• Please review our updated Terms and Rules here

Monitor vertical collapse??

O

OldDustyandBroke

New Member
Joined
May 16, 2026
Messages
6
Hi Forum Members!
First time here…
I’m working on a PET 4016 - it was working, for about an hour - just long enough for me to clean up the keyboard so all the keys work - and then the video collapsed. :-(
Picture attached - does anyone recognize this failure mode? Vertical adjust is responsive, brightness is responsive.
I’m suspecting a cap given the way it stopped working, but I’m sure it could be any number of things. No visual component damage on monitor board or main board, didn’t spot any dry solder.
Thanks!IMG_6369.jpeg
 
OldDustyandBroke said:
Hi Forum Members!
First time here…
I’m working on a PET 4016 - it was working, for about an hour - just long enough for me to clean up the keyboard so all the keys work - and then the video collapsed. :-(
Picture attached - does anyone recognize this failure mode? Vertical adjust is responsive, brightness is responsive.
I’m suspecting a cap given the way it stopped working, but I’m sure it could be any number of things. No visual component damage on monitor board or main board, didn’t spot any dry solder.
Thanks!View attachment 1322047
Click to expand...
Found that the 12V regulator on the video board is only putting out about 6.5V…. Either excessively loaded or gone bad. Will investigate more….
 
This is a very common fault.

The more likely scenario is that it may be overheating due to the deterioration of the heatsink arrangement. Or it is bad.

Dave
 
I ran out of time today, but most likely it’s a bad regulator (now) due to thermal compound that has thoroughly dried out. These devices don’t usually have much of a soft output clamp, although (perhaps) theoretically possible to have a load that pulls the output down that far but doesn’t trigger the output to turn off, it’s not very likely.
The time it worked fits with enough time for the regulator to overhead and damage itself.
Input is approximately 20V, so that’s good.

Thanks for the feedback - to my mind the PET is the quintessential image of early computing, absolutely beautiful design, the Jaguar e-type of computers! I’m excited to get this baby running like it should.
 
These regulators generally just shut down when they overheat.

Most of the time they don't even have heatsink compound, and are just pop riveted to the heatsink. This thermal interface generally fails.

Dave
 
daver2 said:
These regulators generally just shut down when they overheat.

Most of the time they don't even have heatsink compound, and are just pop riveted to the heatsink. This thermal interface generally fails.

Dave
Click to expand...
Thanks - I ordered new one, and we’ll see if it comes back to life when replaced.

Whilst I’m here…. I’ve been doing some research this evening. This PET is a 4016 with a 9-inch monitor. It seems like it might be pretty rare? I’m NOT interested in selling it, I’ve been looking for an affordable 9-inch PET for a long time because it’s fond memories of my childhood when my dad would bring one home from work during the Christmas company shutdown. Probably that was a 2000 series, I only have fuzzy memories - I think it had an accompanying floppy drive. But anyway, now I have a 4016-N with the 9-inch monitor and I’m just wondering how rare it really is?
 
Did you order a voltage regulator with a thick metal tab (old ones) or a cheap and nasty modern one with a thin tab?

Did you order some heatsink compound as well?

The better the heatsink arrangement the better the projected life of the part.

A PET (of any type) is not really rare. They come up on eBay quite regularly. Many of them are stored in lofts or basements and we get quite a few people on VCFED digging them out of storage and trying to get them working again.

We succeed in the large majority of cases. Some PETs fight back though, but we generally win in the end!

Dave
 
Cheap and nasty!
I’ll add some thermal paste.
But…. She’s up and running again.
The best thing to do would be to replace all the linear regulators with form factor correct drop-in switching regulators, maybe I’ll do that once everything is working well.

The display has a fairly pronounced trapezoid – narrower at the top than at the bottom. and a little bit of wavering on the edges, any ideas there?
 
I would leave the linear regulators alone (personally).

Can you post a short video of the image on your monitor?

The usual causes of a 'wobbly' screen are:

1. Smoothing capacitor after the monitor voltage regulator requires changing.

2. A partial fault in the bridge rectifier after the monitor voltage regulator.

3. Magnetic interference from the PETs transformer.

1 and 2 can be checked with an oscilloscope monitoring the monitor DC power supply rail.

3 can be checked by lifting the lid of the PET, or loosening the monitor and adjusting the position of the monitor (in relation to the PETs transformer) to see if that has an effect.

I would need to see an image of the screen to suggest what to do regarding the image geometry.

Dave
 
I kind of call the 9" 40xx PETs "fake 40xx" or "badge engineered 40xx", as they are essentially a 30xx/2001N but with newer ROMs.

My experience is that the 78xx TO220 style regulators tend to cycle off-on-off-on if they overheat. I.E. a bit of hysteresis built in to the overheat protection part. Maybe it's different between different brands/whatnot?
 
We have been over this on other threads. One factor is that the Tab on the regulator package was distorted by the pop rivet they used, this elevated the flat suface around the hole edges on the metal tab because of expansion and pressure on the inside of the hole in the tab. It worked ok from new, but as the thermal paste filling the gap dried out, the regulator goes into thermal shutdown mode. It you put a temperature sensor on the regulator's tab you would have found it was running very much hotter than the heat sink it was riveted to nearby

When it is replaced, it is better to use a screw/nut spring washer arrangement and a good thermal paste like MX-4.

They pushed the regulator hard for dissipation in this application, that is why it has a very large heat radiator.

I agree with Dave, I wouldn't dream of replacing an analog regulator with a switching one in this case, the switching one is much more efficient and runs much cooler than the Analog part, which it definitely will, it is that the Analog regulator has much better internal protection against failure modes (you just saw how it behaved when overheated) also it protects against over-voltage at its output just as well.

Unfortunately switching regulators cannot protect downstream electronics nearly as well against an over voltage failure mode. So it depends on how much you value your vintage computer gear and the special parts they contain. I would never fit a switching regulator to any of my vintage VDU's or into many of my vintage computers that once had analog supplies , just to save on thermal dissipation.

There are exceptions though. If you take a look at the switching supply invented by Zenith for the IBM-5155 computer, you will find an amazing arrary of over voltage and over current detector circuitry built into that. But as time went by, many manufacturers dropped that, so failure modes in their switching supplies could fry the devices they were powering.
 
I forgot to comment a thing:
When replacing 78xx/79xx style linear regulators it's a really good idea to check the capacitors near the regulator. If those fail the regulator might oscillate and send out over voltage.
I think it's probably a good idea to look at the data sheet / application examples for the regulator (preferably the specific manufacturers data sheets) and compare with the circuit and possibly add what might seem like missing decoupling capacitors.
I can't remember if these regulators like a "nanofarad" capacitor in addition to electrolytic capacitors both on the input and output side, or if the "nanofarad" capacitor was only necessary on one side, and possibly only if there weren't other "nanofarad" capacitors somewhere else between the regulator and the load, perhaps?
 
Thanks for the advice and insights everyone!

Here is a video of the display now. It’s very short due to file size limits, just 3 seconds, and the waving is slow, but you should be able to see it I hope.
Maybe ‘severe’ trapezoid is an over exaggeration, but it’s there.
I tried opening the ‘lid’ and it didn’t change anything, so I don’t think it’s interference from the main transformer.
I think it’s either the main supply big-can capacitor by the transformer or the monitor board
supply regulator capacitor… I need to get a scope on them and see.
 

Attachments

Yes, the hum is modulatiing the scan width, you need to scope the VDU's 12V supply but also the H. Drive waveform to the VDU from the computer, in case that has the hum modulation on it.
 
You must log in or register to reply here.
Back
Top