MK1210 chip SMPS based supply - can I fire this up?

[dkneyle]

Power supply used in an Onkyo laser disc player, circa 1995?

Came into the Repair Cafe after another tech had a go and gave up. Only evidence I could see of a repair was C22 (2200uF) cap had been changed and D25 was missing. The power plug had been left disconnected so I figured I wasn't going to apply power until I'd checked everything out. I replaced the usual suspects: the SMPS chip ML1210, all the electros except the main charge caps, the 33V zener, replaced the missing high-frequency rectifying diode. I meticulously checked over remaining components with my meter and all seem to look to be correct for their job.

I reinstalled and put a dim-bulb tester in series. Initially the bulb glowed for a moment as the caps charged, then went out, but after about a second it came back on again. I didn't muck around in removing power just in case, but therefore didn't see if I had any life from the player. I've removed most components to see if the situation changes such as the MOSFET Q1, even pulled those charge caps (which appear perfect). There's almost nothing left in the circuit except MK1210. I did pull the bridge rectifier and there is no glowing bulb when that is out of circuit; I've since put a bridge rectifier back in circuit and the bulb glows.

Now the dim-bulb test lamp is not glowing at 100% - I'd say about 75% and there is some flickering in it.

My question is, am I perhaps being gun-shy? Is the dim-bulb tester preventing the SMPS from starting properly? I've had this before with a TV I was repairing. I took the risk with the TV, and without the DBT it started perfectly. The fact the bulb is not at full brilliance suggests it's not a short (and I don't measure any shorts either). [Why am I gun shy? Couple of weeks ago I was repairing a SMPS in an audio unit. DBT in series lit up, but I risked plugging it directly into mains. I was greeted with a very large bang as the encapsulated fuse blew its top off. But that turned out to be a crook bridge rectifier. In this case I've changed out the bridge rectifier just to be sure.]

I've never come across a SMPS circuit quite like this one.

Would you risk plugging straight into the mains?

[momaka]

Re: MK1210 chip SMPS based supply - can I fire this up?

How many Watts is your bulb rated for? 40W? If so, that may be the reason why. Try 100W if you have one... or several 40W bulbs in parallel (3-4x). If not, step up to a 200-300 Watt Halogen tube lamp (I forgot the socket was called for that one.) If bulb still glows with that: STOP and re-check the PSU. But if PSU works with higher power bulbs, then you're good to plug it in directly.

DBT can only be used to test PSUs with no load or very very light load. A rule of thumb I made up and follow is that the load on the PSU cannot exceed 10-15% of the bulb's power rating - i.e. for a 40W bulb, don't expect to pull more than 5-6 Watts from the PSU. And for a 100W bulb, 10-15 Watts would be about the max you can do. Of course, some PSUs may be able to provide more. And it also depends if the PSU is rated for wide-voltage range (100-240V) or standard 120/240V. The former typically uses flyback topology and is more forgiving with low input voltages (which is what happens when you have the DBT in series and apply a load on the PSU.) But standard voltage input SMPS tend to be somewhat more strict about what voltage range they will work in... so with a DBT, if the input AC voltage falls below 10-15%, the SMPS may cut out or act erratic.

[dkneyle]

Re: MK1210 chip SMPS based supply - can I fire this up?

OK, I'd led myself up a dry gulch. After working my way backwards and removing lots of components I realised the substitute bridge rectifier I'd dropped in had blown (or might have already been defective?) Anyway, I dropped the original bridge rectifier back in and started slowly rebuilding the circuit. I got as far as replacing the MOSFET Q1 when I had full illumination of the DBT and an audible noise coming from the circuit. (Very glad now I have that DBT in circuit!) Suspecting a crook MOSFET I replaced it with a 700V 12A version, but no difference. I pulled Q2 and Q3 plus the PUT but no difference; still in overload mode.
Trying to understand how this circuit actually works is doing my head in! Tomorrow I'll test those transistors and substitute/changeover the PUT and see how that goes?

By the way, there is still no load on the output. The diodes on the cold side have all been lifted. There's not much left to check. Any thoughts/ideas greatly appreciated.

[dkneyle]

Re: MK1210 chip SMPS based supply - can I fire this up?

How many Watts is your bulb rated for? 40W? If so, that may be the reason why. Try 100W if you have one... or several 40W bulbs in parallel (3-4x). If not, step up to a 200-300 Watt Halogen tube lamp (I forgot the socket was called for that one.) If bulb still glows with that: STOP and re-check the PSU. But if PSU works with higher power bulbs, then you're good to plug it in directly.

DBT can only be used to test PSUs with no load or very very light load. A rule of thumb I made up and follow is that the load on the PSU cannot exceed 10-15% of the bulb's power rating - i.e. for a 40W bulb, don't expect to pull more than 5-6 Watts from the PSU. And for a 100W bulb, 10-15 Watts would be about the max you can do. Of course, some PSUs may be able to provide more. And it also depends if the PSU is rated for wide-voltage range (100-240V) or standard 120/240V. The former typically uses flyback topology and is more forgiving with low input voltages (which is what happens when you have the DBT in series and apply a load on the PSU.) But standard voltage input SMPS tend to be somewhat more strict about what voltage range they will work in... so with a DBT, if the input AC voltage falls below 10-15%, the SMPS may cut out or act erratic.

I'm running a halogen 70W 230V globe - which is as close as I can get to a 100W incandescent. You can't actually buy incandescent globes in Oz any longer due to environmental initiatives. As per my most recent post, the DBT seems to be doing its job in saving circuit.

I've repaired many SMPS. This one just seems weird. It appears as if the MOSFET is switched hard on. So what would be causing that? The two transistors check out OK on my transistor tester. I've pretty much manually checked everything else on the hot side. The PUT is about the only thing I don't know how to test so I'll substitute with a known good one.

[momaka]

Re: MK1210 chip SMPS based supply - can I fire this up?

I'll give the schem. a closer look tomorrow and let you know what I think. It's almost 4 AM here, so gotta catch some sleep.

[dkneyle]

Re: MK1210 chip SMPS based supply - can I fire this up?

Well, I got the SMPS going (well, sort of...)

So after spending many hours on it, even changed out the PUT with no different result, I got to the point of replacing all the secondary output rectifying diodes. And that seemed to be it!? It would appear (??) that in this design, if the secondary is not providing the necessary feedback via the optocouplers, then the big FET goes short circuit, and the power supply kills itself (not if you have a DBT in line though).

So figuring I had this sorted I put it back into the Laser Disc Player (which also reads CDs). Initially I got no joy reading discs, but suddenly the CD was turning and appears to be working in that mode. However when launching a laser disc (these are actually quite heavy - never handled one before) the moment the motor starts up it causes the SMPS to shat itself again. Full brilliance on the DBT lamp and buzzing sound from the SMPS. Remove power, and we are back in business again. I assume if I had it directly connected to the mains it would have blown the input fuse each time it seem to go into this overload mode. What a strange design of SMPS; if something goes wrong on the secondary the primary goes into short mode and destroys itself!?

I've noticed the fluro display is also non operational. Can't help but wonder if there has been a significant incident with this unit, such as a power surge??

I'm going to start another thread elsewhere on the forum to see if anyone else has any ideas.

[dkneyle]

Re: MK1210 chip SMPS based supply - can I fire this up?

UPDATE: So after going around in circles I questioned whether the DBT in-line was preventing the SMPS from providing enough initial oomph to get the heavy laser disc turning. I figured I only had a 2 amp fuse to replace (hopefully).

Well, it worked! Plugged directly into 230VAC. Laser disc spins up. But the player still has a couple of other issues for me to sort through. No fluro display and this problem where the discs won't initially load; the sled motor makes noises and it doesn't move into position.

I still think this is a very strange design of SMPS. If something goes wrong on the secondary side the primary MOSFET goes fully short and would blow the fuse. Hence when it couldn't draw enough current to start the spindle motor (with DBT in-line) ii detected that as a fault and would have blown the fuse in normal situation.

[momaka]

Re: MK1210 chip SMPS based supply - can I fire this up?

I still think this is a very strange design of SMPS. If something goes wrong on the secondary side the primary MOSFET goes fully short and would blow the fuse.

Yup, that's a classic 2-transistor self-oscillating circuit right there. They are brilliantly simple in that they don't need any specialty ICs. But that simplicity also comes at the cost of having almost no protections. Technically speaking, a 2-transistor self-osc. design can be made to have short-circuit / crowbar protection, and most should also incorporate self-limiting / over-power protection. In practice, however, very few of these circuits have these protections.

Yours does seem to have both crowbar protection (via optocoupler D4 and Q3 circuit) as well as possibly self-limit / over-power protection (via Zener diode D5 and Q2, along with Q3 with resistors R11 and R12.) But just because the components are there doesn't mean that it will work. 2-transistor circuits are a bit tricky to design... and if one thing goes wrong in them, they often blow up.

