Aeroflex (IFR) Questions

Post by **4MR_NST** » Sun Feb 22, 2009 8:01 am

Greeting,

We're looking to replace an HP8920 used to support Motorola conventional, trunking, and paging - both bench and field. I reviewed the slick sheet for the 3920 and didn't see a mention for POCSAG pager testing. Did I miss it? Is it an option?

Also, does (or will) the 3920 support other digital trunking formats for bench/field work? At $30k, I want to CMA should we need to support digital Mototrbo, digital Passport, MPT1327, Opensky, Tetra, or other protocols that could be available in the foreseeable future.

I know that these digital technologies are out there, but don't know if special options are needed to test/align the radios or infrastructure. I read something about a "STAR" for TDMA but don't know what that means.

Thanks in advance.

Post by **Wowbagger** » Mon Feb 23, 2009 9:34 am

4MR_NST wrote:Greeting,

We're looking to replace an HP8920 used to support Motorola conventional, trunking, and paging - both bench and field. I reviewed the slick sheet for the 3920 and didn't see a mention for POCSAG pager testing. Did I miss it? Is it an option?

Not at this time. The idea is that paging is served by the 294X series. Once again: If you want it, you need to convince rob.barden@aeroflex.com that putting paging into the 3900 would be worth doing.

4MR_NST wrote: Also, does (or will) the 3920 support other digital trunking formats for bench/field work? At $30k, I want to CMA should we need to support digital Mototrbo, digital Passport, MPT1327, Opensky, Tetra, or other protocols that could be available in the foreseeable future.

You must have missed that the 3900 family already supports MotoTRBO and TETRA.

As for digital Passport - we haven't had the sales on the Passport option for the 2975 we had hoped, so that has discouraged us from doing Passport on any other products.

As for Opensky and MPT1327 - once again, rob.barden@aeroflex.com - convince him there is a market.

4MR_NST wrote: I know that these digital technologies are out there, but don't know if special options are needed to test/align the radios or infrastructure. I read something about a "STAR" for TDMA but don't know what that means.[/b/]

Neither do I. Could you be confusing D-STAR for a TDMA system?

Post by **4MR_NST** » Mon Feb 23, 2009 4:57 pm

Wowbagger,

Thanks for setting me straight. I appreaciate you bearing with me on this! I have some follow-up questions for you, if it's OK.

The 3920 can take a POCSAG input from an encoder, correct?

Is "high stability" a given?

Can custom test setups be stored for rapid recall like the HP8920? We use them routinely for testing standardization, such as mts_tx, mts_rx, mod_comp, tp_13, tp_21, etc, etc. That sure is an advantage over, say, the older R2021D.

What spec can be looked at to evaluated RF leakage? Our HP8920 is known to desense a pager on the other bench in the screen room even with the RF output port terminated and output power set at 1uv. We've had it repaired during calibration and it fixed it for awhile, but it eventually came back.

About TDMA signal testing, I don't know enough to ask the right questions, unfortunately. I read that the service monitor must be capable of monitoring TDMA signals and have the capability of displaying a STAR pattern for a visual verification of digital integrity of the signal. If this sounds familiar, can you clarify?

I've contacted Aeroflex to find out more info. I appreciate your help in understanding the unit's capabilities ahead of time.

Post by **Wowbagger** » Tue Feb 24, 2009 8:46 am

4MR_NST wrote: The 3920 can take a POCSAG input from an encoder, correct?

Yes, external modulation is supported.

4MR_NST wrote: Is "high stability" a given?

All 3900 family have an OXCO that is factory calibrated. If by "high stability" you mean anything like a rubidium standard, then no - use an external reference.

4MR_NST wrote: Can custom test setups be stored for rapid recall like the HP8920? We use them routinely for testing standardization, such as mts_tx, mts_rx, mod_comp, tp_13, tp_21, etc, etc. That sure is an advantage over, say, the older R2021D.

Boy howdy can they be stored. The 3900 can save many thousands of setups. Setups can also be stored on a USB memory stick.

