Sound chip on cpc plus (AY38912A/P vs AY38912/P)

[Bryce]

Interesting, I've never tested what parts stop functioning first if you start to dial down the voltage. I would have expected the parts that use the most current (CPU, CRTC, GA) to fail first.

@makinavaja: The reason why the switch can have an effect is because when it gets dirty or oxidised it becomes a resistor. So the more current the computer is pulling, the higher the voltage drop across the switch will be.


Bryce.

[makinavaja]

Hi!

Or just the power switch that need a good cleanup    @makinavaja can you test this CPC using the keyboard/power switch of your other plus ?

I got the same results :-(

@makinavaja: The reason why the switch can have an effect is because when it gets dirty or oxidised it becomes a resistor. So the more current the computer is pulling, the higher the voltage drop across the switch will be.

Thanks for the explanation!


Those ASIC chips are a pain in the ass.  On MSX, the 95% of the computers uses MSX-engine chips for multifunction purposes like sound, cpu , signals.... and of course there are no chip over there for purchase.


Sh**t, so, this computer is dead, I supose u_u;

[makinavaja]

Also, I tested this too: I configured the powersupply to give 6V.  Then, the AY received 5,4V and, also, no sound no keyboard.

For an other thread about supply for GX4000, I've bee testing one down to 4V : sound is the first thing to fail.
Knowing that the  measured supply on the AY is about 4.3V on this plus and the other one is fully working, if the supply provided is indeed 5V, the power switch is one of the possible culprit for the voltage drop.

[Bryce]

Also, I tested this too: I configured the powersupply to give 6V.  Then, the AY received 5,4V and, also, no sound no keyboard.

That was a dangerous move. 6V is enough to fry many of the other chips, but without additional test equipment there is not much else you can do now other than declare it dead.

Bryce.

[Audronic]

Hi All


Has anybody tried to Reflow the solder on the gate array with a a hot air gun ? ?


Hopefully yours


Ray

[Bryce]

Hi All


Has anybody tried to Reflow the solder on the gate array with a a hot air gun ? ?


Hopefully yours


Ray

Why would you use a hot air gun on a through hole part? A soldering iron is the correct tool to do that.

Bryce.

[Audronic]

Hi Bryce


" Has anybody tried to Reflow the solder on the gate array with a a hot air gun ? ?"



The GATE ARRAY the one with nearly 200 pins IC101 AMS40489 or equivalent.
Thats what I was referring to ? ?

Ray

[robcfg]

That would be the ASIC.

[Audronic]

? See Photo ?

[robcfg]

Hold your horses.


If you take a look at the Wikipedia page for Gate Array, it defines them as:

A gate array is an approach to the design and manufacture of application-specific integrated circuits (ASICs) using a prefabricated chip with components that are later interconnected into logic devices (e.g. NAND gates, flip-flops, etc.) according to a custom order by adding metal interconnect layers in the factory.

Now, for the sake of communication, the naming convention around those chips has been the following (taken from our wiki):

Amstrad Custom chips
The Amstrad CPC used one custom chip: the video Gate Array (also called VGA – no connection with the Video PC standard).


Latter CPC cost down series included a "pre-ASIC"-called ASIC to merge the VGA and the CRTC.


The Amstrad Plus included a "second heart" simply referred as the ASIC.


CPC+ ASIC's part number is 40489

So you are right that the part is named Gate Array in the service manual. But it is a type of ASIC. So it is confusing. That's why it's easier to use Gate Array for the 40007, 40008 and 40010, Pre-ASIC for the 40226 and ASIC for the 40489.

[Bryce]

? See Photo ?

Yes, I see a photo of the ASIC, not a Gate Array . But now that we know which chip is meant, my suggestion would however still be the same. A hot air gun puts too much heat into the body of the chip before it has heated the pins sufficiently. Going around the pins with a soldering iron would still be a safer solution. Yes, I've done this on occasions when the solder looked particularly dry and cracked, but it was rarely (if ever) the solution to the problem.

Bryce.

[Audronic]

Hmmnn

[makinavaja]

Hi again everybody


If I manage to get an oscilloscope, would anybody help me saying wich kind of tests do I have to do and wich results will I get?
I never used one before and I plan getting one of those mini oscilloscopes available on internet shops. I supose with this I will learn the basics.
Another option was asking anybody to ship him/her the board (paying, of course) to see if it can be reapaired withouth touching the ASIC.
I'm quite stubborn with this kind of stuff, as I wont ever get quiet until something broked isn't fixed.  Repair repair and repair before purchasing another one...

[Bryce]

Before you go and buy one of those mini oscilloscopes, let us know which one. Some of them are absolute junk and not even worth the few Euros they cost.

You can of course send the board to me any time you like and I will diagnose (and fix) it for you if it's fixable.

Bryce.

[Bryce]

This Plus arrived here last night. I've repaired the sound issue, Q106 had failed. That's the small KTC1815 transistor that drives the output. I haven't looked at the keyboard side of things yet because I have to go out now and it will take a bit longer because the AY replacement wasn't quite as perfect as stated above. There's at least two damaged pin holes and I'll need to go through them all. Job for tomorrow...

Bryce.

[Bryce]

Finished the Plus this morning. Sound and keyboard are now doing what they should. Moving on to the classic 6128 now (also from Makinavaja). For reference, this one displays a blackish screen with flickering grey lines and has already had quite a bit of mods / repairs done to the board. I'll post the progress in this thread too as I don't think Makinavaja opened a seperate thread for it.


Bryce.


Edit: Oh dear! A bit of a thermal Christmas tree going on here. Any chance that this got connected to the wrong voltage or reversed polarity?

[tjohnson]

It's not suppose to look like that then?  Is it showing most of the chips are fried?

[Bryce]

Yup. I've done some diagnosis and almost everything that's connected to Data line D6 is fried. Looks more like a static discharge issue than an over-voltage. Lots of chip changing required I'm afraid. Luckily, the 40010 and the CPU (and possibly the 8255) have survived, but several logic chips, RAM and the CRTC weren't so lucky.


Bryce.

[makinavaja]

Oh my god u_u


I would say this: cpc comunity is SO LUCKY for having hardware masters like Brice, really.


I'm still looking for people doing the same for MSX systems...


Bryce, thank you very much!

[Bryce]

Finally starting to get somewhere with the 6128. It now boots, but the keyboard is completely dead and I can't test the floppy circuitry yet, so I don't know whether that works either.

My first suspicion was that it had got a blast of static discharge, but I now think that it was a mistake I've seen before: High voltage on a data line (in this case D6). The CRTC, 74LS244, 74LS373 and both D6 RAMs had to be replaced to get this far. Last time I saw this type of failure was when someone used an unregulated soldering iron on a computer while it was still connected to a monitor.

Anyway, need to take a break now and get back to it later.

Bryce.

Edit: I don't just do CPC's, I do any retro computer.

[Bryce]

Doh! My fault, I forgot to reconnect one of the pins on the 8255 that I had isolated. The keyboard and disk interface both work fine, the sound too.

Job finished, time for a beer

Bryce.

[makinavaja]

By the way... the fdc controller chip was fake or was good?

[Bryce]

The FDC chip is a real one.

Bryce.

Edit: Aaaand parcel already on its way back to you....