Faulty AKF12 monitor

[Tony359]

Hi all,

The AKF12 monitor of my Acorn 3000 has passed away! The picture disappeared left by a faint high pitch noise. I know that it could mean an issue with the FBT or horizontal circuit.

I checked things briefly and I noticed the below:

- What I think is the horizontal transistor is getting warm (not too much but it's on a heatsink so maybe not expecting it to get warm straight away)
- I have 3 voltages coming out of the power supply board. With nothing connected they are 125V, 25V, 16V. With the rest of the electronics connected I have 48V, 10V and 8V.

I'm thinking of a short and the power supply is protecting itself.

I'm looking for schematics. So far I've been unsuccessful. I know this is a Philips but I looked at the CM8533 and I don't think it's a match as there are many more voltages coming out of the power supply section on the schematics. I have only three + ground.

Can someone point me to the correct schematics for this monitor please? And also please feel free to suggest what I should look at!

Thanks!

[atsampson]

Some variants of the Commodore 1084/1084S were based on the same Philips model, sometimes with different input configurations. Gona's site has schematics for many of them - I think I concluded that page 1, page 2 matched the one I've got, and that has a PSU with 125V/26.5V/15.5V outputs.

(I'm impressed that yours still has the front flap!)

[Tony359]

Thank you so much. That looks like the one! The 240V of the Philips CM8500. At least the output voltages are correct and the case is also identical.

Great, so that seems to be confirming there is a short. Also, I see the horizontal transistor - if I see that right - should not show a short between Base and emitter.

Cool, thanks again, that's a very cool link to have. I hope it's not the flyback, finger crossed!

[Tony359]

Update on this thread, it was the flyback.

I checked all I could before getting a replacement and didn't find anything unusual so I ordered a flyback and a horizontal transistor. Once replaced, it worked fine!

Thanks all!

[PJTech]

Update on this thread, it was the flyback.

I checked all I could before getting a replacement and didn't find anything unusual so I ordered a flyback and a horizontal transistor. Once replaced, it worked fine!

Thanks all!

I somehow missed this thread until now.

I'm curious as to whether you replaced the flyback and horizontal transistor at the same time or if you did them separately?
If you haven't already binned your old flyback would you be willing to give it to me (I'll pay postage) just because I've got a tester I've never used and a faulty one would be useful to try it with.

Where did you get a new one? I thought the Philips ones were getting quite rare and expensive now


Cheers
Jonathan

[Tony359]

Hi Jonathan,

We have had the same idea: I'm also would like to build a ring tester and I've saved the flyback to test it. Once I'm done I'd be happy to post it to you.
What tester do you have?

I've replaced them both at the same time, it's my understanding that the transistor might not like the short so it's a good idea to replace that too.

There is a place in Ireland which sells all those things: https://www.donberg.co.uk/

I found the compatible one on a webpage and then ordered from them.

Thanks
Tony

[PJTech]

I did originally consider building a ring tester circuit from TV magazine but didn't have the right components to hand.

In the end I bought a pair of dedicated HR Diemen line output transformer testers - one for 15khz and the other for 32khz monitors but as yet I haven't needed to test any LOPTs as I've been focused on CUBs which use a much simpler transformer plus a tripler.

Failure of the flyback/LOPT is much more rare than the failure of the Horizonal output transistor, which is quite capable of going pop by itself but also could be killed but a dud flyback or by failures of the Horizonal width capacitors.
Personally I'd be more inclined to replace just the HOT which is a relatively cheap part and possibly also the Horizonal capacitor(s) and only suspect the flyback if the HOT fails again or the transformer falls on a tester.

If you're anywhere near me in High Wycombe bucks then I'd be happy to let you test it with my machine. Equally I'd be happy to share test results with you.

[Tony359]

I'm not so far away - Wiltshire!

The transistor tested fine on a component tester, we decided to go with the full replacement.

Let me find a ring tester to build, then I'd be more than happy to send you the LOPT so you can test as well! Remind me if you don't hear from me in a while! Regardless, I'll show the tester on my channel if you fancy following it! (Also, the LOPT replacement will be in a video coming out this week)

Cheers!
Tony

[Tony359]

I made a quick video on this repair Let me know what you think!

https://www.youtube.com/watch?v=3-LrkMFsfw8

[PJTech]

I made a quick video on this repair Let me know what you think!

https://www.youtube.com/watch?v=3-LrkMFsfw8

That's a nice video Tony - I've wondered about doing some about Microvitec CUBs as I'm repairing lots of those.

