Removed the old Dallas chip in the IC333 slot on the YA board, copied the data from the old chip to a new Dallas chip, installed the new chip and put back the YA board, turned on the G90 and Shazam everything was exactly there as before.

Bit of History:
A few months ago I had to turn off my G90 via its hard switch on the back and had left it off for about one hour. When I went to turn it on again I received a series of error code flashes on the error code LCD window on the back of the G90 in the following order: 88, 88, 10, and then remaining at 11. Remote would not turn G90 on and turning the hard switch off and on again produced the same series of error codes. Did some reading on this forum and determined that the battery in my Dallas chip was failing. At that point in time in order to get the G90 going again I had to do a reflash of the firmware software, which after a few attempts I did successfully.

Here is the post on that:

http://www.avsforum.com/avs-vb/showthread.php?t=933217&highlight=dallas

Here is a more detailed procedure to changing the Dallas chip.

1. Turned on G90 and did a Data save to Service Block – Page 132 of G90 Installation manual. This was done as a precaution to a possible total failure of the old Dallas chip whereby it would no longer be readable (i.e. Battery too weak causing corrupt data, lack of readability etc.).
2. Turned G90 off from remote.
3. Turned G90 off from hard switch on back (I did not unplug my G90 but you can if you want to be extra safe).
4. Unscrewed the 10 screws on the back of G90 near where YA board is installed and removed back cover.
5. Unlatched the YA board black clips at either ends of the board and slid out the YA board carefully. Make sure that you are not carrying any static electrical charge as the YA board is sensitive to static electricity and you can end up damaging the board rendering it useless. (I think Sony is currently charging about 1700 to fix it)
6. Placed the YA board on an antistatic foam sheet. Bought the sheet at a local electronics store measuring 18” x 24” for few bucks.
7. Put on an antistatic bracelet and attached it’s alligator clip to the back of my PC.
8. Removed Dallas chip with my fingers while holding the YA board down near the chip. Be careful in prying it loose here as the chip was solidly plugged into a white socket. It is not soldered. You actually need a fair amount of force to pry it off. I had pulled a bit too hard and accidentally hit part of the Dallas chip wires against my other hand and bent some of them. I carefully rebent the wires back into place.
9. I then set up my EPROM programmer and placed the Dallas chip into the programmer chip holder. Very important to align pin one of the Dallas chip in the correct socket or you run the risk of damaging the chip when reading it with the EPROM programmer. There is a small circular indent (chips pin 1 marker) on the top part of the Dallas chip so it was easy to identify pin 1.
10. I then read the chip and saved data (hex file containing serial number, tube hours and user data – convergence, gray scale etc) on my computer.
11. Pulled old Dallas chip off EPROM programmer and installed new Dallas chip and Programmed/verified data to new Dallas chip.
12. Installed the new Dallas chip to the YA board. Make sure that chip is installed with pin one in the correct socket. The old Dallas chip and new Dallas chip looked identical so I just installed the new chip in same direction as the old chip (identified by the direction of the writing on the chips). Make sure you make a note of the direction when you remove the old chip. Again, be somewhat careful as you need a firm amount of force to fully put the Dallas chip all the way back into the socket and you do not want to bend the Dallas chip’s metal pins in the process. I first aligned the chip parallel to the socket on the YA board, initially pushing lightly to get the pins in as much as possible and as evenly as possible and then applied a much firmer even push to both top ends of the Dallas chip to get if fully in.
13. Installed the YA board into my G90 and reinstalled the back plate and 10 screws back in.
14. Turned G90 on from the hard switch on the back and looked for any possible error codes. No codes present.
15. Turned G90 on via remote and went into service mode and checked brightness, contrast, serial number, tuber hours etc.. All information was exactly the same.
16. New Dallas chip should be good for at least 10 more years.

Other information:

-Old Dallas chip was a DS1245Y - 120 with date inscription on top of chip of Dec 1998 (Chip date code 9852H - showing chip made in 1998 on 52nd week)
-New Dallas chip was a DS1245 - 120+ with date inscription on top of chip of 2007 March (0711M – showing chip made in 2007 on 11th week). The “+” sign refers to RoHS compliant and is Pb-free as required in Europe. This makes no difference and since this requirement was introduced in 2007 I knew I would be buying a “newer chip”.
-The old chip was manufactured with battery connected or engaged so that its 10 yr battery life span would start on the manufactured date. Probably reason why my battery started showing signs of dying in 2008 since it was made and battery engaged 10 years earlier in 1998. The new Dallas chip was manufactured without battery connected/engaged so it life starts from the date that I first programmed it which was today even though it was manufactured in March 2007.
-Since I had copied the old data fully from the old Dallas chip to the new one I did not need to reflash the Sony G90 firmware software again as I had previously needed to do when the Dallas chip failed a few months ago.

