Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

   Home   Help Search  
Pages: [1]   Go Down
  Print  
Author Topic: Measuring Receiver Noise Floor vs. Ingress Noise  (Read 9763 times)
K6ECS
Member

Posts: 29




Ignore
« on: April 22, 2014, 11:27:41 PM »

Our repeater is a venerable Motorola MSF 5000. We love them because they have very good front ends on them. We use Sinclair Q2330e 6-cavity duplexers that have no bearing on ingress noise.

We previously ran this MSF from an 85 foot tower with a side-mounted DB224e antenna pointed at 230°. 1.6 miles behind us at at 333° is a nasty site that clobbers most of the VHF on site with VHF pager noise and who knows what. Most of the VHFers mount their antennas 20 feet off the ground to avoid interference to some extent. This site never seemed to bother us at the old location.

We have since moved closer to this site and are 1.07 miles away using an omni-directional antenna mounted above a new Verizon triangle. Our overall receive distance has increased and we're happy about that. But now I'm getting a popping sound when average signals come in; sometimes they are taken out all together. HT's are tough and they our primary concern. We also have CTCSS popping that I've thought until recently to be kerchunkers. We don't run a courtesy tone and kerchunkers are common.

Now I am left to wonder what tests to run. If I put the service monitor on (a HP 8924C) what do I do next? LOL. Do I test the MSF? Do I check the noise floor first and then connect the antenna? What am I trying to learn by gathering information? I have a dual cavity Angle Linear bandpass set (Chip Angle is well known in the commercial world as well as amateur) and his cavities are known to be quite effective. They create a 1MHz window that he follows up with a 17dB preamp. I have not installed this yet and honestly I don't know how a preamp could be useful if one has already knocked out the world. I guess it is amplifying the frequency that the cavities are tuned to pass. At any rate I see a preamp as a possible addition to the problem.

So then, where do I start?
Logged
K4JJL
Member

Posts: 503




Ignore
« Reply #1 on: April 23, 2014, 07:47:50 AM »

If the popping noise is an outside source, take your RX jumper coming off the duplexer and put on the ANT port of the service monitor.  Put the monitor on spectrum analyze and look for your popping signals there.  Where ever that is, place a notch can in line tuned to that frequency.  Notch can be just shallow enough to push the pop signal below the squelch threshold of your RX.

It's also possible that you've got a noisy antenna or some kind of static discharge picked up from wind.  What kind of antenna and how old?

Also, what kind of coax are you using?  If you're using any variation of Times Microwave's LMR coax, you're going to have problems.  It's not designed for a full duplex application and gets super noisy when used that way (been battling this myself).

Do you have a directional coupler or a network analyzer?  If so, try sweeping the antenna and coax with the service monitor for issues.  If you have a noisy antenna, you'll see your signal jump up and down on the reflected port of the DC.

Another issue I found was loose N connectors.  Went through my duplexer, bulkhead connections, and polyphasers with some channel locks and snugged up connectors.  Made a huge difference in noise.  Ambient heating and cooling of hand-tightened connectors will make them come loose.  PL-259s are the worst offenders.
Logged
K6ECS
Member

Posts: 29




Ignore
« Reply #2 on: April 23, 2014, 08:24:10 PM »

Thank you for the reply. I appreciate it.

The antenna system is new. Brand new. It uses a Commander Technologies Station Commander (Station Master clone) that was sold under the RFS name. The 1/2" Andrews hard line is new with new positive contact N connectors. All connectors are snug. RX strength bothers me at -108dBm or or .891uV after the duplexer. This seems high but we're receiving from quite a distance. I will test it again without the duplexer this weekend.

The popping that I am referring to appears to be stations that are just strong enough to break PL but no audio passes. The only difference is that this is a commercial site with an omni antenna versus a private site with nothing else n the 85 foot tower using a DB224e side-mount antenna fed with 1/2". The winds are calm. The popping of the PL is not frequent enough to sit at a site looking at a service monitor. A watched pot never boils LOL.

I placed the Angle Linear dual cavities inline without the 17dB.

AH HA! I think I've found it. Broadcast band intermod. I happened to be listening to the input and we have a short tail. While the carrier was up I heard an echoy voice that dropped with the repeater. I put a scanner into search and found what I believe it the voice down at the high end of the FM band. 107.9MHz.
Logged
K6ECS
Member

Posts: 29




Ignore
« Reply #3 on: April 23, 2014, 11:09:01 PM »

Chip doesn't so much make ham gear any longer. He is swamped with other work so I am left to look at other options including cavities from Tessco or Talley.

But in that it appears to be an intermod issue from a known FM broadcaster... From my home my repeater a few blocks away up on a hill is -13dBm. I am measuring an unmodulated carrier on 97.2 at -15dBm. That should be enough for something to ride the tail of the repeater for me to hear. I'll have to go to the site and take a reading from there. My input is 147.105MHz. I am hearing 107.9MHz.