You can also program the 3900 to automate tests and store those automated tests as well.

We have several automated tests already developed.

4MR_NST wrote: What spec can be looked at to evaluated RF leakage? Our HP8920 is known to desense a pager on the other bench in the screen room even with the RF output port terminated and output power set at 1uv. We've had it repaired during calibration and it fixed it for awhile, but it eventually came back.

Egads.

The 3900 has passed all industry standard EMI conducted and radiated emissions standards - I don't recall off the top of my head what level of radiated emissions that would be, but it shouldn't be anywhere NEAR that bad. Of course, that does depend upon what cables you have attached - you use a crappy Ethernet cable and you are going to take us out of spec.

4MR_NST wrote: About TDMA signal testing, I don't know enough to ask the right questions, unfortunately. I read that the service monitor must be capable of monitoring TDMA signals and have the capability of displaying a STAR pattern for a visual verification of digital integrity of the signal. If this sounds familiar, can you clarify?

OK, now I understand. The "star" you are referring to is called a constellation pattern, and yes, the 3900 can do a constellation pattern for both P25, MotoTRBO, TETRA, and Motorola HPD.

4MR_NST wrote: I've contacted Aeroflex to find out more info.

Good, that will get you the best possible information.

Post by **4MR_NST** » Fri Feb 27, 2009 11:30 pm

Wowbagger, thanks for the info on the 3920.
Much appreciated.

GlennD · Post by **GlennD** » Thu Mar 05, 2009 6:45 pm

Yesterday I downloaded the new 1.75 update for my 3920. It promptly turned my monitor into a thirty-five thousand dollar brick!

I have to fault Aeroflex for releasing a update that was not fully tested from the download. The support person bricked three test units with the same download, but I must commend them for fixing the problem overnight. Despite my worries the fixed download fixed my unit.

The upgrade has some new neat features. You can now use the unit as a tone remote and for those with access to a time machine it will even spit out modat.

tvsjr · Post by **tvsjr** » Thu Mar 05, 2009 7:21 pm

GlennD wrote:it will even spit out modat.

Ooh. Truly wacktastic!

Post by **Wowbagger** » Thu Mar 05, 2009 7:45 pm

GlennD wrote:Yesterday I downloaded the new 1.75 update for my 3920. It promptly turned my monitor into a thirty-five thousand dollar brick!

I'll give Monte (the project lead) a bit a :o over that

Post by **Wowbagger** » Thu Mar 19, 2009 10:45 am

Well, now that it's being advertised in Urgent Communications I can stop biting my tongue about AutoTest II.

Several folks have asked about automatically tuning a Motorola radio - for some time General Dynamics had a program you ran on a computer, which controlled a GD2670 and the radio (via a RIB) and tuned the radio for you.

Aeroflex has that one better: we don't need the computer. We've finally announced Auto-test II, which can automatically both test and tune many Motorola radios, with the 39[012]0 controlling the radio directly.

http://www.aeroflex.com/auto-test/

----------
Obviously I've know about this for some time, but again - I cannot talk about things under development until they are announced.

ke2d · Post by **ke2d** » Wed Apr 15, 2009 8:02 pm

Quick question for Wowbagger or anyone else who might know -

I've been looking for another COM120B, and have noticed that there are two different looking front panels. One has a bolder, more colorful look, with sold blocks of blue and red surrounding the ANT, RF IN/OUT, and GEN connectors. The other has simple outlined boxes/lines around the same connectors, and the overall front panel labeling looks much smaller in font size.

Is one a newer version than the other? Which is newer? Any other info I should know regarding differences? Both are definitely COM120Bs, and NOT As.

Thanks
Eric

Post by **Wowbagger** » Thu Apr 16, 2009 4:08 am

ke2d wrote:Quick question for Wowbagger

The more colorful front panel is newer. You will also notice the keys changed from the hard plastic keys to a more rubberized key.

