Tek CSA7404/TDS7000 repair project (Upgraded memory/CPU)

[andy2000]

Hook a monitor up to the SVGA port. Do you get video there? I suspect that you will, and see that the MB is reporting a CMOS checksum error.

Jay

That's probably it.  For some odd reason, Tek didn't modify the BIOS to set the PCI video card for the internal LCD as the default display.  Instead, it defaults to the AGP/chipdet video when the CMOS setting are lost. 

[Jwalling]

Hook a monitor up to the SVGA port. Do you get video there? I suspect that you will, and see that the MB is reporting a CMOS checksum error.

Jay

That's probably it.  For some odd reason, Tek didn't modify the BIOS to set the PCI video card for the internal LCD as the default display.  Instead, it defaults to the AGP/chipdet video when the CMOS setting are lost.

Yup. "Gee Wally, what a dumb thing to do." 
The fact that he's getting a floppy access and two BIOS beeps (which is CMOS checksum failure) it's probably prompting press F2 to enter setup. MB uses a standard CR2032 battery.

Jay

[imaurente]

Good evening, I'm sorry but I need help with my Tek TDS 7054, I use the operating system Win98 and I'm not getting the boot TekScope, it crashes. I bought this equipment in the state, was not with any program of Tektronix, only OS and follow the steps reported here, the link downloaded: http://www.tek.com/oscilloscope/tds7254-software/csa7000-and-tds7000 -series-firmware-upgrade and not get hit, always fighting. Can someone help me??? Greetings!

[Litchenko]

Hello all.
I found your Forum and read about the repair of a TDS from Tektronix.
The reason is that we have had in our lab in the TDS7104, who suddenly ceased to make measurements.
Win 98 operating system boots, Tektronix also downloaded the app, but the measurement does not produce.
The PPC Board is booting after switch on from ".8" and count normaly up to "-".
Help revive TDS7104.
Best Regards,
Alexandr Litchenko

[Litchenko]

Hi again.
Maybe someone can share the value of the voltage reference points. Very interested in the Acquisition board.
Best Regards,
Alexandr Litchenko 

[TiN]

Vitaju

Did you check power supply voltages? Given that almost all modules failed with errors - maybe some voltage is off? Try swapping DIMM module (PC100 256MB) on PPC board.
Check visually if any burnt/hot parts around. Your scope seem to have good NVRAM, otherwise it would not get into TekScope app, but worth to check serial log output, like covered in this thread on first page.

[vealmike]

Hi guys (and girl, 'scope repair isn't just for boys is it?)

I have been gifted a broken TDS7404. I know, I'm a lucky "£$%£$.

I was hoping that the fault was going to be a flat NVRAM battery, Windows boots perfectly, but the 'scope application hangs at the splash screen. Sadly it's not going to be that easy. NVRAM is fine. I've hooked up to the serial port and I can see VxWare booting successfully. TopScript hangs at the point it tries to initiate the acquisition board, baulking that there is no response from the Acq. board processor.

So, I guess I'm looking at a fault on the Acq. board, the PPC board, or a power supply unique to the Acq. board (something sent over the front power distribution board, that's not on the rear power distribution board.)

My problem is, I don't know what signals and voltages should be present anywhere in the box. Hence this post. I'm looking for some clues.

Let's start with the PA Bus.
What is a PA Bus? Is this a Tektronix proprietary bus? How is it connected to the PPC? I've googled and found nothing.

Looking at the serial output, the PPC board has a 750. No surprise there. From memory, this processor uses a 60x bus. Presumably there's a bridge chip between the 60x bus and the PA bus.
Does anyone have any details about the PA Bus, or the PPC to PA bus bridge chip please?


Can anyone tell me what power rails are supposed to ship to the Acq board over the front power distribution board please?

If it helps, I do actually know what happened to this 'scope. I was using it at work happily, then accidentally switched it off at the wall socket, rather than using the front panel button. On reboot it was faulty.

Many thanks in advance for any assistance.

[andy2000]

If it helps, I do actually know what happened to this 'scope. I was using it at work happily, then accidentally switched it off at the wall socket, rather than using the front panel button. On reboot it was faulty.