Which leads me to this...

By the way, there is still no load on the output. The diodes on the cold side have all been lifted.

It would appear (??) that in this design, if the secondary is not providing the necessary feedback via the optocouplers, then the big FET goes short circuit, and the power supply kills itself (not if you have a DBT in line though).

Yeah, you should NEVER do this (remove diodes on the secondary side and try to test the PSU without them.) All of the diodes on the secondary side need to be installed (or at least all the ones that generate voltage rails that provide feedback to the primary side), or else you may blow up the primary.

And this doesn't just go for 2-transistor circuits. You can blow up / damage even PSUs with a PWM IC. Generally speaking, current-mode PWM ICs like UC384x series and similar would be pretty good at catching an open feedback signal. But again, if the circuit wasn't designed properly (and sometimes it isn't - go figure ) you will do (more) damage to the PSU.

What a strange design of SMPS; if something goes wrong on the secondary the primary goes into short mode and destroys itself!?

Well, it shouldn't... but in this case it does, probably because the over-power protection limit is set too high... or not at all?

There are different ways to implement it for these 2-transistor circuits. Here is what a 2-transistor circuit for the 5VSB section of an ATX PSU (HEC Orion PD585d) looks like:
https://www.badcaps.net/forum/showth...85d#post935081

^ On the schematic of that circuit, the 34.7-Ohm resistor (lower-left side of the circuit) gets feedback voltage from the two Source current-sensing resistors (2.7 Ohms and 2.2 Ohms in parallel.) When the current across them exceeds a certain threshold so that they generate about 0.65V or more, this voltage will then go through the 34.7-Ohm resistor and turn-On the small PN2222A transistor, which in turn would pull the Gate of the MOSFET (2N60) low and turn the MOSFET Off. The MOSFET won't stay off however, because of the 1-MOhm pull-up resistor on its Gate. Therefore, in the event of open feedback or short-circuit on the secondary side, this 2-transistor circuit will rely solely on the Gate current-sense resistors and the 34.7-Ohm resistor for limiting power.

Now on your circuit, this appears to be done through resistors R11, R12, and Q3. Increasing the resistance of R12 will lower the over-power limit... but may also have other implications in the circuit. So overall, it's better not to experiment with it, unless you plan to do a full set of extensive tests on the PSU to see how it because at various loads and scenarios.

Oh, and as far as the MK1210 chip - that's just an input voltage sensing IC. It's used to automatically switch the operation of the PSU when it is used in countries with 110/115/120V AC. When it detects a 120V system, it will simply connect the "middle point" of the two big input capacitors to Neutral, essentially creating a voltage-doubler circuit. But when the IC is used in countries with 220/230/240V, it breaks that connection and imply runs the AC through the bridge rectifier and then stores it in the two big series caps.
IMHO, that's a bit of a useless feature, since I don't imagine you would bring this player to another country with 120V AC.

I'm running a halogen 70W 230V globe - which is as close as I can get to a 100W incandescent. You can't actually buy incandescent globes in Oz any longer due to environmental initiatives. As per my most recent post, the DBT seems to be doing its job in saving circuit.

Yes, DBT will work with any incandescent / halogen lamp. But again, read my reply in post #2. You can only draw a certain amount of power through the DBT, based on the power rating of the bulb you are using. Generally, it's about 10-15% of bulb's power rating. So for a 72W halogen lamp, 7-10W would be the max load you can expect to pull with an average PSU. Depending on the PSU's design and efficiency, though, that number may be substantially lower.

So that's probably one reason your DVD drive would make the PSU shut down when it tried to spin up the disk - it was asking for too much power, making the voltage on the primary sag and drop out of the range in which the PSU can work with.

[dkneyle]

Re: MK1210 chip SMPS based supply - can I fire this up?

