Aeroflex (IFR) Questions

Post by **xmo** » Sun Apr 02, 2017 5:53 pm

ljohn6 asked: "I have a question about a 2975 as this was one of your baby's I figured you would be able to tell me can the Hard drive be updated to an SSD and is there anyway to back it up or make an image. for that day when our hard drive decides its lived long enough. "
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------

I found the timing of this question to be interesting because I had the same question.

I have several test instruments from various manufacturers that have their operating system on hard drives. Not wanting to have an expensive piece of equipment turned into a boat anchor for lack of a hard drive, I have been backing these up in various ways.

Now that the parents have disowned the 2975 it falls into that category of concern and I planned to tackle it next. My 2975 is Aeroflex branded, serial number 5980027XX. It uses a 40 GB Hitachi Travelstar drive.

First I obtained another hard drive of the exact same part number (cheap on ebay).
Then I opened the instrument (18 screws!) and removed the CPU card to access the hard drive(4 more screws).
Using a USB adapter I connected the drive to a PC running Windows 7.
Using Selfimage I copied the entire drive to a file.
I then swapped the spare drive onto the USB adapter and copied the image file onto the new drive.
I put the duplicate drive into the 2975 and it boots and runs fine.

SO... Mission accomplished. I now have the 2975's original drive as a spare and I have a known working image file and process.

I may consider trying an SSD replacement, but for now I don't have to worry about losing the 2975 because of a drive failure.

Post by **ljohn6** » Sun Apr 02, 2017 7:38 pm

XMO
thats great info, I was thinking of doing the same thing as well, I was also thinking of SSD you can buy 30 - 40 gig drives in an IDE configuration, I did use one in a toughbook and it worked fine and it is much faster but i guess before i experiment i better find a drive that matches mine make a working copy, then try an SSD. but the info you posted very Valuable
Thanks

Post by **Tech255** » Thu Apr 06, 2017 6:43 am

Wowbagger
I have a com120a and i was wondering what is required to get the SSB option is it a board that i could build or does it contain firmware, i presume as the model is no longer supported that the option is no longer available.
Also is the service manual still available or have they stopped selling that as well.
Thanks

Post by **Wowbagger** » Fri Apr 07, 2017 4:07 am

To get SSB you need hardware - the SSB mixer and filter. That's going to be tough because the only way you are getting that would be to find one on the used market, as the whole 120 family is totally out of support now.

Post by **roxbox** » Tue Apr 11, 2017 9:29 am

ka3yan wrote:So no one thinks I'm ungrateful, I did speak to KE7JFF offline. Still no joy finding an inexpensive non-burned in screen, but he thinks he might have a lead.

I'm still looking for any manuals for the COM-120B.

I got a display for the 120b from eldisplays.com/el560-400-lp/ and it worked great but it has been over a year and cost me 995.00 usd.

I also have the service manual in pdf form but the diagrams are hard to read and do not know if it is ok to post that here.

I have 7 units I am working on right now and only one is in working condition at this time. I would like to know if I take pics of the options and sn# if I could get the codes for them as I know of two that have lost them and by the time I get to the others, I worry that they may have too.

Post by **Wowbagger** » Wed Apr 12, 2017 3:29 am

The options printed on the back of the unit don't help me - those were the option numbers sales used. Unless somebody else has a chart telling me what they mean, I have no means to convert them to the software level bit map that I feed into the tool to create the options codes.

The single best thing anybody can do is do a "*idn?;*opt?" over a serial port BEFORE the RTC fails - and note down what IF filters are installed.

