View previous topic :: View next topic |
Author |
Message |
ElTopo
Joined: 07 Nov 2006 Posts: 1608
|
Link Posted: Wed Oct 10, 2018 8:30 pm Post subject: |
|
|
The date code in this thread says '9941H' which means manufactured 1999 in the 41th week
http://www.curtpalme.com/forum/viewtopic.php?t=33484
Mine do have a date code of '17xxx' which were manufactured 2017
_________________ Barco Cine 9 the one and only
|
|
Back to top |
|
|
dave148
Joined: 15 Feb 2011 Posts: 33 Location: Colorado Springs
|
Link Posted: Thu Oct 11, 2018 4:08 am Post subject: |
|
|
For what version of the part? i.e. what is the full part number?
|
|
Back to top |
|
|
dave148
Joined: 15 Feb 2011 Posts: 33 Location: Colorado Springs
|
Link Posted: Thu Oct 11, 2018 4:13 am Post subject: Original scan freeze = bad battery data? |
|
|
Does anyone know where I can find data leading to the conclusion that a bad DS1245 battery caused the scanning to stop?
|
|
Back to top |
|
|
HaydnG90
Joined: 22 May 2006 Posts: 1335
|
Link Posted: Thu Oct 11, 2018 4:51 am Post subject: |
|
|
PM Craig Rounds. He's about the best person nowadays to ask that sort of question. The salient point is you have a 20 year old Dallas chip. You could nuke your tubes to continue running a Dallas that old. Why risk it? The chips are ~$60. If it solves your other issue, great. If not at least you can run the G90 safely while figuring out what's wrong.
|
|
Back to top |
|
|
dave148
Joined: 15 Feb 2011 Posts: 33 Location: Colorado Springs
|
Link Posted: Thu Oct 11, 2018 5:17 am Post subject: |
|
|
HaydnG90 wrote: | PM Craig Rounds. He's about the best person nowadays to ask that sort of question. |
How do I do that?
Did you ever check what happens when you hit the "A" key on your unit?
|
|
Back to top |
|
|
dave148
Joined: 15 Feb 2011 Posts: 33 Location: Colorado Springs
|
Link Posted: Thu Oct 11, 2018 5:49 am Post subject: |
|
|
HaydnG90 wrote: | .... The salient point is you have a 20 year old Dallas chip. You could nuke your tubes to continue running a Dallas that old. Why risk it? The chips are ~$60. If it solves your other issue, great. If not at least you can run the G90 safely while figuring out what's wrong. |
Since I'm an engineer, I like to understand what I am working on.
It sounds like one or more people experienced some loss of scan, and the problem was blamed on the age of the SRAM battery. This is entirely possible, but it is inconsistent with what I currently understand about the system:
1. The G90 theory of operation document indicates this part is only used to store user data, and so far, I haven't found a way for the user to set the scan dimension below a minimum reasonable value.
2. The datasheet for the current versions of this part make it clear that the battery is not connected when the part is fully powerd up, indicating a bad battery won't cause a problem while the part is powered.
3. People in this forum have indicated that a new, unprogrammed DS1245 won't cause any problems, so I can't come up with a scenario where a programmed DS1245 with a dead battery could cause a problem, since data on an unused static RAM is generally defined as unknown. So unless the other chip in the package does some sort of nice initializing of the data the first time this thing is powered up, the data will be even more strange/ random than you get with the corruption of previously written user data. So far, I have found nowhere in the datasheet this other chip is used for th is purpose, only to determine when to connect the battery.
Now don't get me wrong, I've seen many unexpected effects in my career, so I am looking for more data to figure this out, but usually the scan problem you mention would be caused by something like a failure in the high voltage VCO circuit, not by some user data getting corrupted.
If it ends up being the user data causing the problem, it would be an interesting story, so I'm all ears.
Dave
|
|
Back to top |
|
|
HaydnG90
Joined: 22 May 2006 Posts: 1335
|
|
Back to top |
|
|
Curt Palme CRT Tech
Joined: 08 Mar 2006 Posts: 24305 Location: Langley, BC
TV/Projector: All of them!
|
Link Posted: Thu Oct 11, 2018 11:14 pm Post subject: |
|
|
That sounds like Sony trying to cover their butt.
I'm a board level tech, but don't get into reverse engineering stuff like you post about, I just know that I've seen a bunch of spot burned Sony G90s with no other problem short of a bad Dallas chip. It's so rampant that pretty much every Sony G90 owner that's spent time here knows about it.
Now, the Ya board that the Dallas chip is on can develop other problems (error code 88 being one of them) that can cause spot burn as well, but change the chip!
|
|
Back to top |
|
|
dave148
Joined: 15 Feb 2011 Posts: 33 Location: Colorado Springs
|
Link Posted: Fri Oct 12, 2018 12:00 am Post subject: sovled |
|
|
It turns out the 5BNC mode was set on. This is only for using those jacks as an output. When I set it to off, I could select input A.
|
|
Back to top |
|
|
dave148
Joined: 15 Feb 2011 Posts: 33 Location: Colorado Springs
|
Link Posted: Fri Oct 12, 2018 2:21 am Post subject: |
|
|
Curt Palme wrote: | That sounds like Sony trying to cover their butt. |
What do you mean?
