Quantar exciter alignment

This forum is for discussions regarding System Infrastructure and Related Equipment. This includes but is not limited to repeaters, base stations, consoles, voters, Voice over IP, system design and implementation, and other related topics.

Moderator: Queue Moderator

Post Reply
User avatar
515
Batboard $upporter
Posts: 297
Joined: Wed Sep 12, 2001 4:00 pm

Quantar exciter alignment

Post by 515 »

I recently had to perform some alignments on a Quantar after an exciter module was sent in for flat rate repair. When the module was returned, it didn't include any notices to perform any alignments after installation.

A quick test on an IFR 2975 showed the P25 modulation fidelity was way too high (around 25%), so I figured an alignment was in order...

Both the Reference Modulation Compensation and TX Deviation Gain parameters required significant adjustment to get the P25 modulation fidelity down to 2% or so.

The Reference Modulation Compensation adjustment was fairly sraightforward, but the instructions for the TX Deviation Gain adjustment left a little to be desired...

Basically, for the TX Deviation Gain adjustment, the station transmits what sounds like a 1 kHz tone on four test frequencies, and you're supposed to measure the deviation of this signal, and report what you're seeing to the Quantar via the CPS.

The instructions don't mention if the deviation value is supposed to be a peak or average. On the IFR 2975, you have choices of PEAK, AVERAGE, POSITIVE PEAK, and NEGATIVE PEAK. On the R2670, you get choices of Normal, Peak Average, and RMS Average. From trial and error, I found the Normal mode on the R2670 and the PEAK mode on the 2975 gave the values that the Quantar would be happy with...

In addition, the CPS help file states "Make sure the communication analyzer (R2000 series) is set to Wideband mode". The older RSS manual written for the R2001 makes no mention of this requirement. Fortunately I had access to an R2670, where the "wideband mode" is +/-100 kHz, and narrow is +/-5 kHz. The closest IF bandwidth choices on the IFR 2975 to the R2670's "wideband" are 60 kHz and 200 kHz. I found the 60 kHz setting on the IFR produced values similar to the R2670's.

Finally, the CPS help file states:
The minimum for the first frequency field and third frequency field is 1.82 kHz. The minimum values for the second frequency field and fourth frequency field depend on the values entered in the first and third fields, respectively. The maximum frequency for all four fields is 10.00 kHz.
The instructions say to start with the first test frequency, and then do 2, 3, and 4. I ended up doing the tests out of order, otherwise the CPS wouldn't accept my changes. This didn't seem to cause any problems...


So, my questions:

Do the Quantar exciter modules always need to be realigned after a repair? As far as I know, we received the same exciter back, so I thought the tuning values wouldn't need to be changed, or this would have been done at the repair depot.

What realignments need to be done when an exciter is moved from one station to another (same band/range)?

Are the tuning values stored in the exciter or station control module?
User avatar
xmo
Moderator
Posts: 2549
Joined: Fri Oct 12, 2001 4:00 pm

Re: Quantar exciter alignment

Post by xmo »

In the olden days B.C. [Before CPS], the Quantar RSS package came with a real book.

In that manual there is a section on performing post-repair alignment. It definitely says you need to align the mod comp and deviation when replacing either the exciter or the station control modules.

This manual suggest that the 2670 with firmware 5.04 or later should be set to narrow band and the filters set to H.P.>5Hz, L.P.<20 KHz.

The wideband suggestion in the CPS may intend to deal with all 2670's regardless of firmware version. Some versions may have different filter defaults. The real goal is to have the most accurate deviation reading for a particular instrument.

I really never liked the way they designed the Quantar deviation setting procedure. Here are a few hints for tweaking the deviation:

If you lower the numbers that you entered in the TX Deviation Gain Adjust screen when you did the initial station alignment, this will cause the deviation to be higher. Lowering the entered deviation values by about 40 Hz should result in a 0.1-dB increase in deviation.

Changing the deviation by altering the data entered in the RSS deviation screen affects both analog and ASTRO deviation.