Post by **roxbox** » Wed Apr 12, 2017 4:37 am

I believe the option codes on the back are as follows:
O the first letter is an O not zero for Options
01-internal battery
02-ocxo
03-30 khz if filter
04-variable audio generator-2
05-generate amplifier
07-data generator/bit error rate (ber) meter
08- single sideband receive filter
09-rcc signaling formats
11-audio/digital signaling formats (ccir,ccirh,ccirh4,eea,eia,natel,...)
12-spectrum analyzer tracking generator
13-gpib
14-clearchannel ltr
15-amps cellular testing
16-edacs trunking
17-mpt-1327 trunking

the above is from the 120C ops manual
I have one unit with o010204050708091112 that I think may turn on, I will do the "*idn?;*opt?" and make sure later today.

Post by **roxbox** » Wed Apr 12, 2017 8:22 am

got that unit to power on. maybe this will help with the mapping of the bits to the sticker options.
!*IDN?
IFR SYSTEMS INC,COM-120B,8645,0400-0400-0400-0301

!
!*OPT?
28927

Post by **ljohn6** » Thu Apr 20, 2017 7:33 pm

Well my shop 2975 hard drive failed this morning before i was able to get a image Wowbagger what do you think my chances are of getting a new image from Cobham now that they no longer do the repairs? there goes my passport testing, my p25 and my ltr options. Heck i don't even remember all the options it had as it was just transferred to me from another office when they found out no one used it. Wowbagger if you can chime in with your suggestions boy that would be great.
Thanks

Post by **ljohn6** » Fri Apr 21, 2017 8:16 pm

"I contacted Cobham today and no go on a image. I was told as they stopped support at the end of last year they could not help and suggested i call around and see if any of the independent service centers could help. So was not able to find any. now the company has a very expensive boat anchor. I understand the need to move forward and no company can afford to support products for ever. i have to tell customers too that they have a radio that is no longer supported, i get it. I just wish I had made a backup before this.

[you are toeing the line here on what you are saying and asking - we respectfully ask that you do not cross it. -Alex]

Post by **Audiotron** » Sun Apr 23, 2017 12:48 pm

Hi Friend

Some Binary files and hard disk data of some instruments that I have, I am saving just for not having more support in this type of problem, and a pity an excellent instrument stopped for lack of a File

Hugs

Post by **Wowbagger** » Sun Apr 23, 2017 5:48 pm

(speaking of toeing the line - I too have a line that I'm not going to cross)

Have you tried getting a Linux machine, and running ddrescue to see if you can recover enough of the system to work?

Post by **Audiotron** » Sun Apr 23, 2017 6:17 pm

Hello Wowbagger ,

you indicate some linux in particular to perform this type of procedure,

Hugs and a great night .

Post by **ljohn6** » Sun Apr 23, 2017 7:49 pm

I certainly would not ask anyone to violate the law,copy right protections or patents. Nor did i want to insult anyone and if i did, i apologize. I was able to get an image off the hard drive but the software has been damaged i am guessing a platter got damaged, we have preserved the original hard drive as is, so we do have the original. i have not tried to use DDrescue, i will burn the image we have to a new drive i ordered and mount the drive in linux and give it a try thanks for the suggestion, is there a way to install the update software i downloaded from cobham without the machine booting ? the error i receive when i start the machine is undefined symbol: (null) and that error repeats several times and then i get this at the bottom of the screen, loadaoutlib error: can't add '(null)' to system symbol tab
then it stops, it does load the network card but after that nothing.
Again I am sorry to anyone who may have been offended.
John

kb6nzv · Post by **kb6nzv** » Sun Apr 23, 2017 8:49 pm

Hello Wowbagger.

Would you mind confirming which version of VXworks you had chosen for the 2975 service monitors?

Respectfully,
Phil KB6NZV

Post by **Wowbagger** » Mon Apr 24, 2017 3:41 am

You need to use ddrescue on the original drive - the idea behind ddrescue it that it will try to read all sectors, at first skipping bad sectors in order to recover what it can easily recover, then going back and trying to read the the damaged sectors with the tenacity of a terrier going after a mouse under the couch at 3AM.

