Spectratac and narrow banding

[JMINN]

We have a Spectratac voting system consisting of an MTR2000 as the transmitter and one of the receivers and two Micor spectratac receivers. We recently narrowbanded the system. I used a narrowband if crystal filter kit from Comspec for the Micor receivers. I realigned the micor receivers per Comspec directions and made surre all line levels and equalization was the same after narrowbanding. All the units used on the system are Motorola and none have companding enabled. We have had complaints about the audio quality ever since changing. I choke some of it up to less bandwidth, thus less fidelity. I am wondering if anyone else has narrowbanded an analog Spectratac system and what their experience has been.

Thanks

Joe

[Bill_G]

How does it sound to you?

[escomm]

And what are the objective user complaints that are reproducible with normal operations?

[N4DES]

Did you do a post-test to see if the modifications changed the de-emphasis curve?

We just recently narrow-banded our UHF EMS system by repalcing all of the Micor Spectra-TAC's with the Quantar receiver equivalent.
System sounds really good even though it is rarely used as the users prefer to use trunking 800 to communicate with our hospitals.

[JMINN]

Thanks for the responses. I personally think that the system is ok. I admit it sounds different, but so do all the systems that go from wide to narrow. Around here it only takes one person to start an avalanche of complaints. We have a saying around here "change is hard". I wanted to get input from the board in case I am being to closed minded to possible narrowbanding issues.

Joe

[kb4mdz]

FWIW, I have seen four major modes with regards to system changes:

1) No changes made to any part of the system - no one notices, no one complains

2) No changes made - someone 'notices', starts complaining (sub-set here; a) everyone ignores him or b) it starts an avalanche of unfounded accusations)

3) Change the system, some times even drastically, like a major failure; tower site down, etc. - no one notices, no one complains, because 'they didn't know any different' or several other excuses.

4) Change the system, maybe even for the better (expanded coverage area, better audio, more features....) but someone complains, loudly....

Multiple sub-variants in each category.

Always involves sorting fact from opinion (which the holders of those opinions view as facts, anyway) from fiction.

Good luck!

[Bill_G]

If it sounds different, then perhaps the suggestion to revisit the deemphasis curve is spot on. It shouldn't sound different - just lower in volume on an older scanner or wideband radio. I've got customers with mixed systems (12.5kc and 25kc channels) before the push for narrowbanding, and the users could not tell the difference in service though it was audible on a scanner.

[JMINN]

One of the main complaints is that it is not as loud as our other channel which is a narrow band MTR200 repeater. We didn`t have the budget to replace the Micor voter receivers so I install a narrow band if crystal filter kit and readjusted the discrimator for max output with 1kc tone at 2.5 dev full quieting. I can`t hear the difference between the audio from the micor voters and the MTR200 voter receiver. I do however wonder if the audio from the micors might be suffering from the conversion to narrow band even though I don`t hear it or see it on a scope.

Joe

[Bill_G]

Have you got a way to sweep the receivers to look at the response curve? ie: pink noise gen and an audio spectrum analyser?

[JMINN]

Been playing around with an identical system on the bench. I have the MTR2000 configured with low priority repeater ptt to fall back to conventional repeater operation should the spectra tac comparator die. I discovered that with the same level of modulation into the MTR`s receiver that repeater tx modulation is significantly higher that the wireline voter input no mater how the voter line out is set or the mtr tx line level is set. I think that this is why they complain in the field that the conventional repeater is louder that the voter system. I noticed that if the tx mod adjustment of the mtr is set to 4000 instead of 2500 that the average mod level though still less than 2.5kc comes much closer to the level seen when the mtr is set to 2500 as a conventional repeater.

