• Please review our updated Terms and Rules here

Mac SE CRT ghosting/Bleeding to the right

ivannudem

Member
Joined
Jan 20, 2021
Messages
17
Location
Canada
Hi, everyone,
I recently bought a vintage Mac SE (with 2 floppy drives) on eBay. But it had a very bad CRT. I mean I could bear the burn-ins, dimmed screen due to less eletrons available due to age, but I encoutered some very bright bleeding/smearning lines to the right of black frames/edges/lines.
(see attached links)

I don't know what caused these, because the failure mode of CRT that I expected was that it will dim due to less available eletrons from the electron gun's filaments, but not this kind of non uniform bright bleedings to the right. I hasn't recap the AB yet, but a quick search shows that recap didn't help:https://68kmla.org/forums/topic/60642-se30-crt-ghosting-bleeding-to-the-right/
also: https://www.reddit.com/r/crtgaming/comments/d0hnmx/seeking_help_with_a_streaking_problem_on_my_pvm/

I would like to know if this is a CRT problem meaning that this CRT is bad (has internal shorts?) or what caused this? I no longer want to invest in this **** thing since I've spent way to much on it.
Is there a cheap way to repair it without buying a tube (which are hard to find now unless I buy a new SE).

Thanks a lot.
 
It got much sever if there's a long black line/edge ,the longer the black line is ,the more bright ghosting it has to the right. I uploaded another img in this post.
 

Attachments

  • photo61345.png
    photo61345.png
    165.5 KB · Views: 1
The streaking can be caused by several different issues. It could be a problem on the analog board like failing caps, bad solder joints or that the tube is being overdriven. Like if you turn the brightness up too high with the wheel on the bottom left face of the SE.

If I were you, I'd recap the analog board and redo every solder joint on the analog board. I've had to fix three SE analog boards in the past year and all had problems with failing solder joints. Sometimes the hairline cracks are so small that you can't see them, which is why it's best to just redo all of the joints. Another thing that can cause issues is the adjustment potentiometers getting oxidized and crusty. To fix this, you'll need to mark the position of the potentiometers with a sharpie and then hose them down with contact cleaner and spin them back and forth several times. You'll usually see oxidized junk come out of them, wipe this off and hose them down again until the contact cleaner stays clean.

If the tube is dark, you can try to bump the voltage to the tube up a tiny bit with the "brightness" potentiometer on the analog board in conjunction with the brightness knob on the face of the machine. Just be VERY conservative with the analog board brightness control because it can damage the tube if turned too high. I wouldn't go more than maybe just the width of a flat blade screwdriver head in either direction, it is very sensitive. Also be sure to use an insulated plastic tool when adjusting these because there's high voltage all over.
 
Thanks GiGaBiTe for your suggestion,
I'm pretty a newbie in vintage computing, and my soldering skills are really very basic (I only soldered the PiDP8 and PiDP11 before). And I'm afraid of the high voltage part of the AB, CRT, Flyback Transformers... But I think a recap of the AB is inevitable...
I watched some YouTube video about how to discharge the CRT, I just bought some alligator clip wires but I'm really afraid of any spark or sound of it if it ever still retains a charge, is that frightening if I discharge it while it still has a charge (in case the bleed resistor doesn't fully discharge it)?

I saw someone posted a more frightening pic about his/her SE/30 with worst CRT bleedings: https://www.vcfed.org/forum/forum/g...intosh-se-30-crt-ghosting-bleeding#post946493
In that case, they concluded the CRT itself has problem. That will be my worst fear, that my CRT/Yoke is bad. If its only the AB, it might still be fixable.
Is is possible to determine if my problem lies in the AB or CRT? Because I also asked in 68KMLA, they said that recap AB will only make it brighter but cannot heal the bleeding/smearing, and told me that the CRT will degrade over years and result in this kind of ghosting, but how could that happen because as far as I can understand, the filament will burn out little by little so the screen gets darker and darker, how could the Yoke/windings or anything causing this kind of bleeding degrade?