Post by **Audiotron** » Mon Apr 24, 2017 7:56 am

Hi to All ,

I never thought of going against copyright, I just think of saving vital files to the proper functioning of instruments that are now discontinued

Hugs from Brazil to everyone !!

Post by **Wowbagger** » Mon Apr 24, 2017 11:06 am

kb6nzv wrote:Hello Wowbagger.

Would you mind confirming which version of VXworks you had chosen for the 2975 service monitors?

Respectfully,
Phil KB6NZV

5.3.1, with PC hardware BSP V1.2

Post by **xmo** » Tue Apr 25, 2017 1:23 pm

Continuing with the earlier concept of replacing the 2975's hard drive with an SSD, I did try to replace my 40 GB drive with a 64 GB Kingspec SSD.

No joy. I tried to put the image on the SSD with two different PC apps and I also tried a physical cloning hard drive workstation. This pretty much confirms what Wowbagger told us.

He also told us: ".... So for each size drive, I had to work out what lie the BIOS would tell, and update the code in the system to use that lie as needed..." That suggests that there have been several sizes of hard drive used in the 2975 over the life of the product.

Also, I am sure that there are at least two very different processor cards. This suggest that for a third party to support the 2975 they would need a lot of different data.

At this point in the life of the 2975 several questions come to mind.

Has Cobham provided any independent instrument repair firms sufficient technical data to enable them to support the 2975?

If so, which firms are able to assist users now and in the future?

Does Cobham have any path for an instrument owner to enable additional options they might need such as LSM?

Is important information such as the instrument's serial number, enabled options, and/or calibration data stored on the hard drive or in non-volatile memory?

Post by **Audiotron** » Tue Apr 25, 2017 3:43 pm

I own a 2975 without the HDD I would like to find the internal content of it to make these experiences !!

Post by **ljohn6** » Wed Apr 26, 2017 8:56 pm

Hello Wowbagger we did an image with DDrescue and had no luck i checked with a service center in grass valley we do business with and they contacted Wichita who had them direct me to a Canadian Company. So far i have hit brick walls. looking at the file structure the files and folders seemed to be scrambled. but not having a good file to compare it to i have no basis to verify. is there anything else i can try ? Thanks for any help you can provide
John

Post by **Wowbagger** » Thu Apr 27, 2017 3:08 am

I hate to say it, but not really. I don't have a 2975 image, and the FTP server that had the images on them was decommissioned some time ago. All I can suggest is that you try to find another 2975 owner you could meet up with.

Post by **ljohn6** » Thu Apr 27, 2017 5:13 pm

Wowbagger

If i load a image from a different machine. I am guessing my options are gone. which if the machine comes back to life i could live with that. But if my options folder is still in tack which it looks like it is will it recognize it ? what would be my next step.
Thanks Again John

Post by **Wowbagger** » Fri Apr 28, 2017 3:51 am

There is a single file that contains the options information for the unit, and it is tied to the CPU. If that file is not corrupted, and you place that file on the image, you should be fine for your options.
If you still have the original option install files from Aeroflex, once the unit has had the proper serial number reset, you could re-apply the options.
Otherwise, your options will not be valid - I deliberately designed the option system to account for people trying to clone hard disks to allow many units to share one serial number and thus share options.

I also had a thought yesterday while talking about this problem with Carl. The issue with the different hard disks is getting the right geometry data into the boot sector. I realize that it might be possible to write a program in DOS that would read the geometry data from the BIOS (thus getting the correct lie from LBA) and write that to the hard disk's partition table, and thus enable the system to boot from any size hard disk. It would be a bit tricky, as I don't have a working 2975 to test on, but in theory it could be done and that would enable any size hard disk to be used.

Post by **xmo** » Fri Apr 28, 2017 12:33 pm

Here is a crucial question.

Early 2975's have a two piece processor consisting of a carrier card that plugs into the 2975 motherboard and a piggyback computer with a jillion cross-connects.

