What more i can do?

[asis]

I can't imagine how deep we'll go into this topic, it seems crazy to me, but it's time to move on to some fragments of circuitry.
There will be more mutual understanding.
-
I hope for your help.

[222Lab_Test222]

I can't imagine how deep we'll go into this topic, it seems crazy to me, but it's time to move on to some fragments of circuitry.
There will be more mutual understanding.
-
I hope for your help.

I am all in.

[Swake]

@222Lab_Test222

Would you please redo the measurement of OSC3 near the TORNADO chip that you already did in message 252 https://www.eevblog.com/forum/repair/what-more-i-can-do/msg5720533/#msg5720533

In that image the frequency measured by the scope is correct (2.45MHz) but the voltage is very high (over 16 Volt). As someone else pointed out, it might have been because your probe was set to 1x instead of 10x. Let's make sure things are normal here.

[m k]

What say CPU BS8, BS16, BNE0 and BNE1?

Here are signals for Non-Working Board

Those are fine.

Next we can check the combined ROM image and see that F800 has
F800 FA (CLI)
F801 E4 (IN)
F802 35 (port)(irq controller)
F803 A8 (test)
So completely rational start of a code.

How we can perform this?

It's time consuming without a logic analyzer and not very informative yet.
We know that from CPU's point of view the ROM data is wrong, so the above is not happening.

After the High and Low EROMS swap in Non Working Board.

How did you do it exactly, put low ROM to high socket and high ROM to low socket?

[222Lab_Test222]

How did you do it exactly, put low ROM to high socket and high ROM to low socket?

Yes like that.

[222Lab_Test222]

@222Lab_Test222

Would you please redo the measurement of OSC3 near the TORNADO chip that you already did in message 252 https://www.eevblog.com/forum/repair/what-more-i-can-do/msg5720533/#msg5720533

In that image the frequency measured by the scope is correct (2.45MHz) but the voltage is very high (over 16 Volt). As someone else pointed out, it might have been because your probe was set to 1x instead of 10x. Let's make sure things are normal here.

Yeah I confirm it.
The voltage was indeed 2V

[222Lab_Test222]

By the way if you guys know or not,
the current consumption of No Working Board is less than current of Working Board.
Usually Working Board current is 2.8A tentative and No Working Board is 2.3A tentative.

[asis]

Hi,


I have already noticed this (Reply #420 on: December 09, 2024 p.17) and this means that the POST process and the initialization process have been successfully completed and the device is in the working cycle, with the working image displayed on the screen, and all peripheral ports are polled, including KBD, Mouse, PRN, ISA.....and whatever else is there....
And also the SIMM DRAM (SIM2 RAS/CAS) is refreshed and the GPU (VGA) is working, forming a raster.
All this explains the consumption level of I ~2.8A and it can be higher, depending on the tasks.
-
Q1. Do you have the ability to raise the pins pin10(!); pin11 (IC23 uPD4990), or dismantle it, since you still intended to replace it with a new one,
and turn on the device.
-
Q2. Do you have DRAM memory strips (SIM2) in other cases.
Please take pictures from both sides (preferably after removing the stick).


Thank you.

[222Lab_Test222]

Hi,


I have already noticed this (Reply #420 on: December 09, 2024 p.17) and this means that the POST process and the initialization process have been successfully completed and the device is in the working cycle, with the working image displayed on the screen, and all peripheral ports are polled, including KBD, Mouse, PRN, ISA.....and whatever else is there....
And also the SIMM DRAM (SIM2 RAS/CAS) is refreshed and the GPU (VGA) is working, forming a raster.
All this explains the consumption level of I ~2.8A and it can be higher, depending on the tasks.
-
Q1. Do you have the ability to raise the pins pin10(!); pin11 (IC23 uPD4990), or dismantle it, since you still intended to replace it with a new one,
and turn on the device.
-
Q2. Do you have DRAM memory strips (SIM2) in other cases.
Please take pictures from both sides (preferably after removing the stick).