Thanks Momaka for such a comprehensive explanation of how this particular circuit works. Very comprehensive explanation for what I was observing. Answered so many questions, like what that MK1210 was doing there?
On your advice I'll ALWAYS make sure I have everything on the secondary side of a SMPS operational before attempting to fire it up.
Finally spoke to the (actual) owner to get an idea of what had actually occurred. They suffered a "brown out" event and it took out various other HiFi appliances at the same time. I'm assuming when the mains voltage came back there was a surge down the line.
On this SMPS I believe the only component that had failed was D23 (or it could have been D24). Part number described it as a 60V 2A Fast Switching diode. The original repair shop appeared to know nothing about SMPSs. Yes, the faulty diode had been identified, but in the bag of parts left behind inside the unit, I found a new pack of regular rectifying diodes (IN4004 or something similar) and I assume the tech working on it had discovered they didn't work, and he gave up in frustration. The unit was returned to the customer with the comment "it's too old and we can't get the parts". I'm only a hobbyist running the Repair Cafe as a volunteer service to save electronics from entering landfill. But pretty sad to think this is how "professional service centre" staff are with their level of training and knowledge of electronics repair.
Sadly I think the fluro display is kaput. It picks up the AC to drive it from the SMPS transformer. I'm assuming the overvoltage surge translated to an overvoltage on the AC line for the display and took it out. Fairly unique part - doubt I'll be able to get another one.

[momaka]

Re: MK1210 chip SMPS based supply - can I fire this up?

You're welcome!

Finally spoke to the (actual) owner to get an idea of what had actually occurred. They suffered a "brown out" event and it took out various other HiFi appliances at the same time. I'm assuming when the mains voltage came back there was a surge down the line.

Yes, that could indeed be what happened.
Or it could be that the extensive brown-out just blew the 2-transistor circuit. These 2-transistor self-oscillating circuits may not always deal well with low-line AC voltage. Another possibility is that the MK1210 chip got confused from the brown-out and tried switching to 110/120V AC mode for a bit, possibly causing an over-voltage on the primary side, which eventually made the diode on the secondary go bad? Well, hard to say exactly what happened... but in any case, glad to see you got to the bottom of it.

The original repair shop appeared to know nothing about SMPSs. Yes, the faulty diode had been identified, but in the bag of parts left behind inside the unit, I found a new pack of regular rectifying diodes (IN4004 or something similar) and I assume the tech working on it had discovered they didn't work, and he gave up in frustration. The unit was returned to the customer with the comment "it's too old and we can't get the parts". I'm only a hobbyist running the Repair Cafe as a volunteer service to save electronics from entering landfill. But pretty sad to think this is how "professional service centre" staff are with their level of training and knowledge of electronics repair.

I'm not surprised. A lot of repair shops exist simply for the sake of being a business and nothing more. They don't really make a ton of money, though. So most of the time, they'd rather go after the easy money / easy fixes and let someone else deal with the "hard" stuff... or simply just tell the customer the device cannot be repaired and charge the "diagnostic fee" (if any... though usually there is.) In fact, the only way some shops manage to stay afloat is through the diagnostic fees. If they happen to fix anything, that's just bonus. That said, I'm not trying to demeanor repair shops, but only explain how some of them operate, since they often have to pay rent, employees, and etc... and spending a lot of time on a cheap DVD player or other device just won't pay the bills at the end of the day. The only shops that tend to do that are ones where it's a 1-man business, typically running from home or some really really cheap place.

Sadly I think the fluro display is kaput. It picks up the AC to drive it from the SMPS transformer. I'm assuming the overvoltage surge translated to an overvoltage on the AC line for the display and took it out. Fairly unique part - doubt I'll be able to get another one.

I don't have much experience with fluorescent displays, but from what I remember reading, they do tend to be fairly tough. So perhaps check all the lines and voltages going to it. Maybe something is missing?... or at least I'd be surprised if it's totally dead.

[dkneyle]

Re: MK1210 chip SMPS based supply - can I fire this up?

I've managed to buy a complete working front panel for this laserplayer. Actually, it is from the clone of the Onkyo, a Pioneer CLD-D504. I'm hoping that by getting the whole front panel with associated electronics attached I'll be able to get the display going with a simple changeover.
What you say about repair shops is all very true. Here in Australia they are almost non-existent because there is no real way to make money.
For your interest, laserplayers never were a technology introduced into Australia. I had to Google the reason: they were manufactured for the NTSC market, and Oz is (was) PAL. About the mid-90s many TVs being sold were multi-standard, so a few laserplayers were sold in Oz, but very limited numbers and by this time DVDs had appeared, so that was the end of that. Just a few 'enthusiasts' keeping them going in their HiFi collections now. That is why customer is prepared to pay well over A$200 in parts to get this back to full operational condition.

[momaka]

Re: MK1210 chip SMPS based supply - can I fire this up?

