Image Map Image Map
Page 1 of 2 12 LastLast
Results 1 to 10 of 20

Thread: CBM 3032 screen interferences

  1. #1

    Default CBM 3032 screen interferences

    I have a 3032 with the same issue discussed here (shaky screen):

    http://www.vcfed.org/forum/showthrea...een-flickering

    I'm sure that display is shaking due to the electromagnetic field caused by the power supply because if I open the case the flickering is less noticeable, and if the monitor is removed from the case it disappears completely and the screen becomes stable!

    I tried to:

    - Replace the electrolytic capacitors on the monitor board (totally useless: this was before discovering the real cause of the shaking screen)
    - Replace the "vertical height" potentiometer
    - "shield" the power supply using a copper tape (see photo)

    trsf.jpg

    but nothing changed.

    The board of my screen is:
    http://www.zimmers.net/anonftp/pub/c...deo-layout.gif

    I checked the waveforms described in the service manual with the oscilloscope, and they look fine.
    I noticed that some PETs have a metal shield over the power supply like this one:
    https://i.ebayimg.com/images/g/C-MAA...7w/s-l1600.jpg

    mine has none: is this by design?

    Also, the case of my monitor is made of plastic. A friend of mine has a 2001N with a metallic case for the display so maybe that helps avoiding interferences?

    Anyone here had the same problem? I don't really know what I should try to fix it (if it's possible to do it).

    thanks

  2. #2

    Default

    Yes, any magnetic shielding around the power transformer, or the CRT, will help prevent the radiated magnetic field from the transformer modulating the position of the CRT beam.

    One interesting thing is that possibly the radiated flux field from some transformers may increase over time, it appears particularly bad with older 60Hz transformers run off 50Hz because the primary magnetization currents increase as the core is pushed further up its B-H curve towards magnetic saturation.

    Having a copper flux band which you appear to have helps. One thing that really helps a lot, if the transformer is bolted to a steel chassis with steel standoffs, to lift it off the chassis on insulated mounts.

    My SOL-20 has this issue too, if I put a small VDU over the area where the transformer is, or near the right rear of the case, the CRT beam is interfered with from the transformer's radiated field.

    One of the more curious examples I saw of this recently, was in a lift in a hotel that had a CRT monitor to display the floor number. As the lift was travelling, the iron girders in the lift shaft were affecting the local magnetic field and the raster was moving around.

    Interesting notion too, because it is often taught that if you were in a closed container travelling at a very high constant speed, and not accelerating, you would never know. However, if the box you were in passed through an intense enough magnetic field, everything in the box that was magnetic could be torn to pieces.

    The axis or angle of the transformer's iron stack can also have a big effect, and if a power transformer is interfering with a CRT, a position of a null can usually be found by rotating it to some specific angle.

  3. #3

    Default

    Thanks Hugo! My machine at 50Hz, so it could be that case.
    The strange thing is that the copper band has no effect! I tried your suggestion and isolated the whole chassis by putting an antistatic foil underneath, but nothing changed as well.

    Maybe there's something that I can try with the CRT tube?

  4. #4
    Join Date
    Jun 2012
    Location
    UK - Worcester
    Posts
    3,851

    Default

    The key phrase here is “magnetic shielding” as opposed to electric shielding.

    Ferrous materials (metal structures) can become magnetised (as a result, for example, of electric arc welding). In this case, you can either shield the CRT or demagnetise the metal structure.

    In the “olden days” CRTs had a mu-metal shield around them to minimise magnetic effects.

    Yes, some transformers also “age badly” in this respect.

    Dave

  5. #5

    Default

    Quote Originally Posted by Elder0010 View Post
    Thanks Hugo! My machine at 50Hz, so it could be that case.
    The strange thing is that the copper band has no effect! I tried your suggestion and isolated the whole chassis by putting an antistatic foil underneath, but nothing changed as well.

    Maybe there's something that I can try with the CRT tube?
    There must be a very high flux field. Interesting the flux band has no effect, in theory the current induced in it should generate a field that opposes the radiating field, that is if it is a dead short all the way around.

    In some vintage American TV sets I have restored the old 60Hz transformers were diabolical running n 50Hz with a lot of radiation to the CRT. In one example, measuring the primary off load current (core magnetization current) is was about 1 amp. With a new Hammond transformer, rated at the same power, but with modern core material and primaries designed for 50Hz use too, it was only 47mA. The radiation from the Hammond transformer was negligible.

