Hi all, To exclude any other interpretation of external radio noise, which is given in the ITU-R P.372-14 (2019-2020), we can refer to See explanation 1 on page 74 of this document: 'For man-made noise this Recommendation gives the external noise figure. That is, the component of the noise which has a Gaussian distribution. Man-made noise often has an impulsive component and this may be important in affecting the performance of radio systems and networks." Yes, for the VE7BQH Antenna Table, the recommended values of the Gaussian distribution of the man-made noise component were taken. These values are not something constant or strictly regulated. Therefore, an interactive mode was introduced into the table. The user of the table can set any values of the noise temperature (the lowest value corresponds to 1 K). By the way, in the table (by default) for Tsky you can see the minimum possible value of the noise temperature averaged over the entire sky due to galactic noise for a short vertical antenna. For example, for 144.1 MHz, you can see the default 290 K. This value is also not something constant or strictly regulated. Therefore, you can use your own considerations or refer to a more detailed analysis of statistics on galactic noise. This can be foundational work by Blake or others. The work of IDA (1998) is of interest: 'IDA to perform an independent estimate of effective galactic noise temperature. This estimate is based on the most recent and carefully calibrated astronomical data, which were taken at 408 MHz. The effective temperature is defined by IDA to be the mean plus one-sigma value of the temperature of an ideal lossless antenna having a 30-deg beamwidth which is scanned over the entire celestial sphere. This is equivalent to assuming an antenna at a random position on the earth pointed randomly at the sky. If the operating conditions can be constrained to keep the galactic center below the local horizon, then the noise temperature would be significantly less.' I made a small calculator per IDA conception. Look at the attached xls file. Hope you find this calculator helpful. Vladimir. 31 марта 2021, 07:33:24, от "Edward R Cole via Moon-net" &lt;moon-net(a)mailman.pe1itr.com&gt;om>: Earl, One reason I am putting up a shorter tower is to shield the eme array from local noise.  At 52 feet for the center of my array I could see unobstructed horizon 360 degrees of azimuth.  That was great for gaining max Moon time.  But exposed the antennas to a wider circle of ground noise sources.. I am going back up with a shorter tower that will place array center at 32 feet.  Tree tops are about 45-feet.  With trees at 56 feet away at MR, Moon is an elevation angle of 13 degrees when it clears tree top (angle = arcsin(45-32)/56).  At 144 the trees are not opaque to RF but will add some attenuation to the signal.  Thermal noise of the trees is not significant at 144 due to present total noise level.  Note: VE7BQH rates 144 noise as 1600K in rural areas and 5400k in residential areas (I interpret that to be suburban area noise) in his antenna comparison charts.  For city noise 14,500K is used.  The old 144 Tsky=290K is long gone in today's environments.  So when 144 is quiet (like the old days) I figure 1600K so adding 150K tree temperature is somewhat significant.  But at 5400K not at all.  At MR the upper yagis are looking at clear sky and only the lower antennas looking thru trees (so drop thermal effect from 290K to 150K).  A 0.6 dBNF preamp is about 56K.  Super low-noise preamps for 144 don't really improve SNR anymore. Guess I will know more the effect of lowering my tower after it is up and running.  But I also will have changed cable type used for phasing lines so it will be a combination of effects. Not going to change antennas (M2 XP20) .  Rebuilding tower with new concrete base and boom truck service will cost me a lot.  Repair of the one XP20 will be about $200 vs $456 for a new antenna (I paid about $212 in 1998).  14 by 40 foot new gravel driveway to tower site for heavy trucks to service the new tower, included (currently lawn). One advantage of the lower tower is I can use a man-lift for maint.  My Hazer will not be re-installed. 73, Ed - KL7UW At 02:14 PM 3/30/2021, Earl Shaffer via Moon-net wrote: Hi All. Just to clarify a few things on this subject. I have done extensive testing on my gaAsfet preamp and measuring levels going into and out of the preamp. The biggest offenders are two FM BCB signals as well as a 162 Mhz signal. The gaAsfet preamp has unity gain on FM BCB due to selectivity in the LC input filter. There IS some gain at 162 Mhz but perhaps not enough to be a problem. I also added a 2 pole filter ahead of a preamp just as a test. This was before I added the cavity after the premap so perhaps I need to repeat the test of the filter ahead of the preamp. I think it is important to make test's in the correct order since it is very likely a combination of things. If memory serves me correctly, the offending signals coming out of the preamp is something like -50dbm. I don't know what the wideband level is but I have difficulty using an SDRplay at optimal sensitivity due to overload of the hardware from all that pulse noise in a wide bandwidth. What my cavity after my antenna preamp is solving is likely IMD in the mixer of the ORIGINAL IQ+. I want to make sure that is clear because I believe later versions of the IQ+ have a significantly strong mixer.  The issue is the difference in bandwidth when dealing with pulse noise. The Delta44 uses 96Khz bandwidth, but the IQ+ mixer may see 5 Mhz worth of pulse noise which CAN overload the mixer.  