SNV-12 Voter audio issues on a 2.5khz system

[quint 18]

Hello all,
We have recently upgraded our system to a 2.5 kHz vhf conventional repeated system with 2 satellite receivers and a receiver at the transmitter site. The issue I am having is that we are using a jps snv-12 voter as a voter and also a repeater controller, insomuch as the “voter” uses the EAI tone set to cause the voted audio to be repeated. The voted audio will sometimes dropout on the console when units are talking however the audio is repeated out of the transmitter for that channel.
So I have several questions to ask here
1. Has anybody voted a narrow band with a snv-12 voter?
2. Did you use status tone from the radios?
3. How did you determine your levels for audio?
4. Did you notice any issues on the console end of things that were not on the rest of the system?
5. Are there any precautions to be taken with using quantar base stations in this type of configuration?
I have read over the manual for the voter many times, the manual gives good instruction for configuring the voter on a wide band system but is very lacking on a narrowband system. I have also spoken with the Raytheon support folks they have been most helpful for the most part but some of the issues and questions I have they have no answers for.
I know this has been done successfully somewhere and it is my hope that if you can’t answer my questions you may be able to point me in the right direction!

Any help or input is much appreciated
Frederick Kinder

[Wile E. Coyote]

1. Has anybody voted a narrow band with a snv-12 voter?

Yes

2. Did you use status tone from the radios?

most of ours use status tone. Some use COR via E&M signalling through a T1

3. How did you determine your levels for audio?

See section 4.6.7 in your SNV-12 Operations manual.
We inject a signal into the receiver at 60% deviation (about 1.5K deviation on your service monitor for narrow band). this should be about -10dB on the wireline. The green "normal" LED should just begin to light at this level. If it does not, adjust the RX level pot unitl it just begins to light up.

4. Did you notice any issues on the console end of things that were not on the rest of the system?

No. However, aging voters tend to have audio problems when caps in the -12V supply begin to fail. For more info, read this: http://batboard.batlabs.com/viewtopic.p ... lit=snv+12

5. Are there any precautions to be taken with using quantar base stations in this type of configuration?

Not that I can think of.

[FMROB]

Ok, so here is my two cents on this particular operation. I am going through this without looking at my notes, so please bare with me.

I have a few JPS SNV12 units in operation now using astro tac and quantar stations, connected to Avtec consoles. My transport medium varies and includes telco RTNA, carslon microave links, local copper.

Narrow band is not an issue with the SNV unit, your audio levels will just be lower acorss the board, thus you will need to increase the audio adjustments upward. This is the same idea for adjusting deviation of a radio, tone remote etc. instead of 5 kc or audio your dealing with 2.5 or a bit less.

My advice is to downlaod the newest manual, print it, bind, and read it. Make certain that you read the pre-check list for jumper settings, audio levels etc. Pay special attention to the level adjustment sections (which are scattered throughout the manual).

So here goes, forget what you know about setting levels in a voting system. Use the idiot lights on the front of the SNV. On air send 1K tone at 2.5 kc of audio and adjust each line card to the solid yellow light with some peaks of red. adjust all line cards the same way by disabling a RX'r at a time. Now, to fine tune these settings for real world use take a known, good working portable and talk normal. Ocassionaly give a long high FIVEEEEEEE and watch the idiot lights. Again, you a striving to adjust for a yellow light on normal speech with peaks of red on high notes (occasional). So I know that you are all laughing at me, but I tried it the right way and it didn't sound as good as letting the DSP do its thing. The SNV works, and works well.

NOW I will tell you that on the far end side of the system RX and wire lines I made certain that I leveled all my RX sites. I used the old spectra tac philoispohy of a -13db system (espicially if you are using phone lines) as you don't want to hit them with anything near or higher than 0 db. So I leveld all my wirelines back to the SNV and made them as close as possible, obviously adjusting for telco path loss etc. This way I know if I put a meter on a fouled site at the SNV I know what I am looking at and have a solid base to check performance in the future.

Now, there is a special adjustment for the console out. It ships from the factory set to -10db which is where I would feel comfortable. Use a lineman or tims set to see the audio level to the console with a 1kc tone at rated system deviation. Adjust the POT on the from of the CIM to raise the audio. My thought is you may have the audio way too hot and causing clipping in the console. Remember repeated audio out and console RX audio are two seperate outputs and adjustments.

Also, I wouldn't mix E&M COR and status tone in the same mix. I prefere using 2175 status tone. Just my 2 cents.

If you have any other questions please post, also I can e-mail you the manual if you need it.

- Rob

[chartofmaryland]

What kind of console are you using?

Does it use AGC?

If you crank up the voted out to the console does the problem of the console receiving a transmission go away but you now have screaming audio?

Sounds like a level issue to your console.

CoM

I am keeping ALOT of equipment running on 2.3kc with JPS and it sings along. Just have to watch the levels and make sure the speech detection can work properly.

