PSU alternative?

Nagging hardware related question? Post here!
User avatar
Cristian
Super Gold Card
Posts: 506
Joined: Mon Feb 16, 2015 1:40 pm
Location: Veneto

Re: PSU alternative?

Postby Cristian » Tue Nov 07, 2017 7:33 pm

Thank you Martin for your explanation. Learning something is always important for me.


Martin_Head
Gold Card
Posts: 356
Joined: Tue Dec 17, 2013 1:17 pm

Re: PSU alternative?

Postby Martin_Head » Thu Nov 09, 2017 10:56 am

I was looking at the circuit diagrams of the +/- 12 volt supply on the QL and the PSU trying to figure out how they work.

The PSU circuit states that it's a 16v AC winding on the transformer.

On the QL circuit the +/- 12 volt regulators use the 9 volt DC line as a ground/reference?

You can get plus 16 volt, or minus 16 volt from an 16 volt AC winding, but how do you get both? Unless the two diodes by the +/- 12 volt regulators on the QL are some kind of a voltage doubler. (Which I've never really figured out how they work)

But then I would expect the +/- 12 volt rails to be referenced to the 9 volt rail. i.e. +21 and -3 volts.


User avatar
tofro
QL Wafer Drive
Posts: 1447
Joined: Sun Feb 13, 2011 10:53 pm
Location: SW Germany

Re: PSU alternative?

Postby tofro » Thu Nov 09, 2017 12:17 pm

Setting the ground reference of the +/- 12 V lines to the +9V line looks weird, at first, but you need to look into the PSU in order to understand it.

Sinclair has simply saved on one strand of wire, the 14VAC ground. Most probably, the complete transformer is one continuous winding, the input to the secondary coil being GND, the next pickup point the AC for 9V and the last the ~12V. The difference in potential for 14VAC is thus between the +9V and the ~12V output. The AC between GND and VAC is probably much higher that 14V, as it is the AC produced by both secondary transformer coils.

The diodes are there to let the positive sine wave pass into the 7812, the negative sine wave to the 7912.

I guess the whole circuit would make more sense when the PSU circuitry is included in the drawing.

Tobias


ʎɐqǝ ɯoɹɟ ǝq oʇ ƃuᴉoƃ ʇou sᴉ pɹɐoqʎǝʞ ʇxǝu ʎɯ 'ɹɐǝp ɥO
Silvester
Trump Card
Posts: 189
Joined: Thu Dec 12, 2013 10:14 am
Location: UK

Re: PSU alternative?

Postby Silvester » Thu Nov 09, 2017 12:29 pm

Martin_Head wrote:On the QL circuit the +/- 12 volt regulators use the 9 volt DC line as a ground/reference?

It's an error in the original service manual circuit, it is 0V. I checked PCB when doing PDF version.


Martin_Head
Gold Card
Posts: 356
Joined: Tue Dec 17, 2013 1:17 pm

Re: PSU alternative?

Postby Martin_Head » Fri Nov 10, 2017 11:14 am

The PSU http://www.dilwyn.me.uk/docs/hardware/QLperip.zip shows the 16 AC as a separate winding with one end connected to the 0v (GND) line.

If as Silverter says, there is an error in the service manual circuit, So that the + and - 12 volt circuit would make sense. Then the AC signal arriving at the QL would have to swing between about + and - 16 volts. ie. about 32 volts peak to peak around ground.

Thinking about it a bit more, it does makes sense. The 16 volt AC, if measured with a multi meter, would be 16 volts. The meter would half, or full wave rectify the AC signal, so you would only 'see' the upper, or lower half of the AC signal. Whereas if you looked at it with an Oscilloscope, you would see that the signal actually went from +16v to -16v.

So both are correct, The PSU supplies 16v AC, and the QL receives the quoted 44v peak to peak.


User avatar
Pr0f
Gold Card
Posts: 287
Joined: Thu Oct 12, 2017 9:54 am

Re: PSU alternative?

Postby Pr0f » Fri Nov 10, 2017 11:45 am

This is just half wave rectification, Sinclair being cheap again ;-) The current requirements for the +12 and -12v are so light, they can get away with this, and the ripple is easily dealt with by the regulator..


User avatar
1024MAK
Super Gold Card
Posts: 518
Joined: Sun Dec 11, 2011 1:16 am
Location: Looking forward to summer in Somerset, UK...

Re: PSU alternative?

Postby 1024MAK » Sun Dec 24, 2017 8:10 pm

Apart from the "common" 0V / GND / "ground" wire, the 9V DC and the 15.6V AC parts of the power supply work independently.