The mixer in your rig may or may not be able to handle this wide band pulse noise but it might be worth investigation. When noise overloads a mixer, I believe it can raise the noise floor to a degree higher than explained by on frequency noise alone. A simple test might be to try an RF attenuator. If reception improves with the attenuator, you may have uncovered some trouble in your system. On noise: Isn't really all noise pulse noise? It would be hard to imagine multiple sine waves at a lot of frequencies. By that definition they are no longer sign waves. I DO believe we can have different peak to average pulse noise ratios  Galactic noise for instance would be from so many different noise sources, the peak to average ratio might be low while power line noise can have a high peak to average power ratio. I believe when you can not resolve the noise pulses, you simply need to sample at a faster rate. When I evaluated my pulse noise at 5 Mhz bandwidth  I concluded that much of my pulse noise  was at at least 300Khz. That noise when looked at in a 96Khz bandwidth simply looks like an elevated noise floor due to pulse stretching and ringing. Regarding antennas: I think a person would be hard pressed to use a pair of maximum gain yagis on EME in the city. What is needed is low sidelobes and to be satisfied with not hearing anything at low elevations such as 10 degrees. One might get away with a pair of long low sidelobe vertical yagis side by side, but I think city dwellers should be using 4 low sidelobe yagis even if they have to be short. Another trick often used is to place the array as low as practical. A low yagi will have a high radiation angle and that will result is far less low angle signal pickup.  One trick I often use for lower angle moon elevations is to park the array in the first elevation null. In that case it is 25 degrees elevation. The main lobe is then above the horizon  while the first sidelobe is then below the horizon. There is some signal loss, but I get good signals this way down to about 15 degrees elevation. 10 degrees is usable if it is a good 4 yagi station I am shooting for. If this causes me to not be heard, I simply lower the elevation during TX. I used this approach a lot during the ARRl EME contest.  I'm pretty happy with my score especially considering my city location. There are some good low rear mounted rotatable polarization  arrays out there. I'm not about to give up on 2 meter EME in the city. My incremental changes HAVE been yielding positive results. I like the challenge. WB9UWA On Tue, Mar 30, 2021 at 2:22 AM Oliver Dunker via Moon-net < moon-net(a)mailman.pe1itr.com&gt; wrote: Hi Earl! Great to hear that you are onto something regarding your heavy noise problems! When it comes to Preamps/LNAs there still is that misunderstanding "the lower the NF the better" and "take the most modern transistor"..In the lab that looks fantastic,but there are several problems:All of these transistors are designed for ghz commercial applications.Not at all for VHF.When you force them into 2m work,thats a problem by itself.Then also shooting for lowest noise factor makes real real bad IMD figures! Ok,there are somewhat OK designs,but also real bad ones.Like that very popular german brand that uses a stoneage JA... design that was critical from the beginning.What happens is,when you force them to lowest noise you completly loose any kind of selectivity! Now combine that with a city enviroment and that preamp will be saturated 24 hours a day.A wide sweep at a city based 2m antenna will show shocking results there.Like fm radio stations or pagers that are so vvery strong they might light a bike bulb at your coax end. Such a preamp will do nothing against this,quiete often even amplify such signals.At least air airband and pagers,while fm radio gets through 1:1,which is real bad.  I remember that DJ9BV showed us what actually happens then with a GaasFET frontend: One might expect to find birdys or such,but that is not the case.What really happens is a user undetected,somewhat white noise level.Or wide noise level?you know what i mean,lol The usual filtering after the LNA,for best NF cant help there at all.One can say its already too late.  You can get around this,thats what we did,by using super low loss matching in front of that transistor.Like cavitys for example.The matching is the filtering there,thats the problem.Or at least very hi class parts in L+C..Shure that ... brand never did that. I know why i have 2 Landwehrs,he had ervything in silver,and expensive Cs,whole LNA designed with IMD in mind from the beginning and not "the latest transistor"..etc..  Ok,so what to do today? Well thats real easy,since electronics have moved forward so rapidly. Just take a look at the IMD figures of devices like the PGA103 and you will be shocked.Made for full duplex in mobile phones its unreal what they can take!I even tried those in the city with no front end filter at all (except the ant) and it worked! Simply because they can take it,theres no magic noise raising! But if you go ahead and build something like YU1CF,a PGA103 with a very hi Q,super low loss filter,you havve a real winning combination!  Oy yeah:And we also learned from DJ9BV what happens if you(fE) shoot with your GaasFET LNA on a radio or xverter that already HAS a GaasFET frontend:Talk about killing IMD totally!Your neighbours hairdryer might be enough to kill such a system. But i see people doing that,over and over. I disabled/bridged my IC275H frontend,that does the Landwehr preamp.More than enough gain.  I like what you wrote abot array directivity on other than design qrg,very true!A 2m beam will be more like a 4 leaf clover there,so fE fm or pager noise might be much worse than one expects!  The SDRs.Well i think that direct sampling and all the other things are great.But to really work in a tough enviroment we need at least 10 years more tech there(or money?).Just look at the IMD figures of that IC 9700...oh boy,a TS790,IC970,IC 275 etc runs rings around that thing,very easily!And thats 1980s tech...  best 73s and stay healthy! Olli DM2TT ex DJ2TX      -- MUSTANGSERVICE.DE _______________________________________________ Moon-Net posting and subscription instructions are at http://www.nlsa.com/nets/moon-net-help.html -- Earl Shaffer,  WB9UWA  My EME array photos https://www.dropbox.com/sh/bbn2ktonko8e3az/AACHBBOhf6Djk5Pf2c86Br5ja Detailed array photos http://www.gm4jjj.co.uk/WB9UWA/ Facebook http://www.facebook.com/earl.shaffer Linkedin http://www.linkedin.com/pub/earl-shaffer/12/881/735 _______________________________________________ Moon-Net posting and subscription instructions are at http://www.nlsa.com/nets/moon-net-help.html 73, Ed - KL7UW   http://www.kl7uw.com Dubus-NA Business mail:   dubususa(a)gmail.com _______________________________________________ Moon-Net posting and subscription instructions are at http://www.nlsa.com/nets/moon-net-help.html