Tek 2712 spectrum analyzer freezes during boot

[rodd]

Hi,
I have just bought this instrument at Ebay for a reasonable price.
Auction was "as is" but the pictures showed a "working" instrument.
However when it arrived, it would finish the startup sequence.
It "freezes" in the "file cleanup" and does not respond to any button press.
I have not opened yet to check if something was loosened during shipping, but I would like to have an idea of what (or where to look) before I start.
Any ideas/comments will be highly appreciated.
Thanks,
Roger

[Kilo Tango]

Hi, sorry to hear you are having problems.

The service manual is available from Tektronix under the service and support section. I have a 2715, converted to 50R input and it works fine, pretty reliable. Its also portable !.

A clever chap called David Emery worked out that the 271, 2712, 2714 and 2715 family use a common base so parts are interchangable. If you google "tek 2711 silk purse into sows ear" you can find his write-up.

Cheers Ken

[andy2000]

I'd start by replacing the defective NVRAM battery that's causing all those file system errors.  It shouldn't cause it to freeze, but it's a good place to start.

[Mattylad]

Butting in, that's a nice idea, where can you get those batteries? they look extremely obsolete.

[rodd]

Hi,
I have replaced the batteries but the SA freezes at the same point.
I was hoping that I would find a loose connection or similar issue since it is a long way from US to Brazil, and it was "in working condition" when I bought it (with photos).
Unfortunately that was not the case.
I wonder if the problem is related to the front panel since the last line of the error screen says "press W to cont; Z to abort" but I have tried every key and nothing happens.
Is it possible to know if the problem is really a freeze during boot or a defective front panel?
Any other ideas?

[cncjerry]

there are w,x,y,z buttons on the front panel, obviously, and if they aren't working I wonder if the front panel ribbon cable connector is loose.  Have you reseated, replugged them all?  usually keyboards are scanned so maybe you can watch for the scanning with a scope.

[PA4TIM]

First check the power supply (always), I had to replace all 220uF (if I remember well) electrolytic caps in a 2710 I owned (I still have a 2712 but that one is OK) They were not all bad but because a lot were bad I took no risk and replaced them all.  They are stuffed with programmable logic that can go bad. In my 2710 that caused selftest problems.
They are not easy to repair. 

[rodd]

Hi folks,
Thanks for the suggestions.
I will recheck the ribbon cable, but since I can see led's flashing during the initial phase of the startup sequence, I guess it is not the culprit.
So I will move to the power supply.
I am not familiar with this instrument and I was wondering if it is normal to go to a "file system cleanup" during every startup.
This may be a clue that something is wrong.
Can anyone confirm that "file system cleanup" is a part of normal startup sequence?
Thanks.
PS Answering to Mattylab, the batteries are standard CR2450 with terminals attached to it.

[Triton59]

Hello from the France,

I have the same fault on my 2712 Tektronix with TG  .

I changed the 2 batteries as well as all the capacitors 10uF 35V aluminium which had more capacity but nothing done for now.

I also changed a 0, 1uf which was a deposit of carbon C397 on Platinum A9.

I continuous inquiries and follow your subject.

Sorry but my English is not very good, so I go through a translator

Kind regards

 

[guido]

Any PLS logic ICs in these? Those are programmable and loose their content (if only partly) over time. Think there are smd variants in these machines. You can compare them with PAL logic.

I once helped somebody with a 271x with bad PLS logic. We did find the binaries on the net for the 2710. But still very difficult to fix, don't think he succeeded with his machine.
But obviously the fault could be a different one.

[Triton59]

Hello

I noticed that the 100 Mhz reference oscillator had a great influence as well as the switching power supply. To unlock the keyboard simply turn off the Analyzer and turn it on again immediately.

This procedure calls to rebalance all floors.

 

[rodd]

Hi Folks,
I received the batteries and replaced them, but the problem stays the same.
Then I noticed that one of the "new" batteries had only 2.5V!
Since I can't guarantee that these batteries are not form some chinese crap, I ordered other from a diffrent seller.
However while waiting form the smd caps replacement from Mouser, I noticed the the 220uF x 16V caps on the display and processor boars were showing low capacitance (around 50uF and high ESR), so I replaced them.
No luck yet.
I decided to film  the boot processes so I could read the messages on the first screen and I read the following message:
CHECKING NON VOLATILE MEMORY - 249 ERRORS FOUND
DATALOAD SETUP NOT CREATED E205
REF OSC FREQ 99.999 MHz.

Any suggestions?

[Armadillo]

You know how to ask for help.
But you wouldn't response to others seeking from you.
So your spectrum analyser should stay rot. 
Cheers!.

[rodd]

Why do you say so? I am just an “entry level” hobbyist that uses this forum to learn. I don’t remember being asked a question and not answering it.