Thanks
 
Generally, to produce a good video image requires that the bandwidth of the video amplifier, for a TV image at least is about 20 Hz to 5MHz. Good computer VDU's often have a better range of 20 Hz to 7 or 8 MHz, depending on the size of the CRT. And you can see graphics and text with that bandwidth as sharp as a tack, especially on a monochrome VDU.

What happens if the bandwidth gets limited ?

In the case of the high frequency end being limited, the image becomes smeared and after what should be a sharp edge, there is a blur or streak trailing after it from the left to the right. Primarily loss of HF response results in an equivalent of de-focus and loss of high frequency detail in the image.

But, one interesting this is, what happens if the low frequency response of the video amplifier is corrupted ? This is a less discussed problem. It results in the defect you are seeing with the "smearing" over broad sections of the video signal, after a significant video voltage level transition.

So how does the low frequency phase distortion come about ?

Invariably in most video amplifiers there is R-C coupling between the stages. As the capacitors age and become leaky, in this case it can result in excessive low frequency response causing the broad areas of image smearing.

So short smearing is loss of video HF response, long smearing to call it that is low video frequency phase distortion, typically due to faulty coupling capacitors in the video amplifier circuitry.
 
Thanks a lot for the detailed explanation from Hugo. I think now it's time to recap the analogue board. Hopefully my problem is different from the issues of the SE/30 described in the other thread of this forum (In his case, the CRT or yoke itself is faulty which is not easily repairable)... He had large area of ghostings not only for the long black lines/edges, but also shallow grey area also leaked to the right badly. Since mine only has streakings at the end of black lines, I think Hugo's explanations on hte "significant video voltage level transition" makes sense.

I really only have very basic knowledge about analog circultry because I don't have a background in EE. The RC circuitry which still remains in my mind were the ones I learned in high school physics.. lol. But I have an uncle who used to be a TV technician, I may ask him to give me an introductory "course" on the theory of the CRT

Thanks again for everyone's help!
 
Thanks GiGaBiTe for your suggestion,
I'm pretty a newbie in vintage computing, and my soldering skills are really very basic (I only soldered the PiDP8 and PiDP11 before). And I'm afraid of the high voltage part of the AB, CRT, Flyback Transformers... But I think a recap of the AB is inevitable...
I watched some YouTube video about how to discharge the CRT, I just bought some alligator clip wires but I'm really afraid of any spark or sound of it if it ever still retains a charge, is that frightening if I discharge it while it still has a charge (in case the bleed resistor doesn't fully discharge it)?

High voltage is really only present in the upper rear quadrant of the analog board where the flyback is. The large metal fence around the flyback is ground and everything outside of the fence is going to be low voltage. The only other troublesome area is the deflection circuitry for the yoke where it's high frequency, but low voltage (under 35v.)

You'll always want to discharge the tube after it has had power applied to it. There are bleeder resistors, but never trust them to be in functional condition. To discharge the tube, you'll want a resistor in series with the wire to one of the screws on the implosion ring around the face of the CRT. Using a resistor usually prevents arcs, sparks and loud pops. I'd recommend a 3 or 5W 1 Meg Ohm resistor.

and told me that the CRT will degrade over years and result in this kind of ghosting, but how could that happen because as far as I can understand, the filament will burn out little by little so the screen gets darker and darker, how could the Yoke/windings or anything causing this kind of bleeding degrade?

To diagnose the tube, you'd have to swap it into a known working chassis with a known good analog board and power supply. While image smearing or streaking is one sign of tube failure, it's not always the case. I've seen plenty of worn out CRTs in my day that just got too dark to be useful, or one of the cathode guns got weak and the color rendition got all wonky.

If you decide to recap the analog board, you need to do the power supply as well. The SE machines I recapped all had horribly leaking capacitors in the power supply module that started to damage the PCB and had to be cleaned up. No real surprise though since the PSU is at the bottom of the case with no fan except the one on the back, and it definitely doesn't have enough flow to prevent hot spots.
 