Later units have a one piece card. Several different hard drive sizes were apparently used.

Will an image from a 2975 with one processor type and drive size run on a 2975 with a different style processor and/or drive size?

Post by **Audiotron** » Fri Apr 28, 2017 1:27 pm

I believe that with the production time what was changing was the size of the HDD, with one or the other improvements, the biggest current problem when having the internal contents of the HDD would be the physical size of the HDD compatibilize with the BIOS

Post by **ljohn6** » Fri Apr 28, 2017 6:27 pm

Wowbagger if you would like a unit to test with i would be happy to loan mine.

kb6nzv · Post by **kb6nzv** » Thu May 04, 2017 7:56 pm

Hello Wowbagger.

In what year was the 2975 first available?

Thanks,
Phil KB6NZV

kb6nzv · Post by **kb6nzv** » Thu May 04, 2017 8:01 pm

Hello Wowbagger.

As a follow-on to my prior question, I'm curious about your lessons learned on test equipment design over the years. Do your newer designs avoid fragile (spinning) storage media in favor of more easily maintained solid state media?

Thanks,
Phil KB6NZV

Post by **Wowbagger** » Fri May 05, 2017 3:32 am

Things are moving solid state, but I don't know that I would say solid state is much more durable than spinning rust - I have seen SSDs fail from too many writes just as much as I have seen disks fail.
The single biggest lesson learned is to use off-the-shelf as much as possible - buy your CPU boards, buy(1) your OS, buy any libraries, buy tools. Do what you do best and buy the rest. Design in the biggest FPGAs, fastest microcontrollers, largest flash you can, because by the time you go to market, it will be middle of the road rather than "not recommended for new designs". Where possible, make the hardware modular so that as things go obsolete you can replace them with minimal problems. Make the software modular so that it can accommodate the changes in hardware. Security must be a design requirement that management buys into - backdoors are to be banned because they will leak. The only thing that is constant is change - don't fear it, use it.
(1) or use FLOSS, and abide by the licenses.

jry · Post by **jry** » Fri May 05, 2017 5:37 am

just a quick point that all storage media have a life span. SSD is no exception.
Spinning disk can fail in storage from lubricant migration and SSD's will effectively become discharged overtime ( especially the consumer grade MLC " multi level cell " drives )
Even tape has a limit on data retention due to numerous issues with the binder and media.
These are often measured in a number years but the limits are still there.
SO just because you dump an image to some backup media be aware at some point that media will fail as well.
The adage that there is no such thing as too rich, too thin or too backed up applies.
Short version is refresh / test your backups periodically and have multiple copies. Having one backup is usually never good enough for any data of value.

Post by **Wowbagger** » Fri May 05, 2017 12:20 pm

And, when you make a backup image, get some form of checksum on it - MD5, SHA1, SHA256, something - and write it down. When you want to check the image, verify the checksum first. Make your life easier - write down the tool(s) used to make the backup (and the checksum) as well. Your future self will thank you.
Also remember that a backup isn't a backup until it is restored.

Post by **ljohn6** » Fri May 05, 2017 6:06 pm

I would like to add that it is best to have 2 backups and to keep one off site preferably in the cloud.
Cheers

Post by **k7dmk** » Sat Jul 01, 2017 12:40 pm

Wowbagger: I really need your advise. I have a recently obtained 2975 for our Ham club. The unit seemed to be working fine after purchase, but the CMOS battery was dead, and after what we read here, I also wanted a backup of the HD.

The unit was carefully disassembled, CPU board removed, battery replaced and the drive imaged with The SelfImage software. The s/n is 59800281X and it has the two part ETX style Kontron CPU module using a ULV Celeron 400. After re-installation, the bios was entered with an attached keyboard and the F2 key. The CPU temp seemed excessive at over 70 degrees C in the Hardware Monitor section of the bios, so the thermal pads were checked on the processor module and the one on the CPU replaced. That brought the temps down some.