[jwwissman]

I recently narrowbanded an SNV-12 voter using Kenwood TK-880 receive radios, CPI DTP series interface cards (for 600 ohm balanced audio), and a Kenwood TKR-840 repeater. This system was designed and wired by JPS/Raytheon about 10 years ago and has never had any problems during that time. I could not get proper audio levels to the transmitter after narrowbanding. I followed the alignment instructions in the voter manual to a tee, yet users would sound like robots. I could get normal sounding audio, but at the expense of very low transmit audio. After many hours of hair pulling (and unsuccessful calls to JPS tech support), I started checking how the hardware was actually configured. The system came from the factory wired for balanced audio on the receive and transmit side of the voter. After rewiring the voter for unbalanced transmit audio, everything sounded great. JPS was using a single DTP card to create balanced audio for the transmit and receive side of the repeater simultaneously. The transmit audio was bleeding into the receive audio and causing oscillations in the audio circuits (hence the robot voices). There is no need to have balanced audio on the transmit side. Remember, when wiring for unbalanced transmit audio, you must ground the unused tx audio pin on the SNV-12 termination block.

[Wile E. Coyote]

jwwissman, I am very confused by your post.

Narrow-banding, if done properly, should have no effect on your transmit or receive wireline levels, or the operation of the SNV-12 voter. When you program a receiver or base station for narrow band, it should automatically adjust the wireline levels to compensate for the lower deviation.

Also, if things have been working fine for the last 10 years, why would narrow-banding create problems now? What levels were you getting from your receivers before & after the narrow-banding? The transmit dB level going to the base from the SNV-12 should not change.

No offense, I'm just trying to understand what is going on. Can you please provide some more details of how it was set up before, and how you are going about narrow-banding now?

[desperado]

You have to remember that narrow banding will change the S/N ratio. The noise floor doesn't change but the audio levels do.
Noise floor therefore comes up closer to the audio level. This is the reason you loose coverage footprint when narrow banding, the distance to where the S/N ratio can't be deciphered by the receiver gets closer to the transmit site. It applies here to but that's not what the post is about.

The S/N voters are great if you have at least one GOOD signal to choose from. But if you are noisy on all the receivers, it becomes a challenge to the voter to pick out the best one and stay voted to it.

[Bill_G]

I just finished narrowbanding a VHF system with a three site SNV-12 comparator. No problems. +150 user mobiles and portables. MTR2000 rptr with two MTR2000 satellite rcvrs. Read, change, save, drive to next site. The customer preferred to wait to do the fixed end equipment last. As we progressed through the mobiles and portables, dispatch and some wideband users sounded pretty ratty at times depending on who had good mic technique and who didn't. Once completed, the system sounded as good as it ever did.

Narrow Mot and Kenwood rcvrs seemed to handle wide input better than Icom which squelch clamped at times chopping the audio. Once the repeater system was done, it made finding the few remaining wide users easy because they came in with a zit zit zit edge through the MTR rcvrs again depending on who was a boomer. Overall, it was a painless transition. We'll be monitoring for coverage changes, but don't expect any.

[jwwissman]

jwwissman, I am very confused by your post.

Narrow-banding, if done properly, should have no effect on your transmit or receive wireline levels, or the operation of the SNV-12 voter. When you program a receiver or base station for narrow band, it should automatically adjust the wireline levels to compensate for the lower deviation.

Also, if things have been working fine for the last 10 years, why would narrow-banding create problems now? What levels were you getting from your receivers before & after the narrow-banding? The transmit dB level going to the base from the SNV-12 should not change.

No offense, I'm just trying to understand what is going on. Can you please provide some more details of how it was set up before, and how you are going about narrow-banding now?

I was only providing information regarding balanced transmit audio from the SNV-12 to the transmitter. Mine was originally wired for balanced tx audio to the transmitter using a single DTP card. As my original post stated, this was causing very strange problems after narrowbanding. After wiring the system for unbalanced transmit audio to the transmitter, things worked fine.

[gtriever]

Gotta admit I don't understand jwwissman's issue either. In my case, I just finished installing a system that has four sites with four MTR2000 radios per site, linked via a digital microwave system back to four four-channel STARS-enabled SNV-12s. 1950 Status tone used on all radios, output at -12dBm. Rx Audio sent from the station at around -7.5dBm with a full-quieted signal, Tx audio to site at -6dBm from the BIM card, resulting at around -7dBm Tx level from the voter. No problems at all, both console and voted tx audio are clear and clean when looked at on the service monitor's mod scope.

The one thing I noticed during testing is that if you overdrive the audio, you can get that "robot" effect. In my case, that was approaching a 0dBm level. However, we had already been advised that this new microwave system didn't like seeing 0dBm, so we should stay below it.