That was all done as a part of a cost-reduction on the 120 - the rubberized keys greatly reduced the cost to make the front panel.

Other than that, there is no difference where it counts.

Post by **escomm** » Fri Apr 24, 2009 9:36 am

Does the 3500 support autotune?

Post by **Wowbagger** » Fri Apr 24, 2009 9:40 am

escomm wrote:Does the 3500 support autotune?

If you mean the autotune app for aligning and testing Motorola P25 radios - no, the 3500 handheld test set doesn't support that.

Post by **escomm** » Fri Apr 24, 2009 11:34 am

Bummer, that would have sealed the deal. I don't suppose there are any plans for such an option? It sure would make service calls and PM calls infinitely easier.....

Post by **Wowbagger** » Fri Apr 24, 2009 12:29 pm

I cannot comment on unreleased development.

All I can say is "Offer enough dead presidents and it might happen."

Post by **Crack768** » Tue May 05, 2009 7:32 pm

why you cannot comment on unreleased development.? Is there any problem?

_________________
Vending Machines

Q2222A · Post by **Q2222A** » Tue May 19, 2009 5:07 pm

Wowbagger,

I've been using the Autotest II software for XTS5000 today and I find it a great application and product. My 3920 is getting really happy these days. I wonder if there are any scripts that have been written or can be developed to have Autotest II also align the analog portion of the radio with the use of the RLN4460A test box?

This would be a great addition.

Thanks for your thoughts,

Dennis

Post by **Wowbagger** » Mon Jun 08, 2009 10:56 am

Q2222A wrote:Wowbagger,

I've been using the Autotest II software for XTS5000 today and I find it a great application and product. My 3920 is getting really happy these days. I wonder if there are any scripts that have been written or can be developed to have Autotest II also align the analog portion of the radio with the use of the RLN4460A test box?

This would be a great addition.

Thanks for your thoughts,

Dennis

Sorry for the late reply - for some reason this message didn't show up in the "new messages" list.

I'll kick your comment to the folks who can authorize development - as far as I can see there is no reason we couldn't automate that *UNLESS* we weren't given the commands to the radio for analog tuning.

fineshot1 · Post by **fineshot1** » Tue Jun 09, 2009 10:47 am

OK - I've got a IFR 1900CSA and on the T/R port it takes about 2.25mV to open the receiver(its noisy at this point) at minimum squelch.

This cant' be normal - right?

I know this port terminates into the power termination mechanical assembly. When I ran a power(1mW) input test on this port it tested real close
so I'm thinking either I'm doing something wrong(setting error or something else simple) or there is a problem with the Term Assy.

Any checks I should do you can think of?

TIA.....

Post by **Wowbagger** » Tue Jun 09, 2009 12:41 pm

fineshot1 wrote:OK - I've got a IFR 1900CSA and on the T/R port it takes about 2.25mV to open the receiver(its noisy at this point) at minimum squelch.

This cant' be normal - right?

I know this port terminates into the power termination mechanical assembly. When I ran a power(1mW) input test on this port it tested real close
so I'm thinking either I'm doing something wrong(setting error or something else simple) or there is a problem with the Term Assy.

Any checks I should do you can think of?

TIA.....

No, that's not right: that would be about -40dBm, and you should be good to -80dBm easily.

Can you check the levels coming OUT of the port - i.e. generate -30dBm and verify that the level coming out is -30dBm?

Can you check the antenna port sensitivity?

If the power levels measured by the detector in the power term are good, then I'd suspect a bad connection from the power term to the receiver module, or a bad switch on the receiver module.

fineshot1 · Post by **fineshot1** » Wed Jun 10, 2009 5:10 am

Wowbagger wrote:
fineshot1 wrote:OK - I've got a IFR 1900CSA and on the T/R port it takes about 2.25mV to open the receiver(its noisy at this point) at minimum squelch.

This cant' be normal - right?