Enviado do meu iPhone usando Tapatalk

[rodd]

If I did that it was an accident and I apologize for that.


Enviado do meu iPhone usando Tapatalk

[Armadillo]

Hi;

thanks for reply; I think it's accidental.
Do check your previous posting if you have missed anything.

I have no idea with this analyser, otherwise I would have chipped in.
But there are many experts in this repair forum. So, no worries.

Regards;
Armadillo;

[Triton59]

Hi all

My parser is finally fixed  , I did more than to the recalibrated.

The failure was at the level of a 8-bit converter, the circuit is oxidized because of an aluminum capacitor leaking.

[DrJay73]

I had the same problem and while looking for snswers, encountered this forum.
My problem started when my 2715 began having some level problems. I looked in utilities and then the diagnostics section. Show logs seemed a reasonable choice. I selected show all. I believe the analyzer froze at that point. I toggled the machine off then on and that horrible screen appeared. Log errors flash by and then the screen flashes about 10 times right before a log line disappears. Then it freezes. Hit W to continue Z to exit. Nope. Totally froze up.

I had hoped to find a reset code that would bring the boot sequence back to normal. Was not successful in my quest.

In a Hail Mary, last ditch attempt, I held the util button down while I turned the unit on.
Well, Hail Mary! It worked!!

I joined the forum to share my experience. I thought I was staring at corrupted nvram.

Give it a shot. It worked for me. I suspect it is a closely guarded secret. I would be very interested to know if it works for others. I suspect it will!! Good luck!
Dr Jay

[DrJay73]

I made it out of the locked up screen. Since my 2715 has not had the battery replaced thete is a real good chance it is bad. I didn't recognize any symptoms  that would make me look at the battery.

I will be opening her up and doing the basics. Reseating anything and everything and some strategic vibration testing. I'll source a battery before hand.

What symptoms would you expect if the battery was low? Perhaps an entry into the diagnostic logs?

I am reluctant to display the logs after it caused my 2715 to freeze. I did select show all and it showed plenty of errors before locking up. I took pictures of the screens if it should prove helpful.

Thanks,
DrJay73

[joseph1207]

I had/have the same issue with my 2712, also an ebay purchase. to start when turned on would start to boot then a electrical arc could be heard and boot would restart. found carbon trace on power supply board. i cleaned the trace then removed the board <not a real simple task >.  i replaced the caps near the offending arc over and inspected the board quite a few caps showed leakage <puke on the board>spent some time cleaning corrosion from board <some had invaded surrounding components <SCARY> . since i had a supply of lytic caps <and no lcr or esr>, i replaced all caps on the board.
   That was when i got to the  <boot and freeze> with no response at the keyboard. replaced one of the lithium batterys that showed less than 1 volt, replacement showed immediate voltage slump <more bad caps>.  finally got to boot and performed a file cleanup.  It now boots though not reliably and shows some loopback errors. Still a repair in progress.  but occupys my spare, spare time  .     
   Also i have had problems with the 2 phased<?> outputs from power supply could never see on scope what the manual is looking for.
I am just a hobbyist but do have to cut coaxial stubs and phasing lines on occasion with this working should get me closer than the sig gennerator/receiver on a tee that i have been using.  i am really rambling now!!!  but anyone out there with one of these as a project keep posting your findings !!!

[rodd]

Pressing and holding the util buttom during power up did not worked for me.


Enviado do meu iPhone usando Tapatalk

[rodd]

Any PLS logic ICs in these? Those are programmable and loose their content (if only partly) over time. Think there are smd variants in these machines. You can compare them with PAL logic.

I once helped somebody with a 271x with bad PLS logic. We did find the binaries on the net for the 2710. But still very difficult to fix, don't think he succeeded with his machine.
But obviously the fault could be a different one.

Yes Guido,
There are several PLS’s in it.
I found o friend with a working 2711 and swapped the boards, one by one and the problem persisted.
I have also tried my instrument with all the boards swapped ( except rf and ref oscilator) without success.
There are 5 or 6 logic ic’s on the main board which perform some logic operations with the data lines. Between the boards.
I will test them when I have time and post the results.



Enviado do meu iPhone usando Tapatalk

[sprit]

Hi Everyone

Sorry for d*gging this post again I have same model but when power on,  the screen just flash and no signal,  the front panel freezed,  i cant do anything. Im trying to replace the bat but unfortunately its doesn't work at all. Haizz.

[sprit]

Hi everybody.  Wish everyone have a good day.  I have a Tektronix 2712 50R Spectrum Analyzer.  When starting it emits a single beep and then the control screen freezes in the above state.  Nothing appears on the screen.  I don't know where to start.  Tried measuring some Testpoints and it meets the power supply enough.  Does anyone have an opinion on it?  Thank you very much.