Many thanks in advance for any assistance.

That shouldn't have hut the hardware, but it certainly could have corrupted some files on the hard drive.  I'd start by reinstalling the scope application, then possibly the entire OS before digging into the hardware too much. 

[vealmike]

If it helps, I do actually know what happened to this 'scope. I was using it at work happily, then accidentally switched it off at the wall socket, rather than using the front panel button. On reboot it was faulty.

Many thanks in advance for any assistance.

That shouldn't have hut the hardware, but it certainly could have corrupted some files on the hard drive.  I'd start by reinstalling the scope application, then possibly the entire OS before digging into the hardware too much.

Thanks for the suggestion. Yes I have already checked the contents of the Vxboot directory. I have already uninstalled and re-installed the software.

Here is the serial log. As you'll see the PPC is happily attached to the Windows HDD and is loading firmware merrily from vxboot. The issue occurs when the PPC tries to communicate with the acquisition board.

Code: [Select]

UUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUUU

romPost: MPC106  Walking-one passed.

romPost: MPC106 configuration passed.

romPost: First PCI access  passed.

romPost: PC87560 Walking-one passed.

romPost: PC87560 configuration passed.

romPost: DEC21554 configuration passed.


romPost: PERFORMING ROM CHECKSUM TEST...

romPost: ROM checksum passed.


romPost: PERFORMING DRAM CELL TEST...

romPost: Enabling data cache...
romPost: Data cache enabled.
romPost: Enabling instruction cache...
romPost: Instruction cache enabled.

romPost: DRAM Cell test passed.
romPost: Disabling data Cache...
romPost: Data cache disabled.
romPost: Disabling instruction cache...
romPost: Instruction cache disabled.

romPost: PERFORMING DRAM MARCH TEST WITH CACHE OFF...
romPost: Writing address to each *address...
romPost: Reading address from each *address...
romPost: DRAM march test passed...

romPost: PERFORMING DRAM MARCH TEST WITH CACHE ON...

romPost: Enabling data cache...
romPost: Data cache enabled.
romPost: Enabling instruction cache...
romPost: Instruction cache enabled.
romPost: Writing address to each *address...
romPost: Reading address from each *address...
romPost: DRAM march test passed...

romPost: PERFORMING DRAM WALKING-ONE  TEST...

romPost: DRAM Walking-one test passed.

romPost: PERFORMING NV-MEMORY WALKING-ONE  TEST...

romPost: NV-MEMORY Walking-one test passed.
romPost: Disabling data Cache...
romPost: Data cache disabled.
romPost: Disabling instruction cache...
romPost: Instruction cache disabled.
romPost: Enabling instruction cache...
romPost: Instruction cache enabled.


romPost: POWER-UP DIAGNOSTIC COMPLETE.

Hel










                            VxWorks System Boot


Copyright 1984-1996  Wind River Systems, Inc.





CPU: Tektronix HPP - PowerPC 740
Version: 5.3.1
BSP version: 1.0.4.0
Creation date: Nov  2 2000, 11:03:34




Press any key to stop auto-boot...
 0
auto-booting...


boot device          : sm
processor number     : 0
file name            : c:/vxboot/vxWorks.st
inet on backplane (b): 192.168.0.1:ffffff00
host inet (h)        : 192.168.0.2
gateway inet (g)     : 192.168.0.1
flags (f)            : 0x1008
target name (tn)     : tds7000
startup script (s)   : c:/vxboot/topscript.hw

Initializing backplane net with anchor at 0x1e80000... done.
Backplane anchor at 0x1e80000... Attaching network interface sm0... done
done.
Attaching network PC1 dio device driver installed
interface lo0... done.
Attaching DIO client... done.
Loading c:/vxboot/vxWorks.st...1773500
Starting at 0x200000...

Host Name: bootHost
User: target
Initializing backplane net with anchor at 0x16e0000... done.
Backplane anchor at 0x16e0000... Attaching network interface sm0... done
done.
PC1 dio device driver installed
Attaching network interface lo0... done.

Adding 6008 symbols for standalone.


