Philips CM8833-II LOPT failure

[Mitch-W]

I have a Philips CM8833 Mark II which is from an old industrial equipment, that is being used only for about two hours a week now. Monitor began to die after a few minutes with a loud whining noise. After few tries, it got to a point, that it won't start at all. Just makes a loud whining noise.

I inspected the inside and there was just no B+. When removing the LOPT to check it, I found that the monitor starts up fine.

So, I would need a new LOPT. Anyone knows where to get it at a reasonable price?

Also, another question: this equipment uses an RGB/CGA style connection with only GND, R, G, B, H-sync and V-sync used. Anyone had any luck of using it with an VGA LCD monitor? I think that it might just have too low horizontal refresh rate for VGA. We tried it with one of the cheap arcade monitor converter boards and it didn't work.

[nigelr2000]

stick a 60 watt filament bulb across B+ (HT+) and ground to give it a load to drive just to prove it's not a PSU fault before looking at LOPT

[McMurdo]

more likely the PSU can't source enough current for the lop stage. Probably capacitor or dry joint trouble so eliminate that first.

[KeithsTV]

I've had a LOPT failure on one of these monitors several years ago. The symptoms were similar to yours.

Keith

[John123]

Am I right in thinking a bad LOPT will usually take out the output transistor and that in turn will blow nearby fuses/ fusible resistors? I guess if those are fine then most likely it's the PSU, as mentioned.

[cmjones01]

If you're looking for a substitute for a CM8833, you should be able to wire the R, G, B and sync signals to a SCART connector and plug it in to more or less any modern LCD TV set. You may have to select the RGB/AV input manually, or arrange something to assert the function switching and fast blanking signals on the SCART connector. But the horizontal and vertical sync frequencies and polarities are definitely compatible.

You're right, however, that most VGA monitors can't handle the 15kHz horizontal frequency that the CM8833 is designed for.

Chris

[Welsh Anorak]

LOPT failure was very common back in the day on these monitors. No further damage was usual. However sourcing another may well be a problem. Try HR Diemen or Donberg.
Glyn

[jay_oldstuff]

Originally Posted by cmjones01

If you're looking for a substitute for a CM8833, you should be able to wire the R, G, B and sync signals to a SCART connector and plug it in to more or less any modern LCD TV set. You may have to select the RGB/AV input manually, or arrange something to assert the function switching and fast blanking signals on the SCART connector. But the horizontal and vertical sync frequencies and polarities are definitely compatible.

Distant memory but cant the CM8833 accept both analogue RGB like modern scart would expect. But also TTL like CGA? think there was a switch on the rear to select witch. if the eqippment is outputting TTL level CGA I dont think it will work connected to scart of a modern TV. I repared a number of thses monitirs many years ago when scools where getting rid of them. used to sell them like hot cakes to video game enthusiasts. think i had a few with bad lopts but back then youd just nick one from one with a bad CRT, most repairs where for bad caps in the psu tho.

Jay

[Mitch-W]

Capacitors were checked and they are all good. I resoldered all connections on the board, as the solder on these Philips monitors was very bad. This monitor had seen lots of hours of use so the tube has some burn-in. The horizontal output transistor has already been changed and also few capacitors before I got to have it checked.

I spoke to my friend that had TV service shop back in the early 90s and he said that definitely I would need a new LOPT. Even if perhaps it is still good, it would be a nice part to have a spare, as these are getting rare.

I am not that familiar with these TTL and analogue signals, but I took a photo of the back of the monitor immediately after it was delivered. Switches were set as shown on the attached picture:

[John123]

It looks as though the Analogue/ TTL switch is depressed, so it is configured for TTL RGB operation.

If all else fails you could try and source an old Microvitec Cub monitor that was used with old Acorn and BBC Micro computers, as those were configured for TTL RGB input (via internal jumpers). You would have to cobble together a cable with a DIN connector, but shouldn't be too difficult.

[cmjones01]

I wouldn't worry about the TTL thing. TTL-type signals can go straight to a SCART connector's RGB pins with just a resistor (I think 330R or so should do it) in series with each one. People run BBC Micros like this all the time.

Chris

[Argus25]

Originally Posted by John123

Am I right in thinking a bad LOPT will usually take out the output transistor and that in turn will blow nearby fuses/ fusible resistors? I guess if those are fine then most likely it's the PSU, as mentioned.

If there are shorted turns in the line opt transformer, or a heavy load on it from say failed rectifiers or capacitors in tripler circuits or other loads, this normally won't destroy the line output transistor, if it has a reasonable heat sink. The peak collector voltage becomes low.

The main cause of failure in the H output transistor is that it's collector peak voltage rises too high and the junction punches through and shorts out. This can happen if the drive voltage is abnormal and it holds the transistor on for too long (longer than about 30 to 50uS), say more than 60 to 150uS and too much energy is stored in the Lopty / scan coils at the end of scan. THe fly back peaks become very high.

The other thing is a faulty tuning cap on the transformer primary, this limits the peak collector voltage, which is on the transistor's collector circuit normally. The fly back diode in that location , also along with the transistor should be checked for shorts before concluding the Lopty or circuits on its output have failed. But I suspect the OP has checked those already so the transformer looks like the suspicious party.

