Trudie - The Next IDE Board

[exxos]

In light of issues with assembly with basically everything lately.. I have just done a 5.1 version using a lot more SMT which JLC can assemble..

Preview does not seem to be working but all these parts will be assembled at the factory.. Leaving just the main PLD on the top of the board to solder manually...

Capture.JPG

[exxos]

Got these back today.. So the only thing I can balls-up is soldering the PLD :)

IMG_5693.JPG

[PhilC]

Got these back today.. So the only thing I can balls-up is soldering the PLD :)

The only thing? :lol: :lol: :dizzy:

[Icky]

Nice re-spin for SMT assembly @exxos. Will really cut down the time to build. :)

[exxos]

Pasted the board here https://www.exxosforum.co.uk/forum/viewt ... 976#p51976

Just a video showing some progress at least with assembly...

[exxos]

Next to work on the firmware..

IMG_5752.JPG

[exxos]

:sigh:

IMG_5753.JPG

[exxos]

Thats better!

IMG_5754.JPG

IMG_5755.JPG

[exxos]

After checking out the signals on the board I found that I have some 33pF caps which ended up that 65pF & 215pF :roll: I thought these were bad caps but turns out the capacitance is actually on the PCB itself :roll: While I was aware there would be some capacitive coupling I really did not think it would end up at 200pF! :roll:

1.PNG

2.PNG

3.PNG

So I have limited the ground fill on the top and bottom keeping it well away from the caps. I also routed in master slave select line which I can only work out is on pin 28 of the IDE44 pin header.

I added in a second 4pin header as I missed off a ROM CS header :( And the other three are spare IO's which may get repurposed for something in the future but it will also help to have the to use them as debug lines for my scope.

The active LED which comes on when the interface is active, is on for such a short amount of time is basically never visible. I separated the interface active and ID active LEDs as at some point in the past it was useful to work out of the problem was with the interface or IDE card itself.. I think the IDE LED just stays on for a long time inherently to make it visible. I assume there is some delay logic for that to happen or it could just simply take ages to do anything.. But either way that LED is clearly visible and there is no problems with it.. The active LED I used a load of clock dividers and delay chains in the PLD to keep it on for longer, but this uses a lot of chip resources to the point I maxed out the thing :roll:

So I added on a simple RC Delay line which I can charge and discharge to get approximately half a second delay on the active LED.. Which turned out I needed to use a 10uF capacitor which I could not really discharge using a PLD IO pin :roll: So I had to add in a small MOSFET to discharge it, then a series resistor to limit the current flow a little bit. It got a bit more complicated than I really wanted it to be, but I want the LED to stay on longer so.. :lol:

Capture.PNG

[exxos]

Finally had a bit of time today to look at the byte swap issue again... And it does appear to be working now :D But with some odd side effects which I am currently trying to figure out...

So this is what I have on the PC..

IMG_5770.JPG

This is what I have on the ST..

IMG_5771.JPG

And just to verify that this is not HD11 byte swapping active...
(if it was then the speed drops to something like 300 kB a second.. But I am not currently convinced this is working properly)

IMG_5772.JPG

The problem I am currently having is while it looks correct on initial load...

IMG_5773.JPG

... If I load the driver a second time "SANDISK" ends up swapped..
I also see the same in the utility as well but otherwise everything seems to work fine :shrug:

IMG_5774.JPG