 ]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]
 ]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]
 ]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]
      ]]]]]]]]]]]  ]]]]     ]]]]]]]]]]       ]]              ]]]]         (R)
 ]     ]]]]]]]]]  ]]]]]]     ]]]]]]]]       ]]               ]]]]
 ]]     ]]]]]]]  ]]]]]]]]     ]]]]]] ]     ]]                ]]]]
 ]]]     ]]]]] ]    ]]]  ]     ]]]] ]]]   ]]]]]]]]]  ]]]] ]] ]]]]  ]]   ]]]]]
 ]]]]     ]]]  ]]    ]  ]]]     ]] ]]]]] ]]]]]]   ]] ]]]]]]] ]]]] ]]   ]]]]
 ]]]]]     ]  ]]]]     ]]]]]      ]]]]]]]] ]]]]   ]] ]]]]    ]]]]]]]    ]]]]
 ]]]]]]      ]]]]]     ]]]]]]    ]  ]]]]]  ]]]]   ]] ]]]]    ]]]]]]]]    ]]]]
 ]]]]]]]    ]]]]]  ]    ]]]]]]  ]    ]]]   ]]]]   ]] ]]]]    ]]]] ]]]]    ]]]]
 ]]]]]]]]  ]]]]]  ]]]    ]]]]]]]      ]     ]]]]]]]  ]]]]    ]]]]  ]]]] ]]]]]
 ]]]]]]]]]]]]]]]]]]]]]]]]]]]]]]
 ]]]]]]]]]]]]]]]]]]]]]]]]]]]]]       Development System
 ]]]]]]]]]]]]]]]]]]]]]]]]]]]]
 ]]]]]]]]]]]]]]]]]]]]]]]]]]]       VxWorks version 5.3.1
 ]]]]]]]]]]]]]]]]]]]]]]]]]]       KERNEL: WIND version 2.5
 ]]]]]]]]]]]]]]]]]]]]]]]]]       Copyright Wind River Systems, Inc., 1984-1997

                               CPU: Tektronix HPP - PowerPC 740.  Processor #0.
                              Memory Size: 0x2000000.  BSP version 1.0.7.0.
                             WDB: Ready.

KERNEL: THERMAL CONTROL INITIATED
Executing startup script c:/vxboot/topscript.hw ...
dbFsIoctl: invalid code 3
    ################### topScript starts ###############################




     cd "dbfs:c:/vxBoot"
value = 0 = 0x0


    ld < TekScope.o
value = 23666688 = 0x1692000
    ld < TekScopeSM.o
value = 23666048 = 0x1691d80

    pDbDev = ramDevCreate(0, 512, 1152, 1152, 0)
new symbol "pDbDev" added to symbol table.
pDbDev = 0x1693940: value = 23673184 = 0x1693960 = pDbDev + 0x20
    pDBfs  = dosFsMkfs("/db", pDbDev)
new symbol "pDBfs" added to symbol table.
pDBfs = 0x1691bc0: value = 23665696 = 0x1691c20 = pDBfs + 0x60
    copy ("TekScope.DBF", "/db/TekScope.DBF")
value = 0 = 0x0

    overrideFPdefaults()
value = 0 = 0x0

    measurePPCtimebase
value = 0 = 0x0


    registerMemPartitionSizes ( 0x00000000 )
value = 0 = 0x0

    semInitForQueueCheckLogMsg
value = 21818112 = 0x14ceb00

    initSysWaiter()
value = 0 = 0x0




    raxInit
value = 0 = 0x0
    paiInit
done
value = 16983988 = 0x10327b4 = PaiBaseRegisterTable + 0x948


    verifyAcqBoardID
value = 0 = 0x0






     tcsLoadTableToNVRAM "pc:c:/vxBoot/Setup.tcs"
value = 0 = 0x0
     updateAcqProcFirmware("dbfs:c:/vxBoot/acqProc.hex")
0x16987e0 (tShell): Program enable command didn't echo back properly.
value = 0 = 0x0
     cd "pc:"
value = 0 = 0x0

    raxInitializeDevices
done
0x16987e0 (tShell): (Re)starting Acq Processor
0x16987e0 (tShell): Acq processor powerup failure!