I think you ought to include a brief comment on safety at the beginning of any CRT videos - yours starts with the anode cap already removed but I've seen one YouTube video where the guy says he thinks it's some kind of earth and pulls it off without discharging first Fortunately for him he didn't get a Darwin award that day....
The topic of safety is covered in all kinds of articles and videos so it's probably enough just to refer viewers to one.

You mention it being a 1084 from Philips but I've only ever known that number as the Commodore model - most people refer to it as a Philips CM8833 although it may still be a variation on that. Using that model number in the video description ought to help make it relevant to more viewers

In terms of the faults - as we've already discussed I'm not convinced the flyback is faulty so am interested to see how it tests out. I'm going to a wedding somewhere near Marlborough (I think) the weekend after next so I might be passing close enough by that I could lend you my HR Diemen tester to use and perhaps even do a video of your tests

The clicking sounds when you first powered up after the repair sounded like one of the spark gaps on the CRT neck - as the screen came up super bright it was probably the G2 voltage being way too high.

I think it would be really useful to also include part numbers, sources and perhaps even the prices you paid for the components involved. I suspect the flyback may have needed an equivalent compatible one and those mains switches are crazy prices for an original - did you find a cheap compatible switch or cough up about £25 for the real deal? This is all stuff I think viewers may be asking.

I'm hoping this is useful feedback so I'm just doing a bit of a brain dump here - I'm not precious about my opinions and am happy to be challenged or proved wrong!

The use of the light bulb current limiter is probably not of any great benefit in this situation - unless you're diagnosing a problem with the switch mode PSU itself then you're unlikely to see it pull excessive mains current as it will self-limit as with the horizontal output failure. Still, it's a really cool device and I've thought about building one but wondered if the diminishing supply of tungsten light bulbs would quickly render it unusable.

I use a mains isolation transformer when working on monitors although this again is probably somewhat overkill unless getting close to the switch mode PSU because everything else is isolated by the PSU itself... I need to do a bit more research on this because the CUBs I work on are still connected to mains ground so this means that voltages are all still referenced to ground and can (and do!) still deliver a shock, which to my mind rather negates the point of the transformer.

It seems to be a bit disputed as to whether it's better/safer to disconnect mains ground from the chassis whilst working on it because then the voltages are only referenced to the floating chassis ground meaning that the potential difference between a high voltage point and real ground should be zero and therefore no shock

Anyway it's really great that you're producing this content and also helping to keep a museum up an running.
Please yell if you need any help with other CRTs

All the best
Jonathan

[Tony359]

Hello there

Brainstorming is always good, thanks for your feedback!

I totally forgot about the safety warning... I'll be more careful next time.

The parts and manuals and chassis number are in the description, in this case it's chassis CM8500 - at least I found that the schematics match.

The limiter is more of a psychological thing BUT it does also include an isolation transformer in it which is important with monitors. The fact that I can remove power straight away by flipping a large switch makes me feel better than having to pull the power cord from the wall I fitted a "bypass" switch as well which bypasses all the bulbs in case I need more power. Light bulbs are traffic light bulbs. Regular ones cannot be sourced anymore.

Re. ground, I understand that neutral and ground are tied at some point so if your live is coming from mains and you touch it, you get shocked. If your live is coming from a transformer and you touch it, you won't get shocked unless you also touch the "neutral" wire. That's my understanding at least!

Spark gaps: I thought something like that when inspecting a monitor schematics. Scary!

Not sure about prices, the museum curator bought the parts! But I think he got the real deal power switch. I believe the total was £68. And the Flyback is a Diemen one. All PNs are in the description which come from the link which also is in the description

I'm still looking for a Ring Tester I can build myself then I'll make a video about it then you can have the Flyback no worries!

Thanks for the feedback, always welcome!

[Tony359]

Further to this old thread, the monitor failed 2 hours after it was switched on

I only now found some time to take a look. Same symptoms, the PSU rails are 50%, telling me there is a short somewhere.
The horizontal transistor is not shorted. I built myself a ring tester and the flyback tests ok!!