I know this port terminates into the power termination mechanical assembly. When I ran a power(1mW) input test on this port it tested real close
so I'm thinking either I'm doing something wrong(setting error or something else simple) or there is a problem with the Term Assy.

Any checks I should do you can think of?

TIA.....
No, that's not right: that would be about -40dBm, and you should be good to -80dBm easily.

Can you check the levels coming OUT of the port - i.e. generate -30dBm and verify that the level coming out is -30dBm?

Can you check the antenna port sensitivity?

If the power levels measured by the detector in the power term are good, then I'd suspect a bad connection from the power term to the receiver module, or a bad switch on the receiver module.

The antenna port sensitivity is better than spec(-80dBm) at about -92dBm

When I first generated at 0 dBm level from the 1900 T/R port into a GD 2670B it was about +.8dBm but after about an
hour it slowly moved to about -1.5dBm. I have the covers off the 1900 so I am speculating that it is not cooling very
well and may be overheating - not sure about that but all the modules seem very warm.

I have some more checking to do but there does not seem to be any bad connections as far as I can tell.

Post by **Wowbagger** » Wed Jun 10, 2009 5:57 am

fineshot1 wrote: The antenna port sensitivity is better than spec(-80dBm) at about -92dBm

When I first generated at 0 dBm level from the 1900 T/R port into a GD 2670B it was about +.8dBm but after about an
hour it slowly moved to about -1.5dBm. I have the covers off the 1900 so I am speculating that it is not cooling very
well and may be overheating - not sure about that but all the modules seem very warm.

I have some more checking to do but there does not seem to be any bad connections as far as I can tell.

OK, I'd suspect either the connection from the power term to the receiver or the receiver's internal switch - I don't think you've blown anything past that switch in the receiver, nor do I suspect a bad connection within the power term, given what you've said.

fineshot1 · Post by **fineshot1** » Wed Jun 10, 2009 6:10 am

Ok than it must be a bad switch in the receiver cause the cable(marked #4) between them seems to be fine although I have nothing to swap with.

Whats the diff between FM1, FM2 and FM3 in the receiver setup on the screen choices. FM1 & FM2 seem to have better results but FM3 does not.

PS - I'd love to send this in to be properly checked out and fixed but just can not afford it. I will just have to live with this problem.

Post by **Wowbagger** » Wed Jun 10, 2009 9:08 am

fineshot1 wrote:Whats the diff between FM1, FM2 and FM3 in the receiver setup on the screen choices. FM1 & FM2 seem to have better results but FM3 does not.

The IF bandwidth. IIRC FM1 is 15kHz, FM2 30kHz, and FM3 is 300kHz, but I'd have to ask Bonnie or Carl to be sure.

So as you widen the bandwidth you pick up more noise.

fineshot1 · Post by **fineshot1** » Wed Jun 10, 2009 9:51 am

OK - many thanks dave.

Oh - and I keep forgetting to ask this - when I first power up the 1900 there is the one beep and then the screen scrolls its IFR logo.
Then after that there are two beeps and then 4 beeps. Is this an error indication as well?

Post by **Wowbagger** » Wed Jun 10, 2009 11:59 am

fineshot1 wrote:OK - many thanks dave.

Oh - and I keep forgetting to ask this - when I first power up the 1900 there is the one beep and then the screen scrolls its IFR logo.
Then after that there are two beeps and then 4 beeps. Is this an error indication as well?

No, quite the contrary: each of those beeps indicates a part of the boot sequence. IIRC on the 1600 family, it's:
1 beep: Main CPU up
2 beeps: Smart boards up
4 beeps: Application started.

fineshot1 · Post by **fineshot1** » Wed Jun 10, 2009 12:17 pm

Great - that is good news for me. Again - thanks for your help. I do love this model although I hate how heavy it is.

Post by **Wowbagger** » Wed Jun 10, 2009 1:30 pm

fineshot1 wrote:Great - that is good news for me. Again - thanks for your help. I do love this model although I hate how heavy it is.

