| Electronics > Repair |
| Tektronix DPO3034 Oscilloscope Help, Won't Turn On |
| (1/1) |
| tixdave:
Hello, I have a DPO 3034 that is not turning on. When you first press the power button the lightable buttons turn light up but no screen, fan turns on. Looking at the debug data it gets to the DRAM section and stops, fan is still running. --- Code: ---U-Boot 2009.08 (Feb 02 2011 - 10:33:21) Tektronix, Inc. V1.08 CPU: AMCC PowerPC 440EP Rev. C at 333.333 MHz (PLB=133, OPB=66, EBC=66 MHz) Bootstrap Option G - Boot ROM Location I2C (Addr 0x54) Internal PCI arbiter enabled, PCI async ext clock used 32 kB I-Cache 32 kB D-Cache Board: Tektronix Redondo/Metro AMCC 440EP Main Board VCO: 666 MHz CPU: 333 MHz PLB: 133 MHz OPB: 66 MHz EPB: 66 MHz Programming DIA...done I2C: ready DRAM: 128 MB --- End code --- Might just need to send in for repair. |
| tom66:
U-boot usually happens after memory training is done (that would be done in a first-stage bootloader or boot ROM)... so DRAM probably OK. Best guess is that the kernel image is corrupted or the U-boot image is corrupted, or some other hardware issue is present that is resetting the CPU just as things get going, like a power supply problem. If you have some way to reprogram the flash IC with a known good flash image then it may fix the issue. Of course, it is possible calibration data is loaded there, so that may cause other issues. |
| tixdave:
It is consistently getting stuck there so I don't think it is a power supply issue. It has been sitting on the bench for a while and haven't turned it on probably in about a year, but last time I did it was working normally. Probably corrupted image somehow from just sitting? |
| tom66:
Flash memory can do that unfortunately, as the memory cells self-discharge over time. |
| dbator:
Hello, I would not recommend you re-flashing unless you have no other choice. Some of us already made that mistake and now have problems with restoring of their scopes systems. Unless you're very familiar with embedded Linux operating systems and you think you're able to restore such an OS. Thinking more about re-flashing, if a bad solder joint is causing the issue you will probably fail at erasing of the FLASH. Boot log may stuck at "F" (FLASH) and still, issue may be caused by a faulty DRAM. If you did check all supply voltage rails (Vrefs, DDR termination etc.) and clocks I advice you to check thoroughly bus traffic between CPU, FLASH and DRAM. Those scopes are pretty old and additionally if working hours counter in case of your scope is high (in my case - MSO4054 - it was around 27340 hours when a Micron DRAM failed) it may be something wrong with DRAM or any BGA IC solder joint. If in your case there are also Micron DRAMs installed I would say - suspect them first. So, if I was you, I would download at least datasheets of: CPU, FLASH and DRAM and check all command/address and data lines. So, warming up (a bit over) or cooling down (a bit under) room temperature and observe ICs using a thermal camera may also be helpful if you own one. If using a thermal camera, keep the scope powered on for maximum few seconds and make breaks because you don't want the power section of the board affect thermal imaging results (false negative). I confirmed several times that Micron DRAMs failures occur depending on their temperature (they usually work if warmed up and fail if cooled down). It may not be easy, but if you really want to take up the challenge, then please remember ESD precautions like cotton clothes, anti-static gloves if you don't have access to EPA area where you could perform the repair. Good luck. |
| Navigation |
| Message Index |