exxos's Atari, Amiga & retro forum

On a brighter note... R. Czuba is certainly approachable, I'm in contact with him every now and then. Not sure what "firm" you are referring to but he is perfectly reachable on his decade old email address available at his website: http://powerphenix.com ... he is busy but maybe it's not impossible to get him to look on your Falcon.

When it comes to the symptoms, sure, changing the NVRAM IC (Dallas) & PSU is the first thing to try, especially if you didn't do anything "stupid". However having a screen like that can be caused by other issues, too - for instance if there is an issue with the clockpatch resistors (R216, R221, R222) - cut, shorted, damaged, ... all sorts of strange things can happen.

christophe.griard wrote: 17 Apr 2020 19:12 Thanks a lot Exxos for the way to get in touch with Steve : I will contact him to see how to proceed.

Have a nice evening Exxos.

Hi Christophe, as Mikro suggests it is best for you if someone geographically close can look at your Falcon. Try R Czuba firstly as he is in France and is obviously very professional. Then failing this you can try myself . I am happy to help depending on what you choose to do.

Hello,

Thanks a lot Mikro and Steve for your information...

I was about to contact you Steve following Exxos advice.

With Mikro information, I will first contact Mr Czuba on the given e-mail address. I atempted to contact him by the past using the web contact on a website where I was said he worked (I do not remember the name of this company), but probably this information was false...

Before contacting him, I will follow your advice Mikro and will recap the PSU (I will also have in few days in parallel a Falcon PSU in working order) and see if it makes some difference (I checked the voltage both 5v and 12v and they are ok so I don't expect to see any difference but you never know).
After that, I will change the Dallas by the same one bought some months ago by Best Electronics (in fact I planned to change PSU capacitors and Dallas the week after the Falcon went to black screen...).

I will keep you uptodate as soon as I make these changes.

Thanks a lot again to both of you for your help.

Dear All,

As promised, a small update of the search for the breakdown of my FALCON.

I changed the power supply by one that has been recapped (not by myself but by a skill person; I will recap mine later, maybe this afternoon) : the screen is unchanged and presents the same strange green stripes.

I followed your advice Mikro and I checked the 3 resistors R216, R221 and R222: they are very clean (see enclosed picture) and all three have the same value of 33 ohms, which seems to be the right value.

As these first two checks did not identify the root cause of the problem, I will change as proposed, the Dallas chip (even if I am skeptical about the fact that this solves the problem) and will keep you aware about the result.

I wish you a nice end of the day.

Take care and stay healthy.

Christophe.

Hello,

Before changing the Dallas battery, I took out my little oscilloscope to look at some signals on the expansion connector (just above the CPU). I did it with my little oscilloscope on USB port (sampling at 100 M/s and bandwidth of 25 Mhz).

On the Reset side (pin 17 of expansion port), I can clearly see the Reset signal rising when I press the Reset button, then I can see the signal returning to 0 once the button is released, confirming the correct operation observed on the screen (black screen at the start of reset then screen with green bands once the reset button is released).

On the clock side on pin 23 of the expansion connector, I measure a frequency of 8 Mhz (see picture at the end). It is one of the tasks of the COMBO IC to divide the frequency of 32 MHz into different values ​​(16, 8, 4 MHz and 500 kHz). Which would mean that the COMBO IC is doing its job and that it did not burn out.

On the Bus error signal side (pin 19 of expansion port), signal is at 0. I conclude that there is no issue regarding the bus.

More interesting is the behavior of pin 48 of the expansion port (address 1):
- Indeed, with 4 MB of RAM present, the signal is all the time at 0.
- But if I remove the RAM, the screen no longer has a green band but remains black. In this configuration, as soon as the reset is complete (and this for about a second) I have a signal on pin 48 (see middle picture) but about a second after, the signal returns to 0 (see picture in first place) ...

This is where I am with my investigations this evening. I am taking your opinions on these findings to guide me.

Thank you in advance and have a good evening.

While this is for the H4 motherboard, you may get some ideas as the 68030 operates in a similar way. https://www.exxosforum.co.uk/forum/viewt ... =78&t=2362

Also, you may want to check Falcon Service Guide on https://docs.dev-docs.org, there's a nice trouble-shooting guide.

christophe.griard wrote: 24 Apr 2020 22:07 On the Reset side (pin 17 of expansion port), I can clearly see the Reset signal rising when I press the Reset button, then I can see the signal returning to 0 once the button is released, confirming the correct operation observed on the screen (black screen at the start of reset then screen with green bands once the reset button is released).

Hi Christophe,

Can you please check this again? Reset should be active low (I think) so should be high for normal operations. Also I see that your clock is stuck at 8Mhz. With any luck it could be a simple problem with your reset circuit. There are some inverters at U2 and U11. Might be worth checking them.

Also never a bad idea to replace the reset switch... They're pretty easy to find and it's an easy job.

Cheers

Hello Sety,

Thanks a lot for your message and advice : I will check that with attention this afternoon and I will keep uptodate with the results of this investigation.

For the clock frequency on the expansion port, you think I should have found 16 Mhz instead of the 8 Mhz I measured?

Thanks a lot again and have a nice afternoon.

Hi Christophe,

Yep, 16Mhz. That instruction comes from TOS and happens about 1-2 seconds after power up. (Init PSG happens shortly after, so you get that lovely *pop* in the speakers) So if Combel switches CPU clock to 16Mhz then you have a mostly working system as it's loading ROM (or some of it)

Did you say you had bars on the screen? Do you have a substitute RAM board you can try? Absolutely nothing works if RAM is broken. I'll post you one if you don't have a spare.