The CMOS settings were all lost when the battery died. The unit did still seem to be fine with the dead battery, just no RTC. After the work, now the monitor will not finish the boot process. It get's really close, but hangs before entering the GUI. Another drive was installed with the working image and it does the same thing. So I don't thing it's a corrupted drive. It's like the start-up script is waiting for something to happen before proceeding, but it's not happening. All the power rails measure WNL and the led's on the CPU board show the rails are all present. The boards were all checked to be sure they were properly seated as were the coax interconnect cables.

On the other PCB's there are some led's too. I'm a bit concerned about the second to next board from the CPU. It has two led's in front and 4 on the middle top of the board. The two front led's are both on, but of the middle four, the one closest to the front comes on in the boot process, then blinks rapidly, the 2nd led never comes on, and the back two come on and stay on. Could it be that the board is not fully initializing and holding up the boot? Also, sometimes an error message box pops up after a long delay into the hang that says: Error: BBAudio Cmd timeout System=21 Subsys=149. The hang occurs during the "Restoring Instrument State" phase of the script. The YIG cal and IMBE stuff all complete without error and all the progress bars say "done" except the Inst State as mentioned.

I'm not sure I have the bios settings correct. A service bulletin was mentioned here that defines the proper bios settings. Do you have a copy of that. or can you shed some light on whether there are any reserved IRQ's that need to be set or any other specific parameters that need to be configured? I'm guessing it could be some kind of interrupt conflict causing the boot hang.

I have spent many hours trying to get this unit working again. I'm running out of ideas to try, Any suggestions would greatly appreciated. Our club budget is shot for several years and we have a very fancy doorstop right now.

BTW, do you know whether a service manual for the 2975 was ever produced and sold, or was it "internal only"? If so, are they available anywhere?

Thanks in advance,

Dave, K7DMK EVAR Club (K7EVR)

Al · Post by Al » Sat Jul 01, 2017 6:58 pm

Dave,
I purchased the service manual for my 2975 when I bought the unit. Sorry to say there's no really detailed information on hardware that would help you. There's a good system overview, no schematics, nothing on the modules including the CPU except - return to Aeroflex for service.
If you think that I can help with BIOS settings I'm willing to check my bios for the settings that you may have questions about.

Post by **k7dmk** » Sat Jul 01, 2017 7:59 pm

Al: Is there any information in the service manual about the led status indicators on the various PCB's? I'm not sure if they are all supposed to be on after the machine boots or not. A few of mine are off.

As far as the bios, mine has the Phoenix version and I was looking for any specific settings that need to be changed from the default values. Especially some of the stuff under PCI/PNP. You can overide the defaults for the IRQ's for each PCI slot and reserve certain interrupts for the ISA bus. I don't know which bus they may be using for the internals. There also also several devices like LPT that can be disabled or not and could cause conflicts.

Thanks, Dave K7DMK

Post by **k7dmk** » Sat Jul 08, 2017 6:17 pm

Does anyone have a 2975 with the 30G Hitachi Endurostar drive (HEJ423030F9AT00)? I need the Phoenix bios drive parameters for that HD. Is it set to "Auto" or "User" and if "User", what are the Cyl, Hd and Sector parameters that you have?

Thanks,

Dave K7DMK

Post by **k7dmk** » Fri Jul 21, 2017 9:19 am

Well I found that interrupts 10,11 and 15 need to be reserved for the ISA bus in the 2975 Bios. Auto mode seems to work OK for the HD setting. The IRQ settings solved the boot hang problem, but the CPU is still not entirely stable despite my best efforts. I found another identical CPU module (made by Kontron) same part number, it was an expensive purchase, but I can't use it until the serial number file is updated with the new CPU serial. The bootloader checks the CPU serial and stops the second phase of the boot. The unit only has one option, remote cal, so I'm not worried about losing options. I just want to reset the CPU serial. I'm sure there must be a utility to do this.

