+12V unavailable for the drive

[Salvo B]

Hi guys...

During my last experiments for connecting a PC floppy drive to my CPC something went wrong and now I'm unable to get the 12V on the power cable of the drive.

First to be sure that wasn't just a cable/connector issue I've measured the tension on the board in the point where  cable starts...
Next I've looked for the schematics of the board in the service manual.

My first idea for fixing this was to pick the 12V directly from the connector that comes from the external power, but if I've recognized well the circuit there's a transistor and a resistor between the external plug and the actual pin on the drive, so maybe I, should check if this is damaged...

The problem is that seems that the transistor (labeled Q303 on the PCB) isn't listed in the parts list (or maybe is the one listed as W303?)...
When I'll be back at home I'll check for the label on the component itself...

Anyway... Is the option of picking the 12V directly from the external plug connector on the board risky?

Thanks in advance

[Salvo B]

Sorry, I've forgot to mention that I should use the 12V for the original 3"drive and not for the PC drive

[Bryce]

Hi Salvo,
     taking the 12V directly from the connector isn't really risky, but it may make the start-up process a hit or miss affair. The 12V rail is switched on slightly later than the 5V rail to allow all circuits to have reset. You won't have found the 12V transistors in the schematic because Amstrad messed up the original schematic releases and forgot lots of bits. Luckily they fixed this later with the Amendment schematics, which you can find here: http://www.cpcwiki.eu/index.php/Service_Manuals (The sixth link under "CPC Computers"). On page 19 you'll find the 12V circuitry. It's actually a 2 transistor configuration that turns on the 12V.

What happens:
When the CPC turns on the 5V rail (VCC) rises to 5V and the voltage divider R151/R152 biases (switches on) Q104.
This in turn pulls the base of Q103 towards GND biasing Q103 and allowing the 12V rail to be connected.

Quick test method: Check whether there is a voltage drop across R153. If there is (ie: current is flowing through it), then Q104 has switched and all parts before that point (R151, R152 and Q104) are fine, so the broken part is Q103. If not then Q104 is most likely broken.

Bryce.

[Salvo B]

Thank you Bryce, you're reply was very appreciated, on the amendment manual the circuit refers to another version of the board (different components codenames) but I was able to identify the two transistors and the resistances following the rails on the PCB and the codes on the components themselves.

I just have a lame question... how to perform the measurements on R153? You've talked about tension drop, so I should check if the voltage is slightly under 5V? Under 12V? Across the resistance means that I should put the two probes on the two pins of the resistance? Or one in a pin and the other to GND? Where to pick GND?

Ehm... As you can see from the flooding questions I'm not so experienced on this kind of works, even if I've understood the basics of the circuit I don't have the full skills needed to performs proper measurements... I'm just trying to manage it on myself thanks to the help of the forum users that is always appreciated.

Well so if you can write this as I were a 5year old electronics-freak-child will be very helpful to me

Thanks for you patience... (feel free to give up if it bothers you)

[Bryce]

To measure the voltage drop across the resistor you need to put the meter to VDC and place the black and red probes on either end of the resistor (no ground point needed). Which end black and red is connected isn't important, just ignore the +/- of the voltage reading. If the voltage reads 0V then no current is flowing through the resistor.
If there is a voltage, you can calculate what current is flowing through resistor to judge if it is too much or too little:

V = I x R   =>  I = V/R

So if you measured 2V across the resistor it would mean that 2mA is flowing through the resistor (as we know that R=1000). This wouldn't be enough as the KTA950 is being used as a switch and the transistor can handle up to 800mA. I don't have a datasheet for the KTA950, but if we assume it has a hfe (gain) of 100, it would need at least 8mA to fully turn on, so you should be expecting to measure around 8V across the resistor.
However, both transistors should have a small (0.6 to 0.7V) voltage drop across them too, so if the voltage across the resistor is the entire 12V, then that would signify that both transistors have burnt through and are shorted.

Bryce.

[Salvo B]

Very well crafted reply, thank you for your time.
I'll check the voltage across the resistor as soon as possible, meanwhile I've already managed to find the replacement components.

During the school days I've studied all of this things (bjt configurations, and so on) but I didn't get a chance to get this skills in practice so they are almost missing somewhere in my brain

Anyway this is a really basic circuit and I was lucky enough to avoid other damages in other parts of th board since it's very independent (except for the link with the VCC)

Now that everything looks clearer to me, someone had a guess on the reason of this damage?

I've just plugged a 3,5" Pc drive inverting the +12 and +5 pins on the power plug using some breadboard pin cables.

It turned on correctly but since the drive wasn't recognized I've just tried to turn off and on again the machine a couple of times (waiting some seconds between the on and the off).
And then the issue occurred...

Inviato dal mio GT-I9301I utilizzando Tapatalk

[Bryce]

If that's how it broke then I would guess with about 99% certainty that Q103 is the only bit that got damaged. The KTA950 can take 800mA at most. Connecting 12V to the 5V rail of the drive would have caused a short circuit and the amps across Q103 would have gone considerably higher. However, the PC floppy drive will also be damaged, so I wouldn't connect it to anything ever again.

Bryce.

[Salvo B]

I've plugged the drive this way because the plug of the cpc ha the 12V/5V inverted, so I guess that I wasn't wrong

[Bryce]

I've plugged the drive this way because the plug of the cpc ha the 12V/5V inverted, so I guess that I wasn't wrong

Yes, that catches many people out. Unfortunately it usually means having to repair your CPC and having to throw away a now useless 3.5in PC drive.

Bryce.

[Salvo B]

the tension across R153 is null, what does it means so? I've already ordered Q103, should I buy Q104 too?

[Salvo B]

the tension across R153 is null, what does it means so? I've already ordered Q103, should I buy Q104 too?

I think I can reply on myself, If I've recognized correctly since no current is flowing through R153 the 5V trigger from Q104 isn't working so even if Q103 isn't burnt I guess that it can't work anyway.
At this point I'm just tempted to check also if the resistors are OK, but I guess that if they are burnt I should be able to see some physical damages too, am I right?

[Bryce]

If the transistors aren't taking weeks to be delivered, I'd swap Q103 first and see if it then works before ordering Q104.

Bryce.

[Salvo B]

Well, today finally I found the time to swap the transistors, now everything works fine!

Just for the records, both of them was damaged...

Thank you Bryce!

[Bryce]

Cool, glad you got it sorted.

Bryce.