But the question is, do I use the dead on approach and get a cavity for the offending station or the intermod station result or use the well respected DCI-146-4H http://www.eham.net/reviews/detail/271 and brute force the entire FM broadcast band plus the paging stuff above me?
Logged
K4JJL
Member

Posts: 503




Ignore
« Reply #4 on: April 24, 2014, 06:16:29 AM »

What's the insertion loss on the DCI filter?

Also, the offending signal might be mixing somewhere else and getting in your receiver.  We had that problem years ago on Stone Mountain.  No filters would fix it.  Tried everything.  Randomly unplugging cabinets, banging on the tower, tightening bolts.  Nothing worked.  It eventually went away when the paging transmitters disappeared.
Logged
K6ECS
Member

Posts: 29




Ignore
« Reply #5 on: April 24, 2014, 08:10:15 AM »

The DCI unit has a <1dB insertion loss. The published specs appear to show a >90dB attenuation of the FM broadcast band. When our repeater TX's I don't see any new spikes other than the Xmitter itself, so I don't believe we are generating IM. I really think that our RX is just getting blanketed. This is why we chose not to use that site for our repeater and chose the lesser commercial site.

All reports of the DCI indicate that it is well made and does a fine job of knocking down anything outside of its +/-2MHz window.
Logged
K6ECS
Member

Posts: 29




Ignore
« Reply #6 on: May 01, 2014, 11:37:37 AM »

Our entire team visited the site on Sunday. One was the supervisor, one held the ladder, one screwed in the light bulb while three rotated the ladder... LOL.

After a lengthy visit to the site we measured the FM broadcasters on our station antenna up at 85 feet at -1 to -3dBm! We measured VHF paging transmissions at -25dBm. We measured no significant reflected power on the antenna prior to connecting it to the service monitor. Plugging the antenna into the ANT IN on the HP Agilent 8924C I read -76dBm with a span of 2MHz. This is with a 6-cavity Sinclair duplexer and a Angle Linear dual band pass cavity. The symptom is like a blanket has been thrown over the receiver of our MSF 5000 that worked fantastically at its former location. I know enough to run the tests and record the findings but not enough to solve the problem. I honestly don't remember the reading in dBm without anything connected beside the antenna but there were tall spikes all over the screen with a 135MHz span -- we needed to see the broadcast stations and paging transmitters. There appeared to be a small spike on or near our input with this wide span. It wasn't there with a 2MHz or even a 1MHz span.

I received a DCI-146-4H band Pass filter that is quoted at -89dB at 126MHz and the plot shows it out of range for the FM broadcast band. Everything is grounded to a new ground rod with 8ga. multi-strand copper wire. All cables are RG-214 with new crimp N connectors. They average 4' in length and can be shortened once we put in a 19" rack. After speaking with an extremely well respected manufacturer of repeater band pass cavities we learned this morning that we should aim for a 5" length of cable. If that doesn't work we should next try 24".

But are we wasting our time? I simply don't know what more to do. Once adding the DCI band pass filter I will be adding an additional 1.5dB of loss on top of the cable loss, the duplexer and the Angle Linear (he makes good stuff) should I even think of a preamp? The Angle Linear was received with a 17dB preamp and I think that may be too hot. Currently we can hear stations at our coverage borders that previously were workable but have now dropped to marginal at best. Again referring back to a conversation with an extremely well respected manufacturer he suggested that once everything is tuned correctly we should have at least a 20dB return loss. Once achieving this a 17dB preamp is not too hot. First things first.

But the overall point/question remains the same. Should I change all of the interconnect cables to 1/4" SuperFlex; the RG-214 is spec'd at only 95% braid for both Belden and Times Microwave (we have Belden deployed and a roll of Times Microwave RG-214.)

We are NOT running LMR 400 anywhere!!! We're indoors and maybe we should since TM is quoted as agreeing that outside LMR 400 does not hold up because of dissimilar metals versus indoors and no corrosion potential. TM LMR 400 is quoted at >90dB rejection of ingress/egress and >180dB of crosstalk. But lets NOT go down this road. We are NOT running LMR 400. The question is whether or not the interference is coming in by way of the antenna (using Andrews 1/2" hard line (LDF4-50) or if the 95% braid of the inter-connect cables is letting it in. Once again referring to a leading manufacturer LMR 400 is quite usable for inter-connect cabling. Please, please, please, let's not go there Smiley RG-214 is working fine but the length is way off. 5" or 24".

Quite humbly I don't know what to do Sad I am going to follow the directions I have been given and I offer this message as a follow-up with a hand out for more suggestions. I am not too proud to admit that I don't know enough and that I am struggling through.
Logged
K5LXP
Member

Posts: 4536


WWW

Ignore
« Reply #7 on: May 03, 2014, 06:00:53 AM »

Personally I would never run a preamp at a repeater site.  While you have a lot of filtering it's still amplifying signals away from your receive frequency.   First thing I would try is to bypass it and see if it helps any.