[vealmike]

The work continues, but at a pretty slow rate. I have a lot of other plates spinning.

Last night I took a look at the front and rear power distribution boards (from now on 'PDB'). They look like simple 2 sided PCBs, with GND on the outside of the chassis and power on the inside. Nice and easy to probe. I took a bunch of measurements with a DVM and annoyingly left them at home. I'll post them up later.

The rear PDB is good. I see 3.3V, 5V, 12V and I think 9ish V (can't remember exactly.) This is what I'd expect as this board feeds the x86 and PPC boards, all of which appear to be happy.

The front PDB measurements weren't so good. 3.3V was present, but the other power rails were not doing so well. I had a 0.9V, a 0.76V & a 0.09V. I took a quick impedance check of each rail to GND, all were in the tens of ohms range, so the PSU isn't folding due to over current.

I've not pulled the power supply board out to take a gander yet. But TiN has posted a picture of one...
https://dev.xdevs.com/attachments/download/625/PSU.jpg (Thanks TiN  )

Front PDB connector is top left. Rear PDB connector is middle left. Without studying the photo too hard, there's a handy silk screen line denoting the low and high voltage sections of the board. It starts bottom left and terminates top right. Straddling the line are two large power transformers, each roughly in line horizontally with one of the power distribution board connectors. On the RH side of each there are a set of identical TO220s and a heatsinked power FET.

It looks very much to me like there are two more or less discrete SMPS supplies on this board. I'd hazard a guess that the one associated with transformer T7 is dedicated to the Front PDB and has gone pop.

So it's good news, a power fault. This is far better than a fault on the acquisition PCB. That board is stuffed with FPGAs that would be difficult to source and impossible to reprogram. I'm far happier having a PSU to fix. Even better news, as all the rails on the Front PDB are stuffed (I need to determine if the 3.3V that I did get is wired directly to the Rear PDB before I can be sure) I think I can suspect something on the primary side of T7, or something shared on the secondary.

Even better better news, it looks like the two implementations of SMPS are copies of roughly the same schematic. So I can compare working with non working.


OK, enough prose. Is anyone able to help me please by measuring the DC voltages on the front PDB?
Thanks.

[vealmike]

OK as promised, here we go.

Rear power distribution board.
A/B/C 1 GND
A/B/C 2 Measured 3.54V
A/B/C 16 Measured 5.34V
A/B/C 27 Measured 12.75V
A/B/C 30 Measured 8.77V
 

Front power distribution board.
A/B/C 1 GND
A/B/C 2 Measured 0.95V
A/B/C 11 Measured 0.145V
A/B/C 17 Measured 0.768V
A/B/C 20 Measured 0.09V
A/B/C 22 Measured 3.36V
 

[TiN]

I might take the hood off my scope for you, if you can share photo with exact points you want to measure    

[vealmike]

TiN that's an awesome offer.
I typed a detail response and have deleted it.

I was checking the service manual (http://dev.xdevs.com/attachments/download/633/071701102%20CSA7000%20TDS7000%20Service%20manual%20.pdf) for a diagram showing the measurement points, and on page 6-66 (  ) I found a table with the voltages I need.

Many thanks, but no measurements needed.

[vealmike]

  Round 2 

The PSU board came out nice and easy.  And would you believe it, this is what I found nestling next to the switching FETs for the transformer T7 that I suspected above:


(Dear Santa,
Thank you very much for the marco extension tubes wot I got for Christmas. They are very hand and I am having a lot of fun taking photos of exploded electronics!)

The chip is a UC3844|https://www.onsemi.com/pub/Collateral/UC3844-D.PDF a fixed frequency PWM controller.
Pin 5 is GND, pin 7 is VCC, which according to the datasheet is 12 to 25V.
Looks like once upon a time there may have been a component where that crater is now.

Not too far away is the other, near identical (working) channel. Here's a pic, same extension rings, different lens. 

Yup, two SMT ceramic caps across pins 5 to 7 (notice how one of them looks leaky?)
Next thing to do was break out the meter.
Pin 5 on both chips are linked. That makes sense, it's GND.
Pin 7 on the working channel is linked to the blown trace on the faulty channel.

