Beaten by VT100 PSU.

[Mattis Lind]

2016-08-17 11:14 GMT+02:00 Peter Coghlan <cctalk at beyondthepale.ie>:

> >
> > It is a VT100 PSU (H7831). I tested it with dummy loads and it worked
> fine.
> > But when used in the terminal with the Basic Video board and monitor
> board
> > it gives a jumpy picture. Both horizontally and vertically. First I
> thought
> > that it was related to the monitor board but soon recognised that the
> +12 V
> > had a most peculiar waveform on it:
> >
>
> I've seen a lot of VT100s with poor vertical linearity and of those, many
> jumped vertically as well.  I never got to investigate whether this was
> poor adjustment and/or noisy controls or a more real fault.
>

I could have filmed the behaviour but never did. But the size of the
picture was jumping which goes well with the bad input voltage. The
linearity was fine.


> >
> > http://i.imgur.com/d0z0NQS.jpg
> > http://i.imgur.com/gQqmSN5.jpg
> > http://i.imgur.com/P0dt5y1.jpg
> >
> > This waveform is only present on the +12V, not the +5V, not the -12V or
> > -23V.
> >
> > So I connected just the Basic video board and a variable dummy load
> instead
> > of the monitor board. I used a Variac on the input. It turned out that
> > there were no problems now with the +12V. Until I pulled out and
> reinserted
> > the keyboard. Then it was there. If I lowered the input voltage it was
> > impossible to provoke this problem and also if I increased the +12V load.
> > Further testing also gave that putting a few amps extra load on the +5V
> > also made it resistant to this type of failure mode.
> >
> > The amplitude and frequency of this waveform is shifting by +12V loading
> > and AC input voltage.
> >
> > The VT100 SMPSU is a primary switcher regulating the +5V. Then the +12V
> is
> > handled by a secondary switcher which is synchronised with the primary
> > switcher. The other voltages have linear regulators.
> >
> > It looks to me that something in the regulation circuitry is not
> behaving,
> > thus oscillating. But what component has failed (or is out of spec)? I
> > checked transistors. I checked the waveform from the 555 chip and ramp
> > voltage input to the 555. But I cannot figure out what the problem is.
> >
> > I checked the 560uF output capacitor but my LCR meter said it was in good
> > shape. Around 700 uF and very low ESR.
> >
>
> If the waveform is present across the +12V output capacitor and +5V is ok
> and
> and the regulation only monitors the +5V, it sounds like the power supply
> is
> doing it's best and maybe the problem is in the +12V load?
>

Yes. The waveform is present on the +12 output capacitor. There are dual
regulators. The main is only regulating the +5V and then there is a
secondary to regulate the +12V.


>
> Alternatively, could there be an abnormally varying load on the +5V which
> the
> power supply regulation managing to smooth out but in doing so causing the
> unmonitored +12V to vary?
>

The +12V is not unmonitored since there is a switching regulator there
using a NE555 chip and a LM393 regulating the +12V as a secondary side
SMPSU.

The main regulator is a SG3524 chip. The waveforms from the SG3524 and the
output on the +5V looks perfect. It is just the +12V that is bad which
directs me to the NE555 and LM393 regulator circuitry.



> Monitoring the current waveform in the +5V and the +12V might throw some
> light
> on this, maybe by putting the scope across a small series resistance in
> each
> line.
>

That is an interesting idea. I will bring in my current probe.


>
> I guess the 560uF capacitor referred to above is on the 12V line?  The
> corresponding capacitor on the +5V might also be worth checking in case
> the regulation of the power supply is managing to compensate for it
> being bad.
>

I will take a look at the +5V output cap as well just to be sure.

If I put in the VT125 option board the terminal behaves just fine, so the
loading seems to be of importance.


>
> Regards,
> Peter Coghlan.
>

/Mattis