[Bill_G]

Gotta admit I don't understand jwwissman's issue either. In my case, I just finished installing a system that has four sites with four MTR2000 radios per site, linked via a digital microwave system back to four four-channel STARS-enabled SNV-12s. 1950 Status tone used on all radios, output at -12dBm. Rx Audio sent from the station at around -7.5dBm with a full-quieted signal, Tx audio to site at -6dBm from the BIM card, resulting at around -7dBm Tx level from the voter. No problems at all, both console and voted tx audio are clear and clean when looked at on the service monitor's mod scope.

The one thing I noticed during testing is that if you overdrive the audio, you can get that "robot" effect. In my case, that was approaching a 0dBm level. However, we had already been advised that this new microwave system didn't like seeing 0dBm, so we should stay below it.

Since wireline cards were developed for telco wires with their expected attenuation, in a lossless system like a microwave, it is good practice to set everything for -10db. Neg ten out, neg ten in - easy to remember, and easy to spot a problem. Let us know how that STARS system works. It looks interesting in concept, but I've never heard it applied.

[gtriever]

STARS has worked pretty well for us in the past. As a fallback, we do try to have the Tx Default site be as centrally located as possible. I agree neg 10 would be easier to remember, but I don't get to make that decision...

[PhaseLockedLunatic]

I run two systems that are very similar to what you have Quint. And...I'm experiencing the same problem.

In each of my systems the SNV-12 is the voter and repeater controller. Each system has six sites connected to it: four are full duplex stations, two are half duplex base stations that link to repeaters.

1. Has anybody voted a narrow band with a snv-12 voter?

Yes, all are narrowband. We did not use wideband on these voters ever.

2. Did you use status tone from the radios?

Yes, 1950 Hz.

3. How did you determine your levels for audio?

Each receiver is set so that with 1.5 KHz in I get -10 dBm at the voter. Usually this ends up being -7 or -10 at the receiver. I then verify that this -10 into the SVM-2 results in -10 out on transmit. Then the transmitter is adjusted for 1.5 KHz.

4. Did you notice any issues on the console end of things that were not on the rest of the system?

I'm not sure what you mean by this exactly. I've faced audio dropout issues with the Zetron 4000 series before, most recently due to VOX COR being improperly picked off post AGC. Zetron claimed to fix that with a firmware update to the DCC. No matter, I have since wired the console input for COR using the COR output of the SNV-12. It hasn't helped.

5. Are there any precautions to be taken with using quantar base stations in this type of configuration?

Couldn't say. I'm using a mix of MASTR III and Daniels MT-4 that are stock.

I know this has been done successfully somewhere...

Like you, I know there are many others doing this and doing it well. I really want to find out what you and I are doing that is unique, what is making this audio quit.

I am frequently told that units in the field heard everything but my dispatchers either got nothing or got choppy audio and had to request a repeat. That's on both of my systems, so yours is the third.

[Bill_G]

This is where logging at the telco demarc helps with troubleshooting. If your system has a logging recorder punched down on the 4W or 2W point between the console and the transport to the field (whether it is telco, microwave, ip, dsl, or crossconnect wire to voter in the same room, etc), and the dispatchers note the time of the dropout, and the time stamps on the recorder are accurate, you can review the "tapes" to see if the dropout came from the voter & transport direction, or happened within the console. If the logging is punched down on a logging port from the console, it may still have occurred in the console.

Your only option for this is to put in a line monitor - ie: plain old tone remote control set - for dispatchers, or you, to manually determine if the dropout arrived at the console input. If it does come from the voter, then you get to move the line monitor to the other end of the line (again whether it is telco, microwave, ip, carrier pigeon, xcon wire, etc), and patiently wait for it to happen again. Hopefully, if all the equipment is in the same room, you are on your game enough to have checked crossconnects between the voter and console input already. The assumption is that if the field units do not hear the dropout, the connections between the voter, the base station(s) and receivers are good.

Failing that, it might be prudent to reseat the CIM (console interface module) in the voter, or swap in a spare to see if it changes the problem. I have had SNV-12 voters post line faults in the log that were resolved by reseating the individual SVM card.

[chpalmer]

Multiple SNV-12s
Zetron 4048
Status tones. (By the way, on some e&m ports the cor will beat the audio by as much as 100ms. CISCO E&M and Harris Syncrocast are both guilty of this. This gives the snv-2 card full quieted audio for that period and can cause it to vote incorrectly. To prove this turn on a status tone on the radio and leave the snv2 set to cor. If you hear the tone chirp through at the beginning of the unsquelch- you have a problem.)
Zetron set to COR.
MTR2000's using 1950. Over Harris Syncrocast 3 transport. MTR set to 6dB with tone 13dB down. At the TP on the front of the SNV2 card your peak audio should measure +5dB. However if your repeat audio is fine you then most likely have a console connection problem. Like others have said- AGC. That has to be correctly set on the DCC (Zetron)
With the Zetron DCC you have to connect with the serial port on the card and tell it that it is a COR connection if you use that.
No problems at all here.