So these two chips share power and GND. Good news! One of the channels works, so there can't be a power fault.
I think that this was just a smoothing cap that's gone pop. Judging by the leakage on the other channel's cap, they are either not sufficiently over rated, or VCC had a transitory high, I'd have expected more damage in the latter case.

If we assume a cap with a DC rating too low, then the SOIC near the crater is probably good. I just need to clean some of the carbon off and reconnect it, looks like the cap fire melted the traces. I hope the chip is good, the datasheet says its obsolete.

Oh, more good news, this looks like a simple 2 layer PCB. So I don't think there will be any internal damage.

Here's another gratuitous macro ring crater shot:

[andy2000]

There's nothing in a ceramic cap to leak.  That residue looks like flux to me.  It's not unusual for SMD ceramic caps to short circuit due to a defect, or physical damage.  If there's enough current available, they will get hot a burn.  I'd pull one of the good one to measure its value.  The datasheet for that IC should give you an upper limit for the voltage.  Be sure to remove every trace of carbon to avoid future trouble. 

[vealmike]

Andy, you are of course correct. That'll teach me not to post when I'm tired!

The 3844 was removed and the PCB traces cut short of the charred PCB. The charred PCB was sealed with nail varnish and the traces were yellow wired back. Sadly the 'scope still didn't work.

Whilst On Semi don't make the 3844 any more, lots of other vendors do. A £1 chip was acquired from Mouser (£12 postage) and the 3844 was replaced. The scope was booted and volia! All is working.

I'd love to keep this, I don't have a DSO at home. But sadly I can't afford to keep a piece of kit this valuable for home use, it'll have to go up for sale.

Many thanks for the assistance offered and rendered here.

[valerieflyer]

I went through this wonderful blog thread last year to fix my TDS7104 - replaced battery on NVRAM.  Now, it is dead again.  Replaced battery with giant external battery pack.  Still ran that down in a couple of months.  So, I will be replacing the NVRAM and battery this time.  I've had this happen before over the years with Dallas Semi NVRAM increasing power usage after > 10 years.  Don't really know why the quiescent current goes up so much. 

[STDio]

Greetings.

I recently obtained a TDS7104, and this forum really help alot 
getting these old stuff, is really like betting a lottery , as stuff might fail.
The su810 nlx really is a old board but, I didn't even bother upgrading the firmware, ,
and the NVRAM battery cap lifespan is unknown, it might need change anytime soon
Really hope these machines run for years , so far I upgraded a newer IDE diskdrive with win xp and a celeron 1.2 tualatin
My spec is

TekScope version 2.5.5
Option 3M: 16 Meg Max 4M/Channel
SM Serial Mask testing
Disk Drive Measurements 2 (Option 48)
USB 2.0 Test package (Option 50)
Communication Pulse Measurement 2
App 54 Avail : not installed
Power Measurement 3
Ethernet Compliance test software
Jitter analysis Advance.
Jitter analysis Essentials
Can and Lin Timing and protocol decode software.


Anyone knows what software is for the, App 54 Avail : not installed?

[STDio]

Hi TIn

how can i know the ram on the PPC board?
is there a log in vxboot folder?

Thanks

[vtp]

Anyone knows what software is for the, App 54 Avail : not installed?

App 54 is the same as option number 54, that is, PW2 power measurements. Since you have PW3 (option #62) it does not matter. Your scope has been upgraded sometime from PW2 to PW3.

[Jwalling]

App 54 is TDSRG - I think this is report generator.

[vtp]

App 54 is TDSRG - I think this is report generator.

Maybe it is, maybe it is not.

I went back to check my sources and PW2 came to this key:

TDS5104  B020149  9WSFG-QYL5H-993AA-GFXED-VD4GB-Z  option  2M ; 2A ; PW2

So far the bit positions have been 100% consistent across the whole range of windows based TDS scopes.

I still think the last sentence of my previous reply is correct explanation for the missing app.

[Jwalling]

App 54 is TDSRG - I think this is report generator.

Maybe it is, maybe it is not.