Could be worse: the military version is the box you have, plus a second pizza-box on the bottom for radio interconnect, plus a second pizza-box that is a many-hundreds-of-watts power supply for the UUT - even the Marines consider it a "two-man lift".

GlennD · Post by **GlennD** » Tue Aug 04, 2009 11:35 am

I just downloaded a new version firmware for my 3920. I am afraid to install it until I hear some good news from others. The last one turned my unit into a brick until they posted a fix.

misawatech · Post by **misawatech** » Wed Aug 05, 2009 9:03 pm

Thanks for your offer and assistance.

My 2975 is having problems. It takes quite a while to boot so I finally sat with it and watched the screen. After some time it scrolls a number of error messages about a file i/o error, but finally boots and is usable. I've been talking to someone from Aeroflex and they tell me the hard drive is having issues again and that they've seen these issues with more than one of the relpacement drives they've been using. They say the error is at the part of the drive that records the setup files and they prefer I wait for them to find an older drive whcih they will send me with preloaded software. They say the bios is affecting the new drives as it is not designed to recognize drives larger than 30GB. As a side business I fix computers and know anything smaller than 60 GB is hard to find now. If you do find one there is very little chance it is good or will last much longer.

Before I go too much off track here are my questions. What is the largest drive the system can handle without losing information?
Would I be able to ghost the drive to another, smaller, drive? I have a couple 26 and 30 GB drives at home that still test good.
What other options would I have for getting this back in working order?

I don't want to go around the person I have been working with but I just had the drive upgraded last year. The first drive had many other boot issues so they sent me a replacement. Now I am having problems again. The rig was working fine until I asked for the upgrade to facilitate addtiions of DES/AES capabilitiies and autotesting.

Thanks for your help

Post by **Wowbagger** » Thu Aug 06, 2009 3:30 am

The reason this isn't being solved quickly is that we are still working on trying to figure out all the nuances ourselves - in fact, while we know what is happening, we don't know why, we don't know when (as in, we cannot reproduce it on demand), and so it is very difficult to work out the details.

I wouldn't recommend Ghosting the drive: the file system that VxWorks uses is a bastardization of the FAT format, and I'm not sure Ghost knows enough about it to make the needed changes to the partition table and file systems when you do the copy.

What I'd suggest you do is offer to send one of the smaller drives in, and let us do the install to that drive. I'm not saying Customer Service will take you up on that, but....

This problem has been a real :o, and is one of the many reasons I wish I'd not selected VxWorks for the 2975 (and why we are moving to Linux on future projects like the 3900).

Post by **alex** » Thu Aug 06, 2009 7:11 am

Wowbagger -

Could he not use DD to copy the drive? I don't know if you can get a prompt on a 29XX monitor or not.

-Alex

Post by **Wowbagger** » Thu Aug 06, 2009 8:26 am

alex wrote:Wowbagger -

Could he not use DD to copy the drive? I don't know if you can get a prompt on a 29XX monitor or not.

-Alex

1) DD would NOT adjust the sector counts for the various partitions - that would corrupt the file systems.
2) As I said - the 2975 runs VxWorks, not Linux - and VxWorks doesn't have DD.

Post by **alex** » Thu Aug 06, 2009 8:31 am

Wowbagger wrote:
alex wrote:Wowbagger -

Could he not use DD to copy the drive? I don't know if you can get a prompt on a 29XX monitor or not.

-Alex
1) DD would NOT adjust the sector counts for the various partitions - that would corrupt the file systems.
2) As I said - the 2975 runs VxWorks, not Linux - and VxWorks doesn't have DD.

Gotcha. I'm not super familiar with VxWorks. I guess where I was going is if you have a 30gb HD and another 30gb hd and you want to make a copy of it - you can (in theory) use DD to accomplish that task.

-Alex

Post by **Wowbagger** » Thu Aug 06, 2009 8:41 am

alex wrote:Gotcha. I'm not super firmiliar with VxWorks. I guess where I was going is if you have a 30gb HD and another 30gb hd and you want to make a copy of it - you can (in theory) use DD to accomplish that task.