I hope the above assists in alleviating anyone’s concerns in regards to their G90 and Dallas chip. All you need to do is buy a chip and programmer and you are set. I would suggest changing the chip upon seeing first signs of failure since if the chip becomes too unstable and unreadable then the data might not be transferable. At the very least save your user data to the service block periodically so you at least have a back up there.

I have a couple extra Dallas chips I acquired just in case. Should anyone want to send me their old Dallas chip to make a computer back up and program onto a new chip just let me know.

Dallas chip at Newark:
http://www.newark.com/88K4025/semiconductors-prototyping/product.us0?sku=MAXIM-DS1245Y-120

Below are some pictures of the procedure.

More pictures.

Additional pictures.

There's something unusual about this. I just had issues with a G90 I purchased and when I got it the error 11 code was coming up and the unit wouldn't come out of standby.

This machine had a new Dallas chip installed. Still having the old Dallas chip I swapped back to that with the expectation that the firmware would no longer need to be upgraded. To my surprise I still got the error 11 when switching the machine back on.

So I proceeded with the emergency firmware upload, with the old Dallas chip in place. This worked and then the machine switched on. The next day I wanted to try the new Dallas chip again (the one that had originally displayed the error 11) so I swapped it back into the unit with the expectation that I'd need to again do an emergency firmware update. Well, I swapped the chip back but the unit didn't require the firmware update. So I'm thinking, it's not the Dallas chip that stores the firmware, or something else is happening here?

There's something unusual about this. I just had issues with a G90 I purchased and when I got it the error 11 code was coming up and the unit wouldn't come out of standby.

This machine had a new Dallas chip installed. Still having the old Dallas chip I swapped back to that with the expectation that the firmware would no longer need to be upgraded. To my surprise I still got the error 11 when switching the machine back on.

So I proceeded with the emergency firmware upload, with the old Dallas chip in place. This worked and then the machine switched on. The next day I wanted to try the new Dallas chip again (the one that had originally displayed the error 11) so I swapped it back into the unit with the expectation that I'd need to again do an emergency firmware update. Well, I swapped the chip back but the unit didn't require the firmware update. So I'm thinking, it's not the Dallas chip that stores the firmware, or something else is happening here?

Yes that is strange. When you installed the new Dallas chip the last time did you turn the G90 off from the hard switch?

Could it be possible that somehow the battery on your new Dallas chip is near the end of its life too possible causing erratic behaviour?

Yes hard switch every time I turned it off Frank. I don't think it's the battery in the new Dallas chip, I think there's something else going on here, but I'm not sure what.

It seems that maybe something else on the YA board stores the firmware, and also maybe there's a battery elsewhere on the board too.

I had the unit off at the hard switch (cord unplugged) for several days after that and it still didn't lose it's firmware.

Dallas chip?

http://us.movies1.yimg.com/movies.yahoo.com/images/hv/photo/movie_pix/columbia_pictures/the_fifth_element/bruce_willis/orangetank.jpg

Frank, excellent step by step post. I think that I would feel comfortable getting the Dallas chip out and installing a new one, but do not know that I could do the eprom programming. When you say: "I would suggest changing the chip upon seeing first signs of failure since if the chip becomes to unstable and unreadable then the data might not transferable," what do you mean by the first signs. I'm assuming you mean - error codes. My g90 is at 10 years now, so i suspect the days on my Dallas chip are numbered. I really appreciate your offer to program a new chip and may be taking you up on your offer in the future. Thanks again for your great post.

Dallas chips have a 10 year rated life so it would be a good idea to change it now before it starts giving problems as it may be too late by then.

hear hear. Great write up.

Frank, excellent step by step post. I think that I would feel comfortable getting the Dallas chip out and installing a new one, but do not know that I could do the eprom programming. When you say: "I would suggest changing the chip upon seeing first signs of failure since if the chip becomes to unstable and unreadable then the data might not transferable," what do you mean by the first signs. I'm assuming you mean - error codes. My g90 is at 10 years now, so i suspect the days on my Dallas chip are numbered. I really appreciate your offer to program a new chip and may be taking you up on your offer in the future. Thanks again for your great post.

By first signs I mean:

1. You have a power failure in your house and the Dallas battery is too weak causing lack of readability of Dallas chip by the G90. At this point your G90 will not turn on and will go into firmware emergency upload mode.

2. You turn off your G90 for whatever reason from the hard switch on the back. End result is similar to point 1 above.

3. If you know your G90 is getting close to its 10 year life and it is important for you to keep the same information i.e. serial number, tube hours, convergence information etc..

4. If you recently had the power to your G90 turned off and on again without issue (indicating the battery still has good power in it - although this can change very quickly if the battery is close to its last leg) than you can read the date on your G90 Dallas SRAM chip and if it is getting close to its 10 year life and it is important for you to keep the same information i.e. serial number, tube hours, convergence information etc.. To do this you have to turn off the G90 from the hard switch on the back first and then pull out your YA board and read the top of the chip. The date code should be a 4-digit number followed by a letter.


Just remember if you wait too long you run the risk of the battery completely ding off and possibly corrupting the data at which point it may not be possible to make a good copy.

