• Please review our updated Terms and Rules here

IBM 8513 Monitor width

Maybe I don't understand something about resistors. I got 10 x 10W (5W was unavailable) 3.9J (5%) resistors, and all of them measure 4.5 Ohm (15%, three times more than 5% allowed). Is it ok? Maybe they are supposed to loose resistance when they warm up or something, or are they just bad, manufacturer decided to put more resistance for the same price?

BTW, adding it in parallel yelding, if I understand correclty 1/(1/4.4 + 1/4.5) = 2.2 Ohm (measured, it is 2.2) and it did not affect the width in any way.

On the plus side, my fiddling probably dusted off something somewhere and now it is at OK-ish width (1/4 '' each side) at 115V B+, so I am good.

Thanks everyone for help!
 
It is possible that the VDU model was arranged to be a little under-scanned. It appears that we have not identified any specific fault.
Certainly, in the case of EGA, with a very short flyback time, to see the first and last pixel on a line, it really does need to be an under-scanned raster.
If you have it in EGA mode from your vintage computer's EGA signal plot a box in BASIC, using the first and last pixel and you will see what I mean. The edge of the box will be very close to the raster edge.

This works (the EGA screen in BASIC is screen 9):

LINE (0,1)-(637,349),15,B
 
Last edited:
Nope, I was to quick, now focus disappeared, previously I was able to adjust a screw on the flyback to perfect focus, now it is blurred to even more blurred. Is the flyback replacement the only cure?
 
Maikudou said:
Nope, I was to quick, now focus disappeared, previously I was able to adjust a screw on the flyback to perfect focus, now it is blurred to even more blurred. Is the flyback replacement the only cure?
Click to expand...
Is the focus control having any effect at all ?

Have you adjusted anything else such as the screen voltage ?
 
It adjusts from bad blurry to totally blurry. I did not touch screen voltage. B+ adjustment does not affect focus. I thought maybe screen voltage might affect it, but did not want to turn it before I know if it makes sense.
 
The resistance of your test probes can easily add 0.5 ohms. What do you read when you short the probes together? Subtract that from the reading for low value resistors like this.

A focus problem is very likely to be caused by a bad flyback transformer. I think the 8513 was known for flyback failures.
 
Probes have less than 0.1 ohms of resistance, it does not add up.
The flyback was ok two days ago. Does it break abruptly and not degrade over time?
I think it was replaced at least once, there is flux around its solder joints that does not show on other joints on the pcb.
 
I don't know why the Flyback transformer always seems to get the blame for most faults. They can fail, but this one would not be typical.

The flyback transformer itself is obviously ok as the H scan is good, as it the EHT.

If the fault resides in the flyback transformer at all, it would be in the resistive divider and pot, inside the body of the flyback, that divides the EHT voltage down to a around 3 to 5kv for the focus electrodes in the CRT.

But that would be jumping to conclusions, if you cannot make some measurements, it requires a probe like the Fluke 80k-40 to measure the focus voltage, without significantly loading it, with & without the CRT's base & socket connected to this high voltage supply.

The resistive divider and the focus preset in the flyback body are very high value resistances, therefore the source resistance of the feed to the focus electrodes is high. It doesn't take much electrical leakage to load it down, including from dusty spark gaps that are often built into the CRT's socket.

Generally in these sets, the collector voltage on the H output transformer peaks to around 1kV. That could be voltage multiplied to make a defacto focus voltage supply for the CRT, in the case that the resistors in the flyback body had failed.
 
Last edited:
If to assume the devider in the flyback works ok, can we suspect that current or voltage coming in the flyback is wrong? Or it would lead to some other issues? Now I see only focus issue.

Two days ago somewhere in the middle of the pot position the focus was perfect and turning it either side would blur it. Now it is least blurry in on extreme position and more blurry in another. So assuming focus grid in crt has not failed (unlikely?) it means it either devides it wrong and yes, resistors are bad inside or what it gets as input is wrong. But again wrong input of the flyback might cause the whole device to malfunction? It seems ok except for focus.

I don’t have appropriate equipment to test such things unfortunately.
 
Maikudou said:
If to assume the devider in the flyback works ok, can we suspect that current or voltage coming in the flyback is wrong? Or it would lead to some other issues? Now I see only focus issue.

Two days ago somewhere in the middle of the pot position the focus was perfect and turning it either side would blur it. Now it is least blurry in on extreme position and more blurry in another. So assuming focus grid in crt has not failed (unlikely?) it means it either devides it wrong and yes, resistors are bad inside or what it gets as input is wrong. But again wrong input of the flyback might cause the whole device to malfunction? It seems ok except for focus.

I don’t have appropriate equipment to test such things unfortunately.
Click to expand...
There is nothing to suggest the flyback is defective (except possibly the resistors in the focus chain inside it). If anything the slightly smaller than usual sized H scan width is associated with a higher EHT voltage. This value is proportional to the peak collector voltage on the horizontal output transistor. This value is affected by the tuning capacitors on the transformer primary C418, C419, C420. These are also normally special quality parts to withstand the peak voltage there, which can be in the order of 1kV. If the capacitance drops, the peak voltage increases.

