Falcon clock patch V4 series

[exxos]

- measure the clock signal on the bus with the CT60e
- measure the clock signal on the bus without the CT60e (what implies the use of the old power supply)

I think I forgot to mention it, but I did not really see any difference on the expansion connector with or without the CT60. Signal is reasonable on the V4 output, of course trace length to the expansion connector is where the ringing mostly generates from. I don't know if the CT60 has any sort of termination on the clock input or not though anyway.

[Swissfalcon]

ok. But I understood right, on the CT60e, without the 68R you measure this:

IMG_3520.jpeg

and with the 68R you measure this:

IMG_7108.jpeg

correct?

it’s look like a pretty good improvement overall. My understanding is that the first signal is pretty bad, in this case it’s maybe enough on my mainboard to make the CT60e crashy, who know….. Testing at 66 Mhz and then with 68R will help to understand better.

[exxos]

correct?

Yes.

it’s look like a pretty good improvement overall. My understanding is that the first signal is pretty bad, in this case it’s maybe enough on my mainboard to make the CT60e crashy, who know….. Testing at 66 Mhz and then with 68R will help to understand better.

The original Atari clock patch looked like this.. It's way worse than what the V4 is doing.

sdma.png

So it does not really make much sense why the Atari one would work better with the CT60, indeed it doesn't work well, which is why Joy fails.

[Swissfalcon]

So it does not really make much sense why the Atari one would work better with the CT60, indeed it doesn't work well, which is why Joy fails.

my motherboard didn’t had any clock patch, it had the rev 0 Atari modification (0 Ohm resistor on R221 and C208 removed). My motherboard could’n run Joy

The motherboard of my friend got the Atari clock patch version 1.3 as mentionned on Mikro’s webpage, we didn’t touched anything. this motherboard run Joy and the CT60e stable.

What I will do is to measure the signal on the expansion port on this motherboard too, to have a better view.

[Swissfalcon]

I made some measurements.

First one is on the motherboard of my friend, the one with the Atari V1.3 patch

IMG_3481.jpeg

CT60e fitted, measured on this pin as explained on your pic:

IMG_2434.JPG

This is the result:

CT60 clock pin Falcon with Atari patch Benoit.jpeg

Then my motherboard, with V4 patch, without the CT60e, original power supply, measured on the same pin (but this time on the motherboard, obviously)

view.jpeg

I got this

without CT60 on the bus my falcon.jpeg

Finally, I fitted the CT60e on my motherboard with V4 patch and measured on the same pin on the CT60e

view2.jpeg

I got this:

with CT60 on the bus my falcon.jpeg

I have the feeling that my signal on the V4 is worst that my signal on my friends MOBO, or I'm mistaken. What do you think?

I don't have any 68R resistor, I have to find one, this will be the next point.

And the bandwidth limit is activated

IMG_3534 3.jpeg

[exxos]

The V4 has sharper rise and fall times as expected. It has more ringing but resistors combat that. But the logic levels look good anyway.

Your results look better than mine did in fact. So while the 68R resistor will help with ringing, I don't think that is going to help with your issue anyway.

I guess if you sent me your CT60 for a few days I could try it in my falcon. It may give indication if there is some problem with your CT60 or your falcon. I don't have any other suggestions.

EDIT:
I would suggest unplugging netusbee as people have had all sorts of strange problems with that..

[Swissfalcon]

The V4 has sharper rise and fall times as expected. It has more ringing but resistors combat that. But the logic levels look good anyway.

Your results look better than mine did in fact. So while the 68R resistor will help with ringing, I don't think that is going to help with your issue anyway.
I guess if you sent me your CT60 for a few days I could try it in my falcon. It may give indication if there is some problem with your CT60 or your falcon. I don't have any other suggestions.

well, I lowered the 060 frequency to 66 MHz, and Falcamp hanged after 6m22 seconds. so, no luck at all with a lower frequency.

I Still have to try the 68R and the V3 patch before sending you the boards :-)

I lowered the 060 frequency to 77 MHz, and Falcamp hanged after 6m22 seconds.

I Still have to try the 68R and the V3 patch before.

EDIT:
I would suggest unplugging netusbee as people have had all sorts of strange problems with that..

REally? What kind of problems? I realised that sometimes, the loading of the system hanged while the DHCP was releasing an address. Is this a know problem?

[exxos]

There's been various problems, I don't remember them as I don't even own one. Use the search on the forum.. I think mostly it would crash some machines and some people went to great lengths to improve things.

Some examples..

viewtopic.php?p=34359#p34359
viewtopic.php?p=89483#p89483
viewtopic.php?f=23&t=5838&p=92291&hilit=Netusbee#p92291

So I would leave it unplugged while doing tests to make sure that's not adding to your troubles..
Maybe some other members can mention odd problems with it also..

[exxos]

Also earlier you said you was using a mATX PSU but your image shows I think a Pico PSU ? While people use them, I've never had good experiences with them. During tests you should use a proper ATX PSU to rule out power supply problems.

Also be aware if your CPU fan is blowing downwards, your blowing hot air from the CPU over the CT60 chips and it will heat up the inside of your falcon when enclosed.

[Swissfalcon]

now I have a big problem. I tried the CT60e at 77 MHz, still crashing. And then I tried to select 66 Mhz. And the CT60e won't boot anymore. I see now that the minimal speed for the CT60e is 66.6 Mhz. How can I put the value back? The cpx in 030 mode tell that I have to hook a Xilinx programmer on the machine...

[edit]
solved, following this procedure

Capture d’écran 2024-07-19 à 00.45.43.png