Use of non-screened transformer in old monitor

[jms...@gmail.com]

I have partially repaired an IBM mono monitor whose mains transformer had failed.

http://minuszerodegrees.net/oa/OA%20-%20IBM%20Monochrome%20Display%20(5151).pdf

Driven from a genuine PC of the day which drives another such monitor perfectly well the picture now sort of moves around but at a rate of 1-2Hz. Turning up the brightness shows a wavy left/right edge implying a few hundred Hz effect. But the effect is also vertical.

Note I say moves. It does _not_ contract or expand which would indicate insufficient headroom after the rectifier before the 15v regulator. I am happy this isn't happening.

Now I notice the original transformer was a solid metal screened thing with an extra grounding wire.

I have twisted and moved the secondary wires away from the tube and also moved away and screened the transformer temporarily to no effect. Various things to try next include ferrites on primary/secondary wiring, a UPS and maybe some extra caps and also jumpering across from a working monitor.

The environment is terrible being a retro museum on an industrial estate with a railway depot over the road.

Though generally experienced this baffles me somewhat so I'd appreciate your thoughts.

Is the problem likely to be common mode noise ?

Ideally I'd like to keep this transformer and not have to seek out a screened one.

Thanks.

[Pimpom]

On 3/12/2018 7:41 PM, jms...@gmail.com wrote:
> I have partially repaired an IBM mono monitor whose mains transformer had failed.
>
> http://minuszerodegrees.net/oa/OA%20-%20IBM%20Monochrome%20Display%20(5151).pdf
>
> Driven from a genuine PC of the day which drives another such monitor perfectly well the picture now sort of moves around but at a rate of 1-2Hz. Turning up the brightness shows a wavy left/right edge implying a few hundred Hz effect. But the effect is also vertical.
>
> Note I say moves. It does _not_ contract or expand which would indicate insufficient headroom after the rectifier before the 15v regulator. I am happy this isn't happening.

Did you check to see if the regulator does have sufficient
headroom? Maybe the replacement transformer has a lower output
voltage or lower current capacity.

>
> Now I notice the original transformer was a solid metal screened thing with an extra grounding wire.
>
> I have twisted and moved the secondary wires away from the tube and also moved away and screened the transformer temporarily to no effect. Various things to try next include ferrites on primary/secondary wiring, a UPS and maybe some extra caps and also jumpering across from a working monitor.
>
> The environment is terrible being a retro museum on an industrial estate with a railway depot over the road.
>
> Though generally experienced this baffles me somewhat so I'd appreciate your thoughts.
>
> Is the problem likely to be common mode noise ?
>
> Ideally I'd like to keep this transformer and not have to seek out a screened one.
>

My first suspect after the power transformer would be a filter
capacitor that had become dry or leaky.

Considering that signal levels in a monitor are much higher than
in sensitive analog systems like audio amplifiers, I doubt that
the symptoms are caused by stray fields from an unshielded
transformer.

[tabb...@gmail.com]

Ferret beads aren't going to do anything at the frequency of interest.
Have you checked all the psu rails for hum? Tried some grounded steel between trafo & tube? Put grounded steel over the whole monitor to see if it's an external field?


NT

[Jon]

I did try doubling up the 4700uF with another which yes I checked the capacitance of first.

Also what are R602 and D605 for ?

Jon

[Jon]

The 4700uF cap sits at 28V which given a measured consumption of 1.4A is fine giving a ripple of the order of a hundred millivolts. And as I said doubling it makes no difference.

[Fox's Mercantile]

There's a reason they put a shielded transformer in there
initially.


--
"I am a river to my people."
Jeff-1.0
WA6FWi
http:foxsmercantile.com

[tabb...@gmail.com]

On Monday, 12 March 2018 16:30:21 UTC, Jon wrote:
> The 4700uF cap sits at 28V which given a measured consumption of 1.4A is fine giving a ripple of the order of a hundred millivolts. And as I said doubling it makes no difference.

Typically there are other smoothing/decoupling caps in there for low current rails.


NT

[Jon Elson]

jms...@gmail.com wrote:

> I have partially repaired an IBM mono monitor whose mains transformer had
> failed.
>
>
http://minuszerodegrees.net/oa/OA%20-%20IBM%20Monochrome%20Display%20(5151).pdf
>
> Driven from a genuine PC of the day which drives another such monitor
> perfectly well the picture now sort of moves around but at a rate of
> 1-2Hz. Turning up the brightness shows a wavy left/right edge implying a
> few hundred Hz effect. But the effect is also vertical.
>
> Note I say moves. It does _not_ contract or expand which would indicate
> insufficient headroom after the rectifier before the 15v regulator. I am
> happy this isn't happening.