However there is a resistor, R476 which is on the HV line, which gets to silly temps, like 160C. Cannot be like that.

I was advised to disconnect the anode lead and keep it under a glass bowl while powering up. Then the PSU came back to life - even though I can see it's still struggling. The idea is that the HV diodes in the flyback are shorted - hence the ring tester cannot see that - and removing the HV lead is alleviating the problem.

Upon further inspection I think the top of the flyback is deformed, chances are the diodes shorted and it massively overheated until the PSU decided it was time to stop.

We are going to ask for a replacement flyback, what do you think?

[PJTech]

That doesn't look too happy does it!

I would suspect you might also be wise to replace the high voltage non-electrolytic capacitors c467, c468 and c470 as they can break down in ways that are hard if not impossible to test and could theoretically have led to excessive current or voltage going in to the flyback.
It could also just have been a duff flyback as some of the aftermarket ones were very cheaply made
That resistor sounds bonkers at that kind of temperature

Good luck!

[Tony359]

Thanks for the advice!

What would be the symptom of one of those capacitors being faulty? Happy to swap them all of course if it might help.

The resistor is still ok - that was another clue that it's not supposed to be so hot, the PCB would be discoloured and the resistor charred! It might be wise to replace it as well though.

I think this time I'll test it a bit more before I consider it fixed

[PJTech]

Thanks for the advice!

What would be the symptom of one of those capacitors being faulty? Happy to swap them all of course if it might help.

The resistor is still ok - that was another clue that it's not supposed to be so hot, the PCB would be discoloured and the resistor charred! It might be wise to replace it as well though.

I think this time I'll test it a bit more before I consider it fixed

I think the theory goes that these really high voltage capacitors which govern the width of the horizontal pulses (and I suspect in this case it's primarily the 2kv one) handle extremely high voltage spikes and this can eventually lead to internal breakdown and change the characteristics at the higher voltages whilst potentially still appearing to be okay when tested using the much lower voltages testers would normally use... This could lead to a variety of failures which may ultimately lead to stress on be Horizontal output transistor or the flyback.
It's very important to replace them with the exact same capacitance - not always easy to find although I've found that Mouser seems to be able to supply a good range.
Of course it's still entirely possible you simply got a bad flyback but these caps aren't terribly expensive and would be the next thing to look at if you're getting repeated failures.

With your ring tester did you also test the old, failed flyback?

I'm curious to find out if my fancy HR Diemen tester can detect failed diodes now. It's supposedly a complete tester so I hope it can.

[Tony359]

With your ring tester did you also test the old, failed flyback?

I'm curious to find out if my fancy HR Diemen tester can detect failed diodes now. It's supposedly a complete tester so I hope it can.

Yes, it does read a low number so the other one was shorted indeed. I also have a working flyback from a different monitor and that also reads good so the ring tests seems to be ok.

For the diodes, I'm told they might have a high voltage drop so might be difficult to test. Also, they might misbehave when there is high voltage running. Not sure.
Thanks again about those caps, I'll see if I can source them!

[PJTech]

.

This is terribly short notice Tony but I'm planning to visit the museum in Swindon tomorrow (Sat 20th) - not sure if that's definitely the place you frequent and where this monitor is from but if you just so happen to be there as well to i could bring along my HR Diemen transformer tester.

Did you ever get any further with your latest replacement?

[ScurvyGeek]

Hello both,

I'm just about to undertake a similar repair. I've ordered a replacement flyback transformer, and was hoping to replace the horizontal transistor plus c467, c468 and c470. The unit is a long time discharged, but I'll discharge the high voltage under the anode cap and PSU capacitors just to be careful. As I'd prefer to carry out the repairs all in one go - do you by any chance have the part numbers for the horizontal transistor and aforementioned high voltage capacitors?

TIA.

[PJTech]

I'm afraid I don't but at least for the horizontal transistor I think you really need to look at the one that's in your monitor as they can vary and it's best to try and replace like for like where possible.
You should be able to safely view it without discharging the tube - it's attached to the big heat sink next to the Flyback and you'll just need to pop off the little metal clip so you can read the number.