Note: To save time you may only need to change two of the values on the TX Deviation Gain Adjust screen. The TX Deviation Gain Adjust screen asks you to enter the deviation that you measure at four different test frequencies. These frequencies represent the end points of the two VCO ranges. If the base station only operates on one frequency you only need to adjust two of the fields. Determine the normal operating frequency of the base station and find which two of the four test frequencies it falls between. You only need to modify those two deviation settings.

With your 2975 you can verify the modulation fidelity with the station passing real digital traffic. Alternatively the RSS/CPS can cause the station to transmit a test pattern for measuring transmit BER which you can do with either the 2670 or 2975.
User avatar
xmo
Moderator
Posts: 2549
Joined: Fri Oct 12, 2001 4:00 pm

Re: Quantar exciter alignment

Post by xmo »

Here is some additional information on the subject from Motorola:


The ASTRO simulcast system deviation for "Wide Pulse" is 3.00 kHz ±5% and for "Narrow Pulse" it is 2.83 kHz ±5% as measured using the standard deviation test pattern (utilized in the "ASTRO TX Align and Test" mode of the station). This is not the same as the over-the-air ASTRO signal.

Note: There is no reason to align TX deviation in the field unless the exciter or station control board has been changed. TX deviation is aligned in the factory to an accuracy of 1%.

Pre-Alignment Checklist:

You must set the deviation utilizing the service mode of the RSS. Setting deviation using off-the-air signals may cause inaccurate system deviation which may result in missed subscriber receive audio.
A strong signal is required to get an accurate deviation measurement. It is recommended that you have a minimum of -10 dBm signal strength input to your Communications System Analyzer.
The Narrow Bandwidth-NB filter should be enabled on the R2670. This is done to reduce the amount of noise inherent to the wider IF of the R2670. Previous versions of Motorola service monitors did not require setting the bandwidth to narrow.
The High Pass filter should be set to 5 Hz and the Low Pass filter set to 20 kHz in the R2670’s "Special Function Menu" post filtering section.
The R2670 firmware version should be 5.04.xxx or newer. The newer version includes a new function which provides "Deviation Smoothing" in bar graph mode with the "expand" command enabled. This will allow a more accurate reading of deviation. (See Service Information Bulletin )
All channels on the system need to be measured with the same R2670 service monitor to ensure accurate deviation in overlap areas. If multiple R2670s will be used, measure the difference between all pieces of test equipment and apply the appropriate correction factor to your measurements.
Interpreting Test Equipment Results

Based upon system deviation tolerance and service monitor accuracy the chart depicts what error can be allowed with the deviation measuring equipment. The chart lists the allowable tolerance necessary to achieve the 3.00 kHz ± 5 % specification, using various communication analyzers.

Lower Limit Lower Reading
Measurement Specification Upper Reading
Measurement Upper Limit
System Tolerance ± 5 % 2850 Hz 3000 Hz 3150 Hz

1 % Instrument Accuracy 2880 Hz 3000 Hz 3120 Hz
2 % Instrument Accuracy 2910 Hz 3000 Hz 3090 Hz
3 % Instrument Accuracy 2940 Hz 3000 Hz 3060 Hz
4 % Instrument Accuracy 2970 Hz 3000 Hz 3030 Hz
5 % Instrument Accuracy 3000 Hz 3000 Hz 3000 Hz

Due to the accuracy of the R2670 (±5%), the final value must be as close as possible to the specification of 3000 Hz (We recommend an error of no more than ±20 Hz).

Procedure:

1. Verify that the "Digital Modulation Compensation" is adjusted properly using the procedure as outlined in the Quantar RSS manual "Aligning Reference Modulation Compensation, F7 "Align REF MOD2"

2. Perform "TX Deviation Gain Adjustment" in RSS Service Mode using the R2670 settings as described in the pre-alignment checklist. Input the highest of the +/- displayed deviation from the bar graph with smoothing enabled.

The goal is when step 2 is performed with the standard deviation test pattern enabled in the RSS that 3000 Hz be displayed as close as the R2670 allows.