Dave, K7DMK

Post by **Wowbagger** » Sat Jul 22, 2017 3:43 am

No, the bootloader does not check the serial number. The serial number checks happen when the application is up. You don't have the disk parameters set correctly.

Post by **k7dmk** » Sat Jul 22, 2017 10:29 am

I have the drive parameters set to auto and lba mode in the bios and the original cpu module and drive boot fully now, but I could experiment with other settings. When I was having interrupt assignment problems, I did try multiple HD bios settings. It does seems to want lba mode enabled in the bios. I tried "user" mode using the auto discovered parameters with the same result. This particular drive is a 30G Hitachi HEJ421030G9AT00. It says 16383 Cyl, 16 hd, 63 sectors on the label.

The main problem I'm having is that the unit sometimes crashes during warmup, after it's already booted and into the main application. When the CPU crashes, the white LCD backlight comes on without any graphics, and it sometimes reboots, sometimes just hangs. I suspect the original CPU mezzanine card may have a bad BGA connection under one of the large IC's that becomes intermittent as the chip warms up. Generally, once the monitor has be on for a while, it's more stable. I have examined the mezzanine and carrier boards carefully for other bad solder joints and looked at the CPU inter-board connectors under a microscope and cleaned them with IPA.

With the new processor module installed, it stops at the blue screen before the main loader application starts. Not sure where the bootloader stops by definition and the 2nd phase starts, but that's what's happening with the same bios settings. It never gets to the white screen that shows the progress bars loading the software.

I have a bootable mem86 memory tester floppy and the new processor module will run the memory test for hours without crashing, but it won't fully boot. The original CPU module generally crashes running the same memory test, usually after a few minutes. That's what made me suspect the CPU module is bad.

What needs to be done with the software to use the new CPU module? The new module is exactly the same Kontron part number as the orignal board and has a low voltage Celeron 400 CPU with an Intel 815 chipset.

Thanks, Dave K7DMK

Post by **Darel** » Sat Sep 02, 2017 2:46 pm

I recently was given a very nice IFR Com 120A and an looking for a little guidance if someone would be so kind! Sometimes the front power button works fine, sometimes it won't and if I press it the srceen freezes and I need to turn it off from the rear. Other times when in either receive of gen mode the screen flashes every second or so. I am also measuring only 5 volts on the red/black wires that go to the battery ( since removed ). I have tried to read up and research these problems and the "baseband tray" keeps coming up ...where is it? I can see nothing labeled as such. Any help and guidance would be much appreciated!

Darrell / W6VHK

Post by **Darel** » Sun Sep 03, 2017 8:05 am

I managed to find the baseband tray ( cable laying on top of the label ) and found several leaky caps ...will replace them all. Still wondering why only 5V on the battery wires.

Darrell / W6VHK

jry · Post by **jry** » Mon Sep 04, 2017 8:09 am

The COM-120A has multiple boards on which the capacitors leak.

The baseband just happens to have a lot of them.

Replacing the caps are just the start since the leakage affects the surrounding components. Just replacing the caps by themselves may not fix all of the issues.

It also sound like the primary DC supply has failed as well. The battery charging runs almost directly off of the main supply so you should see a much higher voltage there. Maybe you got lucky and just the charging circuit failed.

You may want to run off an external DC supply via the rear connector and see if the voltage comes up or some of your power issues get resolved

The VICOR modules that are used in the primary supply are not cheap so I would get the main unit working on an external supply and save that repair for last.

And on top of that you may have some other failures mixed in with the capacitor failures. Most shops tended to run these into the ground and "minor" failures were often tolerated until the unit was almost completely non-functional.

The Real Time Clock is another failure point which can cause multiple issues including the loss of the installed options. The RTC fails mostly due to the internal battery going bad after almost 20 years of use.

After the HW is repaired and functional it then needs to be re-calibrated to be of much use.