Do all the Sony projectors from that lineup (G90, G70, D50) use the same chip?

Do you have a model/part number (or name) for the eeprom programmer you used? I searched online and there are bunches of different kinds but I'm having a hard time knowing/finding one that will:

A) Work with a Dallas chip and
B) Has integrated software that will run on XP

Thanks-good post!

Do all the Sony projectors from that lineup (G90, G70, D50) use the same chip?Yes, same kind of NOVSRAM chips but no: different brand and/or reference (i.e. DS1235Y/smaller memory size on G70/D50).
3. If you know your G90 is getting close to its 10 year life and it is important for you to keep the same information i.e. serial number, tube hours, convergence information etc.. PJ serial number, tube hours, chassis hours can be added in a blank chip using the programmer Hex editor. However, convergences/MG focus data will be losted when using a new NOVSRAM chip. John

Hello John,

So if I am reading this correctly, once you install a new Dallas chip and even if you copy all of your data using a programmer to the new chip, you will still lose your convergence/focus settings??

How would one backup all of their user/service/expert/pro settings including conv, focus, mg focus, grey scale tracking, position, size, brightness, contrast...etc..etc..etc..

In other words, backup everything so that in the event that one of these Dallas chips fail, it is simply a matter of installing a new one and you are good to go??

On the NEC XG series, they have a software program that you can backup everything without having to take the unit apart. Having owned my G90 now for over a year and having had a few NEC XG series projectors, I can honestly say that the NEC has a MUCH easier way of protecting/backing up all of the user/service data without having to resort to pulling circuit boards and purchasing a eprom reader/programmer just to accomplish backing up this vital data.

FrankD, EXCELLENT step by step detailed explanation of your procedure!! This should definately be a reference post to save for all G90 owners...


Yes, same kind of NOVSRAM chips but no: different brand and/or reference (i.e. DS1235Y/smaller memory size on G70/D50).
PJ serial number, tube hours, chassis hours can be added in a blank chip using the programmer Hex editor. However, convergences/MG focus data will be losted when using a new NOVSRAM chip. John

So if I am reading this correctly, once you install a new Dallas chip and even if you copy all of your data using a programmer to the new chip, you will still lose your convergence/focus settings??No Brian, it seem's that I was not clear enough :o:o I mean that if you simply use a blank NOVSRAM chip (because you completely loosed the previous Dallas chip content), you can use this blank chip straight on the YA board (it will ask for for the NOVSRAM formating process, once) then, you can still fill-in the PJ serial, tubes and chassis timers values using a programmer. But because you loosed the ability to backup your convergences/MG focus setting with the death of the previous Dallas IC, you can't recovers these settings from scratch unless :
- You've stored these adjutments to the projector service block before (flash IC334),
- you redo the convergences and MG focus adjustments from the beginning and store them within the new Dallas chip.

I hope I'm clearer now :o
In other words, backup everything so that in the event that one of these Dallas chips fail, it is simply a matter of installing a new one and you are good to go?? Yes, you are right it'll work this way :)
John

Thanks John for the clarification.. I am seriously considering the purchase of a replacement Dallas chip along with the programmer to backup my settings. I have spent way too much time getting everything dialed in to lose it.




No Brian, it seem's that I was not clear enough :o:o I mean that if you simply use a blank NOVSRAM chip (because you completely loosed the previous Dallas chip content), you can use this blank chip straight on the YA board (it will ask for for the NOVSRAM formating process, once) then, you can still fill-in the PJ serial, tubes and chassis timers values using a programmer. But because you loosed the ability to backup your convergences/MG focus setting with the death of the previous Dallas IC, you can't recovers these settings from scratch unless :
- You've stored these adjutments to the projector service block before (flash IC334),
- you redo the convergences and MG focus adjustments from the beginning and store them within the new Dallas chip.

I hope I'm clearer now :o
Yes, you are right it'll work this way :)
John

Thanks John for the clarification.. I am seriously considering the purchase of a replacement Dallas chip along with the programmer to backup my settings. I have spent way too much time getting everything dialed in to lose it. Me too Brian :eek: I tried to order some of these chips at Newark and they just stated me that they refused to export those ICs outside USA :mad: I tried to find them in Europe and this is very difficult (found one reseller at $120 :eek:).
I need to change and backup four of my G90 customers here in France ;) just to be honest with them...

John

Do you have a model/part number (or name) for the eeprom programmer you used? I searched online and there are bunches of different kinds but I'm having a hard time knowing/finding one that will:

A) Work with a Dallas chip and
B) Has integrated software that will run on XP

Thanks-good post!

A) I think that just about any Willem EPROM programmer will work. Just make sure the Dallas chip is listed in the EPROM programmer's programmable chip list.

B) I have XP on my computer and it worked with the EPROM programmer software no problem.