    The peak magnetic flux that the transformer core experiences is directly proportional to the applied voltage and inversely proportional to the frequency. So if you lower the line voltage by a factor of 50/60, then you would be back to a similar situation where the transformer was running on 60Hz. I'd suggest run the computer on a Variac, drop the line voltage until you just have a little headroom on the DC voltage feeding the regulators, most regulators (unless LDO types) required a voltage about 2.5V higher at their input than their output voltage. In any case, if you lower it too far, hum bars will suddenly appear in the video signal and you will know you have gone too far. This is at least one way to reduce the magnetic radiation, because the effect of core saturation is non linear and there can be a good effect lowering the line supply voltage , if you can get away with it.

    Mu-metal shields are easy on scope tubes , but due to the size and shape of a yoke on a magnetically deflected CRT, very awkward.

    I could have mentioned, that if it gets to the point where you are forced to replace your power transformer, go for a toroidal core type, they have very low radiated fields.
    Last edited by Hugo Holden; June 7th, 2020 at 04:46 AM.

  6. #6

    Default

    I assume the copper loop is soldered some place. Heated bare copper that is soldered can make a higher resistance connection over time. You can find the solder joint and resolder it to see if that helps. I doubt that is the problem but simple enough to try. When I've seen it before, it was at 150C burnin temperatures.
    Dwight

  7. #7
    Join Date
    Feb 2009
    Location
    Southern California, USA
    Posts
    2,926

    Default

    Quote Originally Posted by Hugo Holden View Post
    One thing that really helps a lot, if the transformer is bolted to a steel chassis with steel standoffs, to lift it off the chassis on insulated mounts.
    Hi Hugo, if our OP tries this, he had better make sure the four long screws that hold the transformer to the chassis are torqued very tight to keep the laminated plates together, or there will be a lot of 50/60 Hz hum. I know this from experience.

    Also, I wonder if the metal braid that attaches the lower chassis ground to the top chassis ground at the back hinge is in place. Would a bad connection here be a problem?

  8. #8
    Join Date
    Mar 2012
    Location
    Augusta, Georgia, USA
    Posts
    764

    Default

    Generally when I see this on arcade monitors, I replace caps and transistors related to the vertical sync, and retouch solder joints all around. This cures it 99% of the time.

  9. #9

    Default

    Quote Originally Posted by rittwage View Post
    Generally when I see this on arcade monitors, I replace caps and transistors related to the vertical sync, and retouch solder joints all around. This cures it 99% of the time.
    If there is any kind of AC ripple in the power supply, it will cause ripple in the raster and/or in the video signal, though this sort of thing, due to electrical causes is quite distinct from the radiated field from the transformer interfering with the CRT beam directly. In this latter case and if there is any confusion in determining the cause of the problem, the CRT and the transformer can simply be spaced physically apart from each other and if it is direct radiated magnetic field from the core, the interference will disappear. If this is the cause, then no amount of repair, changing components or anything done to the electronics will have any effect on the problem at all.

    The OP said:

    I'm sure that display is shaking due to the electromagnetic field caused by the power supply because if I open the case the flickering is less noticeable, and if the monitor is removed from the case it disappears completely and the screen becomes stable!

    So this suggests it is nothing to do with power supply ripple or any problem at all in the electronic components, soldering etc. Though since the flux band doesn't seem to be working (to tell that it would have to be un-soldered or removed) I agree with Dwight's suggestion of checking the soldering, for it to be any use, it needs a low resistance join.


    It could also be worth, as I mentioned, changing the orientation of the transformer, eg, tipping it 90 degrees and altering its mounting position, that sort of thing in conjunction with a reduction in line voltage (if possible) may get it to a level where the interference is tolerable.

  10. #10

    Default

    I randomly discovered that if the transformer is rotated in this way and touching the metal chassis in the upper and left part and the board heatsink on the right, the shaking on X completely disappears. Only a minor shake on Y remains, so I suspect that this could be also related to a grounding issue. If the transformer is not touching in one of the 3 points the shaking on X comes back.
    Position is -somewhat- involved, because if i move away the transformer (even 5 cm down from the original position) the screen starts to shake less.

    The metal braid is correctly connected on the bottom of the case!

    photo_2020-06-08_01-04-08.jpg

    also:

    Quote Originally Posted by Dwight Elvey View Post
    I assume the copper loop is soldered some place. Heated bare copper that is soldered can make a higher resistance connection over time. You can find the solder joint and resolder it to see if that helps. I doubt that is the problem but simple enough to try. When I've seen it before, it was at 150C burnin temperatures.
    Dwight
    Yes, it's soldered on one side but I applied the copper recently, so I don't think that it needs resoldering!
    Last edited by Elder0010; June 7th, 2020 at 03:30 PM.

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •