Hi all,
you may have seen in a different thread, that MrLou was having issues with his DDI-1 / HxC setup, so he sent it over to me. The problem was that he had connected it the wrong way around and his DDI-1 instantly became an ex-DDI-1. Now normally this shouldn't have happened, I have often connected floppies and even the HxC the wrong way around to my 6128, and other than the hassle and wasted seconds of my life having to re-connect it, nothing more happened.
Yesterday I discovered that with the DDI-1 it's a completely different story...
Firstly, MrLou had connected it using a setup that still fed the 5V through the flat cable, which I thought was the most likely problem. After inspecting the DDI-1 I saw that both IC108 (74LS38 - Output inverters) and IC103 (74LS240 - Input inverters) had been destroyed. This was to be expected, as they are the first lines of defence in the circuit.
However, after replacing these and getting the DDI-1 back working with an FD-1, I then made the 5V conversion so that the DDI-1 gets its current from the CPC and those lines were diconnected from the flat cable. This should now be exactly the same as the 6128 (and as robust), but alas I stupidly re-connected the HxC, forgetting that I hadn't reversed the connector yet. Once again the DDI-1 died, but this time it doesn't seem to be the two ICs mentioned above that died, they're fine? It was then getting late, so I haven't found out what's broken this time, although I suspect it could be the µPD765. It will be swapped tonight if that's the case.
I'll report back with more as soon as I have more info. Both the HxC and the FD-1 survived all of this, which is good :)
I just thought I'd post this as a warning to others who might consider connecting a HxC to a DDI-1 through "trial and error" - Warning: Don't!
Bryce.
Edit: I should probably mention, the error that now occurs is that the red FD-1 LED is always on, the drive turns, but reports disk missing. The AmsDOS ROM still works fine and the CPC doesn't crash, so much of the DDI-1 is still functional.
Ok, an update. The DDI-1 wasn't actually broken the second time. The problem was that I was using my FD-1 to test it. For some reason the bridge that gets added to supply the DDI-1 from the CPC instead of from the FD-1 (required for 3.5in or HxC connection) stops the FD-1 from working. I'm not sure why just yet, there are many things that could go wrong: Brown-out, noise interference, timings at start. I'll have to investigate it later when I have time. At the moment you would need to add a switch if you intend using a DDI-1 for both FD-1 and HxC. But maybe I can work out a switchless solution to the problem, when I have time to find out what's the real problem is.
Eitherway, the DDI-1 is alive again and on it's way to MrLou for his gaming evening :)
Bryce.
Yay!!!
The community is really lucky to have a wizard like you among us Bryce.
Thanks a lot for taking time to look into this!
By the way. I've now experimented a little further. Connecting the HxC in the wrong direction only blew the DDI-1 because of the 5V on pins 21,29,31 and 33 of the flat cable. These are the wires that the FD-1 uses to supply the DDI-1 with power normally. After these were disconnected, the DDI-1 doesn't get damaged by connecting it reversed (as would be expected).
Bryce.
Like I always seen on tutorial about external CPC drive :
- always check the ribbon cable to not plug it reverse. (check pins number or Vcc/Gnd lines)
- never connect 21, 25, 29, 31, 33 pins to avoid destroying parts on your CPC if the cable is reversed.
I don't understand how it's possible to plug so expensive device on a precious computer w/o checking that...
mr_lou is lucky to get you for fixing the hardware. By chance, no +12v here. :D
Pin 25 is Drive Select 0, so it should be connected, the others are all 5V and should definitely be disconnected.
The problem is really that Amstrad strayed way off from the standard, but used the standard connector. So it is an easy mistake to make and you can't really blame the user for thinking that it would work. On any other computer or setup it wouldn't be a problem, just on the DDI-1.
Bryce.
Quote from: TotO on 08:38, 07 September 12
I don't understand how it's possible to plug so expensive device on a precious computer w/o checking that...
It's easy. It goes like this everytime (with me at least):
You connect it the right way. But it's not working. Puzzled by this, you first try to disconnect and re-connect other pieces, just to make sure they weren't loose. (E.g. the DDI-1). When things still aren't working, you consider that maybe you connected the cable wrong - and that's where you make your mistake.
I haven't got a clue about electronics, so it's easy for me to mess things up. Usually I do connect everything right at first though.
The new connector has a keyway, as does the HxC, so it's now impossible to get wrong :)
I also added ParaDOS, with a switch for AmsDOS / ParaDOS while I was at it :)
Bryce.
I like happy endings to horror stories!!!
the first thing I did with my DD-1 was to disable the +5v on the cable and have the device take 5v from the CPC464.... oh and lumped in a freshly burnt Parados ROM =)
when I plugged it in and the 464 stated to have a fit.... I thought I had borked something - out with a multimeter - and the better part of an hour later found nothing wrong.... it turned out it was the contacts for the port.... it was the filthiest slot this side of.... well.... lets not get to colourful with the language ;)
A siff bit of card, IPA and a clean cloth sorted out that problem, and she is now my mainstay Arnold! with a 3.5" Drive.
which reminds me I must put my spare FD-1 up for sale.....
Edge connectors were the worst idea that the 8-bit designers (not just Amstrad) ever came up with. They're not suitable for regular connection/dis-connection and they don't take kindly to dust, dirt, large expansions, movement, un-even desks, sneezing in the same room and so many other things. I'm thinking of soldering a header connector onto my 464 to solve this. The Schneider was a million times better with the Centronics connection, but they didn't do this for the sake of the user. They were forced to use a connector to comply with German EMC laws.
Bryce.
Heh, I always love Bryce's detective stories... :D
There's more to come. At the weekend I intend fixing MaVs Spectrum 48K, although with the weather heading towards 30°C again, I may postpone it for a few days :)
Bryce.
A Speccy? More like a b-movie... :D