Thank you.

1. I have done that in last time, do you want me to do again, I can do it no worries. BUT last time i did it did not run.
2. i do not understand, this. I mean recently i send the DRAM images already. Can you clarify more to it.

[asis]

Ok.
Let's leave it (1)...
I'm just haunted by output pin 9 (MOSFET).
What does it block?
-
You had SIM 1 strips with DRAM chips in other x 8 cases.
-
I don't rule out that in addition to the current date and time (IC23), Flash RAM (SIM2) is polled, where the profile of the current parameters of the power unit itself can be stored, i.e. the polling algorithm during initialization includes two consecutive stages.
-
Sorry, I made a mistake:
DRAM - "SIM 1"
FRAM - "SIM 2"

[Swake]

In one of the pictures there is an issue with DRAM slot, see attached.
I cannot say for sure if contact is made correctly or not. Certainly worth checking out in detail.

[asis]

If it's "SIM 2" - next to CPU, then it's pin 70 of SIMM72 (FRAM).
This is PD3 (Presense Detect) determining the full available FRAM volume, with all the ensuing consequences.
-
@Swake
Please give the post number to view the content.

[222Lab_Test222]

It is indeed worth checking those Flash sticks BUT the problem still same.
Why?
Because in working board, when given 5v there is no beeping even if both flash and DRAM is removed.

[222Lab_Test222]

Ok.
Let's leave it (1)...
I'm just haunted by output pin 9 (MOSFET).
What does it block?
-
You had SIM 1 strips with DRAM chips in other x 8 cases.
-
I don't rule out that in addition to the current date and time (IC23), Flash RAM (SIM2) is polled, where the profile of the current parameters of the power unit itself can be stored, i.e. the polling algorithm during initialization includes two consecutive stages.
-
Sorry, I made a mistake:
DRAM - "SIM 1"
FRAM - "SIM 2"

I am also still haunted by RTC things, the OP and TP is weird, all input as well as osc seems good, i even changed RTC still same issues.
8 cases /2 cases, all seems similar types of DRAM. I just think it is some version difference only.
Here are images

[Swake]

The issue is with the DRAM SIMM (not with the FRAM). From a datasheet this pin 70 is presence detect 4 (PD4). Nothing to do with addressing of data, this is used to provide size and speed information of the SIMM.

Code: [Select]

X = connected to GND
o = open

Pin      70    69    68    67       Size           Speed
------------------------------------------------------
         o     o     o     o	    Free / Invalid
         o     o     o     X	    1 MB	   120 ns 	
         o     o     X     o	    2 MB	   120 ns	
         o     o     X     X	    2 MB	   70 ns	
         o     X     o     o	    8 MB	   70 ns	
         o     X     o     X	    16 MB	   70 ns	
         o     X     X     o	    2 MB           80 ns On older machines
	 o     X     X     o	    32 MB	   70 ns On newer machines	
         o     X     X     X	    8 MB	   80 ns	
         X     o     o     o	    Reserved   
         X     o     o     X	    1 MB	   85 ns	
         X     o     X     o	    2 MB	   85 ns	
         X     o     X     X	    4 MB	   70 ns	
         X     X     o     o	    Reserved   
         X     X     o     X	    1 MB	   100 ns
         X     X     X     o	    2 MB	   100 ns	
         X     X     X     X	    4 MB	   80 ns

[Swake]

In a working machine no one is swapping out these DRAM SIMMs, so this bend pin is unlikely to be the root cause of the initial failures. But as these machines and boards are used in difficult environmental conditions maybe contact corrosion is at play here. Except for cleaning everything with a contact cleaner this is very difficult to diagnose.

Are the boards also subject to mechanical stress, like shaking? In this case I would also visually check all solder joints.