This is the place where I purchased my programmer:
http://www.mcumall.com/comersus/store/comersus_viewItem.asp?idProduct=3199

So the recommendation is to save the current convergence, geometry, colorimetry etc data to a service block as a precaution. Assuming the Dallas chip completely fails and is replaced with a new chip, will the data in the service block transfer either automatically or maybe moved manually by recalling service data?

Is there any chance the data on the older but functional chip could be corrupted during the reading process?

I use the Pocket Programmer 2 for these Dallas Chips and they work fine.

Lewis certainly you could acidentially "glitch" the chip if you don't follow the programmers instructions, but the chance of that is remote.

Terry

So the recommendation is to save the current convergence, geometry, colorimetry etc data to a service block as a precaution. Assuming the Dallas chip completely fails and is replaced with a new chip, will the data in the service block transfer either automatically or maybe moved manually by recalling service data?

Is there any chance the data on the older but functional chip could be corrupted during the reading process?

Well it definitely will not hurt to save the service data from time to time.

If the Dallas chip completely fails you than would have to upload the G90 firmware via computer and RS 232 port on the back of the G90. The next step would be to replace the Dallas chip and save the service data back to the user data on the new Dallas chip. I have not tried this but I imagine it should be easy. You may end up losing your tube hours and serial number this way. As John pointed out you can always edit the hex file using a hex editor but then again you have to make sure you know exactly where the hours and serial number is located and also know how to update it.

There is always a chance the data can get corrupted during the reading process but than again thats why you save your data to the service block. Assuming your Dallas chip still has decent battery power in it than it would most likely not be corrupted during the reading process.

On a different note I have read a few posts where some have indicated that they may have experienced spot burn to their CRT phosphors due to a weak/failing Dallas battery (which may be just guessing on their part). Although I do not believe that the Dallas chip battery weakness is associated with this spot burn I cannot say for certain. However besides having a desire to save my serial number, tube hours and user data, this just gave me another reason to change my Dallas chip.

Let's say that one saves their data to the service block of the G90. If the Dallas chip fails before I had time to back it up, would I be able to simply purchase a new chip, install it in the projector and then copy the service block data back to the new Dallas chip?

If this is the case, where is the service data information being held? I assume that the service data includes all of the settings, conv,focus, astig, contrast,brightness...etc.???





Well it definitely will not hurt to save the service data from time to time.

If the Dallas chip completely fails you than would have to upload the G90 firmware via computer and RS 232 port on the back of the G90. The next step would be to replace the Dallas chip and save the service data back to the user data on the new Dallas chip. I have not tried this but I imagine it should be easy. You may end up losing your tube hours and serial number this way. As John pointed out you can always edit the hex file using a hex editor but then again you have to make sure you know exactly where the hours and serial number is located and also know how to update it.

There is always a chance the data can get corrupted during the reading process but than again thats why you save your data to the service block. Assuming your Dallas chip still has decent battery power in it than it would most likely not be corrupted during the reading process.

On a different note I have read a few posts where some have indicated that they may have experienced spot burn to their CRT phosphors due to a weak/failing Dallas battery (which may be just guessing on their part). Although I do not believe that the Dallas chip battery weakness is associated with this spot burn I cannot say for certain. However besides having a desire to save my serial number, tube hours and user data, this just gave me another reason to change my Dallas chip.

Brian,

The service data (along with the factory data) is being held in the chip that is in the IC334 spot on the YA board. The Dallas 1245Y 120+ chip that is in the IC333 spot on the YA board is holding the user data.

Not to cause panic or anything, but if your set is going on 10 yrs old, it's time to change the dallas chip.
There have been three instances now (one in NJ, one in the mid west and the most recent in Utah) where users suspected that they had Dallas Chip failures (they were getting the warnings on the LED) and in those three cases there was a full collapse of the deflection causing all three tubes to be destroyed.
This third one in Utah posted here on AVS about that a few weeks ago that he was going to send me his G90 so that I could try and duplicate the failure and share that info. But alas after we made contact to get me the machine to me, I never heard back from him. Hopefully he is just busy.

Terry

The service data (along with the factory data) is being held in the chip that is in the IC334 spot on the YA board.Exactly : IC334 is a "flash" memory technology of chip (same kind as IC234 which hold the main projector firmware, but it is smaller in memory size). Using this chip to backup user data into service data is always safer than leaving user data only inside the Dallas NOVSRAM (and it cost nothing for the user ;) ).

Terry, I seriously doubt that the dallas chip failure can cause the three tubes burn spots but if we can be sure that in ALL the three instance cases, theirs dallas chips date code/PJ year of manufacture were found prior to '98 : we should conclude that the effect has found the cause even this seem's not logical in the electronic point of view...

John

Terry, I seriously doubt that the dallas chip failure can cause the three tubes burn spots but if we can be sure that in ALL the three instance cases, theirs dallas chips date code/PJ year of manufacture were found prior to '98 : we should conclude that the effect has found the cause even this seem's not logical in the electronic point of view...

