MSX computer stuck on bootup screen

[Sodius]

Hi.

First time here, but long time EEVBlog Youtube watcher : )

I have an MSX computer that I cannot make boot up properly.
The machine is a Sony HB-T7. Some technical specs here:

https://www.msx.org/wiki/Sony_HB-T7

The main problem is that when I turn on the computer using the ON/OFF switch, the machine starts up, but only shows a black screen.
But if I press the "reset" button, the machine restarts and reaches the boot up screen (blue background with MSX logo). This only works after the machine is ON for some time and then I press "reset". If I press it right after turning it on, it doesn't go to boot up screen.

I measured a few voltages and I found that in the CPU, a Z80, the voltage in the CLOCK pin (pin 6) against ground when I start the computer starts at 1.8volts. Then in the next 2 or 3 seconds, it decreases to 1.6volts, and finally, it jumps to 1.84 ~ 1.85volts and stays there.

Is this a normal behaviour? I measured the same in another working MSX and the voltage of CLK is always stable at 2.5volts from the very moment I press ON/OFF switch.


Another thing that I found is that the boot up screen, besides the MSX logo on blue background, only shows the available VRAM (128Kb), but doesn't show the USER RAM at all, but it should. So I am not sure if this is a RAM problem that makes the whole system to go nuts or a CPU problem that doesn't reach further than testing VRAM and then stops working. Or it could be a problem from a faulty crystal.

Anyway, my first question now is if these voltages and behaviour for a Z80 CPU CLOCK pin are normal.

Thanks a lot.

[austfox]

What is the reset pin (pin 26) doing when booting up from cold? I believe it should be held low for a few hundred milliseconds and then go high and stay at 5V.

[Ice-Tea]

I associate behaviour "like boots better the second time" with dried out caps. Even without the association, electrolytic caps are the first thing to check when working on (really) old gear.

[shakalnokturn]

It does look like a typical case of electrolytics, by your description of the voltages measured on CPU CLOCK, I'd bet that +5V is very noisy during the POST to the point when the CPU hangs then the current draw, and ripple lower.

[Sodius]

Thanks everyone for the assistance. I will start by replacing the biggest electrolytic caps.

What is the reset pin (pin 26) doing when booting up from cold? I believe it should be held low for a few hundred milliseconds and then go high and stay at 5V.

Yes, it behaves like that. Low for a few milliseconds then stable at 5v. I will replace a few caps, then I will see what happens.

[TK]

Hi.

First time here, but long time EEVBlog Youtube watcher : )

I have an MSX computer that I cannot make boot up properly.
The machine is a Sony HB-T7. Some technical specs here:

https://www.msx.org/wiki/Sony_HB-T7

The main problem is that when I turn on the computer using the ON/OFF switch, the machine starts up, but only shows a black screen.
But if I press the "reset" button, the machine restarts and reaches the boot up screen (blue background with MSX logo). This only works after the machine is ON for some time and then I press "reset". If I press it right after turning it on, it doesn't go to boot up screen.

I measured a few voltages and I found that in the CPU, a Z80, the voltage in the CLOCK pin (pin 6) against ground when I start the computer starts at 1.8volts. Then in the next 2 or 3 seconds, it decreases to 1.6volts, and finally, it jumps to 1.84 ~ 1.85volts and stays there.

Is this a normal behaviour? I measured the same in another working MSX and the voltage of CLK is always stable at 2.5volts from the very moment I press ON/OFF switch.


Another thing that I found is that the boot up screen, besides the MSX logo on blue background, only shows the available VRAM (128Kb), but doesn't show the USER RAM at all, but it should. So I am not sure if this is a RAM problem that makes the whole system to go nuts or a CPU problem that doesn't reach further than testing VRAM and then stops working. Or it could be a problem from a faulty crystal.

Anyway, my first question now is if these voltages and behaviour for a Z80 CPU CLOCK pin are normal.

Thanks a lot.

You cannot measure a CLOCK signal with a multimeter.  But in this case, the fact that you are measuring different voltage, it probably is showing some fault in the clock generation circuitry.  You should use an oscilloscope for a better view of the CLOCK signal.