The good news is that there is no CRT defect I know of or scanning yoke defect either, that can cause this problem. It is a problem of the frequency response in the video amplifiers. You would be able to see it too, with a scope looking at the output of the video amplifier feeding the CRT.

There is no need for this fault to have to go under the anode cap of the CRT to discharge the CRT. You only would do that if you were changing out the CRT or LOPT. The charge on the CRT bulb is stored safely there and you won't get a shock from it unless you go under the anode cap. If you do want to go under the anode cap for some reason, and discharge the CRT, use the resistor as recommended. A direct short of the CRT, results in peak discharge currents initially and briefly of over 100 Amps, can even damage pcb devices with high peak currents depending on where the earth clip goes, without a resistor to limit the current. So it pays to use the resistor when its done. The CRT's base pins are completely isolated from this stored charge.
 
There is no need for this fault to have to go under the anode cap of the CRT to discharge the CRT. You only would do that if you were changing out the CRT or LOPT. The charge on the CRT bulb is stored safely there and you won't get a shock from it unless you go under the anode cap. If you do want to go under the anode cap for some reason, and discharge the CRT, use the resistor as recommended. A direct short of the CRT, results in peak discharge currents initially and briefly of over 100 Amps, can even damage pcb devices with high peak currents depending on where the earth clip goes, without a resistor to limit the current.
No offense, I know that you know a lot about doing repairs, but what you wrote here is wrong and plain dangerous.

First, "100 Amps" is completely nonsense. There is high-voltage, but with little to no energy. You are claiming energy of about 1 gigawatt being there! Discharge current is only a few milliamps. It's like an electric fence: high voltage with very low energy.

Second, while a 9" b/w CRT can not kill you anyway, your advice is still very wrong. You always discharge the tube when working on the insides, period. Not doing so is either lazyness or hubris - things that are prone to result in injuries. It's not correct either that you can only get shocked when going under the anode cap. Seriously, I can't understand how someone with your knowledge can even write something like that..?

If I see a person saying: "I'm pretty a newbie in vintage computing, and my soldering skills are really very basic (I only soldered the PiDP8 and PiDP11 before). And I'm afraid of the high voltage part of the AB, CRT, Flyback Transformers...", I would not even try to give him advice how to do the repair. You don't start with repairing a monitor if you have no skills. Find one who can do it for you instead.
 
Last edited:
Second, while a 9" b/w CRT can not kill you anyway, your advice is still very wrong.

This advice is dangerously incorrect. A 9" CRT can MOST DEFINITELY cause death, but you'd have to be incredibly unlucky and/or doing something very dumb for it to happen.

All it takes is about 15-20 mA to cause respiratory paralysis, and there's definitely that available at the anode cap from the flyback and the capacitive nature of a CRT.
 
This advice is dangerously incorrect. A 9" CRT can MOST DEFINITELY cause death, but you'd have to be incredibly unlucky and/or doing something very dumb for it to happen.

All it takes is about 15-20 mA to cause respiratory paralysis, and there's definitely that available at the anode cap from the flyback and the capacitive nature of a CRT.

My last post is still being moderated.

Perhaps there is some confusion. A powered set can definitely cause sufficient electric shock to risk death. What we are talking about here is whether or not, with the set de-powered, there is enough charge stored on the CRT bulb to cause a problem.

When the set is turned off at least after some minutes, the voltage elsewhere dissipates, though it is always worth waiting until the large capacitor in the switchmode psu have discharged fully,The bulb charge is not accessible unless you go under the anode cap, which is not required for general servicing, unless you want to remove the CRT and/or LOPT. So advising people to go under the anode cap, to perform the discharge is not good advice, when generally it does not need to be done.