For your interest, laserplayers never were a technology introduced into Australia. I had to Google the reason: they were manufactured for the NTSC market, and Oz is (was) PAL. About the mid-90s many TVs being sold were multi-standard, so a few laserplayers were sold in Oz, but very limited numbers and by this time DVDs had appeared, so that was the end of that. Just a few 'enthusiasts' keeping them going in their HiFi collections now. That is why customer is prepared to pay well over A$200 in parts to get this back to full operational condition.

Thanks for sharing that!
I didn't grow up in the US in the 90's, so I didn't even know what a "laserplayer" is. Very interesting. It's like the predecessor of DVDs but with CDs. That's kind of neat, actually. For me, the 90's (and even some part of the early 2000's) was mostly just VHS. The DVD "era" didn't seem to last very long once internet speeds picked up where I lived. Though Netflix and RedBox probably helped it a good deal.

[dkneyle]

Re: MK1210 chip SMPS based supply - can I fire this up?

Well, I should have listened to Momaka's advice a bit more closely about VFDs.
A VERY expensive lesson learned, after importing a replacement front panel at extraordinary shipping cost from USA and currency conversion. There was no change with the replacement panel so I had to go back to analysing what could be going on. I knew next to nothing about VFDs and assumed they were driven by high voltage AC. WRONG. It's only about 10volts to the filaments, and even though it was high frequency AC in this case I was able to measure enough to realise it was probably OK. A bit more self-education on VFD via YouTube and I realised there is another opposing very important voltage which runs at about 30volts typically. In my case it is the -29 volt supply. Why I hadn't checked this first up I don't know? Difficult to measure at the power supply itself, but I could have measured it at the front panel PCB without too much trouble. I guess I'd assumed once I'd got the SMPS operational then all would be good with its various outputs?
Turned out to be R29, 68 ohm resistor had gone open circuit. (Although it was actually fitted with a 82 ohm resistor). Why had this resistor failed when the power surge occurred and took out one of the secondary rectifying diode as well in a separated circuit? The mysteries of electronics I guess?
Momaka: You might be able to answer this. The -29 volt rail uses a physically small diode, type AG01Z-V0, described as an 'Ultra Fast' diode [200v, 0.7A]. When I first thought I was perhaps going to have to replace that diode I was wondering what substitute I could use? Can you see any reason why this would be a 'special' diode and why a general fast switching rectifying diode would not substitute?

[momaka]

Re: MK1210 chip SMPS based supply - can I fire this up?

Momaka: You might be able to answer this. The -29 volt rail uses a physically small diode, type AG01Z-V0, described as an 'Ultra Fast' diode [200v, 0.7A]. When I first thought I was perhaps going to have to replace that diode I was wondering what substitute I could use? Can you see any reason why this would be a 'special' diode and why a general fast switching rectifying diode would not substitute?

Well, I have very limited knowledge of VFD circuits... but from a power supply perspective, I don't see anything special about the AG01Z diode. Actually, it looks pretty lousy with a 1.8V V_f rating at rated current... though in this case, it probably doesn't matter too much due to the series 68-Ohm (or rather 82-Ohm) resistor after it. Therefore, I think as long as you match the reverse/DC blocking voltage, current, and reverse recovery time ratings, all should be good. Actually, you don't have to match the recovery time - you can use a diode that has lower (better) recovery time too. For voltage, use the same rating or higher. Same with current. And of course, it's also generally a good practice to use a diode with lower V_f for better efficiency (which should be pretty easy here, as most FR diodes only have 1.3V V_f)... though again, that probably doesn't matter too much due to R25's (or is it R29? I can't quite tell from the schematic) series resistance.

That said, since this resistor was burned...
- check C32, the small 35V 100 uF cap. Probably even a good idea to up-size that to a 50V cap. Moreover, it looks like the -29V rail is not monitored by the PSU, so that's even more of a reason to use a higher voltage cap. When the PSU is loaded down (or experiences a surge again - hopefully not :\ ), no telling how high that rail can go.
- check Zener diode D29 (8.2V as far as I can tell from schematic)... and any other semiconductor downstream not found on the schematic that could be causing a short-circuit.

*EDIT*
I forgot to add... for -reverse- current rating of the diode, smaller is better. But again, AG01Z is kinda lousy, so it shouldn't be too hard to find a substitute.

Looking through what I have in my junk bin in no particular order that would match or exceed this diode's ratings:
BYT11-600
BYV28-200
HER202
MUR-420

Probably will work OK even with something like a common FR10x series (FR103 or higher).