-Alex

Yes, IF the two drives had the same geometry (# sectors, reported # of heads/cylinders/sectors per track), then it would work.

VxWorks's IDE interface routines don't play well with LBA - but since the BIOS really wants to use LBA to boot, it gets tricky.

firemedic · Post by **firemedic** » Thu Aug 06, 2009 8:55 am

Would an IDE to CF style adapter with a 32gb CF card work in the unit? Just wondering.

Post by **Wowbagger** » Thu Aug 06, 2009 9:23 am

firemedic wrote:Would an IDE to CF style adapter with a 32gb CF card work in the unit? Just wondering.

Possibly - we'd looked at that at one point.

You'd still have to get the software loaded onto it - you wouldn't want to try to copy it from another drive.

Post by **nucknfutz** » Sun Aug 16, 2009 11:21 pm

Ok, so I called Aeroflex the other day, as I have a Com120a with some issues...the unit was one of our spare units that rarely got used....To date it has less than 100 hours on it and has started to fail miserably. Our other 120a has closer to 4000 hours and is failing dev/distortion meter tests the odd time. The unit with 100 hours on it is also intermittantly failing these tests, and is also intermittantly giving a power term temp alarm. I pulled the unit apart, and found that the baseband board has leaky capacitors all over it, and is causing malfunctions due to this. I ordered all replacement smt capacitors for the baseband board, and have the preliminary maintenance manual but it doesn't get into specifics. I was also told by a service center that used to repair IFR that the Com120's had a faulty run of capacitors on the boards, and most leaked in all units that had the defective caps. Upon phoning Aeroflex directly they told me that they no longer sell a service manual for the 120 series, and that i could purchase schematics at $25 per page.....at roughly 7 pages per module..ie the baseband board...that is rediculous, and makes me not want to consider buying another aeroflex unit, as this problem was apparently known by ifr at the time, and they did nothing to let people know that their batch of leaky caps was ruining $20k+ pieces of test equipment...so if anyone has schematics of the baseband board let me know, and also for the guy who wanted to repair his ac power supply, I repaired ours once, there are 2 power "bricks", and the unit runs some weird modules..at least in the 120a...it takes 120v and ups it to 300vdc via 1 module, and then theres a second module that takes 300vdc in and drops in to 15 or 12 and fires the supply I believe, its been a while since i worked on ours...and never had any manuals either...but in my opinion is a rediculous setup..and Aeroflex wants over 1000 for both modules...so you may just want to keep running it off of 12vdc...

Post by **Wowbagger** » Tue Aug 18, 2009 5:16 pm

I sympathize with your opinion. I've repeatedly told the Marketing/Communications manager and the Service Manager that we should put the manuals for older gear up on the web site.

However, there are four things working against that:

1) there is the perception that people will pay money for those manuals - as you've run into.
2) There is the perception that there might be "value" to our competitors in those manuals - that our competitors might learn techniques from those manuals that they may not already know.
3) The Marketing guys don't WANT those old units on the market - they don't want our new gear to be competing with our old gear. That's why they are doing everything they can to get those old units off the market, such as offering trade-ins and crushing the gear.
4) Inertia, plain and simple. We never did it in the past, so we won't start now.

The leaky caps were not deliberate: many manufacturers got nailed by those - in fact I had that same problem on my COM-120B. It's a lot like a car recall - unless the issue at hand is a safety issue (as in "somebody could get killed and we would be liable") car manufacturers won't exactly go out of their way to notify car owners that there is a fix that they probably should get. I'm not saying that's right or wrong, but that is business.

I'll say the same thing I've said before: If this is something that would prevent you from considering buying Aeroflex gear in the future, then write a polite letter, stating that fact and detailing (truthfully!) your average budgets, and send it to
Rob Barden
Director of Product Marketing
Aeroflex Wichita
10200 West York Street
Wichita, KS 67215-8999

I just design the gear, I don't get to make decisions about Sales, Support, or Marketing - my job title is Principal Software Engineer, not Director of <mumble>.

