Falcon clock patch V4 series

[Swissfalcon]

Reading the thread carefully, it seems to me that the way of measuring is important. As I'm not really used to use an oscilloscope, can you give the exact parameters to select, to be sure to make the measurement correctly? I'm going to a friend's house to do the installation, so I won't be able to easily take measurements afterwards. I plan to take a measurement before and after installing the patch.

[exxos]

You need a x10 probe with less than 5pF capacitance. X10 selected on the scope and bandwidth limit disabled . You also need to have a scope with at least 50MHz / 1Gas/S. Measure across the 47pF cap solder points.

I still have no idea why you guys are getting different results even with the original patch. Though the important thing is that the logic levels are being met.

[Swissfalcon]

this will be very helpful, thanks. I report when I managed to meet with my friends for the installation.

[Swissfalcon]

Well, I received the clock patches (V2 and V4) very quickly, thank you Exxos! Yesterday, my friend installed the V4 patch in my falcon and we did some measurements.

I'd love to hear your comments and opinions on the results of the measurements I've made.

first, a measure done on the SDMA pin on the chip directly before installing the patch:

measure befor patch applied.jpeg

witouth patch (Atari V0 mod).jpeg

Then mounting of the V4,

V4 patch fitted.jpeg

including C208 fitted back (wasn't present, as discussed earlier in this thread),

C208 back.jpeg

FPU caps (forgot to make a pic)

and SDMA mod:

SDMA mod.jpeg

Then measuring through the 41 pF like you explained:

measure after resistance.jpeg

I got this result:

patch V4 after resistor.jpeg

and if I measure just before the resistor that sits on the SDMA line, there:

measure before resistance.jpeg

I got this:

patch V4 for resistore.jpeg

What puzzle me is that the result after the resistance looks more like what Steve and the others have, and if I measure before the resistance I have more something like you @Exxos have.

I did then 3 other measurement on the V4 patch itself:

measure V4 R216.jpeg

patch V4 R216.jpeg

mesure V4 R221.jpeg

patch V4 R221.jpeg

Mesure V4 R222.jpeg

patch V4 r222.jpeg

The results are mixed: in 030 mode (CT60e deactivated), everything works fine. I was able to run the "Joy" demo 50 times (yes, I did that). Before the patch, Joy would crash randomly (sometimes it worked 5 times in a row, sometimes not at all).So, yes, for a stock Falcon, looks that the V4 patch works well.

On the other hand, with CT60 activated, the machine is less stable than before. It freezes after a few minutes of use, which wasn't the case before, and Falcamp sometimes crashes right away, or when I'm lucky, after 15 minutes max, whereas it could last several hours before. Before I could let mxplayer play all my MP3 for 48h without problem, now after less that 2 hours it freezed totally the machine.

So I think I'll try the V2 patch.

I now have the following questions:

- are the measurements I've taken correct, do they make sense?

Has anyone else already tested V4 with the CT60e and found it less stable?

Has the V3 been tested with the CT60e, and is it stable?

[Steve]

are the measurements I've taken correct, do they make sense?

My comments: The SDMA signal looks beautiful after the installation. The other signals look very odd with strange voltages (one over 6v??) I wonder where you are grounding. You should be grounding the probe to the closest ground point, which on the clock patch would be the ground leg on the chip it's soldered to. Where you ground will effect the reading considerably.

Has anyone else already tested V4 with the CT60e and found it less stable?

My comments: I found it perfectly stable in my CT60e machine. (When bus boost disabled)

Has the V3 been tested with the CT60e, and is it stable?

My comments: Yes I run a V3 in my CT60e (Because V3 supports the bus boost, which my Falcon has installed)

[Swissfalcon]

My comments: The SDMA signal looks beautiful after the installation. The other signals look very odd with strange voltages (one over 6v??) I wonder where you are grounding. You should be grounding the probe to the closest ground point, which on the clock patch would be the ground leg on the chip it's soldered to. Where you ground will effect the reading considerably.

I took the ground on the chassis directly. So, that's mean that all my measurement are worthless?

My comments: I found it perfectly stable in my CT60e machine. (When bus boost disabled)

Ok. Then I'm out of luck here. Can you do me a favour? Can you try to launch falcamp playing some MP3 in loop and see how long it work?

My comments: Yes I run a V3 in my CT60e (Because V3 supports the bus boost, which my Falcon has installed)

ok. I think that it's my only option left now, try to switch on V3 patch. Thanks for the feedback.!

[Steve]

I took the ground on the chassis directly. So, that's mean that all my measurement are worthless?

I'd like another expert opinion from maybe @exxos but yes as far as I am aware, and in my experience, you need to ground at the closest point to the signal for a reliable reading. When I did my clock patch measurements, If I didn't ground close to the signal, the results were horrendous.

Ok. Then I'm out of luck here. Can you do me a favour? Can you try to launch falcamp playing some MP3 in loop and see how long it work?

I cannot test V4, because as stated previously, I am not using a V4 in my machine. I use a V3 because I need bus booster support (SDMA delay)

[Badwolf]

mesure V4 R221.jpeg

patch V4 R221.jpeg

That almost looks as though you lost the ground connection during that measurement. Certainly doesn't look much like the far end (patch V4 for resistore.jpeg).

BW

[Swissfalcon]

I cannot test V4, because as stated previously, I am not using a V4 in my machine. I use a V3 because I need bus booster support (SDMA delay)

yes, that's what I'd like you to test :-) I'm curious to know if with the V3 you can run Falcamp in loop for some times. Because I'm actually not sure if Falcamp isn't simply buggy. Can you make it on the V3?

[Swissfalcon]

That almost looks as though you lost the ground connection during that measurement. Certainly doesn't look much like the far end (patch V4 for resistore.jpeg).

BW

can good be, yes.