• Please review our updated Terms and Rules here

AUVA JUKO Baby XT BXM-10

If you can get a scope on A12 that would be good just to confirm you have a good logic signal. Wouldn't hurt to check others.
Use TEST5077 from [here].
It exercises all of the address lines to the 8KB ROM.
It does not not need the RAM to be functional.
It runs for about 2.5 minutes.

During TEST5077, the different address pins will have different signals, but they all change state at some point during the test.
For example, A12 is normally HIGH, and it will temporarily go LOW about every second.
For example, A6 is normally HIGH, and it will temporarily go LOW about every 2 seconds.

Verify that every ROM address pin is changing state.

( A logic probe is good to do this check. )
 
BTW. Unintentionally, I discovered that this motherboard is of the type that does not use the POWER GOOD signal from the power supply.
 
As suggested, I checked all the PINS on the U10 (2764 Rom Bios Socket)
Everything is the same as the Modem7 board except,

"Pin 26 (not connected) has continuity to NOTHING. "

But on my board Pin 26 connects to U11, Pin 26 of "27256", and U13 Pin 14, and on the Flatpack somewhere in the Pin-7x's (counting anticlockwise from the dot, 30 pins, + 20 pins + 30 pins + 20 pins).

Most of the pins also have a connection to the flatpack.

How many times can a EEPROM be reprogrammed ? my poor chip has been backwards and forwards.
 
As suggested, I checked all the PINS on the U10 (2764 Rom Bios Socket)
Everything is the same as the Modem7 board except,

"Pin 26 (not connected) has continuity to NOTHING. "

But on my board Pin 26 connects to U11, Pin 26 of "27256", and U13 Pin 14, and on the Flatpack somewhere in the Pin-7x's (counting anticlockwise from the dot, 30 pins, + 20 pins + 30 pins + 20 pins).
I just now brought out my motherboard. I checked U10 pin 26 again. I made a mistake. Like pin 27, pin 26 is connected to +5V. U10 pin 26 is not a concern to me. On both 2764 and 28C64 type chips, pin 26 is unused.

It must be just another difference between my motherboard (640K max. motherboard RAM) and yours.

Out of curiousity, I looked at U11:
- Pin 26 on U11 (A13 on a 27256). That connects to U13 pin 14, and a pin on the flatpack.
- Pin 27 on U11 (A14 on a 27256). That connects to U13 pin 4, and a pin on the flatpack.
 
BTW. Unintentionally, I discovered that this motherboard is of the type that does not use the POWER GOOD signal from the power supply.
I didn't think I could see any connection to P8-1 on the power connector. That might also explain why rapid on-off-on produces some random results.
 
Use TEST5077 from [here].
It exercises all of the address lines to the 8KB ROM.
It does not not need the RAM to be functional.
It runs for about 2.5 minutes.

During TEST5077, the different address pins will have different signals, but they all change state at some point during the test.
For example, A12 is normally HIGH, and it will temporarily go LOW about every second.
For example, A6 is normally HIGH, and it will temporarily go LOW about every 2 seconds.

Verify that every ROM address pin is changing state.

( A logic probe is good to do this check. )
I loaded up TEST5077 - it looks like it runs all the way through all 11's no 99's - then it loops back to the start counting up E0 to FF again.

I ran out of time tonight to check the ROM pins with the scope.
 
Excellent, but is that consistent? Lack of consistency has been a problem.


A bug that I fixed yesterday. The latest TEST5077 will stop after [FF][11] is displayed.
yes, since I but in bodge wires it has been consistent - I added a switch to the power on off. Power on proceedure is now - (1) turn on psu integrated switch, (2) turn on the in line switch. This has helped with consistency, my guess is that without a true power good, the rise time of a cold boot psu is a little too long (?).

at the end of the sequence it stops at FF 11 !

now to measure the pins as in #62 above.
 
DE217B1E-4D16-49F5-8BDD-D008E02D65F9.jpeg

I made a logic probe. ;) (Because I’m too impatient to wait for Amazon]
It actually works on some rom pins I get high (red), then a little flash of green about every second.
But not on all Address line pins. Some light up both dimly ! And some lines dont flash!
This is also giving me a lot of hangs - the post card stops counting when I press down on the pin, but doesnt if I use the same pin in the adjacent socket (U11).
The oscilloscope probe was also causing system hangs.
Note: The same channels that were behaving well with my logic probe also gave pulses on the scope.
I’m going to stop here and remove both the corroded 28 pin sockets, examine the damage on the board. Fix the known issues (and get a real logic probe)
 
Power on procedure is now - (1) turn on psu integrated switch, (2) turn on the in line switch. This has helped with consistency, my guess is that without a true power good, the rise time of a cold boot psu is a little too long (?).
A circuit diagram of the reset circuitry of my very-similar-to-yours motherboard is at [here].

Cold start:
1. C60 is, at the instant of +5V being supplied to the motherboard, in a discharged state.
2. Therefore, at that instant, 0V is gong to the flatpack chip, and consequently, the chip will assert the RESET line.
3. C60 will start charging via the resistor - the voltage to the flatpack chip starting to rise.
4. Eventually, the voltage will reach a threshold point where the flatpack chip de-asserts RESET.

In this way, RESET is asserted for at least the minimum required RESET time of the 8088/V20 CPU. See note 1 at the bottom of [here].

This RESET period depends on the value of C60, the value of the resistor, and what threshold voltage that the flatpack chip uses.

Perhaps verify that your C60 is still good.


(After power-off, before then reapplying power, enough time needs to be waited for C60 to discharge, before reapplying power. )
 
Some light up both dimly !
If the line is switching state very quickly, with roughly a 50%/50% ratio of HIGH/LOW, then both LED's will reduce in intensity. It is one of the reasons why logic probes have a 'pulse' LED.

And some lines dont flash!
That's odd.

the post card stops counting when I press down on the pin, but doesnt if I use the same pin in the adjacent socket (U11).
That is a bad sign.

I’m going to stop here and remove both the corroded 28 pin sockets, examine the damage on the board. Fix the known issues (and get a real logic probe)
Sounds good.
 
The oscilloscope probe was also causing system hangs.
Note: The same channels that were behaving well with my logic probe also gave pulses on the scope.
A high impedance scope probe should not impede the signals at all.
By pulses I assume you mean signals with edges that look square?

If you have any address lines that dont have very fast square edges, I would trouble shoot that.

Work back through the bus drivers to the cpu. See in you can identify a bad signal.
 
Bios rom Sockets removed. Some vias / pin holes are dark grey & missing a lot of metal. I found a new connection problem from the big chip 8237 (DMA controller). , next to the rom.C0970B15-657B-4A32-B0B0-1A2263E46E4A.jpeg
 
Last edited by a moderator:
0970F866-CF8F-4919-BFE2-3833AFB06EC9.jpeg
Got some work to do to fix the damage. Red circles show where there is no continuity.
This is the first time I’ve tried to fix traces. I’m going to try and put some copper bridge pieces directly on the topside of the board.
07FF82FB-E718-47D9-B74A-C479E5C99D42.jpeg
 
This is the first time I’ve tried to fix traces. I’m going to try and put some copper bridge pieces directly on the topside of the board.
If you are unsuccessful, consider doing what I usually do, which is to bypass on the solder side using wire-wrap wire, per the photo at [here].
 
Back
Top