Mark K5LXP
Albuquerque, NM
Logged
K6ECS
Member

Posts: 29




Ignore
« Reply #8 on: May 04, 2014, 09:22:44 AM »

The only difference with this site are 1: a new omni-directional antenna, 2: 100' New LDF4-50 with L4TNM-PSA N connectors, and 3: the insertion of an Angle Linear dual band pass cavity with 90dB of rejection. It creates a 1MHz window.

But now we have piss poor reception.

A team member spoke with Angele Linear the other day and was given some awesome information that I cannot wait to try, like using 5" or 24" jumpers as the "magic length". But Mr. Angle also insisted that there is nothing at all wrong with using a preamp, that's why he includes it.

You see, the Angle Linear dual cavity band pass filter creates a 1MHz window with a sharp roll-off +/- 500KHz. By placing the preamp after the cavities it only amplifies what is inside of the window.

With that said, and please don't shoot the messenger, he recommends any quality cable to include LMR 400. Not RG-400, LMR 400 with its dual shield and foil. The full duplex problem comes into play when using it as the feed to the antenna. I ran LMR 400 before from the duplexer to the antenna and never experienced desense. But to be safe it is to be used on the inside inter-connecting cables.

I am using RG-214 to avoid the argument that ensues when discussing LMR 400 which has half the loss. I think that we need to look closer at our Sinclair Q2330e 6-cavity BpBr duplexer. I think we may have as much as 10dB loss, which means that it is mistuned. This is where I am weak. I don't know how to tune the duplexer.

I've been told the best way is to use a Return Loss Bridge and tune for the highest return loss with 46dB representing a 1.0:1 SWR. I never knew that SWR was an issue on receive. He recommended that when I am done tuning the cavities that I inert a dummy load at the antenna port and transmit through the cavities. I should have a nearly flat SWR. Only then should the effective receiver sensitivity test be run. I know that when I test the repeater without any fitration I get a workable -124dBm and a reliable squelch opening.

Now that I know enough to be dangerous...
Logged
K5LXP
Member

Posts: 4536


WWW

Ignore
« Reply #9 on: May 04, 2014, 10:03:31 AM »

I tune my duplexers with the antenna connected, receiving a weak signal over the air.  Tuning duplexers and filters with a network analyzer or spectrum analyzer with a sweep generator gets you close but I've always found everything shifts a little bit when connected as a system with less than perfect 50 ohm resistive terminations.

Coax with the foil shield is known for having PIM issues.  Hardline solves this problem.  But I know LMR isn't a factor here.

You've introduced a lot of new variables here.  One test I would try is to use an injection tee and see if you have an antenna issue, or determine if there's filter loss or desense.  In order for a repeater in a good location to have "piss poor" performance means something is measurably broken - bad antenna or significant insertion loss. 

I would still bypass the preamp as a test.  You may find it's not a problem but it's a variable readily taken out of the equation for now. 

Mark K5LXP
Albuquerque, NM

Logged
KA5IPF
Member

Posts: 1044


WWW

Ignore
« Reply #10 on: May 04, 2014, 07:50:19 PM »

You keep running in circles and chasing your tail. Can't catch it so give up.

You have an excellent service monitor that will do all you need in regards to receiver and transmitter testing. It will also allow you to align cavities. You need to learn how to use it.

Start by measuring the sensitivity of the repeater itself, no cavities, straight into the rx port. The service monitor will allow you to measure the 12dB sinad point, do it and note it. Then put the Sinclairs in line and do it again. You should be within 1-3 dB of the first reading if the cans are tuned correctly. Then add the other cans and do it again. In other words figure out where your loss is, don't guess. Do all of the cans inline meet their spec as far as loss?

If all is OK then to the next step. Note that nothing has been said about the antenna, that's the last thing.

The 8924C is capable of doing full duplex testing. Open the manual and figure out how. That will give you the desense the transmitter is causing. Again reference everything to the 12dB Sinad level for a constant reference. Once you have all those numbers you can start looking out into the "real" world via the antenna. Until the repeater by itself and with it's cans is operating properly into the service monitor you don't know where the problems are.

If you find excessive loss thru the cans in rx or by desense, retune them. Easy to do with the 8924. Just do it by the book and make sure all ports are terminated. A final touchup may be needed once the entire system is operational, note "may" be needed. That would come under optimization, not basic setup.

Nothing has been said about a Circulator, do you have one inline? They are mandatory at most commercial sites and may help your problem.

Start attacking the problem by making sure ALL of your equipment is operating properly and up to spec before hooking up the antenna.
Logged
Pages: [1]   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!