[mhennessy]

When I had a failed LOPT in one of these - over 20 years ago - it was a short from the top end of the primary winding to the core and hence the ground connection. The net result was that the PSU went into current overload tripping, but with no other damage. I didn't take it apart to work out exactly how that could have happened, but it struck me as "strange but fortunate" at the time. A new LOPT went in, and it went back into service - there wasn't time to worry too much about such things...

If the tube is still good, then it's worth fixing them. If you find a source of LOPTs after all this time, it would be good to know about it.

I've got one of these in the attic with less than 100 hours on it. In fact, I tried to sell it for less than a LOPT would cost if (you could find one today) last year, but no-one was tempted. I decided to keep it, as the kids are enjoying using the old BBC micro.

Any old domestic TV with an RGB scart will replace it - using 330 ohm resistors if the source is TTL, as Chris says. They only did 15.625kHz, so many non-PAL monitors won't lock that low, as you've found. But, there's plenty of modern TVs with scart sockets. It sounds like you might have to mix the H and V sync signals, but that shouldn't be too hard. I did it in reverse many years ago - I took an old CGA/EGA monitor that required separate H&V and derived those from composite. It was a lovely monitor, but I couldn't bear the 15.625kHz whistle it had - that was 20 years ago, so I should've kept it as I'm sure it would be fine today!

[Mitch-W]

I'll research some more about these TTL to Analogue signals, and might possibly come up with an adapter.

The HOT is good, I took it out and tested it on a component tester, and is passes with flying colours. All the surrounding capacitors (even Poly./Ceramic types) were also tested, they are all good.

Currently, I am not home, so I will have a look in Monday. The odd thing was that it produced HT, so I always needed to discharge the CRT when I removed the anode cap, and it produced quite an arc. So then I figured, I should try inserting earthed screwdriver underneath anode cup, and powering the monitor up. Immediately upon poweron, it arced once, then the monitor started to whine and after I turned it off, it arced a few times more. That was also observed on B+ voltages, as they came up for a few hundreds of a second and then collapsed to 0V, but when turned off, they came up again until capacitors discharged.

Will keep you posted.
Thanks

[Maarten]

Why would you insert an earthed (whereto? I hope to the aquadag) screwdriver and then power up the monitor? This is a good way to blow up various semiconductor circuits. It also seems to me that you have a reasonable chance that the LOPT is still working, since the most usual faillure modes won't produce a healthy level of EHT.

Collapsing within a few hundreds of a second sounds like protection mode. I don't have the schematic handy (what is the exact model number? I seem to remember something similar to CM11xxxx/xxx) but I would first look at which modes of protection are included in the power supply. Did you test with a 60W light buld on the +B already? Are all ouput voltages in spec?

[Mitch-W]

I know that this practice is dangerous and it can blow up HOT. I grounded it to the aquadag. Actually, it wasn't making direct contact with the anode, just a few millimetres underneath the cup, so it arced when it got enough EHT. I wanted to test if it will produce EHT only on powerdown or also upon powerup.

I forgot one thing, immediately after I got it to check, there was a slight arcing or ticking noise from the anode cup. As the monitor was quite dirty, I proceeded to clean this area and it helped a lot, however, I was still able to hear slight ticking, possibly from the LOPT itself.

I had an old Macintosh compact 9'' computer that had a bad LOPT, but it still produced EHT, however, it cooked the HOT in a matter of seconds.

The exact model is CM11342/10G. I will proceed with further testing on Monday. All voltages come up if there is LOPT removed, but with it present, they are all at near 0 volts.

[jayceebee]

I've never come across a MKII but repaired several MKIs which the majority were LOPT failure but a word of warning though, some so called "equivalent" transformers were not suitable. We obtained several from CPC I think in the 90s and it was found they would appear to work with normal EHT, A1 volts etc., but would not produce a display.

I found that the so called equivalent had an internal link between the bottom of the EHT overwind and a ground connection of one of the secondary windings. CPC or whoever supplied them took them back and after fitting the genuine but more expensive Philips part normal operation was restored.

I hope you have luck in your search for a replacement.

John.

[mickash]

I had a few of these where the rubber wedges under the scan coils had gone conductive and shorted to the windings of the scan coil damaging them.
This can be seen easily if the coils are withdrawn from the CRT neck and can be repaired if you are very careful.
The affected damaged turns need isolating and insulating from each other. The rubber wedge also needs to be insulated from making contact with the repaired area when replacing the scan coils onto the CRT neck.

[Sideband]

The standard check on these was to disconnect the main HT supply to the line output stage (some had a plug with an HT link to the line output stage on the scan coil plug and you just pull this out) and connect a 60W filament bulb across the HT supply from the power supply. If the lamp glows then the power supply is OK. Faults in the Line output stage were usually confined to the LOPT, the line output transistor or the associated tuning capacitor.

[Maarten]

I found a schematic on elektrotanya.com and had a quick look. The power supply has a crowbar protection circuit, connected to all sorts of circuits. If the power supply itself works correctly with the mentioned light bulb load but when connected activates the protection thyristor, the source is probably the EHT/beam current/rontgen protection (if fitted). In that case the LOPT is suspect as is the resistor on the cold end.