3. Perform Check "Astro TX Align and Test" in RSS Service Mode with "Standard Deviation" enabled.

Step 2 allows you to confirm that the station "Wide Pulse" deviation is set to 3.0 kHz ±5% and "Narrow Pulse" deviation set to 2.83 kHz ±5% as displayed on the R2670 service monitor in "bar graph" mode with "Deviation Smoothing" enabled via the "expand" command. If the deviation is not adjusted properly, you must go back and readjust the deviation in step 1.

Note: If you lower the numbers that you entered in the TX Deviation Gain Adjust screen when you did the initial station alignment, this will cause the deviation to be higher. Lowering the entered deviation values by about 40 Hz should result in a 0.1 dB increase in deviation. Due to variations in the stations, a 40 Hz change may not always be sufficient. If no change is observed, you may need to increase the value slightly. Re-verify in the "ASTRO TX Align and Test" mode in the RSS that the desired nominal deviation has been achieved.

For additional information please contact SSC-System Support Center at 1-800-221-7144 and reference bulletin # S-0044

Information on R2670 service monitor:

Service Information Bulletin 003 February 16, 2001

Communications Test Equipment Sales Information Bulletin - "Deviation Smoothing"

In response to numerous customer requests, a new feature has been added to the R2600C, R2670A, and R2625A series of communications system analyzers. This feature makes aligning radios faster, easier, and more accurate by reducing the display fluctuations inherent in wideband deviation measurements. These fluctuations are typically caused by high frequency noise, and are most pronounced when making relatively small deviation measurements in the wideband RF mode.

The feature is automatically invoked when the analyzer is in expanded bar graph display mode. The bar graph display mode simulates an analog instrument and is useful when making equipment alignments and adjustments. It graphically presents the data contained in the RF Display meter area of the screen, including Monitor Deviation, Frequency Error, and Input Level. When in the EXPANDED screen, a special algorithm is implemented that effectively averages the readings and updates the measurement display approximately once every second.

All units manufactured after November 1st, 2000, including the new color display models has this feature as standard equipment. To verify a particular unit, check the firmware version number. You can access this information via the SPF function key. Highlight the VERSION label and select the DISPLAY TABLE softkey. The STANDARD version number should be 5.04.xxx or higher.

Units that do not have this feature can have it added during a standard calibration procedure or normal repair by the Motorola Test Equipment Service Depot. Please contact them directly at 800-323-6967.
User avatar
515
Batboard $upporter
Posts: 297
Joined: Wed Sep 12, 2001 4:00 pm

Re: Quantar exciter alignment

Post by 515 »

That's some great information. Thanks!

It would have been nice if the R2670 did the P25 modulation fidelity measurement. That's quite a bit simpler than getting the Quantar and service monitor set up exactly right. Without an instrument that does modulation fidelity, it seems shooting for as close to 2.83 kHz as possible would be preferred over the BER test.

I would think with the BER test alone, it would be possible for the alignment to be a little off, but not enough to cause any bit errors, and could go undetected.

I did notice the CPS help file advised to do the full RF PA power alignment before anything else was done, but I skipped that step... It sounded like they were worried the PA would put out more than the 125W the service monitor was rated for. I knew the PA was already properly running at it's programmed level, and just kept a close eye on the service monitor's power reading during the tests. The power during the mod comp and deviation tests was around 60W, or 50% of the PA's rated max, just like the instructions said.

I haven't done an RF PA power alignment on a Quantar yet. Any additional advice on that?
User avatar
Wowbagger
Aeroflex
Posts: 1287
Joined: Tue Dec 10, 2002 10:46 am

Re: Quantar exciter alignment

Post by Wowbagger »

I rather surprised Motorola hasn't updated their procedures for the 2975 and 3900, especially since the signal measurements are more accurate.

<SNIFF>I feel unloved</SNIFF>
This is my opinion, not Aeroflex's.

I WILL NOT give you proprietary information. I make too much money to jeopardize my job.

I AM NOT the Service department: You want official info, manuals, service info, parts, calibration, etc., contact Aeroflex directly, please.
User avatar
515
Batboard $upporter
Posts: 297
Joined: Wed Sep 12, 2001 4:00 pm

Re: Quantar exciter alignment

Post by 515 »

Yeah, their proceedures are pretty outdated. The old paper book RSS manual shows an R2001! Wasn't the R2001 close to end-of-life by the time the Quantar was introduced? The most up to date stuff seems to be in the CPS help file. What XMO provided above is better still...

