I'm very happy with the STE DAC upgrade

[exxos]

I think I might know what that problem is..

Looking at the STE circuit which is pretty incomprehensible...

ste.PNG (54.72 KiB) Viewed 1151 times

The STE bit order looks "wrong" .. I thought it was..

BIT3 = 12K
BIT2 = 6.2K
BIT1 = 3K
BIT0 = 1.2K

Hard to read but looks like But the order is actually


BIT3 = 12K
BIT2 = 1.2K
BIT1 = 3K
BIT0 = 6.2K


So looks like bits 2 and 0 are swapped

On the ST BIT0 is 15K, but BIT0 on the STE looks different order again at 1.2K . So the bits are swapped and the order looks backwards compared to the ST..

ST.PNG (64.64 KiB) Viewed 1151 times

But I can't load @sporniket Kicad files for some reason to check better and there doesn't seem to be a PDF.

[sandord]

I think I might know what that problem is..

So looks like bits 2 and 0 are swapped

But I can't load @sporniket Kicad files for some reason to check better and there doesn't seem to be a PDF.

Alright, that makes sense. What is the outlook for a patch, lift two chip legs and cross-connect them?

[exxos]

Alright, that makes sense. What is the outlook for a patch, lift two chip legs and cross-connect them?

You would have to reorder the header row that's soldered onto the ST motherboard as those are the colour bits with bodge wires so you can easily cross them over.

[sandord]

The STE bit order looks "wrong" .. I thought it was..

BIT3 = 12K
BIT2 = 6.2K
BIT1 = 3K
BIT0 = 1.2K

Hard to read but looks like But the order is actually

BIT3 = 12K
BIT2 = 1.2K
BIT1 = 3K
BIT0 = 6.2K

So looks like bits 2 and 0 are swapped

On the ST BIT0 is 15K, but BIT0 on the STE looks different order again at 1.2K . So the bits are swapped and the order looks backwards compared to the ST..

But I can't load @sporniket Kicad files for some reason to check better and there doesn't seem to be a PDF.

This may help:

https://easyeda.com/editor#project_id=9 ... ba77f26b81

(found at https://oshwlab.com/sporniket.studio/ki ... 300780-001)

It reads:

Code: Select all

- B3 - R411: 12K
- B2 - R412: 1K5
- B1 - R414: 3K
- B0 - R415: 6K2

Code: Select all

- G3 - R416: 12K
- G2 - R417: 1K5
- G1 - R418: 3K
- G0 - R419: 6K2

Code: Select all

- R3 - R422: 12K
- R2 - R424: 1K5
- R1 - R426: 3K
- R0 - R428: 6K2

So your assumed correct order seems right.

[alexh]

Hard to read but looks like

BIT3 = 12K
BIT2 = 1.2K
BIT1 = 3K
BIT0 = 6.2K

Bit2 is 1.5K not 1.2K

The MegaSTe schematic floating around is slightly clearer.

[sandord]

Alright, that makes sense. What is the outlook for a patch, lift two chip legs and cross-connect them?

You would have to reorder the header row that's soldered onto the ST motherboard as those are the colour bits with bodge wires so you can easily cross them over.

I've been working on getting the PCB removed but I'm having a difficult time removing all the solder. I used a solder sucker (the Engineer one, such a great product) but I still need to remove more solder because of the tight holes. For that I've used wick and flux but I can't seem to get it all out.

[Badwolf]

You would have to reorder the header row that's soldered onto the ST motherboard as those are the colour bits with bodge wires so you can easily cross them over.

I've been working on getting the PCB removed but I'm having a difficult time removing all the solder. I used a solder sucker (the Engineer one, such a great product) but I still need to remove more solder because of the tight holes. For that I've used wick and flux but I can't seem to get it all out.

Are you able to get in at an angle, cut the (6?) pins needing attention and then desolder them one by one (taking them out the bottom of the board)? resoldering the wires to perform the cross overs should be feasible with the rest of the board still in place.

BW

[sandord]

Are you able to get in at an angle, cut the (6?) pins needing attention and then desolder them one by one (taking them out the bottom of the board)? resoldering the wires to perform the cross overs should be feasible with the rest of the board still in place.

BW

I whish I could do that but unfortunately, the DAC board is only a few mms away from the shifter socket. I can't get my pliers to reach the pins because of the socket.

[sandord]

Today I'm going to try to desolder the header pins from the bottom of the main board. I hope that'll be easier than de-soldering the DAC board from the pins themselves.

[sandord]

Since the pins were fitting so tightly in the holes, even after extensive wicking, the board still wouldn't come off.

tight-pins.jpeg (235.78 KiB) Viewed 1024 times

So I decided I had to cut all of the legs to get it removed. Although I couldn't reach legs that had to be reordered before because of the shifter socket, I managed to cut the legs on the other end of the DAC PCB so I could tilt the PCB in order to reach the pins close to the shifter socket.

tilted-dac-board.jpeg (494.18 KiB) Viewed 1024 times

tilted-dac-board-underneath.jpeg (405.93 KiB) Viewed 1024 times

Then I managed to cut all pins, de-solder the remains from the main board and from the DAC PCB as well.

The next step will be fitting new header pins, except for the connections that have to be swapped, which I will connect using bodge wires.

mainboard-removed-dac.jpeg (481.77 KiB) Viewed 1024 times

desoldered-dac-board.jpeg (273.36 KiB) Viewed 1024 times

Unfortunately, due to the tightly fitted pins, I accidentally ripped off two via solder pads. One of them is no problem I think (the trace is on the other end) but the other one (the third one from the left on the picture), I will have to fix by soldering a wire to the corresponding DAC chip pin.

I realise now that I could have refrained from removing the DAC board entirely and only cut the wrongly connected (now bent) pins instead of all of them. But now that's not an option anymore