Batboard

I did some topic searching but never really found the specific answer.

I am trying to align an XTS3000 Uhf that had the RF replaced. The deviation settings were all screwy with the new deck and it wouldn't properly transmit ASTRO.


My question is with the deviation setting and alignment process.

In the first setting, the Transmit Deviation Balance (Compensation) Alignment screen with the low and hi tones, what is the deviation setting I am looking for to begin with? It says to match both the hi and low as best as possible but what is the figure I'm shooting for to begin with? How many Khz? It doesn't mention 2.83 khz like it does in the help section of the deviation limiting screen.

Also, I am using and IFR1900 BSA for a test set. I am new to deviation measurements. The IFR gives readings in + and - such as
+1.65 -4.91 when receiving an ASTRO transmission from my old astro saber 1. Which figure represents the deviation setting I want to work with? +, -, or a factor of both?

It's not really the actual deviation level per se that you're concerned with. What you'll find is that one of the tones will probably have greater deviation than the other...adjusting the value one way will cause one of them to become greater, while the other becomes smaller. Aligning the other way will do the opposite. So, as long as you set it so both give equal deviation, you're good to go.

Can't help you with the IFR question, as I've never used the 1900.

Todd

I've noticed what you are saying but it seems like I'm going to cook the damn pa by the time I find the setting and the dev meter settles.

I kind of went about it a backward way so far. I went in to the deviation limiter screen and set each freq. so they show as close to 2.83 as possible. Then I went to the dev compensation balance comp screen and stared at a low setting. I then went to the transmitter test pattern screen and selected the voice test. I listened on a P25 radio until I crept the setting up in the balance comp screen to the point the receiving radio decoded. Once I was in the ballpark and the other radio was hearing it I watched on the meter again and adjusted that comp setting until the digital signal showed as close to 2.83 as possible.





make any sense ?

While you may get the system working the way you did it, it's definitely the wrong way.

As Wavetar said, you adjust the compensation/balance so the deviation is the same with the low and high frequency tones, or on the various test frequencies. The actual value of total deviation would be set elsewhere. You're trying to get a flat or equal frequency response across the audio range and across the operating frequency range of the radio.

Yes, the radio will get hot. There shouldn't be too many alignment frequencies to test it on, and you only need to key the transmitter for 5-10 seconds at a time to adjust the compensation. Key it up, note the deviation, press the function key to change frequency, read the deviation, adjust it up or down so it matches the first reading, then go back and check it once more. You can unkey between readings if you want. If you're really concerned, set up a small fan to blow air on the heatsink to keep the radio cool.

I would expect you to get equal deviation measurements for positive and negative excursion. It almost seems like you're reading the actual frequency excursions of the transmitter, not just the relative amount for deviation purposes. In other words, if your carrier should be on 455.000 MHz and it's low by 2 kHz and you're deviating +/- 5 kHz, you'd get a positive peak of 455.003 and a negative peak of 454.993. If everything was on freq, it would be 455.005 and 454.995. You might have to try a different deviation setting on your IFR to give you just the audio deviation, not the carrier frequency shift. Both are useful but you need to know which one you've got and either adjust the carrier so it's dead on frequency (thus giving you proper deviation values) or choose a different setting to eliminate the carrier freq error.

Bob M.

Yea, I was looking at the actual carrier.

I went in to the RMS devation test screen and life was much easier.


HOWEVER, things still don't add up.

I went through and did the test exactly as described. I started at the first test freq and noted the deviation (low setting) most all were at 1.25 khz. I then noted the hi one after testing the low. I adjusted it so it matched the low exactly. Again, most settled around 1.23 to 1.27 khz.

I made sure the hi and low for each test freq. were as close as possible. Most were exact when I was done.


I then went to the limiter screen and adjusted each to 2.83 khz. Almost every single one was right on within .01 khz each way when I was done.

I re-checked the RF power levels and then exited the tuner application.