John

John,

While I agree with what you say and beam deflection failure makes no sense to me from a Dallas chip point of view, we can show three instances. On one of them I have the YA board here and have found a short on it from pins 36-38 on the Boot chip. But since that user didn't really know what they were doing when they attempted to change the dallas chip, I don't know if this short is user induced or another failure they have.
The schematics don't show anything about the boot chip that I have been able to find so tracing down this short is very slow work.

Terry

Since there is a slight chance that a 'novice' could screw up his Dallas chip swap and/or fry the YA board is there any professional willing to offer this a service either by sending the complete YA board (preferable) or the old Dallas chip? I know Frank kindly offered but I would prefer to send it to someone in the US or Canada.

Since there is a slight chance that a 'novice' could screw up his Dallas chip swap and/or fry the YA board is there any professional willing to offer this a service either by sending the complete YA board (preferable) or the old Dallas chip? I know Frank kindly offered but I would prefer to send it to someone in the US or Canada.

Just to let you know I do live in Canada, near Toronto, Ontario.

My apologies Frank. For some reason I thought you lived in Australia. Just sent you a PM.

Guys, sorry about these novice type questions but I want to clairfy something. I have only one memory (21) that I use on a regular basis due to my Lumagen output being 1080P @ 59.94. I have saved this memory to my service block location. Does this mean that if my Dallas chip craps out on me one day, I will still be okay because I saved that input to the service block?

On a seperate note, althouth the G90 has not been made in years, I can not believe that Sony did not provide a simpler method to back up the memories in case of the Dallas time clock failure. Anyone that owns a CRT projector knows how much time and effort it takes to get them tuned in and calibrated to perfection. To lose all of this just because there was not an easier way to back up this data was a major oversite IMO on Sony's part....







Exactly : IC334 is a "flash" memory technology of chip (same kind as IC234 which hold the main projector firmware, but it is smaller in memory size). Using this chip to backup user data into service data is always safer than leaving user data only inside the Dallas NOVSRAM (and it cost nothing for the user ;) ).

Terry, I seriously doubt that the dallas chip failure can cause the three tubes burn spots but if we can be sure that in ALL the three instance cases, theirs dallas chips date code/PJ year of manufacture were found prior to '98 : we should conclude that the effect has found the cause even this seem's not logical in the electronic point of view...

John

Brian,

Yes, however any change to the user settings since your last save to the service block will not be saved.

Also when you install your new Dallas chip you will end up losing your serial number and tube hours (unless you can figure out how to use a hex editor to get the information back in) and you will have to do a firmware upload when installing the new Dallas chip (via serial port on the back of the G90 and a computer). You will also have to use the G90 to format the new Dallas chip. These steps can be avoided by copying the data from the existing good Dallas chip to a new one.

Add to the above the spot burn issue (that may or may not be directly or indirectly related to the Dallas chip failure) and the fact that the chip's battery contains a documented 10 year life span than for what it costs to change/program the chip I think it is a good idea to have the chip eventually changed and probably sooner then later.

Terry, I seriously doubt that the dallas chip failure can cause the three tubes burn spots but if we can be sure that in ALL the three instance cases, theirs dallas chips date code/PJ year of manufacture were found prior to '98 : we should conclude that the effect has found the cause even this seem's not logical in the electronic point of view...

John

The CRT protect circuit (1K) for the neck boards, are sensed at the neck boards and controlled by the "YA" board. The "YA" board is what blanks or mutes the neck boards.

The 1K sense from the neck boards, are also involved in the Error protect circuit (IC861).

And based on the schems, they have designed a rather sophisticated protect circuit (timing) to mute the video from the "YA" board, which is a good idea, which is how they also mute the video when there's an error code being displayed.

The spot burn problems has been consistent with a particular error code reading. And since the problem seems to be more related to time than a defect in the circuits, I would place the Dallas chip as being the key suspect here. if it fails, it can cause any of the circuits associated with the "YA" board to not function as they should.

Even on the G90 that I have here. Once I had replaced the Dallas chip, I did manage to get a whole different readout on the display. To include being able to make the main error code problem I was having go away. So the Dallas chip is directly related to what goes on with the Error codes. And I'm sure it's the cause of the spot burn.

Why Not Mike,

If Sony has developped a really good CRT protection design, why they did not check the whole chain (YA failure) then ? (AMDEC). I'm very surprized that Sony did not thought that wrong Dallas data readout can let the YA board circuit to badly react and completely bypass all this CRT protection design... If this is true : this is a real weakly side (destroying)effect! To include being able to make the main error code problem I was having go away. Do you mean that your "88" red error code you've got on the YA board completely dissapeared when you've changed the Dallas IC ?