The numbers are often abbreviated so if it has one letter followed by three or four numbers then you may need to prefix that with 2S in order to find it online. If it starts with 'BU' then it's probably the full number already.

[Tony359]

This is terribly short notice Tony but I'm planning to visit the museum in Swindon tomorrow (Sat 20th) - not sure if that's definitely the place you frequent and where this monitor is from but if you just so happen to be there as well to i could bring along my HR Diemen transformer tester.

Did you ever get any further with your latest replacement?

Sorry, I missed this message! I need to check the notifications settings

I've just released a "diagnose video" for this monitor so you can take a look at my progress. The owner of the monitor asked for the monitor back before I could finish the repair though so you won't see the outcome - I don't know what happened myself, which is disappointing.

https://www.youtube.com/watch?v=-wmmdAwenZw

[wiggs1979]

I've recently acquired a AKF12 monitor which was working great for a good few hours, but now doesn't seem to supply any power to the tube. There is a faint whine. I was actually away from it as it failed, but when I came back to it, it was warm and had a familiar hot electronics aroma! I've yet to open it up to investigate. It'll be my first proper CRT repair, but I also have an Amstrad CTM644 that needs my attention with a very similar fault, apologies if that's a swear on here!

I'd only just recently replaced the common failed power switch, having scavenged a similar one from an old PC AT power supply.

Anyway, I was curious to know if you ever found out what the issue was with the monitor the second time around? Some great videos by the way.

I may be new on here, but I have been a long time lurker and it's a treasure trove of information, thank you.

[wiggs1979]

I've just done some initial checks and I'm seeing the same behaviour with the PSU. All fine when disconnected, but when connected to the main board the voltages all drop to around half. I don't have a thermal camera and I'm not going to poke my fingers in there to check to see what's getting warm!

[PJTech]

These monitors are very prone to dry joints which might have led to the problem you have now.
The most likely cause of the voltage being pulled down is a problem in the horizontal circuit. If you're lucky it's just a shorted horizontal transistor but it could also be a failed line output transformer.

If you're able to solder then take it out the Horizontal Transistor (usually a BU508 - large one screwed to a metal sink next to the Flyback) and test to see if it's completely short. If it is then replace it and retest. If you can find the right part at a reasonable price it's worth buying 2 in case the first one pops due to another problem (likely to be a faulty transformer if so).

Hope this helps

[wiggs1979]

If you're able to solder then take it out the Horizontal Transistor (usually a BU508 - large one screwed to a metal sink next to the Flyback) and test to see if it's completely short. If it is then replace it and retest. If you can find the right part at a reasonable price it's worth buying 2 in case the first one pops due to another problem (likely to be a faulty transformer if so).

Thanks for the reply. Reading this thread, it does sound like that transistor is a common culprit. I know which end is the hot end of a soldering iron. Joking aside, I've a reasonable amount of experience with repairing and maintaining a pretty extensive collection of retro computers, not CRTs though, so I'll pop that transistor out and check it for shorts. I've also got one on order, a 2SD1577, as I had a few other bits to get for some other repairs and you can never have too many spares when trying to maintain this old kit. Thanks for pointing out the 2S prefix above, that was handy to be sure it was definitely the correct part when Googling.

I've given the board a once over, looking for anything obvious like dry joints, but it all seemed pretty good and very clean, unlike my Amstrad screen which seems to have been home to numerous insects over the years, as well as looking like it's got pretty hot in places.

Thank you again. I'll let you know how I get on.

[wiggs1979]

There doesn't appear to be a short on the horizontal transistor having removed it and checked it over with a multimeter.

I'm guessing there's no easy way to test a flyback without a ring tester or a LCR meter? That's not to say there isn't a short elsewhere in the circuit of course, are there other common faults to look for on these monitors? Will otherwise just work through it methodically.

[PJTech]

If you happen to be anywhere near High Wycombe in Bucks then I've got a proper HR Diemen LOPT tester.
You just might be able to test for a shorted winding with a multimeter but a ring tester or full blown 15khz simulator would be more conclusive.

It's worth checking for shorted rectifier diodes on the voltage lines as well. Whilst the HOT and LOPT both seem to be common faults, they're note the only things that could cause the voltage drop you're seeing