philips cdi 210/40 psu repair?

[2takt]

hi guys,
i got me a philips cdi, IR sensor is not working and the cd drive but i will take on this later on.

when i first plug the cdi in and power it up, the console did boot and i got to the main menu. then after a couple of minutes i switched it off and back on again, now i get only blackscreen( tv did flicker on switch on, so something must happening). i then let the cdi get cold and it worked again, after power cycle black screen again.

i tryed testing the output voltages of the psu but i only got 0.14V. this error must be on my side, philips service manuals are not my friend

i have it appart now the psu has some heat damage where the big resistor is solderd in, there was also a lot of gung on the pcb. on the undersite i notice the solder of the main filter cap did look wierd, like melted in some way. i desoldered the cap and i think it did leak fluid and the cap is AOE branded, the others are all matsushita.

what components should i repair/check?

thanks for your advice daniel

[2takt]

last pic

[TheMG]

The heat mark on the PCB is not something I'd worry about. It's clearly around a resistor which does get pretty hot under normal operation (that's why the resistor leads are long in such a way as to elevate it from the board).

That type of PCB material does change color when exposed to moderate heat over long periods of time. This just means the unit has quite a lot of hours on it.

Check the voltages again. Compare when the unit functions normally to when it doesn't (you say it appears to power up but just doesn't show video?). It's possible the fault has nothing to do with the power supply.

[Haenk]

Check solder joints, there are a couple of visible bad/cold joints on the PSU, maybe even more not-so-visible ones. Use a microsope or just resolder them all (probably faster).
Check the main PCB as well.
PH often has very lousy soldering, I wonder why...

[2takt]

ok good thing to know about the heat mark.

the cdi did show a picture on cold power on, but when i switched it off and immediatly on again i get the black screen. it seems it takes 10minutes cooldown to show the menu again. i did try only 2 times. also the mainboard is in really good shape, the bad solder and heat spot on the psu catched my eye.

i already did resolder everything on the psu i think thats the faster way too

today i did my testing on the mains cap and resistor. resistor is good, mains filter cap has 84 uF when rated at 100 uF.

how long can i leave the psu switched on when no load is connected(for testing)?

edit: the black ntc (resistor?) has about 2mm where black material is broken around the legs

[Haenk]

Did you check the main PCB? Such a thermal failure is almost always connected to bad solder joints. When warm, try putting some pressure on different components and areas on the main PCB, with something non-conductive of course. In fact, you might be able to provoke the error when the system is cold and normally runs.
Also do check the power rails (output), i.e. measure voltage when running and when failing, if the voltage is still OK, the main PCB is faulty. Maybe try this first, but I guess it's some bad soldering nevertheless.

[coromonadalix]

check all solders quality and redo them if there is a doubt, be sure the psu has the main capacitor(s) discharged,  maybe after 5 - 10 minutes unplugged

i would check the electrolytic capacitor near the resistor who heated the pcb,  it can create startup problems, normally you should put 105 degree temps capacitors there

I dont think the psu will blow if its not loaded / connected,  but  theres must be a power on signal somewhere

[2takt]

i have checked the main pcb i have found 2 caps with the same bad solder joints as the mains cap.

the cdi is corently completly apart but i will test the voltages and components when i have it back together.

nope doesnt discharge on its own

i will deffinetly change this cap( and i have contacted someone to give the cdi a complete recap), the only error i have found was a sign of fluid leakage and my tester give me 84uF when rated at 100uF

ok thats good to know, i can do my testing in 1-2mins so i should be save, the power on switch is directlly wired to the psu(the two separate wires in the pictures).
on the mainboard connector there are the 3 red and 3 black wires there are another 2 wires that goes to 2 separate voltage regulators.

thanks for your help!! a virtual beer for you all i will get the cdi back together do my testing and post you the results.

[2takt]

hi,
i have found the issue with my DMM reading to low, somehow sometimes it says A on the display when i have selected volts. i will ignore my measurements till now.

when my dmm worked all voltages are in spec.

i resoldered the 2caps on the main pcb near the psu connector and resoldered the connector itself.

i first test it on the weekend, first try with powercycle i got the blackscreen did some more on off routines, at some point i was able to not get a blackscreen again(befor this point i have noticed the time it took to error and function again did shorten).

today i did another test(found the dmm error) and i was not able to get a blackscreen again. i would assume as an amateur, one of the capacitors got "reformated" and works now again. would do you guys say?