GlennD · Post by **GlennD** » Tue Aug 18, 2009 6:28 pm

Well, after hearing no screams!, I gathered up my courage and tried to upgrade to the latest firmware. My 3920 does not see the flash drive on any port. A USB mouse works fine.

The other 3 units we have have not been upgraded to the last release and they see both flash drives just fine. Since they are not normally turned on due the EXCESSIVE fan noise, the users will not let me install the latest release.

Did the 1.75 release do something to the USB ports? The other units are running the release before 1.75 and they see the drives with no problem.

Post by **Wowbagger** » Wed Aug 19, 2009 3:50 am

I know of no reason that would happen - that's pretty much handled by the Linux kernel, and we don't make a lot of changes there.

I'll ask Monte about it when I get in this morning.

ke2d · Post by **ke2d** » Sun Nov 08, 2009 9:16 pm

Two quick questions -

1. Can a COM120A power supply be installed into a COM120B? I have a 120A that is dead. It starts up, the ant and gen port leds blink, I see the screen for a split second (and I mean split!!), then it powers down. I also havea 120B that works perfectly, other than the fact that the AC side blows the fuse, but I have been running it just fine off of the DC input for years. Now that I have the dead 120A for parts, I'd like to swap the supplies if they are interchangeable.

2. Anyone have a schenatic for the power supply module, or at least the pinout and voltages on the pc style multi-pin jack?

Looking at the supply, it looks like one board has converter modules, then simply has one output that feeds the densely populated board with the multi pin connector. Any idea what the voltage is that leaves the converter board and enters the more complicated board via the two-pin anderson plug? Since my supply is working fine on 12vDC input, the problem has to lie in the converter board, as there is only one output leaving that board and feeding the second board. Is the 12v dc being upconverted to some higher level on the more complicated board?

Oh, to have a diagram of this damn thing!

Eric

Post by **Wowbagger** » Mon Nov 09, 2009 4:30 am

You should be able to exchange the two power supplies - the system voltages didn't change between the A and the B.

The actual design of the power supply did change, naturally - we went to the more off-the-shelf converter modules when they became available.

(I've tried to dig through the archives to find the power supply information myself, but they haven't been pulled into the digital documentation system, and I don't have the time to go digging through the paper archives).

ke2d · Post by **ke2d** » Mon Nov 09, 2009 8:29 am

Well,

looks like I have two bad supplies, albeit different issues with each. I guess my next step is to try and make one good one out of the two.

I'm sure you are quite busy Wowbagge, but if you ever find a moment near that archive of paper manuals, I would greatly appreciate any help you could send my way in regards to some sort of ps documentation.

I did notice that the module was labeled CEI St. Petersburg FL, but a web search for CEI was fruitless.

Thanks
Eric
KE2D

Post by **hdr55** » Mon Dec 14, 2009 11:06 pm

Greetings from Western Australia,
We received a 3920 a few weeks ago, and I had a play with the smartnet option, testing a Motorola XTL1500.
During the tests the 3920 would loos it's system ID and revert back to default eventough it was stored, after that it would lock-up, e.g. could not switch it off.
Another strange thing is when testing in "trunked simulation", on TG 1, using a VM modulated with1 kHz, the RX audio in the UUT would sound fine, but on any other TG it would pulse on a 2 Hz interfal, as if the connectone is switch on and off.

ke2d · Post by **ke2d** » Wed Jan 13, 2010 6:37 pm

Well,

I was finally able to troublshoot and diagnose the failure of original bad COM120B power supply. The Vicor DC-DC converter module which accepts the 300Vdc out of the Vicor AIM module has failed. The design as is as follows: A Vicor AIM module converts 100-250VAC into a 300VDC source, which is then fed into a Vicor VI-262-EU DC-DC converter, which is rated 300vDC 250w input to 15vDC 200w (13a) output. The 15V output then passes thru a pair of relays, which select which power source feeds the rest of the supply module - optional internal battery, external DC via the powerpole jack, or the internal AC supply's 15v output.