However, the stored energy, depends on the size of the CRT (its capacitance, and the previously applied EHT voltage) Generally for a 9" CRT this will be under 100milli-joules. So while this will give an unpleasant zap, which is not a continuous current over 15mA, it is extremely unlikely to be harmful. The main risk is dropping a charged CRT while carrying it. As noted generally electric fences run capacitors charged to 10 times or more this amount of energy. Though with a very large color CRT the stored energy is higher.

So I'm not advocating to put one's fingers under the anode cap after switching off the set to see what it feels like, and what I'm saying is, the charge on the bulb is isolated there and won't affect general pcb work & testing unless you deliberately go under the anode cap or remove it, or disconnect the external aquadag (which forms one of the plates of the charged capacitance) or LOPT to come into contact with that charge.

I think what happened here is that somebody who did not understand the Physics & design of a CRT, that the charge on the CRT bulb was totally isolated from the CRT's base, falsely concluded that you could get a shock from the CRT's base pins, from the residual charge on the CRT bulb.Then came up with the notion it was not safe to work on a VDU unless the CRT was discharged. And, as I have pointed out before, the probability of getting a low energy uncomfortable shock from the stored charge on the CRT's bulb increases , if you go under the cap to discharge it, especially for a novice. So it is better simply left alone.
 
Last edited:
No offense, I know that you know a lot about doing repairs, but what you wrote here is wrong and plain dangerous.

First, "100 Amps" is completely nonsense.

If a capacitance, is charged to 10kv and discharged by a conductor R shorting it out, the initial current is 10,000 / R. So even if the internal and external aquadag coating, that form the plates of the capacitance (and the CRT bulb being the dielectric) presented a series resistance of 100 Ohms (its usually less), the initial peak current is 100 Amps. Also, if a spark appears before there is direct physical contact of the shorting conductor, it has a low resistance and doesn't serve to greatly limit the peak current.

In the early days of transistor TV's it was discovered how a discharge of the CRT's charge to an earth point on the pcb could damage transistors and there were recommendations in some manuals not to do it. If a series resistor is used, say 1 Meg Ohm, the initial peak current is limited to 10mA.
 
Note: Could the moderator please delete my duplicate posts listed on my view as posts #10 & #11 that were flagged as spam due to editing, as I answered the remarks again in post #13 & #14.
 
Last edited:
Thanks for everybody's help. I will proceed with extreme caution and discharge as it is advised. I have watched several youtube video and understand where to hook the clip to. I'd better put in series a MΩ resistor to keep the initial current low enough not to damage the ground. Anyway, to recap the Analog board, I'd need to discharge it too, I really want to get this streaking problem fixed because I've spent too much money and my time into it, cleaning the Floppys + cleaning the outer shell took me totally 8-10 hours, don't want to abandon it. But I live in a relatively small city in Canada, where there're not a lot of vintage computing enthusiasts, let alone CRT TV repair shops, so it's difficult to have someone fix it, I'll have to do it myself. hopefully the SE doesn't have SMT capacitors, so recap should be within my skills.

Thanks
 
Thanks for everybody's help. I will proceed with extreme caution and discharge as it is advised. I have watched several youtube video and understand where to hook the clip to. I'd better put in series a MΩ resistor to keep the initial current low enough not to damage the ground. Anyway, to recap the Analog board, I'd need to discharge it too, I really want to get this streaking problem fixed because I've spent too much money and my time into it, cleaning the Floppys + cleaning the outer shell took me totally 8-10 hours, don't want to abandon it. But I live in a relatively small city in Canada, where there're not a lot of vintage computing enthusiasts, let alone CRT TV repair shops, so it's difficult to have someone fix it, I'll have to do it myself. hopefully the SE doesn't have SMT capacitors, so recap should be within my skills.

Thanks

Yes good, use the resistor if you feel compelled to discharge the CRT , for example if you have to remove the anode cap or disconnect the CRT's external aquadag connection from the set's ground.