[Haenk]

one of the capacitors got "reformated" and works now again. would do you guys say?

Niemals.

Probably still one or more cold solder joints, pure luck it rattled into having contact.

[2takt]

ok i will let it sit and test again tomorrow and check the mainboard again

[Gyro]

My guess (and it is a guess, but based on other PSU startup / warm-up problems) is the small electrolytic cap right next to the switching FET on the primary side heatsink. It is the reservoir cap for the switcher IC supply. These caps are small and tend to dry out with time and high temperature, causing all sorts of intermittent behaviour.

Replace with a 105'C part if possible.

[shakalnokturn]

I totally second Gyro's advice for that one capacitor.

If you're dead certain the problem is the PSU and you're running short of ideas for components to change (I don't think you've looked far enough yet) I've had at least one "experience" with the Philips transformers of that era that you may want to know of:

Some of them, I think the one on your PSU is of those, used aluminium foil as the windings. The aluminium is spot welded or crimped to the transformer leads in the transformer, this connection can break loose from thermal cycling.

[2takt]

error came back today on 3rd try but i did noticed something. i let it cool down and was near the cdi, when i noticed a sound that pitched(dont ask me if the sound did pitch up or down) after realysing the sound and turn it on and blackscreen was gone. it took about 1-2mins.

voltages where fine with error and without.

ok i will look at the small cap, sounds like that could it be, i mean near the heat of the resistor.

Some of them, I think the one on your PSU is of those, used aluminium foil as the windings. The aluminium is spot welded or crimped to the transformer leads in the transformer, this connection can break loose from thermal cycling.

from your explaination i can gues the break would be immediatly after the pin? so i can desolder the transformer and check for continuity, right?

[Gyro]

I wouldn't desolder the transformer unless you really have to - you could cause the very problem that you're looking for. You should be able to check continuity with it in place. The winding resistances are low at DC, compared to anything around them.

[2takt]

ok i will post the results

[Real-Time]

Just something to think about, the CDI uses a Dallas timekeeper ram which will eventually fail. There's nothing vital in there, but with the battery dead it'll be full of random trash which can prevent the console from booting. While that powersupply doesn't look great, I'd look into the timekeeper ram as a possible problem preventing boot. Black screen is a symptom of battery failure on some models.

There is a mod where you cut open the IC and solder in a battery holder, or alternately you can replace the chip for 20ish USD.

[2takt]

sorry for the delay but iam low on free time.

i soldered a new cap(near the ic)in after a couple of trys i got black screen again.
i did a quick check of the transformator, couldnt find any broken ones but some pins are soldered to the same "rail" on the pcb.

remote is working(i had the old model which does not work with this cdi)
i got the cd drive partialy working following this guide:
https://www.diyaudio.com/forums/digital-source/118902-stripdown-clean-cdm12-4-a.html

real-time:
i did already cut out the battery, but something is strange: 2 times i got the error ram full, which is expected from a low battery. but the battery was already gone. i measured the voltage on the pins from the timekeeper ic and there was some. i cant remember exactly but it was 0,5v or 1.5v.

roadmap for me is now: check the solder on the mainboard again and check the voltage on TK when the error happens and finaly fit the battery to the timekeeper.
thanks

[2takt]

so the other day i checked the voltages on the battery pins of the timekeeper chip. i measured 0,1V-1,5V with no pattern that i could see.

today i did try it again, 0v with no wall power, plug in wall 0,05v, system on 0,2v. then i realised i have only blackscreens(did it for the first time). ok i thought and soldered in a battery holder with 3v. blackscreen was gone and the mashine booted. give it a couple of trys and worked all the time.

shouldnt be the battery completly isolated from the cdi circuit? minus had a connection to minus on the cdi and i cant explain the voltage increase,i found no plus connections to the pins of the timekeeper ic.
could this voltage on the plus side of the pin make the non chargeable coin cell batter leak or gone bad in another way?