When I was done astro did not work. It would transmit the modulated carrier but it did not decode. When it ocassionally did it sounded robotic and garbled.

I went in to a test freq near 446 in the tuner's limiter screen again and cranked it up to about 65 from around 20 where it was after I finished testing. The radio would then transmit a signal that could be decoded.


So, I must be looking at something wrong. If I follow the directions for the tuning procedure everything gets out of whack.

OK, first a few points:

First: The deviation meter on the 1600/1900 can be set to either read plus and minus peak, or +/- peak ( (plus peak - minus peak)/2), or RMS. In plus and minus mode, any frequency error will show up as an imbalance between the plus and minus - so if the radio is generating a 2 kHz deviation sine wave modulation pattern and is one kHz off frequency, you will see a +3/-1 kHz or a +1/-3 kHz deviation (depending upon the sign of the frequency error).

Second: be careful when doing the modulation balance alignment - you need to have a wide enough IF setting on the service monitor or you will get screwy readings (I had to explain this to one of our FAE's). The high balance setting is a 3kHz tone at about 3 kHz deviation, for a total bandwidth of 2*(3+3) = 12 kHz bandwidth. Add in any overdeviation and you are looking at a signal wider than 12.5 kHz - so don't use the narrowest setting on the 1900, use FM2 or FM3.

Third: The goal of the modulation balance is to get the shape of the waveform correct - and to do that you need to get the deviation of the low and high tone to match. Not to be any given value, just to match the other tone. (personal opinion follows, please read my sig) This is the most bass-ackwards way to do things, and it would be much better if you could tell the radio to generate a proper P25 waveform and measure the modulation fidelity error while making the adjustments - but the Motorola 2670 wouldn't do that so they had to come up with a way they could do the alignment that the 2670 could do. Moreover, if the Motorola radios would generate the first IF digitally and then upconvert it, like a lot of their competition did, you would not need to do this at all, as the waveform's accuracy would be controlled only by how accurately they did their math when they designed the DSP code. (end personal opinion, again, please read sig.)

Thanks for the pointers. I used FM3 and it appears to be more accurate. I used the tuning values from a friend's radio and that got it in the ballpark. Once I set the meter to the + - /2 setting the deviation readings were in a more reasonable range and my adjustments made sense from that point forward.

So, right now it appears it's aligned properly.

once you get compensation correct, go back and make sure deviation is still at 2.83

when i align, i go back and forth till the values stabalize, especially if the radio has had something drastic done to it (ie RF board changeout, or S-record frankenflashing)

That's what I ended up doing. It shows almost exactly 2.83 khz on all points now.

I suppose if could have been the fact that the initial adjustments were somewhere around 2.5 or so and using the RMS meter (actual level X .707) that could account for the low readings I was initially seeing.

I noticed that the manual calls for a signal level of 4.0 uV to be injected at the antenna port for the receiver front end tuning. Doesn't that seem kind of high? When I ran the meter up to that level the sensitivity curve seemed to really plateau in the middle. It was difficult to narrow down just what the middle of the range actually was. I dropped it to around 1uV and it narrowed things down quite a bit to where I was seeing a more drastic difference when making smaller adjustments.

i use an ifr2975 and its quick on its settings, one step can make a huge difference.

as for the RX, i do the same thing, start high then lower the RF into the radio until i have a very narrow range that the radio unmutes on. Usually i can get a little bit better than what the specs call for.

A lot of our XTs3000's coming back from the new repair depot in
Mexico are coming back screwed up. Bad repairs/bouncers - template wipes and misalignments. Even simple repairs like front housing replacements are coming back misaligned - Dev Bal / Dev Lim and TX-RF . Bounce it back to Moto repair. They are readily admitting that they have a quality issue with their outsourced repairs.
I use General Dynamics AutoTest- it sure makes testing faster and easier.

Also check the PTT for clicking noise and use the squeeze test to check for noise that could be causing digital artifacts and access denials This is a big problem with XTS3000's