John

Just a little more FYI here on this Dallas chip mystery.
A while ago there was someone on here who thought he was having a Dallas chip failure in a new machine from NJ. He had asked for help in replacing his Dallas chip and then when he did get the machine to fire up briefly had spot burn on all three tubes.
What happened immediately after that is an unknown to me but he machine got stuck in orange 88 which it is attempting to "boot".
So it had a catastrophic failure but I don't know what was done to the machine after that and I don't know if the scan failure happened furing the first failure of the second one that lean to a boot failure.
I did just recently find that the Boot Chip (IC233??) had a short on pins 38 and 38.
Tracing that down (it's data lines) I found the short to be on one of the SRAM chips (IC235). I have to get someone to put some of the sm chips back on to get the board to fire back up but in my mind I think I found the orange 88.
What I am wondering is why the SRAM would fail??
Something that had bugged me about the G90 is the lack if circuit protection under the area of the ROM and RAM momemtary switches on the back center of the YA board and whether we could be having problems associated with these or some of the other traces being shorted when the switches are hit??

Terry

Yes, same kind of NOVSRAM chips but no: different brand and/or reference (i.e. DS1235Y/smaller memory size on G70/D50).
PJ serial number, tube hours, chassis hours can be added in a blank chip using the programmer Hex editor. However, convergences/MG focus data will be losted when using a new NOVSRAM chip. JohnHi John,Could you please tell me the exact name of this dallas chip and part no used in the G70QM is it Dallas DS1235Y. Because I had my G70 unpluged for about A week or so, and when I pluged it back into the wall in standby mode(never turned it on) the error code 88 showed up so I unpluged the projector for 24 hours then when it was plugged back in the error was gone.But last night when I started my G70 up to calibrate the error code 10 stayed on(abnormal input signal) but the projector worked ok,I never ran my G70 when showing the error 88 in the LED window....
thanks....

Thanks Frank.. It looks like I am going to have to purchase a new Dallas chip and programmer then. From what you and others are saying, it is cheap insurance to replace it now before it kicks the bucket!! :)

Since I have never used an EProm programmer before are there any hints or tips that you could provide? I have read the first post here which you documented your step by step procedures which are excellent by the way...




Brian,

Yes, however any change to the user settings since your last save to the service block will not be saved.

Also when you install your new Dallas chip you will end up losing your serial number and tube hours (unless you can figure out how to use a hex editor to get the information back in) and you will have to do a firmware upload when installing the new Dallas chip (via serial port on the back of the G90 and a computer). You will also have to use the G90 to format the new Dallas chip. These steps can be avoided by copying the data from the existing good Dallas chip to a new one.

Add to the above the spot burn issue (that may or may not be directly or indirectly related to the Dallas chip failure) and the fact that the chip's battery contains a documented 10 year life span than for what it costs to change/program the chip I think it is a good idea to have the chip eventually changed and probably sooner then later.

Terry,

Is this the unit that was being donated to you for research or has that one not arrived yet?

Ya know it's kind of ironic. Projectors are given to you guys. However, you are only accepting them so you can give BACK even MORE in return.

It's guys like Mike, John, Curt and you that keep this hobby at the forefront of performance and reliability.

CRT will never die as long as the "Final Four" are still around!!!!!!

:)

Thanks for all you guys do!!!!

Doc

Hi John,Could you please tell me the exact name of this dallas chip and part no used in the G70QM is it Dallas DS1235Y. On the G70, the NOVSRAM chip reference is M48Z128Y-120 (from "ST" IC manufacturer for instance). This chip is 128Kbx8 (1Mbits). I said something wrong : this size of NOVSRAM in the G70 is the same as the Dallas/Maxim DS1245Y-120 used in the G90 :o Sorry for the mistake. This means that that this M48Z128Y-120 of a G70 can be replaced by Dallas/Maxim DS1245Y-120 reference ...

Here is where and how the informations (PJ ser-num, timers infos) are recorded the G70/G90 NOVSRAMs:

- Projector serial number is stored in [B0-B5] memory addresses by simply convert PJ label serial number in hexadecimal format (i.e. serial 2000420 -> #0000001E8624).

- Chassis timer is stored in [82-85] memory addresses by simply convert chassis hours in 1/100' format in hexadecimal format (i.e. 8633hrs=3108016324 (1/100sec.) -> #B94090C4).

- Blue tube timer is stored in [86-89] memory addresses by simply convert blue tube hours in 1/100' format in hexadecimal format (i.e. 635hrs -> #0DA028C0).

- Green tube timer is stored in [8A-8D] memory addresses by simply convert green tube hours in 1/100' format in hexadecimal format (i.e. 635hrs -> #0DA028C0).

- Red tube timer is stored in [8E-91] memory addresses by simply convert red tube hours in 1/100' format in hexadecimal format (i.e. 635hrs -> #0DA028C0).