The 15.6V AC from the transformer is referenced to the 0V connection. So on the positive half cycle of the AC waveform, the instantaneous voltage is positive with respect to 0V GND and a diode connducts and allows the "smoothing" capacitor for the 78L12 +12V voltage regulator to charge up. This capacitor then allows the regulator to continue to function during the negative half cycle.

During the negative half cycle, the instantaneous voltage is negative with respect to 0V GND and another diode connducts and allows the "smoothing" capacitor for the 79L12 -12V voltage regulator to charge up. This time, to a negative voltage with respect to 0V / GND. This capacitor then allows the regulator to continue to function during the next positive half cycle.

And yes, what you have overall, is two back to back half-wave rectifier circuits.

My earlier statement that two transformers would be needed, is true only if you are using off the shelf "standard" transformers. Of course, if you are happy to pay extra, there are companies that will custom wind transformers. Then you can have both windings on a single transformer.

Mark


QL, Falcon, Atari 520STFM, Atari 1040STE, more PC's than I care to count and an assortment of 8 bit micros (Sinclair and Acorn)(nearly forgot the Psion's)
User avatar
Cristian
Super Gold Card
Posts: 506
Joined: Mon Feb 16, 2015 1:40 pm
Location: Veneto

Re: PSU alternative?

Postby Cristian » Tue Sep 18, 2018 8:22 am

I found the following sentence in another forum:

"The 15.6V AC can be replaced with anything 14 - 17V AC at about 2A. The 9V DC can be anything from about 7.5V to 12V DC, 1.5A on an unexpanded QL and 2A on an expanded QL."

Is it true? Or will this overheat the 7805 regulator or other parts?


User avatar
tofro
QL Wafer Drive
Posts: 1447
Joined: Sun Feb 13, 2011 10:53 pm
Location: SW Germany

Re: PSU alternative?

Postby tofro » Tue Sep 18, 2018 10:42 am

This is not so easy to answer - it depends a lot on the "stiffness" of your power supply under load.

Power supplies are typically reted at a voltage that can be maintained under a typical load - Note, what you would measure on an unknown PSU is the open circuit voltage which is typically much higher than under nomainal load. People unfortunately tend to mix these terms up.

A 7805 voltage regulator needs an input voltage of at least 7V (more ideally, 8V) under all conditions in order to properly work - Everything below that will make the output voltage unstable and your QL reset. Everything above that, the 7805 will simply convert into heat. Even if it is specified for input voltages up to 25 Volts, the QLs heat sink will not be able to sustainably dissipate that heat at too high input voltages and the 7805 will simply die in the long run.

So, if you got a "stiff" power supply that can maintain 8 Volts at, say, 3 Amps, you should be perfectly fine with that (and be able to operate the regulator at ideal conditions). On the other hand, a PSU that will drop its output voltage significantly under load should be constructed with a bit more reserves.

This all assumes you want to retain the original 7805 for whatever reasons - There are, however, modern and very good replacement switched mode regulators as drop-in replacements for the 7805. These guys work much cooler than the oriinal.

Tobias


ʎɐqǝ ɯoɹɟ ǝq oʇ ƃuᴉoƃ ʇou sᴉ pɹɐoqʎǝʞ ʇxǝu ʎɯ 'ɹɐǝp ɥO
Nasta
Gold Card
Posts: 265
Joined: Sun Feb 12, 2012 2:02 am
Location: Zapresic, Croatia

Re: PSU alternative?

Postby Nasta » Tue Sep 18, 2018 4:47 pm

7805 needs 8V DC at the input with less than 0.3V ripple to properly regulate.
The 15.6VAC line certainly does NOT need 2A, rather 0.2A. It's only used to power the RF modulator and the serial port drivers.
The simplest way to get a replacement PSU is to buy a 9V SMPS module, 2-3A would be ideal. They are available at many electronic parts retailers, Meanwell does a nice small sized open frame series, which can also be adjusted about +-10%. In our case adjusting it to -10% will give us a good 8.1-8.2V supply, which will also leave some slack for losses on the connecting wire.
Next, get a small 15VAC secondary mains transformer, 3-5VA is enough - from the same place as SMPS.
Connect +8.2 and GND to appropriate wires in the QL PSU cable, and then the small transformer output between the 0V and the 15V line.
The inputs to SMPS and transformer (AC mains) go in parallel.
No doubt the same parts dealers will have a suitable enclosure :)



Return to “Hardware”

Who is online

Users browsing this forum: No registered users and 1 guest