I assumed from previous posts that all failing boards are failing in the same way, that seems to point to a single type of failure, but maybe this is sand in the eyes and there are several different causes and potentially also boards with more than one issue.

[Swake]

I'm just haunted by...
I am also still haunted by....

Let me tell you that I'm also a bit haunted... by the OSC3 signal.
It shows as 2 V, that seems very low for a 5 V system. Too low. But I'm no expert in this, I don't really know if 2 V would be enough or not to make it work.
Is this OSC3 used for the Tornado chip? Is this Tornado chip running on 3 V or 3.3 V ? Then maybe the 2 V signal is barely good enough. We don't have the datasheet for the Tornado chip, this document can tell us if the 2 V is enough or not.

I think it was measured with the scope, the scope has a 1 MOhm impedance input. Is this too much of a load for the crystal?

Sooo many things we can still explore on this board. I wouldn't say it is a Christmas present but it comes close for people that like a little challenge

[asis]

@Swake
You yourself provided a photo Reply #510 on: Today , and the location of FRAM (FRAM = "SIM 2").
And the fact that the design uses a SIMM72 connector does not mean anything.
The connector layout and functionality may differ from the standard applicable
to SIMM72 DRAM ="SIM 1"

For example: why does FRAM need RAS/CAS functions (pin's: 33,34, & 40~45)?

[asis]

I have a plot of events developing over many years:

During the operation period, equipment failures occurred.
But in order not to cause downtime, entire controllers were replaced in time pressure.
What was removed was put on the back burner, then fell into someone's hands and was randomly mixed up (PCB & FRAM, DRAM), so to speak, at the "repair" stage.
In the end, all this stuff got to the respected 222Lab_Test222.
And we are all writing this wonderful, exciting novel.
We need to take a breath and start again.
-
Continue...

[Swake]

Forget about OSC3 for the moment. Reason: It is a 2.4576 MHz Crystal and that frequency is a very standard one used for UART purposes. Probably nothing to do with the Wacom Tornado ship or anything that starts the board.

[222Lab_Test222]

Forget about OSC3 for the moment. Reason: It is a 2.4576 MHz Crystal and that frequency is a very standard one used for UART purposes. Probably nothing to do with the Wacom Tornado ship or anything that starts the board.

Not only that OSC, all other OSC has voltage similiar to 2V

[Swake]

The connector layout and functionality may differ from the standard applicable

Maybe, however everything I could check against the 'standard' is exactly as in 'normal' designs. The SRAM picture shows a SIMM with normal layout and the identification tab is in the right place. My believe is that the design of these boards is not using tricks. But again, I'm not an expert here.

We can doubt the placement of the SRAM vs FRAM SIMMs, or doubt that those SIMMs are still working as expected. But 222Lab_Test222 has at least one working board and from what I understand it is confirmed these SIMMs and also the ROMs are functional and in the right place.

I trust 222Lab_Test222 has not mixed up good and bad boards and SIMMs. Anyhow in case of doubt it is easy to startup a 'good' board and confirm all those component are still ok.

[222Lab_Test222]

I assumed from previous posts that all failing boards are failing in the same way, that seems to point to a single type of failure, but maybe this is sand in the eyes and there are several different causes and potentially also boards with more than one issue.

Yeah from what i have gather info from all of you guys and from other resources, I think the main issues of beep is BIOS. If BIOS is performing ok then there is no beep.

But there are lots of causes for BIOS to not working, sometimes IC failure, maybe poor joints or capacitor, there are too many things that may cause issues to this BIOS.
Also if the volatge provide to motherboard is less tha  4.5V, there is also beep, also it maybe be due to BIOS again as there is no sufficient voltage for BIOS to perform.

[222Lab_Test222]

SIM 2 is Flash RAM
SIM 1 is DRAM or SRAM

If i interchange this, even the working board doesnt work.

[m k]

How did you do it exactly, put low ROM to high socket and high ROM to low socket?

Yes like that.

Why did you scope only half of the data?