HB-F9S v9938 partially dead?

Pagina 4/6
1 | 2 | 3 | | 5 | 6

Van Grauw

Ascended (10066)

afbeelding van Grauw

02-01-2016, 16:50

Could it not also be the M1 line that has a problem?

Van flyguille

Prophet (3028)

afbeelding van flyguille

02-01-2016, 18:17

blackmsx wrote:
sd_snatcher wrote:

- Check if there's a 1K resistance between the pin-24 (/WAIT) and the pin-11 (Vcc) of the Z80A.
- Check if the pin-38 (/WAIT) of the S1985 is connected to the pin-24 of the Z80A
- Check if the pin-32 (/M1) of the S1985 is connected to the pin-27 (/M1) of the Z80A

Both pins (38 and 32) from the S1985 are connected to the CPU. There is a 1k resistor (R121) connected between CPU pin 24 and VCC but the resistance measures about 700 ohms.

make sure that when you measure resistence you are not in the the BEEPER/DIODE scale of the multimeter, because each chip I/O pin, any chip, has two protection diodes one going to VCC the other to Ground, those diodes are connected in reverse, so that way both protects the I/O pin against overloads (static electricity).

When the tester is in ohm scale, it uses lower voltage to measure resistence, lower than the diodes threshold, so unable to conduct through them, and only will measure the pasive compontents, in your case, the pull up resistence.

Van blackmsx

Expert (77)

afbeelding van blackmsx

02-01-2016, 19:18

flyguille wrote:

make sure that when you measure resistence you are not in the the BEEPER/DIODE scale of the multimeter, because each chip I/O pin, any chip, has two protection diodes one going to VCC the other to Ground, those diodes are connected in reverse, so that way both protects the I/O pin against overloads (static electricity).

When the tester is in ohm scale, it uses lower voltage to measure resistence, lower than the diodes threshold, so unable to conduct through them, and only will measure the pasive compontents, in your case, the pull up resistence.

Ok, I did not know that. I was using the "2K/DIODE" scale...Now I get the right measurements:
1k with R121 in place, nothing without R121.

I also follow the /WAIT line along the board and everything seems fine.

Anything else I could do before build the 7474 circuit?

Van flyguille

Prophet (3028)

afbeelding van flyguille

02-01-2016, 19:38

blackmsx wrote:
flyguille wrote:

make sure that when you measure resistence you are not in the the BEEPER/DIODE scale of the multimeter, because each chip I/O pin, any chip, has two protection diodes one going to VCC the other to Ground, those diodes are connected in reverse, so that way both protects the I/O pin against overloads (static electricity).

When the tester is in ohm scale, it uses lower voltage to measure resistence, lower than the diodes threshold, so unable to conduct through them, and only will measure the pasive compontents, in your case, the pull up resistence.

Ok, I did not know that. I was using the "2K/DIODE" scale...Now I get the right measurements:
1k with R121 in place, nothing without R121.

I also follow the /WAIT line along the board and everything seems fine.

Anything else I could do before build the 7474 circuit?

Ok the next thing, is to observe the line with a osciloscope with /hold function or a logic analizer, the WAIT/ line is normally HIGH state, and only by tiny periods are LOW. The tool can be homemaded with some ttl logics.

Van blackmsx

Expert (77)

afbeelding van blackmsx

02-01-2016, 20:37

I have access to a Hantek USB Ocilloscope (I dont know if it has a HOLD function) and also a USB logic analizer. What kind of measurements I should do?

Van flyguille

Prophet (3028)

afbeelding van flyguille

02-01-2016, 20:41

blackmsx wrote:

I have access to a Hantek USB Ocilloscope (I dont know if it has a HOLD function) and also a USB logic analizer. What kind of measurements I should do?

you needs to check the presense of those LOW pulses when scroolling you will see lot of them in a burst.

So it must hold on the first low after the connection is made.

Van flyguille

Prophet (3028)

afbeelding van flyguille

02-01-2016, 20:42

flyguille wrote:
blackmsx wrote:

I have access to a Hantek USB Ocilloscope (I dont know if it has a HOLD function) and also a USB logic analizer. What kind of measurements I should do?

you needs to check the presense of those LOW pulses when scroolling you will see lot of them in a burst.

So it must hold on the first low after the connection is made.

another question, the firmware is the original of the machine, right?

Van flyguille

Prophet (3028)

afbeelding van flyguille

02-01-2016, 20:51

Sincerelly, you needs to check there is pulses don't matter when, IIRC the v9938 don't has a /WAIT line output, so it don't advise or slowdown the z80 in any way, I don't remember if the s1985 has one for outputing to the z80, I thinks if it is s1985 based, the /WAIT of the z80 is connected only to slot expanders, and maybe the s1985 is signalling too for some internal operations, not related to general I/O, buy maybe related to writing RTC or its internal SRAM, or so.

In that case, the z80 /WAIT is not directly ANDed using discrete components, it must pass by a TTL logic, AND logic, so if any of its inputs are 0, the z80 /WAIT gets 0, and those input would be pulled up aswell.

In models with v9958, the v9958 provides the /WAIT signal for the z80, but in most mods, it is not used/connected, so the firmware must run ok with or without /WAIT.

I needs to check the s1985.pdf to see what its /WAIT pin does really, and for what it is.

Van blackmsx

Expert (77)

afbeelding van blackmsx

02-01-2016, 21:50

Yes, the firmware is the one that came with the machine, which I belive is the original. The IC 13 was missing and so I used the socket to install the 512k ram.

As I said before, the scroll problem is present ONLY when I plug a Sunrise IDE cartridge I borrough. Can the Sunrise cause somehow the clock problems?

Here are some oscilloscope captures. Upper channel is /WAIT signal, lower M1 signal:

This is the msx with nothing plugged in:

and with the Sunrise plugged but no CF card inserted:

But with when the card gets detected things look quite different:

See how the /WAIT signal has changed.

At other thread some people suggested me that can be a power supply problem... but the wierd thing is that the problem only appears when the CF card is detected....

Van PingPong

Prophet (3764)

afbeelding van PingPong

02-01-2016, 21:52

a load problem of the lines? (fan in fan out issue?)

Pagina 4/6
1 | 2 | 3 | | 5 | 6