It is not trivial to resurrect a COM-120A with the type of problems you are seeing.

I suspect you may now understand why they gave the unit away even if it was mostly used on the bench.

Unless you have the time and skills you may want to consider selling it for parts and acquiring a working service monitor or at least one more repairable.

The COM-120's are nice units when they are working but are a often a challenge to repair and support by most individuals.

Post by **Darel** » Mon Sep 04, 2017 9:06 am

Jry,

Thanks for your response and advice! This unit was actually a fellow hams pride and joy for quite a few years, I received it after he passed away. When I apply 12V to the rear the "applied" light is on in the front but the unit won't power up. It will power up from the rear switch with A/C applied and seems to boot up just fine but if I touch the front power button the unit freezes up and I have to turn it off from the rear. Funny, a few days ago the front power button started working and lasted for a day or so, now it is back to doing it again .....Very hard to diagnose with no service manual or schematics! I found it mentioned in another thread that the front power button is hard wired straight to the PS and noticed a couple of relays inside the PS, possible one is going bad?

Thanks,

Darrell / W6VHK

Post by **Darel** » Mon Sep 04, 2017 3:52 pm

On/off switch problem now repaired, turned out to be a leaky 120 uf / 50v electrolytic in the power supply module. Now on to replacing all the rest of the caps .....

Post by **Fump** » Tue Sep 19, 2017 12:11 pm

Greetings Folks

I have a question for Wowbagger about my 2975. It boots up until "Hardware init".
A small screen pops up "Error in Tel Script" "Error: missing close-brace"
Unit stops right there.
Any ideas???

Thanks for taking the time to read and respond.

Fump

MattSR · Post by **MattSR** » Tue Sep 26, 2017 4:50 pm

If nothing else, this thread has convinced me to stay away from purchasing a secondhand Aeroflex / IFR. Dropping $10k plus on a machine that then dies because of a faulty hard drive and can't resurrected sounds quite painful!

Post by **Audiotron** » Tue Sep 26, 2017 6:08 pm

unfortunately it is a negative point for an excellent instrument ...

Post by **Audiotron** » Fri Sep 29, 2017 3:28 pm

Hi ,

the intention never to go against copyright, bad would be more important that there was more information for fundamental data recovery of HDD and perhaps a fixed topic with relevant information to repair this model of unit.

Post by **hulkinator** » Sun Oct 15, 2017 4:26 pm

va3ext wrote:I just picked up a used A-7550 for parts, lid & soft case for my 1200S.

It starts up & goes past the boot screen to the spectrum screen, then after a few seconds freezes up. Can't do anything.
If I hit the Menu button just after the boot screen, it brings up menu, but then freezes.

Any ideas where to start looking?

Scott

Does it mysteriously work pretty much all ok (except a tad off freq) with AUTOCAL disabled? (jumper installed ON JB1 on interface board) ? But normal A-7550 operation is AUTOCAL enabled (jumper JB1 off)...jumper JB1 installed should only be used during calibration procedure...

I had picked one up for cheap. Unit did just that. Recapped supply even with low ESR on all caps...lithium 3v cell was 0.4 or something...so desoldered it...but didnt replace it right away...service manual doesnt say its critical...just it saves memory...

i troubleshooted this for hours..read sm inside out...so odd unit was working pretty much ok with AUTOCAL disabled...with AUTOCAL disabled navigating menus (running out ideas) I noticed accessing STORE function I would lock up unit...more and more was looking like a cpu-memory issue...could that lithium 3V play an important role and needs to be there and good?...does the firmware keep important AUTOCAL or other values in the backed up ram...?

It does!...i installed a plastic lithium cell holder (tight fit with a resistor network close) and a new 3v cell...a couple of ON OFFs and unit is working fine now...with AUTOCAL disabled...100Mhz ref now right down center

Not a unit for serious work anymore but fine for occasional ham & hobby use!