|
|
Back to top |
|
|
dave148
Joined: 15 Feb 2011 Posts: 33 Location: Colorado Springs
|
Link Posted: Fri Oct 12, 2018 2:30 am Post subject: |
|
|
Curt Palme wrote: | ...I'm a board level tech, but don't get into reverse engineering stuff like you post about, I just know that I've seen a bunch of spot burned Sony G90s with no other problem short of a bad Dallas chip. It's so rampant that pretty much every Sony G90 owner that's spent time here knows about it. |
Maxim warns in multiple documents against negative voltage spikes. In the FAQ it mentions negative voltages causing data corruption. Two other possibilities outside the conventional wisdom are:
1. A negative supply glitch freezes the high voltage VCO, and causes corruption in the user data at the same time.
2. A problem with the VCO actually causes a negative voltage in the supply.
Neither of these scenarios has the DS1245 actually causing the problem, but being a victom of the problem.
|
|
Back to top |
|
|
AnalogRocks Forum Moderator
Joined: 08 Mar 2006 Posts: 26690 Location: Toronto, Ontario, Canada
TV/Projector: Sony 1252Q, AMPRO 4000G
|
Link Posted: Fri Oct 12, 2018 11:18 pm Post subject: |
|
|
dave148 wrote: | Curt Palme wrote: | ...I'm a board level tech, but don't get into reverse engineering stuff like you post about, I just know that I've seen a bunch of spot burned Sony G90s with no other problem short of a bad Dallas chip. It's so rampant that pretty much every Sony G90 owner that's spent time here knows about it. |
Maxim warns in multiple documents against negative voltage spikes. In the FAQ it mentions negative voltages causing data corruption. Two other possibilities outside the conventional wisdom are:
1. A negative supply glitch freezes the high voltage VCO, and causes corruption in the user data at the same time.
2. A problem with the VCO actually causes a negative voltage in the supply.
Neither of these scenarios has the DS1245 actually causing the problem, but being a victom of the problem. |
When Mac was still here he found that the G90's chip would get looped like the old satelite boxs. Wonder if that's caused by a voltage spike.
_________________ Tech support for nothing
CRT.
HD done right!
|
|
Back to top |
|
|
dave148
Joined: 15 Feb 2011 Posts: 33 Location: Colorado Springs
|
Link Posted: Sat Oct 13, 2018 3:58 am Post subject: |
|
|
AnalogRocks wrote: | dave148 wrote: | Curt Palme wrote: | ...I'm a board level tech, but don't get into reverse engineering stuff like you post about, I just know that I've seen a bunch of spot burned Sony G90s with no other problem short of a bad Dallas chip. It's so rampant that pretty much every Sony G90 owner that's spent time here knows about it. |
Maxim warns in multiple documents against negative voltage spikes. In the FAQ it mentions negative voltages causing data corruption. Two other possibilities outside the conventional wisdom are:
1. A negative supply glitch freezes the high voltage VCO, and causes corruption in the user data at the same time.
2. A problem with the VCO actually causes a negative voltage in the supply.
Neither of these scenarios has the DS1245 actually causing the problem, but being a victom of the problem. |
When Mac was still here he found that the G90's chip would get looped like the old satelite boxs. Wonder if that's caused by a voltage spike. |
Excuse my weak vocabulary: "looped"? Do you mean it was in a ground loop?
which G90 chip?
|
|
Back to top |
|
|
HaydnG90
Joined: 22 May 2006 Posts: 1335
|
Link Posted: Tue Nov 06, 2018 8:03 pm Post subject: |
|
|
Get this all figured out?
|
|
Back to top |
|
|
Curt Palme CRT Tech
Joined: 08 Mar 2006 Posts: 24305 Location: Langley, BC
TV/Projector: All of them!
|
Link Posted: Tue Nov 06, 2018 8:28 pm Post subject: |
|
|
AnalogRocks wrote: | dave148 wrote: | Curt Palme wrote: | ...I'm a board level tech, but don't get into reverse engineering stuff like you post about, I just know that I've seen a bunch of spot burned Sony G90s with no other problem short of a bad Dallas chip. It's so rampant that pretty much every Sony G90 owner that's spent time here knows about it. |
Maxim warns in multiple documents against negative voltage spikes. In the FAQ it mentions negative voltages causing data corruption. Two other possibilities outside the conventional wisdom are:
1. A negative supply glitch freezes the high voltage VCO, and causes corruption in the user data at the same time.
2. A problem with the VCO actually causes a negative voltage in the supply.
Neither of these scenarios has the DS1245 actually causing the problem, but being a victom of the problem. |
When Mac was still here he found that the G90's chip would get looped like the old satelite boxs. Wonder if that's caused by a voltage spike. |
No, this was the design/production problem of the IC421 that plagued the G90 a bunch of years back.
|
|
Back to top |
|
|
dave148
Joined: 15 Feb 2011 Posts: 33 Location: Colorado Springs
|
Link Posted: Tue Nov 06, 2018 9:01 pm Post subject: |
|
|
HaydnG90 wrote: | Get this all figured out? |
Yes, it was the 5BNC mode setting.
Evidently, this implies an output on those connections, so once I turned it off, I could seleect input A.
|
|
Back to top |
|
|
HaydnG90
Joined: 22 May 2006 Posts: 1335
|
Link Posted: Tue Nov 06, 2018 9:10 pm Post subject: |
|
|
Well that was simple. Glad you got it figured out.
|
|
Back to top |
|
|
|
|