The VI-262-EU module is almost impossible to find stateside unless I special order it from Vicor @ $185 a piece. I have sourced a few thru Asian parts brokers for around $100 shipped, so that will be the worst case scenario, but here is my question -

The COM120B draws about 6 amps or so when I feed it via my Astron bench 12vdc power supply. The original Vicor module was rated at 13A. I can find a VI-262-CV, which is rated at 150w (10A), stateside for a reasonable price. Do you think this 10A module will adequately handle the load over time (I'm thinking 10A should be enough since the monitor only draws 6A from my external supply)? Or should I bite the bullet and import in an original size 200w module? The smaller rated module is a drop in replacement (physically).

Eric

ke2d · Post by **ke2d** » Wed Jan 13, 2010 6:41 pm

Anyone have any issues with power-up of their 2975?

Here's what I notice with mine -

When I first plug it in and turn on the rear power switch, the display will turn on all white, and the rear fan starts, but the motherboard never starts. While leaving the unit plugged in, if I simply turn the rear power switch off, then back on, the unit boots up just fine.

I can then turn the unit on and off just fine using the front panel soft power button.

If I then turn off the rear switch, then a day or so later return and turn the rear switch back on, the unit does the same thing (fails to start the first time). The cycling of the rear power switch off then on just once makes the unit start just fine once again.

When this happens, it is almost as if the motherboard failed to recognize the power-up sequence, yet the display powers up. Almost like a motherboard pc whose bios is set to "last known state" in the power management settings, and is then plugged in or turned on via power strip switch, without hitting the boot button. Your monitor turns on, but the pc doesn't start...

I replaced the CR2016 coin batter on the motherboard with no change. I even made sure to set all of the bios settings as per the Aeroflex service bulletin regarding the battery replacement.

Anyone else notice this strange behavior?

Post by **Wowbagger** » Thu Jan 14, 2010 4:33 am

hdr55 wrote:Greetings from Western Australia,
We received a 3920 a few weeks ago, and I had a play with the smartnet option, testing a Motorola XTL1500.
During the tests the 3920 would loos it's system ID and revert back to default eventough it was stored, after that it would lock-up, e.g. could not switch it off.
Another strange thing is when testing in "trunked simulation", on TG 1, using a VM modulated with1 kHz, the RX audio in the UUT would sound fine, but on any other TG it would pulse on a 2 Hz interfal, as if the connectone is switch on and off.

Could I get you to report that to Customer Service officially? I'll call this to the project lead's attention, but it would be a WHOLE lot easier if we had a formal report.

Post by **Wowbagger** » Thu Jan 14, 2010 4:38 am

There are 2 reasons for the hefty supply:
1) to support the OCXO module, if installed
2) To support turn-on currents (steady-state load may be 6A, but the turn on current is higher).

It's your unit, and you are free to do with it what you want, but ask yourself this: can you replace your 120 for $185? In other words, even if you get your lower spec Vicor free and save $185, if that module fails and takes out the rest of the unit, have you really saved anything?

In fact, one of our techs took his 120 and put in more recent Vicor modules that were rated for 25 amps, and modified the system to feed DC out the Powerpole connectors - so he now has a built-in UUT supply for radios of 50W or less.

Again, I'm not recommending doing that, just noting it.

Post by **Wowbagger** » Thu Jan 14, 2010 4:41 am

ke2d wrote: When I first plug it in and turn on the rear power switch, the display will turn on all white, and the rear fan starts, but the motherboard never starts. While leaving the unit plugged in, if I simply turn the rear power switch off, then back on, the unit boots up just fine.

It sounds to me like the power supply isn't sequencing correctly, so the backlight on the display comes up, but the main CPU isn't coming up - hence the white display.

My suggesting would be to get the PSU replaced - that is definitely NOT normal behavior.