Joe

[JMINN]

Ok I think I have figured out the method to Motorola`s madness. According to the brief explaination in the wire line tx adjustment page, what ever this input level is set to will yield 60% of full modulation, therefore if the transmitter is set for 2.5kc mod level, the maxium the wire line can drive the transmitter is 1.5 kc mod. When used as a repeater the receiver level to exciter level is pass thru so 2.5 in nets 2.5 out. The transmitter dev setting allows setting for 2500, 4000, and 5000. 60% of 4.0 kc = 2.4kc. therefore if the transmitter is set for 4000 max dev. the max dev. from wireline would be 2.4 kc. I quess this is why the have a setting for 4000 dev. The percieved difference in levels would be more pronounced an a narrow band vers wide band system. For a communication company Motorola is not very good at communicating.

Joe

[Bill_G]

Oh ho ho. You discovered the MTR's unique and faskinating reverse polish notation method of wireline level setting. In the alignment menu under tx wireline, there is a dialog on the screen explaining everything you need to do to execute the procedure. Super simple. And when you're done, it will be wrong.

Off the top of my head, it's something like "inject a 1K tone at -10db, and click okay". You do that, and deviation drops. So, what you do is lie. Inject -16db, click okay, reset the station, and see where you land. Just keep going lower until the average deviation is where you want it. Button it up. PITA, but they didn't invite us to that meeting. We get to live with it.

You definitely want to keep the average high on narrowband systems. If you thought Officer Mumblemouth was hard to hear before, wait until you hear him on a NB radio. K2RF makes a dynamic speech limiter that does a great job of boosting the low voices, and knocking down the loud ones so that the overall average remains high but undistorted. It's a thing of beauty.

[kcbooboo]

Speaking of MTR2000 deviation, I had a similar experience trying to align the deviation settings. The MTR generates its own tone; you measure it and enter the deviation into the screens. When you're finished, the station, even though set for 5000 Hz BW, maxes out at 4000 Hz deviation when driven very hard. So what you have to do is lie about the numbers and enter values that are about 20-25% lower than what your deviation meter indicates. When you're done, the station will now deviate to 5000 Hz like you expect.

Bob M.

[JMINN]

Dear Santa

Please bring me a pot I can put a screwdriver on.

Joe

[Bill_G]

The new screwdriver comes with a qwerty keyboard and an lcd display. Resistance is futile.

[JMINN]

I don't so much mind the laptop. I just wish the software writers new alittle something about radio.
Joe

[d119]

I don't so much mind the laptop. I just wish the software writers new alittle something about radio.
Joe

As far as I know, that radio (MTR 2000) originated in Europe. Ever tried to do anything with a radio made by RF Technologies, Tait, Kyodo West, etc?

Here's a desk, help yourself to banging your head on it.

I'm about to do a complete re-optimization of a VHF system that is all MTR 2000's. This is invaluable information for me! Thanks guys!

[psapengineer]

Ah yes, the 'ol MTR2000 level setting challenge. We've had to do similar, inject a level well below what you want to drive the base at, lie to the MTR software about what level is arriving (on the order of 6 or more db higher), and then drive the station "hotter" than the setting by a few DB. Just watch your composite deviation on a service monitor so you don't go over legal limit.

PS: The problem on a repeater is exacerbated by the fact that its narrowbanded twice; one on the incoming path and once on the outbound path.

Simulcast Solutions is offering a NBL4 to boost and limit the audio in-between the voter and the Tx. We're intending to use them in our final NB system later this coming fall.

[Bill_G]

Simulcast Solutions are great people. Good resource for a lot of cool LMR products. the NLB4 limiter looks like something I would want to try. It solves the problem of limiting tx audio without affecting keying tones. The Convex audio bridge is a super panel for dialing in simulcast audio within a few parts of a db. It's the overlap killer.

[JMINN]

According to the alignment software what ever you set the tx line level to it will produce at max 60% of whatever you set max deviation to. That why I suggested setting max dev to 4000 instead of 2500. 60% of 4.0kc is 2.4 kc.

Joe

[N4DES]

Also watch for your PL/CTCSS level. If your like me, I like to operate 300 to 400Hz of PL in wide-band and reprogramming it might just drop it in half. In doing this it might cause the level to drop below the radios usable window which usually drops out below 200Hz. You might have to turn it up a little so the decoder is happy.