Odd Read Error Issue

General discussion forum - for all that doesn't fit in any other category.
Post Reply
SainT
Posts: 7
Joined: Fri Apr 11, 2014 9:16 am

Odd Read Error Issue

Post by SainT »

I’ve been mounting an hxc with screen in a CPC 464plus. It was all working fine, I removed the lcd header to solder the lcd cable direct and managed to bridge a ground via and the pad next to it on the lcd header and powered on before I noticed. Didn’t boot, quickly powered off, found the issue and sorted it. Booted fine, but getting an intermittent read error.

I found touching the floppy cable fixed the problem. I assumed I’d just managed to damage the connector / cable with repeated removal. Put a new icd connector on, no different. Decided it must be the floppy socket. Resoldered all the joints on the floppy socket. No better, if anything, worse!

I still find touching the floppy cable fixes the problem. Also I’ve found touching around pin 34 on the back of the floppy connector or even touching the pcb around this area fixes the read error. Weirdly even if my finger is VERY close, but not actually touching, it can also fix the problem.

So there’s something really weird going on which the stray capacitance of my finger near the trace sorts it. I’m not sure how this relates to the grounding of the lcd pin, or if it’s actually unrelated.

Any thoughts, Jeff? Have you seen any weird behaviour like this?

SainT
Posts: 7
Joined: Fri Apr 11, 2014 9:16 am

Re: Odd Read Error Issue

Post by SainT »

Looking at the lcd controller pinout I think it was the r/w pin grounded, so I’m not convinced it was related. I can’t remember if you wait for response with those lcd’s, but if so that would explain why it didn’t boot.

Would leaving pins 2-6 and 12-14 floating be an issue? As this is coming from the CPC 26 pin internal connector. I’m not sure if the DS / DCD lines floating could be the issue....

SainT
Posts: 7
Joined: Fri Apr 11, 2014 9:16 am

Re: Odd Read Error Issue

Post by SainT »

I think I’ve got to the bottom of this - it’s the /SIDE1 select line. Its driven by a 74ls38 open collector nand without any pull-up anywhere that I can see, the pin goes direct to the mcu. Should this pin have an internal pull-up which is somehow not functioning? Or is this input expected to be *driven* high?

Adding a pull-up should fix this, but I’d like to know why this problem exists. It previously just worked by luck - I expect the proximity of the lcd display actually caused this pin to pull high!!

Jeff
Site Admin
Posts: 8092
Joined: Fri Oct 20, 2006 12:12 am
Location: Paris
Contact:

Re: Odd Read Error Issue

Post by Jeff »

Are you talking about an SD HxC ? There is anyway an pull-up resistor on the HxC on the /SIDE1 line.

SainT
Posts: 7
Joined: Fri Apr 11, 2014 9:16 am

Re: Odd Read Error Issue

Post by SainT »

It’s the Rev c. I’ll have a proper trace today, it definitely wasn’t behaving like there was a pull-up! Unless it’s somehow missing on my board.

SainT
Posts: 7
Joined: Fri Apr 11, 2014 9:16 am

Re: Odd Read Error Issue

Post by SainT »

Just checked, and yep, it's R3. Getting a reading of about 10M!! The resistor has blown.

This board is second hand, I'm wondering if this was blown before I got it... :?:

Post Reply