This can help some people that have losted these infos because of a dead (unreadable or blank) Dallas chip ;);) This is also usefull for people that want to exchange their YB board (for G70) or YA board (G90) that failed with another one from another spare unit (en recover their previous settings/records)

John

A while ago there was someone on here who thought he was having a Dallas chip failure in a new machine from NJ. He had asked for help in replacing his Dallas chip and then when he did get the machine to fire up briefly had spot burn on all three tubes.
What happened immediately after that is an unknown to me but he machine got stuck in orange 88 which it is attempting to "boot". Interresting story Terry : you've been a witness of this rare event ;)
So it had a catastrophic failure but I don't know what was done to the machine after that and I don't know if the scan failure happened during the first failure of the second one that lean to a boot failure. Arggg, this is no 'luck' : we can't conclude exactly in the events chronology :o
I did just recently find that the Boot Chip (IC233??) had a short on pins 38 and 38. Do you mean pins 37&38 ?
Tracing that down (it's data lines) I found the short to be on one of the SRAM chips (IC235). I have to get someone to put some of the sm chips back on to get the board to fire back up but in my mind I think I found the orange 88.
What I am wondering is why the SRAM would fail?? maybe IC latch-up that falls in short between adjacent pins ?

John

- Chassis timer is stored in [82-85] memory addresses by simply convert chassis hours in 1/100' format in hexadecimal format (i.e. 8633hrs=3108016324 (1/100sec.) -> #B94090C4).
Why in God's name did they think they needed to store chassis time in 100ths of a second!?

Doesn't this cause a slight problem? Locations 82-85 are 4 bytes. FFFFFFFF = 4294967295 100ths of a second. Which by my calculations works out to just over 11930 hours !!???

Apparently they didn't expect Marquee-like longevity out of these beasts!? :)

Terry,
Are you still coming to Dallas for a G90 callibration "tour"? If so, maybe I should get my Dallas chip replaced. My G90 was bought new only 5 years ago with a late serial number and I have just less than 1000 hours on the unit. I would hate to have it fried due to an old chip.
Kevin

Might be worth waiting to schedule the calibration tour until the new Mike Parker mods are available for the G90.

Terry,
Are you still coming to Dallas for a G90 callibration "tour"? If so, maybe I should get my Dallas chip replaced. My G90 was bought new only 5 years ago with a late serial number and I have just less than 1000 hours on the unit. I would hate to have it fried due to an old chip.
Kevin

If you only bought your G90 five years ago and it has a late serial number than you may still have a few years of life left on that battery. If you want to know the exact date of chip production then remove the YA board as discussed in my initial post. Look at the top of the Dallas chip and look for a four digit number followed by a letter. From that you can tell when the chip was manufactured. Older Dallas chips where manufactured with battery engaged so that the 10 year clock started from its manufactured date. Newer Dallas chips are manufactured without battery engaged. The battery becomes engaged when you first apply voltage to the chip (ie. when you first program the chip etc.).

Why in God's name did they think they needed to store chassis time in 100ths of a second!? Probably because the PJ software simply read the PJ RTC (Real Time Clock) registers which are stored in 1/100 sec. and because they have plenty of available space within NOVSRAM, they don't care ;) Doesn't this cause a slight problem? Locations 82-85 are 4 bytes. FFFFFFFF = 4294967295 100ths of a second. Which by my calculations works out to just over 11930 hours !!??? This is explaining why these PJ hours counters roll over is 10 000 hours !

John

Thankyou JohnHWman for your info I realy appreciate it....

Just a little more FYI here on this Dallas chip mystery.
A while ago there was someone on here who thought he was having a Dallas chip failure in a new machine from NJ. He had asked for help in replacing his Dallas chip and then when he did get the machine to fire up briefly had spot burn on all three tubes.
What happened immediately after that is an unknown to me but he machine got stuck in orange 88 which it is attempting to "boot".
So it had a catastrophic failure but I don't know what was done to the machine after that and I don't know if the scan failure happened furing the first failure of the second one that lean to a boot failure.
I did just recently find that the Boot Chip (IC233??) had a short on pins 38 and 38.
Tracing that down (it's data lines) I found the short to be on one of the SRAM chips (IC235). I have to get someone to put some of the sm chips back on to get the board to fire back up but in my mind I think I found the orange 88.
What I am wondering is why the SRAM would fail??
Something that had bugged me about the G90 is the lack if circuit protection under the area of the ROM and RAM momemtary switches on the back center of the YA board and whether we could be having problems associated with these or some of the other traces being shorted when the switches are hit??

Terry

I have a feeling that gentlemen didn't switch his G90 off at the mains - he thought it being in standby was enough and he pulled the YA board :(

Which led to the YA failure on that machine and then the spot burn.

The YA board is such an Achilles heel.

I have a feeling that gentlemen didn't switch his G90 off at the mains - he thought it being in standby was enough and he pulled the YA board :(
Which led to the YA failure on that machine and then the spot burn. In fact, the WHOLE YA board is powered (CPUs are all running) during the standby mode of the unit (to be able to catch the wake-up IR code from the IR remote). If you pull out the YA board from its slot in Standby mode, it may cause unpredictable results on the rest of the electronic. However, since the rest of the PJ is OFF (including the powers that drives the tube deviators and the G1/G2/K), I don't think that this could cause spot burn on the tube phosphore...

I rather think that this can damage the YA board itself and then when you re-insert it inside the unit, this failure may cause the spot burn ?