A VERY expensive lesson learned, after importing a replacement front panel at extraordinary shipping cost from USA and currency conversion. There was no change with the replacement panel so I had to go back to analysing what could be going on.

Yeah, from the little I know, VFD's are pretty tough. Usual way they die is either from running too bright all the time (thus burning out their phosphors) or physical damage like cracked glass, broken/corroded pins that have compromised the vacuum inside, and etc... or perhaps if grossly over-driven with voltage.
Well, at least now you have a spare ... or could try to sell it again.

[dkneyle]

Re: MK1210 chip SMPS based supply - can I fire this up?

Yep, all other components downstream were checked. As a matter of course I changed out all the electro caps on the secondary and primary side (except main charge caps) to ensure longevity and upped the voltage where that meant the cap could still physically fit.
I sometimes wonder about designers of circuits? Do they just go looking through catalogues and find some exotic component and decide "I'll give that a go". You confirmed what I thought: most fast recovery diodes should do the job. This designer has used three different types of FR diode in this circuit.
Anyway, so glad to have this going again. I might have VFD displays come through again, and I'll know what to do next time.
Thanks so much for your advice. Perhaps it will help others who find themselves on this journey too.

[dkneyle]

Re: MK1210 chip SMPS based supply - can I fire this up?

I’ve continued to watch YouTube videos where VFDs have failed in different types of HiFi gear. They always seem to use a very similar circuit, and referring to the VFD Manufacturer recommended circuit it appears they have all adopted it.

There is an electrolytic cap connected to ground after the resistor in question. If it develops poor ESR then excess current flows through the resistor and it fails. Nearly always the solution in every case I’ve watched is this resistor with a value of about 50 ohms.

Hopefully this might help others trying to solve this same issue.

[sam_sam_sam]

Re: MK1210 chip SMPS based supply - can I fire this up?

I’ve continued to watch YouTube videos where VFDs have failed in different types of HiFi gear. They always seem to use a very similar circuit, and referring to the VFD Manufacturer recommended circuit it appears they have all adopted it.

******removed********

Hopefully this might help others trying to solve this same issue.

This is probably because some engineers are sometimes very lazy and want to take the easy route even if the design is poor to begin with I am not surprised by this at all

[dkneyle]

Re: MK1210 chip SMPS based supply - can I fire this up?

The Onkyo laser player bounced. Loss of audio after about 15 minutes. I think I’ve repaired that, and I’ll detail it here, because it might help others fixing these with similar issue.

It was obviously a heat sensitive issue. When I ran without the cover, I couldn’t get it to fail. Using a hairdryer as a heat source applied to various sections I was able to track it back to the power supply. Using freezer spray seemed to indicate it was in the vicinity of the 4 motor drive power transistors, although freezer spray suggested it wasn’t actually those parts. Reference to the circuit diagram and the physical location on the board suggested it was +-12 volt supply failing. This provides both the audio drive circuit and the spindle motor drive. In fact, if I kept running the player after the audio failed I eventually got a motor drive error on the display. So I was pretty sure I was losing the 12 volt supply.

Now this laser player originally came in with a failed SMPS due to a power surge. One of the Schottky diodes in the 5V rail had failed. I suspect the surge had damaged other rectifying diodes but not destroyed them. Under sustained load and when heating up they started to fail. To be sure I changed out all those rectifying Schottky diodes, and just to be sure (since I was already working on it) I replaced the four power transistors, and four snubber diodes associated with that part of the circuit.

I think the lesson learned is if a SMPS has suffered a power surge and you find at least one rectifier diode has failed, then replace them all just to be sure.

By the way, anyone looking to replace those motor drive transistors might struggle to find a substitute. But there's nothing very special about them other than you want isolated tabs. I used old faithful MJF2955G and MJF3055G (from Mouser) and they work just fine.

[dkneyle]

Re: MK1210 chip SMPS based supply - can I fire this up?

Well, disregard most of what I said in Post #18. After quite a few hours of heat soaking the problem returned. Some careful application of heat from a hairdryer and cooling with freezing spray would seem to indicate R33 (which coincidentally in 33 ohms) is the culprit. It actually reads as 27.5 ohms, which is outside the expected 5% tolerance. When I can get my hands on a proper high wattage resistor I'll change it out and confirm if it the source of the problem. Interestingly, it is connected to D23, which was the original rectifying diode that failed.

I'm not quite sure of the job of R33? Any hints are appreciated. Momaka?