enough photos to show how much āline of sightā I have around our trees.
Generally speaking - 5Ghz can tolerate āZEROā trees. Thatās our guidelines to our installers. There are always exceptions, and sometimes skimming slightly through a couple tree tops is possible, if they are thin and far away. But one (1) tree can totally kill a 5Ghz link.
SO - if youāre having problems linking with 900 Mhz equipment because of trees or earth obstructions, then itās almost a certainty that 5Ghz wonāt work from the same locations.
Anyway - my lower back is acting up, but Iāll see about dome drone photos.
Hi. Never did get any pics from you. Still interested is seeing your path.
Weāre testing ePMP 4500 for nNLOS and itās working very well. We have 1.8mi urban link with multiple leafy and needle trees in the way. ePMP 4500 (36dBm EIRP) + 4525SM (50ishdBm EIRP), 20MHz channel, high noise environment, RSSI of -86dBm, around 20mbps up and down.
Prior to this we tested:
- PMP450 900MHz 7MHz 17dBi yagis on both ends, max TX power, aggregate data rate of around 25-30mbps
- cnRanger LTE 2.5GHz licensed w/Tyndal 201 SM, 20MHz channel, around 40mbps aggregate.
- PMP450m 3GHz CBRS & 450b HG, 40MHz 51dBm EIRP, around 80-90mbps aggregate
So, that being said, if you were to take 2x ePMP 400c radios and paired them with, letās say, an RF Elements 27dBi ultradish, (23dBm + 27dBi = 50dBm EIRP) you could probably burn through some trees pretty well, based on the above figures, and clean spectrum, Iād guess youād probably get around 50+ mbps (up/down) bandwidthā¦ possibly more if youāre able to use a clean 40MHz channel.
1 Like
Need to get you a new dictionary if you think that is urban. 
But then again it looks like a metropolis compared to where I live.
Itās unfortunate that you had to replace the 3 GHz link but the 4500 seem impressive considering the received power levels.
I will consult LinkPlanner and see what it predicts.
Trees, wind, difficulty accessing radios, birds and the occasional misbehaving smart meter are the demons I struggle with.
Iām still using the 450 3GHz/CBRS linkā¦ and thatās something you might want to consider as wellā¦ a 450b high gain 3GHz PtP allows for EIRP up to 49dBm, the band is usually very quiet in rural areas and because itās lower down in the band, will have slightly better nNLOS propagation. That being said, the ePMP 4k series leverages OFDMA, which has a whole raft of improvements that help it deal with noise and nNLOS links, so that might cancel out any benefits that 3GHz/CBRS has.
We have a couple NLOS shots that were using Motorola PTP500 radios for several years with low capacities. Switching them to Force 400C radios have been a massive improvement. Iāve mostly been using them with the 2ā RF Elements Ultradish with the ePMP 2000 twistport adapter.
Part of this one was probably due to going from the 5.8 band to the 5.1, but even on the same Frequency I saw a 6-7x improvement.
Another one was getting around 10-20 Mbps aggregate at a -76 to -79 depending on time of year, the Linkplanner is this. 4ft dish on the larger tower and a 2ft on the small one:
Test result:
2 Likes