I went back to check my sources and PW2 came to this key:

TDS5104  B020149  9WSFG-QYL5H-993AA-GFXED-VD4GB-Z  option  2M ; 2A ; PW2

So far the bit positions have been 100% consistent across the whole range of windows based TDS scopes.

I still think the last sentence of my previous reply is correct explanation for the missing app.

This picture may do the trick.
Note the position of TDSRG between TDSCPM2 and TDSPWR3.
Then note this Ebay listing where it is missing:
http://www.ebay.com/itm/TEKTRONIX-TDS7154B-Oscilloscope-4CH-1-5GHz-20GS-s-5M-SM-J2-USB-CP2-PW3-ET3-JA3-/292060163891?hash=item4400246f33:g:cUwAAOSwB-1YzaUw

EDIT: Forgot the picture.

[vtp]

This picture may do the trick.
Note the position of TDSRG between TDSCPM2 and TDSPWR3.
Then note this Ebay listing where it is missing:
http://www.ebay.com/itm/TEKTRONIX-TDS7154B-Oscilloscope-4CH-1-5GHz-20GS-s-5M-SM-J2-USB-CP2-PW3-ET3-JA3-/292060163891?hash=item4400246f33:g:cUwAAOSwB-1YzaUw

I enabled application 54 on my scope and even after re-installing PW3 I do not see TDSRG option, just a note that app 54 is available but not installed.

From TDSPWR3 readme:

Code: [Select]

Product:  TDSPWR3 (Power Measurements 3)
Version:  1.8.1
Date   :  14-Jul-05

...clip...

* This application installs Report Generator. Report Generator
is a common feature shared by several applications. If you install another
application which includes a newer version of Report Generator, then Report
Generator will be updated to the newer version. You can check the version from
Report Generator Help>About.
Perhaps different scopes with different firmwares show TDSRG on place of app 54. Or app 54 is PW2 that has been uninstalled when installing PW3 and TDSRG is shown by different firmware before TDSPWR3 regardless of app 54 bit.

Do you have installation files for plain TDSRG?

BTW, that ebay link you provided is a funny one. $2k for left, top and right plastics and a HDD bodged in a scope that was previously a rack mount unit without HDD and DVD. Miraculously it has got some options too, compare to what is shown on back panel sticker. Seems like there are some people trying to get financial gain from enabling options on scopes that do not have those. I wonder how long it takes Tek to take notice.

[Jwalling]

This picture may do the trick.
Note the position of TDSRG between TDSCPM2 and TDSPWR3.
Then note this Ebay listing where it is missing:
http://www.ebay.com/itm/TEKTRONIX-TDS7154B-Oscilloscope-4CH-1-5GHz-20GS-s-5M-SM-J2-USB-CP2-PW3-ET3-JA3-/292060163891?hash=item4400246f33:g:cUwAAOSwB-1YzaUw

I enabled application 54 on my scope and even after re-installing PW3 I do not see TDSRG option, just a note that app 54 is available but not installed.

From TDSPWR3 readme:

Code: [Select]

Product:  TDSPWR3 (Power Measurements 3)
Version:  1.8.1
Date   :  14-Jul-05

...clip...

* This application installs Report Generator. Report Generator
is a common feature shared by several applications. If you install another
application which includes a newer version of Report Generator, then Report
Generator will be updated to the newer version. You can check the version from
Report Generator Help>About.
Perhaps different scopes with different firmwares show TDSRG on place of app 54. Or app 54 is PW2 that has been uninstalled when installing PW3 and TDSRG is shown by different firmware before TDSPWR3 regardless of app 54 bit.

Do you have installation files for plain TDSRG?

BTW, that ebay link you provided is a funny one. $2k for left, top and right plastics and a HDD bodged in a scope that was previously a rack mount unit without HDD and DVD. Miraculously it has got some options too, compare to what is shown on back panel sticker. Seems like there are some people trying to get financial gain from enabling options on scopes that do not have those. I wonder how long it takes Tek to take notice.

Sorry, I don't have TDSRG - I've just seen it before on a scope.

That Ebay item is not mine... It's possible that the options were purchased after the original ship date. Dunno.