My TDS7154 came from Taiwan, lots of dust inside but otherwise in very good shape in and out. On power up, all front face LEDs are lit, fans spinning and nothing else was going on. By accident I discovered that W2000 boots up after 5 minutes or so. Power PC (PPC) 7 segment display was solid on .8 no activity on console port. I went through all eevblog posts related to TDS7104 as originated by TiN, learned a lot but couldn’t make meaningful start. I contacted andy2000 because symptoms on his TDS7154B were similar to mine. He responded but couldn’t help at the time, he helped me greatly later with opening the world of options. Big thanks. Because I learned a lot, or better to say everything about TDS scopes from this forum I thought it would be fair to share my experience and potentially help somebody new like me. In retrospect, all the answers were already in the forum trail, thanks TiN. I just didn’t read them right and understood things correctly. Reading back the same posts as my understanding of the problem became better was the path to the solution.
In summary my TDS7154 needed PPC fix, hardware, new DC/DC converter PT6409P and Dallas NVRAM DS1245Y – 120 chip and, software, restoring NVRAM data using service port and directing vxWorks boot sequence to copy serial number (SN) form hidden directory c:\vxBoot\.sn file to the NVRAM.
There are folks in the forum that mentioned DC/DC converter on the PPC board and I initially didn’t pay much attention to it. Nevertheless, after checking voltage as per the service manual table (all values were good) I checked DC/DC chip output too and it was 0.4V instead of 2.5V. I struggled removing it from the board, could not desolder 3 ground pins in the middle. In the end I had to swing it left and right until all pins broke, then I hot air blew the stubs out. Pins on the removed chip were still accessible so I hooked up 5V to input and ground and indeed the output was missing. I thought what might have caused the failure, quick resistance check on the PPC board did not show any shorts. I attributed the failure to the chip itself. Removing Dallas NVRAM was a breeze, version I have is older DIL with encapsulated battery. I thought of grinding off the case to expose the battery, chop off the original and solder new one on top but gave up for fear of damaging the PCB. I’m not sure that it was even defective, it was reasonable to assume the battery is dead given the vintage of the instrument. I thought of installing a socket, unfortunately didn’t have one in the shop. I reassembled everything, on power up all was good except for the missing SN on the scope Help/About page.
I celebrated too early, restoring the SN turned out to be more challenging than I anticipated. I spent some time following eevblog folks’ steps (i.e. removing the # in # NvramClearDb = 1 line of vxBoot/topScript.hw script and making sure that .key and .sn files are there and have correct content). I also used HyperTerminal on console port (old RS232 ribbon cable was perfect interconnect between RS232 cable and console port, it was difficult to find one) to see if there are any errors while booting but all looked normal and terminated correctly when scope application is loaded and running. Not that I have a need for it (as a matter of fact I don’t have any need for the scope itself except for the challenge of fixing it) or the options, it is just part of the whole adventure. I Ghost imaged existing HD to 2 SSD disks, both work correctly so I’ll be more confident playing aggressively with software artifacts.
I reread again eevblog posts related to vxWorks boot sequence, did some Google work to educate myself on vxWorks subject and I realized that I didn’t read the eevblog posts carefully enough to see important details as outlined in the initial TiN post. With new understanding using service port and HyperTerminal I stopped vxWorks boot sequence by hitting any key at the right time. Then used “p” to see/record initial parameters the most important being flags (f): 0x1008. I used “c” command parameter to force vxWorks boot to copy *.sn from HD location by setting flags (f): 0x1000, and other (o): nvfs=0x1000. I kept startup script (s) :c:\vxBoot\topscript.hw, used “p” again to verify that I entered the right thing, then “@” to reboot and the whole script flew, 7 segment went through P then blinking ‘-“. I didn’t see Help/About updated until I shut down the instrument and rebooted again. Voila, all was good now, S/N was correct and options were listed. Since I was in vxWorks mood I tried different .sn (not equal to the S/N printed on the back of the scope and inside on the chassis) but I was not able to install any options, either I don’t know what I’m doing or the S/N is hard coded somewhere in the PPC board or elsewhere. I’ll have to read eevblog TDS trail again; I remember somebody was talking about this.
Big thanks goes to TiN and all other folks who contributed and helped me to fix the device. In my view, the biggest help is to know that there are still good and competent people somewhere in the universe, so the hope still remains despite all the buffoonery we witness every day!