Most likely it is magnetic radiation from the transformer. osme
transformers have steel shields, others have copper shorting straps to
reduce the external magnetic field. It REALLY doesn't take a lot of
external field to affect the picture. You might be able to help by changing
the orientation of the transformer, or getting it farter away from the neck
of the CRT. Mu-metal shields might be fashioned to help, but if you bend
the mu-metal, you have to re-anneal it.

Jon

[whit3rd]

On Monday, March 12, 2018 at 7:11:49 AM UTC-7, Jon wrote:
> I have partially repaired an IBM mono monitor whose mains transformer had failed.

> ... the picture now sort of moves around but at a rate of 1-2Hz. Turning up the brightness shows a wavy left/right edge implying a few hundred Hz effect. But the effect is also vertical.

So, the 60 Hz magnetic field leakage from the new transformer is beating against the 62 Hz (or whatever)
vertical sweep. Magnetic field from the transformer reaches through the
glass into the electron beam path..

A conductive belt around the new transformer (can be made from copper tape, and soldered
at the seam) would be a good start. Reorienting the transformer (if possible) might help.
If the original transformer was shielded, now you know why...

[tabb...@gmail.com]

Rewinding the original might be what's needed.


NT

[Phil Allison]

Jon wrote:
------------

> I have partially repaired an IBM mono monitor whose mains transformer
> had failed.
>
> http://minuszerodegrees.net/oa/OA%20-%20IBM%20Monochrome%20Display%20(5151).pdf
>
> Driven from a genuine PC of the day which drives another such monitor
> perfectly well the picture now sort of moves around but at a rate of
> 1-2Hz. Turning up the brightness shows a wavy left/right edge implying
> a few hundred Hz effect. But the effect is also vertical.
>
> Note I say moves. It does _not_ contract or expand which would indicate
> insufficient headroom after the rectifier before the 15v regulator.
> I am happy this isn't happening.
>
> Now I notice the original transformer was a solid metal screened thing with
> an extra grounding wire.
>

** That tranny has a very low external 60Hz field. Besides the steel box, it may have a copper flux strap around it and use more than usual primary turns.

The rest of your post is pure, wishful thinking.

Maybe the whole 15V supply can be placed at some distance from the monitor, in its own box.




..... Phil

[Phil Allison]

Jon wrote:

----------

>
> I did try doubling up the 4700uF with another which yes I checked the
> capacitance of first.
>
> Also what are R602 and D605 for ?
>

** They de-couple the two electros, so the voltage on C604 does not follow the ripple on C603, allowing it to hold a higher average DC value.

Using your 1.4A load figure for the PSU with a 4700uF smoothing cap - the p-p ripple voltage works out at 1.9 volts.

I = C dv/dt ( where dt = 6.5mS )



... Phil

[Jon]

We're 50Hz here as is the IBM MDA frame rate so beating is likely.

[mako...@yahoo.com]

>
> So, the 60 Hz magnetic field leakage from the new transformer is beating against the 62 Hz (or whatever)
> vertical sweep. Magnetic field from the transformer reaches through the
> glass into the electron beam path..
>
> A conductive belt around the new transformer (can be made from copper tape, and soldered
> at the seam) would be a good start. Reorienting the transformer (if possible) might help.
> If the original transformer was shielded, now you know why...

+1
you may have a magnetic field problem, not a circuit problem.
Try moving and re -orienting the power xformer relative to the CRT.
Often you can find a null point.
Otherwise you need mu metal shield or copper strap bucking turn.

m

[jurb...@gmail.com]

This is not a complicated transformer. About 19 volt secondary ot thereabouts, not even center tapped.

Maybe you can find a toroid somewhere which should have less flux leakage.

Or maybe a wall wart.

[Phil Allison]

mako...@yahoo.com wrote:

------------------------

>
>
> you may have a magnetic field problem, not a circuit problem.
> Try moving and re -orienting the power xformer relative to the CRT.
> Often you can find a null point.
> Otherwise you need mu metal shield or copper strap bucking turn.
>

** Mu-metal is effective against *weak* magnetic fields, like the earth's field. It is used to house small signal transformers, cover scope tube necks and the like.

However, a mains transformer has a *strong field* at close range so a copper band right round the winding and core is effective. The band ends must over lap and be soldered.

http://richardschurmann.com.au/Other/Electronic_Burrow/wp-content/uploads/2014/04/Transformer-21.jpg

Don't be tempted to feed the band round the winding inside the core - then it becomes a *shorted turn* and the whole thing will get damn hot.


.... Phil

[Jon]

It is now fine with a toroidal transformer.

It looks like the electrostatic screening of the original was less important than the magnetic screening.

Jon

[jurb...@gmail.com]

So case closed. Literally :-) if ya know whaddImean...

[Jon]

Thanks for that.