Products > Test Equipment
Repairing Tektronix TLA7012's failing flat panel display [SOLVED]
bracecomputerlab:
My second posting here.
Around the same time I ordered, what turned out to be a defective (Acquisition error code 321 issue), Tektronix DPO70804 oscilloscope, I also ordered Tektronix TLA7012 portable mainframe logic analyzer.
The discussion about TLA714, TLA715, and TLA7012 about 6 years ago is part of the reason why I ended up purchasing TLA7012.
Thank you for that great discussion.
It was really helpful in making the decision to buy TLA7012.
At this time, TLA7012 costs only around $500 on eBay including shipping and sales tax.
It also came with one Tektronix TLA7AA4 module for free, and this one appears to be fully functional (passes diagnostics tests except the BNC related ones).
The only "catch" of this TLA7012 unit is that LCD always go blank after about 2 minutes, particularly when Windows XP is loading.
This was fully disclosed by the seller, so I do not have an issue with the seller.
I plan to give a positive rating.
A workaround suggested by someone else also selling the same unit (the same seller that gave me the advice to return the defective DPO70804) on eBay with the same LCD issue is to use an external DVI capable monitor.
I tried it, and it perfectly works around the display issue for now.
In other words, my TLA7012 appears to be fully functional except the built-in LCD.
Looking at how TLA7012's LCD fails after a few minutes, the failure mode seems eerily similar to how an LCD monitor with busted aluminum electrolytic capacitors displays something for a few minutes, and then quits.
If I shutdown the LA and turn the power back on, LCD does come back, but it goes bad again when booting Windows XP.
Does anyone know how to fix this issue?
Currently, eBay has several TLA7012 with the same issue, so it appears to be a fairly common issue.
Any suggestions or advice will be appreciated.
nctnico:
It could be a problem with the backlight inverter or the CFL tubes being worn and draw too much current.
However, if you don't need the logic analyser to be portable, you are better off using it remotely because a bigger screen size really helps where it comes to using a logic analyser. Your PC probably has a much bigger screen compared to the TLA7012. I have my TLA715 high up a shelve with the probe cables reaching to the desk level; I use it remotely exclusively. IOW: the failing LCD panel doesn't have to be a showstopper.
colorburst:
When you say the LCD goes blank, do you mean the image goes to a black screen or does the backlight turn off as well?
If you pause the boot process (e.g. BIOS setup or F8), does the display stay on? Or does it turn off after two minutes regardless? Could be a driver issue.
bracecomputerlab:
nctnico,
I saw your comments on TLA715 and TLA7012, and also about TLA7BB4 from 2017 to 2018.
Those comments along with comments from TiN and lukier were instrumental in my decision to purchase a TLA7012, 2 units of TLA7016 (currently in transit to two different locations), 3 units of TLA7AA4 (varying state speed and memory capabilities), and 2 units of TLA7S16 PCI Express Gen 1/ 2 serial analyzer.
I also bought P6716S PCI Express x16 interposer card kit so that I can finally see how PCIe devices work in practice.
This is really my first "probes" I purchased.
I brought the price down from $800 to $500.
Of course, the original price is closer to $10K or maybe more.
The kit comes with cables that attach to a PCIe interposer card.
I do not currently have any real probes at this point, including fly lead type probe / cable for TLA7AA4.
I think TiN is totally correct that probes cost more than the mainframe / analyzer module, especially P7500 series Tri-Mode active probe family for my ill fated purchase of DPO70804.
Off topic, but I once got someone working for one of the Big 3 (Keysight, Tektronix, or Teledyne LeCroy; I do not remember which one.) to demo me at their expo booth to see how a multi-lane PCIe device initialize itself.
This happened some years ago at a tradeshow called DesignCon, almost always held around end of each January at Santa Clara Convention Center (i.e., Silicon Valley).
Many years ago, I implemented a few early LTSSM (Link Training and Status State Machine) states on my own using PIPE PHY for PCIe.
I always wondered how a multi-lane PCIe device synchronized each lanes, so someone kind from one of the Big 3 demoed it for me.
Basically, what happens is that each lane is not really guaranteed to be synchronized at a power up, but when a special packet set called SKP OS (Skip Ordered Set) is sent across all lanes (still not exactly at the same transmission timing at this point), all the lanes get synchronized against one another.
From here on, all the lanes are transmitting TS1 or TS2 (Training Set) OS continuously at the same timing with an occasional SKP OS.
The thing is, the PCIe base spec does not really specify how to handle this type of a situation, and without being to see a real implementation running, it is virtually impossible to test against this situation in one's verification environment (testbench) on an HDL simulator.
From the previous Conventional PCI era, I have known that specs often leave out potential real world behavioral pitfalls, so visually being able to see how the spec is actually implemented in a real world hardware is really crucial in developing highly compatible hardware implementation.
This type of a situation of having inconsistency between the spec and a real world hardware implementation is perfectly normal, unfortunately.
It is still within the spec.
I got a little carried away discussing my very few occasions that I got to witness how PCIe is working in the real world, but finally I own the hardware to see it for myself how it works, albeit with somewhat dated PCIe support (Gen 1 / 2).
Sorry for the long OT blurb.
I will reply to your comments about the LCD in another reply.
bracecomputerlab:
--- Quote from: nctnico on November 17, 2024, 12:54:45 am ---It could be a problem with the backlight inverter or the CFL tubes being worn and draw too much current.
However, if you don't need the logic analyser to be portable, you are better off using it remotely because a bigger screen size really helps where it comes to using a logic analyser. Your PC probably has a much bigger screen compared to the TLA7012. I have my TLA715 high up a shelve with the probe cables reaching to the desk level; I use it remotely exclusively. IOW: the failing LCD panel doesn't have to be a showstopper.
--- End quote ---
Yeah, I do not think this type of an equipment will casually use aluminum electrolytic capacitors like regular consumer grade electronics hardware.
For this type of an equipment, I often see rather expensive surface mount tantalum capacitors being used fairly extensively , so "recapping" will likely not happen.
I assume that when Windows XP starts up i915 graphics device driver, it switches from VGA upscaling mode to its native LCD panel mode.
Likely the current draw is greater, hence, the outcome is a disappeared screen.
Of course, the screen comes back instantly when I power cycle TLA7012 by shutting down the box and turning it back on right away.
Based on this, I assume the CCFL still has some life left in it, so I suspect an issue with the inverter circuit.
A rather old 19" Dell monitor with DVI input is sitting on top of my TLA7012 at this time as it's working monitor.
I also ordered TLA7016 that should arrive in a few days.
I plan to revive a little used Windows 7 Pro 64-bit that has not been turned on for about 4 years.
I never dreamed going back to the Microsoft land in my life (I am myself sort of a hardcore Gentoo Linux user.), but I have no choice.
Navigation
[0] Message Index
[#] Next page
Go to full version