exxos's Atari, Amiga & retro forum

Ouch, that CPU looks like a bit of an arse to solder. I guess since the pin dimensions fall inside a PLCC socket, that means a socket for the CPU is a no go?

rubber_jonnie wrote: ↑Mon Apr 09, 2018 11:24 pm Ouch, that CPU looks like a bit of an arse to solder. I guess since the pin dimensions fall inside a PLCC socket, that means a socket for the CPU is a no go?

Not sure I follow... Whole idea of this booster is to avoid the PLCC socket for the CPU..

exxos wrote: ↑Mon Apr 09, 2018 11:26 pm
Not sure I follow... Whole idea of this booster is to avoid the PLCC socket for the CPU..

Sorry, I'm being a bit dumb, we've probably talked about this, I'm just a bit tired and confused at the minute....

rubber_jonnie wrote: ↑Mon Apr 09, 2018 11:29 pm Sorry, I'm being a bit dumb, we've probably talked about this, I'm just a bit tired and confused at the minute....

Welcome to my world

How it looks so far...

rubber_jonnie wrote: ↑Mon Apr 09, 2018 11:24 pm Ouch, that CPU looks like a bit of an arse to solder. I guess since the pin dimensions fall inside a PLCC socket, that means a socket for the CPU is a no go?

I take it, you are talking about a socket on the booster board, not on the mainboard... IMHO the pads for soldering a PLCC SMT socket are identical to those for direct soldering, so putting a socket on it should be no problem. A socket is actually much easier to solder, since you can pop out the lower pane of the socket for easier access to the pads. Then you simply do drag soldering and everything is fine. I did so a couple of times with my STE that had a SMT CPU and no socket.

See my adventure here: https://www.exxosforum.co.uk/forum/viewt ... f=18&t=358

Direct soldering is probably easiest when using an oven, I had mediocre results with a hot air station. But as long there is room around the the chip, you can reach the solder points with a fine tip and resolder, if necessary.

Yes, I was talking about the booster board, and not the mainboard.

The booster has been on for a few hours now about to tick over to loop 60. So I think it is safe to say this is all stable. I seem to be mostly having socket problems the dual tos board

Will have a look and see if I can find some slightly higher speed oscillators to try next...

UPDATE:
some reason this booster really does not like a oscillator I cannot even get it stable with a 25MHz one. The original V design did obviously work with a separate oscillator as I had it working easily. As to why this board misbehaving I really do not know. Maybe it is down to the revision of motherboard or something.

The clock switching on the CPU looks perfectly fine. Some of the oscillators do seem to have a 1 V DC offset made things a little bit more complicated. Do a few more tests but as is misbehaving right from the start I think using a external oscillator is probably a bad idea .

UPDATE 2:
I removed the capacitors from the 8 MHz line and then machine booted up the 25 MHz oscillator. So then I tried a 35 MHz oscillator, but it mostly crashes just after the logo appears. It could be CPU simply maxed out, looking at the clock line, there seems to be a 1V offset when it switches to 35 MHz. I have removed the series resistors try and gain a little more voltage but seems to buffer simply cannot work at this speed

So I was just about to basically give up with a external oscillator, but I tried some other ideas and now I am running at 32.768MHz.


And this is the normal 32 MHz.. Please ignore the statistics for comparison as this image below is the previous version of GB6..

Though the only real interesting one I suppose is integer division..




UPDATE:

Now 35MHz



UPDATE - 40MHz


Now I need a 42MHz osc

tuxie wrote: ↑Tue Apr 10, 2018 5:55 pm Never found an 42Mhz OSC my TT is working with 40Mhz but not with 48Mhz so when i can get something between 40 and 48 so i think i can archive this..

A RSO may help ? I used them, variable to whatever.