advertisement


Philips CD160 not working

Well, a chance conversation with @Mike P has put me in a much better position. We were discussing the tda and Mike asked if I'd done the coupling modifications to the WS, DATA & CLOCK connections between the SAA & TDA chips. @martin clark has also mentioned this in the past too. I had done something along these lines some time ago, but I'd used 100R in the CLOCK line and then had 240R with10p to earth on the WS & DATA lines. Mike told me that the resistors were too low and that with 10p caps, 1k0 resistors are required. He'd said that this was recommended by 'ex Philips engineer Henk ten Pierick over on diyaudio'. Changed the resistors and bingo. Much better clarity and coherence. The upper mids and low treble have more detail and the middle image is more solid. Looks like this 'masks' other improvements in the system. Need to give it a few sessions to see how it settles as there is a very slight 'edge' to it, but experience tells me (some won't agree with this!) that 'newly soldered joints' can have this effect.
:)
 
I also took some measurements of the tda -5v and -15v supplies as I'd recently read (on AOS) that the differential between these is quite important. Apparently if this differential is 9v, the sound is 'sat upon', 9.5v is better and 10v is apparently the sweet spot. I'd been suffering a sort of incoherence at the beginning of listening sessions where the right channel almost sounded 'too forward' and falsely projected, which seemed to settle down after about 1hr. Listening to a different source in my last session confirmed that this wasn't (entirely) due to the amplifier coming on song. Measuring the reg outputs (that i'd previously installed) showed that the -15v was going from about -14.8v at switch on and drifting to about -14.9v whereas the -5v started at about -5v and drifted down to about -4.9v so that the difference between the two was changing from 9.7v to 9.9v. I've now changed the reference in the -5 reg from 3xLEDs to a single 5v6 zener, and from two 7.3v zeners with diode to a single 12v zener and 2 LEDs in the -15 reg in the hopes that the -ve tempco of LEDs counteracts the +ve tempco of the 12v zener whilst the 5v6 should have a relatively neutral tempco. Ran for 2 hrs last night in isolation and the differential changed from 10.0v to 10.08v so looking much better. Will have a listen and re-measure at the weekend.
 
Changed the cap on the SAA7220 from a Nover 220uF to an Oscon SH 100uF 16V, and on the tda1541 -15V supply from same to a Panasonic FR 220uF 25V. I also reconnected the earth at the 0V end of the SAA7220 to the top earth plane, as I'd disconnected this and made it via the cap leg at the opposite end, so now it is connected to the earth plane at both ends. So far, seems much better tonal balance and better focus, so we'll see how we go. :)
 
a.palfreyman said:
Changed the cap on the SAA7220 from a Nover 220uF to an Oscon SH 100uF 16V, and on the tda1541 -15V supply from same to a Panasonic FR 220uF 25V. I also reconnected the earth at the 0V end of the SAA7220 to the top earth plane, as I'd disconnected this and made it via the cap leg at the opposite end, so now it is connected to the earth plane at both ends. So far, seems much better tonal balance and better focus, so we'll see how we go. :)
Click to expand...
Oscons are great for the SAA7220. You can still get the old purple ones at good deals on ebay sometimes.
 
Well my modded CD160 is playing up. After about 2.5hrs of working fine, it still plays, but when you press stop, you get quite a lot of noise which sounds like a combination of rectifier noise with a sort of a whooshing sound and a hiss, left channel only. Pressing pause, the noise changes as it does if you open the loading tray. Ok through digital output into off board dac. I'm wondering if it is related to the mute transistors on the output? Need to investigate...
 
3hrs today with lid off on bench connected to oscilloscope. No unusual noises, so I suspect it could be one of the chips getting too hot with lid on. Will see if I can get a heatsink onto the saa as a moistened finger applied to it is painful after a few seconds; and to the tda and decoder chips which are hot, but not as bad as saa. Oh, and I noticed that the ram is only getting 4.3V. The 4R7 (yellow/purple/gold/gold) in line with the supply is dropping 0.8V which I thought was a bit much as it suggests about 160mA and the chip was barely warm. However, resistor measures 45R so about 18mA. I need to get this above 4.5V as data sheet say 5V +/-10%.
Edit: 33R across the resistor in place gives 18R and 4.75V to the ram chip.
 
Last edited:
Fitted heatsinks to saa, tda and m4804a chips with HY910 paste:
IMG-20220923-105857-467.jpg

I cut a 40mm square heatsink into three pieces.
See how we go...
 
Opamps have cap multipliers on PSUs, saa and tda have their own supplies with -15 from a spare winding on the tx fitted for a flea/clock. Not complex, (nor that well implemented ;)) :D
I have it running now to see if the noise problem returns now I've added heatsinks...
 
7hrs with lid on. No nasties seen in track gaps via a DVM. Checked last off with scope and all looks good. :)
Might have to put a miniature fan in to improve cooling as the extra supplies increase the heat and the position of the regs at the front of the motherboard block the cooling somewhat.
 
Quite the read! Lots of good work you have done there. I read this because I have a Philips CD450. Was my 1st CD player bought new for about $350 (CAD) back in the early 80's. I had put it away about a year ago, got tired of the "err" message. Dug it out the other day, took it apart, checked a few voltages, nothing unusual found, tried it....and it worked on every CD I tried. It's now in my upstairs system and still working fine. I don't get it, maybe needed a long rest? BTW, I have never done anything to it except resolder some of the mainboard many years ago. Kind of good to have it back although I fully expect it to act up again soon.
 
You must log in or register to reply here.


advertisement


Back
Top