I essentially used the 2975 for the whole alignment, and it worked fine. I just used the R2670 to verify the deviation values, since I didn't know all the correct peak/average settings. Once I determined that the 2975 needed to be set for 60 kHz BW and the deviation meter set to PEAK, everything else was pretty straightforward.
User avatar
Wowbagger
Aeroflex
Posts: 1287
Joined: Tue Dec 10, 2002 10:46 am

Re: Quantar exciter alignment

Post by Wowbagger »

I ought to see if we have the Quantar in the Autotest II auto-alignment program.

Nope, not yet.

So, if you guys would like to see the Quantar added:

mike.fortna@aeroflex.com

(and while you are at it, if anybody were to put that on a truck stop rest room wall, well, I wouldn't look disfavorably on that. I owe Mike some grief.... ;) )
This is my opinion, not Aeroflex's.

I WILL NOT give you proprietary information. I make too much money to jeopardize my job.

I AM NOT the Service department: You want official info, manuals, service info, parts, calibration, etc., contact Aeroflex directly, please.
User avatar
xmo
Moderator
Posts: 2549
Joined: Fri Oct 12, 2001 4:00 pm

Re: Quantar exciter alignment

Post by xmo »

Escort coordinator.

Discount for new accounts.

For pictures of our talented staff, contact:
User avatar
Wowbagger
Aeroflex
Posts: 1287
Joined: Tue Dec 10, 2002 10:46 am

Re: Quantar exciter alignment

Post by Wowbagger »

xmo wrote:Escort coordinator.

Discount for new accounts.

For pictures of our talented staff, contact:
You ever have the pleasure of meeting Mike?
This is my opinion, not Aeroflex's.

I WILL NOT give you proprietary information. I make too much money to jeopardize my job.

I AM NOT the Service department: You want official info, manuals, service info, parts, calibration, etc., contact Aeroflex directly, please.
User avatar
xmo
Moderator
Posts: 2549
Joined: Fri Oct 12, 2001 4:00 pm

Re: Quantar exciter alignment

Post by xmo »

I haven't met any of the IFR/Aeroflex folks in person.

When we bought the 2975, I talked by phone to Rafael.

He seemed quite agreeable to selling features that you were still writing the code for ;-)

So does Mike need his chain pulled?
User avatar
Wowbagger
Aeroflex
Posts: 1287
Joined: Tue Dec 10, 2002 10:46 am

Re: Quantar exciter alignment

Post by Wowbagger »

xmo wrote:I haven't met any of the IFR/Aeroflex folks in person.

When we bought the 2975, I talked by phone to Rafael.

He seemed quite agreeable to selling features that you were still writing the code for ;-)

So does Mike need his chain pulled?
Well, Mike's one of those guys who will break something, and when you ask him what he did - "I don't know."

Grrrr.

And he has, on occasion, let internal builds get out to customers.

But seriously - he is one of the people helping champion things like AutoTest/AutoAlign, and really, if you can convince him to fight for it, it will make it easier for folks like me to put it into the software.
This is my opinion, not Aeroflex's.

I WILL NOT give you proprietary information. I make too much money to jeopardize my job.

I AM NOT the Service department: You want official info, manuals, service info, parts, calibration, etc., contact Aeroflex directly, please.
mogman
Posts: 127
Joined: Tue Aug 05, 2008 2:25 pm

Re: Quantar exciter alignment

Post by mogman »

Old thread I know, any suggestions on setting up the Aeroflex 3550R for Quantar alignment.
I have not dealt with Mike that I can remember but I have with Jeff McClain and Sam Harris, great folks!!
User avatar
515
Batboard $upporter
Posts: 297
Joined: Wed Sep 12, 2001 4:00 pm

Re: Quantar exciter alignment

Post by 515 »

I've never used a 3550R, but I would try and set it for measuring "peak" deviation with the IF bandwidth set between 60 kHz and 100 kHz for the deviation measurements.
Post Reply

Return to “Base Stations, Repeaters, General Infrastructure”