Talent DPC-200 Power Supply Failure

Page 5/6
1 | 2 | 3 | 4 | | 6

By tsMKG

Resident (48)

tsMKG's picture

15-05-2018, 02:49

Oh, yeah, I was busy to do anything but I have new questions

First, I've checked that's the pin-out according to the voltages I measured with my multimeter and the documentation. I assumed that somehow they flipped it, so I said that. But it's been a long ago so I don't know if I'm correct.

Second, how should I build the IDC cable?
Top side with top side and bottom side with bottom side, is that right?

And third, I'm not able to make the PCB, like, it's always a headache trying to these lines not get crossed. And single-faced it's even worse. IDK what to do Sad (I try to make the PCB using a male IDC-50)
It's there a better way to make the PCB?

By Patrick

Supporter (4)

Patrick's picture

26-10-2019, 09:08

Hello,
I just got a Yeno DPC-64, french version of your Talent DPC-200. It is not working, so I will begin to check the PSU.
Do you have finished tour repaire?
Can you give a link to the documentation where you found the PSU schematics?
Thanks.

By Patrick

Supporter (4)

Patrick's picture

26-10-2019, 10:28

Hum, bad news. I just measured the voltages from CN1: F and G are perfect with +12V and -12V but E gives +8V. I think that is is too high for +5V and I fear that many ICs of the motherboard are destroyed.
In facts, IC1, the STR9005 doesn't regulate anymore. Its input at pin 4 is 8V (mains is 230V AC in France) and it ouputs more or less the same at pin 1.
I will order a new STR9005 and repair the PSU.
I have a working Olympia PHC-2. I will try to power the Yeno's motherboard with the Olympia's PSU. Let see what will happen.

By Patrick

Supporter (4)

Patrick's picture

26-10-2019, 11:06

I just did the test and I have dark blue screen and border, without anything else. Don't know what it means: working VDP? Bad CPU, bad ROM, bad RAM?
First, repair PSU, then motherboard.

By Patrick

Supporter (4)

Patrick's picture

29-11-2019, 15:11

The PSU is repaired. With a new STR9005 it is working well.
Now I have a dark blue screen and border, like my previous test with the Olympia PHC-2 PSU.
I made another test with a RAM extension cartridge. No change.
So there is a problem with the motherboard. With 8V supplied, we are beyond the maximum voltage allowed for PPI and Z80. I didn't check the RAM datasheet, but it is certainly the same thing.
Now, I have to figure out which components are destroyed.

By Alexey

Guardian (2537)

Alexey's picture

29-11-2019, 16:40

RAM is sensitive to over-voltage. I recommend to replace it first.

By Wild_Penguin

Hero (580)

Wild_Penguin's picture

30-11-2019, 17:11

IIRC the memory in a cart might not get used, if it is in a higher (or was it in a lower?) slot than the main RAM. So, in case the BIOS finds the RAM on the MB (but it is bad), changes are the BIOS will still try to use that RAM, and hence the MSX will crash. I believe the RAM and BIOS routine at boot-up is documented somewhere.

You could try to disable the RAM on the MB and then use the expansion cart, as in that case I believe the BIOS should pick up the RAM on the expansion more reliably (as it has no other choice). How to do this in a reversible way is left as an exercise and would benefit from a service manual - and would only be sensible as a diagnosing step.

(p.s. I had a voltage hike on my NMS2855 back ... 10 years ago? There is a thread about it on this forum. The culprit was a broken regulator as in your case. But in my case the voltage was a lot lower - <6V, and only momentarily ~7.4V (during diagnosing / measuring). In any case, after changing the regulator, everything worked, and has up until today =) - good luck with the repair !)

By tsMKG

Resident (48)

tsMKG's picture

11-01-2020, 03:50

Hello! Big smile I'm here again after not being here for too long. So... I've resumed my MSX repair.

Regarding the Expansion Bus on the Talent DPC-200, I've compiled some images of me testing the connectors to find out which pins are which. I've even put a pic of the ribbon cable connected to the MSX (above view). the polarizer key would be on the up side as shown by the microfiber marking.

https://cloud.disroot.org/apps/gallery/s/7DnZATt7HiSdJFz

And yeah, gdx, you were correct about it vertically being flipped in the 2018 photo notes, but what I'm asking now, is it normal to be the pins horinzontally flipped from what is shown at https://www.msx.org/wiki/Expansion_bus_slots :-? Cuz that's what the meterings on the photos seem to imply. And, yeah, from the other side (the original positions in the wiki) seem to give low voltages such as 1.00somethingV but from the flipped side seems correct (5 and 12V). I've included the pic of the ribbon cable just in case I did a mistake with it :P

By Wild_Penguin

Hero (580)

Wild_Penguin's picture

15-01-2020, 17:03

Your observation is correct and completely normal.

That Wiki page you have lionked is dangerous. OTOH it says there is no standard for expansion bus (connectors or pinouts), and describes two different keying. The problem is it gives just one pinout, and as such the page might give the impression there is only one pinout (EDIT: I've put a small note on the page about the pinout).

However, this is not the case. This page lists the expansion bus as proprietary connectors (which they are): https://www.msx.org/wiki/MSX_Cartridge_slot. Here the correct impression will be given: it is up to the user to check the pinout (however a fair assumption has been made that a certain brand should have the same pinout). However, IFAIK there are actually only two common pinouts for the expansion bus (which does not mean there could not be others).

I believe these pinouts are find roughly in 50:50 ratio: the 1) pinout is same as in the cartridge slot and 2) every odd<->even pin function has been swapped (not accounting for the keying - which could, in principle, be in either side of the connector - and of course changing the pin functions further!).

Also, saying pins are "vertically" (or "horizontally") flipped can be a bit confusing. First, the pin numbering of (a male connector) is the same, and it should be noted which pin numbers functions have changed, as I've done above (that way there is less room for confusion). Also, without keys (as most hobbyists IDC connectors are symmetrical and without keying) they can be flipped 180° while being connected to the MSXes (the original, brand-name, proprietary expansion units would have been keyed). Also - unless I'm mistaken - if you take any X by Y matrix, you will notice that mirror + rotation (by 180°) is equivalent to just mirroring (by an axis perpendicular to first case). What I'm saying here is vertical IS horizontal mirroring for these purposes, in the case we don't need want to / it doesn't make sense to account for keying! Point being: talking about vertical/horizontal flip can be confusing / ambivalent (we can not know what kind of connector the user has/had).

In your case I can see the connector is keyed, so you can not flip it 180°.

By rderooy

Hero (605)

rderooy's picture

15-01-2020, 19:27

If the Daewoo systems have a different pin-out we should add another table for them. From the other non-Daewoo manufactured systems listed under the two polarising keys section (Mitsubishi, Sanyo, Sony), I checked the Sanyo MPC-100 and its pinout listed in the owners manual seems to match the table on that page. So I think we can presumably the other Sanyo systems will also match. That leaves the Mitsubishi and Sony systems for which we have no data to my knowledge, so perhaps we should list their pin-out as unknown.

Page 5/6
1 | 2 | 3 | 4 | | 6