If you come across SMT capacitors, the easiest way to remove those, without other special tools, or risk to the pcb tracks, is with two soldering irons, one in each hand, melt the solder on each connection at the same time. Adding extra fresh solder often helps, then clean up the pads with solder wick & IPA or similar cleaner.
 
Anyway, to recap the Analog board, I'd need to discharge it too

There is enough parasitic resistance on the analog board to discharge all of the caps on it. Plus there's no high voltage besides what's inside the fence around the flyback transformer, so steering clear of that you will be fine.

You also MUST recap the power supply, it's in its own metal enclosure screwed to the bottom of the analog board. The capacitors in it can get very bad and cause all sorts of headaches.

hopefully the SE doesn't have SMT capacitors, so recap should be within my skills.

There are no SMD capacitors in either the SE or SE FDHD. But you need to be VERY careful when recapping the logic board because the through-hole VIAs that the axial capacitors are soldered to are very delicate. It's super easy to pull a VIA out of the board and then you have a very difficult problem to fix. Axial capacitors are also getting pretty hard to find these days, so you may want to use radial capacitors instead. You can just fold the legs out at 90 degree angles from the capacitor body and another 90 degree angle to go into the board.

Recapping the entire SE (analog board, PSU and logic board) is quite a number of capacitors. Looking at the BOM for the last SE I did, it was about $60 in parts, the most expensive being the PSU's main filter cap, which is an odd size and hard to get. I ended up having to modify the mounting points on the PCB with a drill because the original spade type cap was no longer available. The other hard to find cap is the 3.9uF horizontal deflection capacitor, this is a high frequency rated capacitor and is unobtanium. You'll have to substitute a special high frequency metal film capacitor like this one:

https://www.mouser.com/ProductDetail/667-ECW-F2395JA
 
Hi, Just an update, I recapped the problematic Mac SE's analog board (changed all electrolytic caps bought from digikey), but the problem still persists.

https://imgur.com/a/gJ4QRkP

Recap the analog board doesn't make any changes, except making the max brightness a little bit brighter... But under the max brightness, I did find the horrible phenomenon that even the whole large area to the right of a vertical black line will bleed as gray color to the right, which is the symptom of the unfortunate SE 30 mentioned on this forum and 68kmla (where he found the problem was the CRT)
https://68kmla.org/forums/topic/60642-se30-crt-ghosting-bleeding-to-the-right/

https://imgur.com/a/ROddjRf

I don't know what to do, but recently I picked up another SE FDHD from a friend, the CRT was perfect without problem and after recap of the analog board with the same batch of new capacitors. So i'd assume that it was the problem of the CRT. But I'm just curious, how could a CRT itself has such a failure mode?

I did find that the CRT's neck has a pinkish indicator which indicates that it has some leakage? After the same recap process, this troublesome CRT was dimmer than the FDHD's when both are at max brighness, so does this has anything to do with aging, i.e. less electrons resulting in the bleeding to the right?

I also touched some old solder joints on the AB if I find that they are old and oxidized, but no apparent soder joint cracks, and used IPA to clean the tuning pots (of brightness and focus)

Any suggestions?
Thanks
 
I also find something strange about this troublesome CRT, when powered up (to the screen prompting inserting floppy disk (it doesn't have hard disk)) , the right edge of the screen is more dark/bluish than the left part, it's hard to catch by a camera because my phone seems to automatically adjust the exposure of the img.

https://imgur.com/a/AzPfwNN

But the good one (FDHD) i got seemed to be OK, the contrast of this FDHD's CRT is also different than the bad SE, it seemed more blackish than bluish, where can I tune the tone of the screen?

https://imgur.com/a/Sdr9zn4

I also noticed some other compact macs on sale on ebay which have this kind of bright streaking and black smearing problem, this might have been a common problem of this CRT,
https://imgur.com/a/Rc2GNuE

Apart from recapping the AB, reflowing suspicious solder joints (which have been performed), how can I improve the streaking problem of the SE without buying a new used CRT? Does this means that it has approached its end of life?

Thanks
 
Back
Top