exxos's Atari, Amiga & retro forum

I always see the caps like an accumulator in a hydraulic circuit, which helps keep the system pressure regulated and smooths out drops in pressure.

So if the caps are not up to spec they can do neither very well.

exxos wrote: 10 Dec 2021 13:15 On or off load ? How much you loading it with ? Is there a schematic ?

On-load. And the ripple is 160mV.

As for the load, I don't know. I'm looking at ways that I can lash my bench power supply up to the 5V line.

As far as I'm aware there is no schematic for this particular Aztec supply but I will look.

I've worked out a way to hook up the bench power supply.

The current draw is 2.9A (and the PSU is rated at 4A on the +5V rail).

Have you try a lighter load ? Also just slapping a quality 4,700uF across the 5V output on the PSU ?

Problem with a lot of specifications as they rarely are actually met in the real world. A lot of the power supplies we used to work with, you was lucky to get 50% of what they actually said they were capable of :roll: Also total maximum output wattage is another way of making the power supply sound better than it really is.

With the I/O board, which is far lighter powered, the voltage drops by about 0.03 volts, similarly the floppy drive. With no load it's dead on 5V. It other words, it seems as if the regulation isn't working.

It's an Aztec BM41011 unit, which I can't find any documentation for anywhere.

I really would try more uF like 4,700uF on the 5V output and see if that solves the problem. I know you say you cannot see any spikes on the scope but normally when power supplies under regulate it's because the feedback is seeing "spikes" where then it thinks that is the output voltage, so the "RMS" output voltage ends up being lower. But those overshoots are generally down to inadequate output capacitors.Beyond that it could simply just be a bad designed power supply which is is not so easy to fix. It could be possible the regulation feedback is done on the another rail like 12V rail ?. ST power supplies have the reverse problem that the regulation is only on the 5V rail.

OK. I'll wait on the 220uF/25V caps arrival, swap those and see what happens.

Anyway... Let's forget the main unit for a bit.

The replacement transistors for the "mouse" arrived yesterday so I thought I'd fit one. The originals aren't being manufactured any more so I had to get a compatible equivalent. Well, I wanted 1 but I had to buy 200!

To cut a long story short, I fitted it and it showed the same fault as the previous transistor. The LED is always lit and the base is at about 2V. So, I guess the inverter chip (CD4069) must have a slight fault which is leaking power onto the output when it should be outputting a low value. It's the only explanation. Anyway, I removed the transistor again as that's the least risky option and it does work with only one LED running.

Anyway, I thought people would like to see the insides... Have you seen this much brass in a "mouse" before?

Actually, I've discovered the manual for the 2.5A version of the power supply. Exactly the same layout: http://binarydinosaurs.co.uk/Astec_AC8151_rm.pdf


As you can see, this is the same layout PSU as in the Osborne 1 and various other machines.

From the spec, 4.77V is actually within spec. for this PSU near full load, so maybe it's working correctly? I really can't believe that Apricot designed the machine to run at such a low voltage.

stephen_usher wrote: 10 Dec 2021 21:01 From the spec, 4.77V is actually within spec. for this PSU near full load, so maybe it's working correctly? I really can't believe that Apricot designed the machine to run at such a low voltage.

Would seem so :roll:

What are you getting on the 12V rail ?