In working on VDU's is does pay to have a scope and also a x100 probe rated to 2kV, so as to avoid any damage to the scope making measurements on the transformer primary side. Ideally too an EHT probe, to check that the EHT and focus voltage are in speck, but from the appearance of the raster, the EHT will be, or close, because the size of the raster scan is inversely proportional to the square root of the EHT Voltage.

There are other reasons why the focus can appear to go off. Including excessive g1 grid voltage with respect to the cathode. The focus should be checked at low brightness levels and set with about medium brightness.
 
Last edited:
I'm not quick to condemn a flyback. Flybacks are pretty reliable, but anything with high voltages can fail. It's not that unusual for problems to develop shortly after resurrecting something that hasn't been used for many years. Failure modes I have seen are shorted turns, arcing, or failure of the focus divider. The high value resistors in the focus divider can drift While it's true that the transformer itself is still working, since the resistors are molded into the transformer body, the whole thing would need to be replaced.

Most CRTs require the focus voltage to be about 20% of the anode voltage. There's not much outside the flyback that can affect that ratio. Even if the HV is off value, the focus voltage would be proportionally off value, keeping the picture in focus. Leakage in the CRT socket, or the CRT itself could cause a focus problem.
 
Maikudou said:
Maybe I don't understand something about resistors. I got 10 x 10W (5W was unavailable) 3.9J (5%) resistors, and all of them measure 4.5 Ohm (15%, three times more than 5% allowed). Is it ok? Maybe they are supposed to loose resistance when they warm up or something, or are they just bad, manufacturer decided to put more resistance for the same price?

BTW, adding it in parallel yelding, if I understand correclty 1/(1/4.4 + 1/4.5) = 2.2 Ohm (measured, it is 2.2) and it did not affect the width in any way.

On the plus side, my fiddling probably dusted off something somewhere and now it is at OK-ish width (1/4 '' each side) at 115V B+, so I am good.

Thanks everyone for help!
Click to expand...
Click to expand...
What kind of meter do you have? I typically don't trust most multimeters under 10-20 ohms... If you have an ammeter and voltmeter you can trust, you can put a known current through the resistor and measure the voltage, then apply Ohm's law.
 
Now that I think about it, there was a known issue with some 8513s having defective flybacks which caused poor focus.

I think this is the part you need, but double check:

www.tedss.com

TFB-185A

TFB-185A byab transformer flyback 2023000668
www.tedss.com www.tedss.com
 
Yep that’s the one, arriving Friday. At this point I am just curious what exactly is wrong, I won’t be surprised if it does not help. But would be great if it does.
 
Replacing the flyback indeed returned focus back. Still a bit narrow though, but I guess I have to live with it.
Just to show what am I talking about: IBM screenshot is from my other 8513, it has 1/3 inch gap at the sides, the Norton Commander is this one, with flyback just replaced, it has 1/2 inch gap. I guess it's ok-ish?
 

Attachments

  • IMG_2995.jpg
    IMG_2995.jpg
    2.1 MB · Views: 9
  • IMG_2994.jpg
    IMG_2994.jpg
    376.8 KB · Views: 8
Interesting, as Modem 7 noted, there must have been a problem with the focus chain resistors inside the original flyback transformer bodies.

Dividing a voltage down from a very high voltage supply is much trickier than it looks. If you calculate the power dissipated in the resistor chain, you can put in resistors that are x5 power over rated, but they can still have a penchant for failure because on power up when there is a rapid rise in voltage, the resistors in the upper part of the chain have higher currents due to the charging of stray capacitances. It is usually the resistors in the top end of the chain that fail. Ohmite make some special high voltage withstand resistors.
 
Maikudou said:
Just to show what am I talking about: IBM screenshot is from my other 8513, it has 1/3 inch gap at the sides, the Norton Commander is this one, with flyback just replaced, it has 1/2 inch gap. I guess it's ok-ish?
Click to expand...
I really think the 8513 was meant to have large black borders all the way around the active image area. The 8513 that you've just fixed would look better if you reduced the vertical height a bit to preserve a 4:3 aspect ratio.

I've dug up a couple of photos from Byte magazine's July 1987 issue that feature the 8513 showing what it must have looked like fresh from the factory. While it's possible to get an 8513 that's in excellent condition to just about achieve a full bezel-to-bezel active raster (by adjusting the horizontal width coil to its absolute maximum) that was never the intention of the designers... keep in mind that the picture tube has a lot of curvature. There's a lot of distortion near the edges of the screen in 80x25 text mode if you max out the horizontal and vertical sizes.
 

Attachments

  • Screenshot_2024-03-25_09-32-03.png
    Screenshot_2024-03-25_09-32-03.png
    822 KB · Views: 7
  • Screenshot_2024-03-25_09-32-xx.png
    Screenshot_2024-03-25_09-32-xx.png
    986.2 KB · Views: 7
Yeah, it seems it should be this way. I've made all the adjustments, keeping squares square and circles round across the picture. With perfect geometry Its margins are slightly smaller at top to bottom than left and right. But as you mentioned perhaps it's the way they intended.

Thanks! At least I am not afraid of replacing flybacks now.
 
You must log in or register to reply here.
Back
Top