I've got a french guy here in France that did the same on his new G70 : he pulled out the YA and YB boards while the unit was in Standby : he broke the YB board (only :rolleyes:) but no spot burn (fortunatly :cool:).

John

I rather think that this can damage the YA board itself and then when you re-insert it inside the unit, this failure may cause the spot burn ?
John

I think your observation is spot on, forgive the pun!

Interresting story Terry : you've been a witness of this rare event ;)
Arggg, this is no 'luck' : we can't conclude exactly in the events chronology :o
Do you mean pins 37&38 ?
maybe IC latch-up that falls in short between adjacent pins ?

John

Sorry John, it's 36 and 38

Terry

Well, I have been convinced that I need to change the Dallas chip and thus have ordered the chip and eprom programmer. Hope to have both in about a week. I am hoping that based on Frank's step-by-step process that I will be able to successfully perform the operation. Once done I will be happy to share my Eprom programmer with other G90 owners for the price of round trip postage.
Frank, I ordered the one you used/recommended - I assume that it comes with the software I will need to transfer the data from the old chip to the new chip. This should be a good test to see if a layman can successfully change the Dallas chip. Stay tuned.

Me too Brian :eek: I tried to order some of these chips at Newark and they just stated me that they refused to export those ICs outside USA :mad: I tried to find them in Europe and this is very difficult (found one reseller at $120 :eek:).
I need to change and backup four of my G90 customers here in France ;) just to be honest with them...

John

I ordered mine here :-

http://uk.farnell.com/1285407/semiconductors/product.us0?sku=MAXIM-DS1245Y-120&_requestid=260841

Farnell also have a French site so I am sure you would be able to get them in France.

http://fr.farnell.com/1285407/semiconductors/product.us0?sku=MAXIM-DS1245Y-120&_requestid=260841

What is considered A safe time to wait after turning off the hard switch on G90/70 before pulling out the YA board G90/YB board G70 when changing the Dallas chip,I have had no problem when pulling the YA/YB board after waiting about 5-10mins there is A little clicking sound (relay) after 1min after turning off hard switch. I think after that click it would be safe to pull board??....

I believe the capacitors will still hold voltage for quite some time. I would suggest leaving it off overnight just to be on the safe side before removing the boards.

What is considered A safe time to wait after turning off the hard switch on G90/70 before pulling out the YA board G90/YB board G70 when changing the Dallas chip,I have had no problem when pulling the YA/YB board after waiting about 5-10mins there is A little clicking sound (relay) after 1min after turning off hard switch. I think after that click it would be safe to pull board??....

Turn it off with the mains power, unplug the unit and then hit the "power on" button on the remote a few times to make sure everything is discharged. It should be fine then.

Terry

With regards the lithium ion power source in the chip itself. Is it rechargeable and if yes, how long will it continue to provide power to the memory blocks. I'm planning to send my Dallas chip away for copying and want it to arrive with all its stored parameters intact.

Turn it off with the mains power, unplug the unit and then hit the "power on" button on the remote a few times to make sure everything is discharged. It should be fine then.

TerryThankyou Terry for your answer....

I believe the capacitors will still hold voltage for quite some time. I would suggest leaving it off overnight just to be on the safe side before removing the boards.Thankyou Lewis for your answer....

Well My G70's old NOVSRAM M48Z128Y-120 has now been replaced with A new DALLAS/MAXIM DS1245Y-120+ as of today, all working well thankyou G90 Richardc of Australia your time and help is appreciated ....

Anytime Jason
Richard

Jason I also have your .bin file for the NVRAM in case of an emergency, I will send it to you shortly
Richard

With regards the lithium ion power source in the chip itself. Is it rechargeable and if yes, how long will it continue to provide power to the memory blocks. I'm planning to send my Dallas chip away for copying and want it to arrive with all its stored parameters intact.

From Dallas
SRAM has a self-contained lithium energy source and
control circuitry which constantly monitors VCC for an out-of-tolerance condition. When such a condition occurs, the lithium energy source is automatically switched on and write protection is unconditionally
enabled to prevent data corruption.

and

FEATURES

10 years minimum data retention in the
absence of external power

Data is automatically protected during power
loss

So from what I understand it is just a battery, that is "switched" into the circuit the first time VCC reaches 5VDC to insure "freshness of the battery. I don't think they are charging the battery nor do I believe we are ever using the batter so long as there is a VCC being applied to the chip. But when the unit is turned off (I'll have to check if VCC is there is you don't turn off mains power) or unplugged, the data is being retained via the battery. This battery is internal and I don't see any way of measuring it on any of the pins.

Terry

anybody know what the difference between DS1245Y-120 and DS1245Y-100 is? I ordered the DS1245Y-120 but received DS1245Y-100. Can the "100" chip be used instead of the "120"?

Should be fine :)

One is 120ns the other is 100 - you can usually use the faster spec'ced chip in place of a slower one - but not the other way around