Removal of SAA7220

[tubee]

Hi there

I wanted to get a CD640 in non-os, it works, with Tent XO to Xin of SAA7220(pin11) and WS, CL , DATA from 7210 to 1541

I want to use this player to try and test periferal gear to select the best Dac's, try Tent clock vs Kwak clock, try opamps, discrete or tube I/V's and maybe more.

But when i removed the 7220 (to test if without 7220 give less pollution on PS) the player plays 5 minutes, stops and cannot read disks anymore. After some power-down time it starts again and same story again: no reading of CDs

I have put the XO in pin19 of SAA7210, and disconnected the transistor with collector on pin 19 and emitter to Gnd. The base is connected by a cap to +5V supply and a diode to Gnd.(When this transistor stays connected to pin19, the XO will not be recognized, no CD reading and the CDM is spinning fast)

What connection have i overlooked? Has it something to do with master/slave function of 7210 and 7220? Or is EFAB the problem?

Thanks

[guido]

Originally posted by tubee
But when i removed the 7220 (to test if without 7220 give less pollution on PS) the player plays 5 minutes, stops and cannot read disks anymore. After some power-down time it starts again and same story again: no reading of CDs

I have put the XO in pin19 of SAA7210, and disconnected the transistor with collector on pin 19 and emitter to Gnd. The base is connected by a cap to +5V supply and a diode to Gnd.(When this transistor stays connected to pin19, the XO will not be recognized, no CD reading and the CDM is spinning fast)

What connection have i overlooked? Has it something to do with master/slave function of 7210 and 7220? Or is EFAB the problem?

Thanks

Never seen such a transistor on the clock line. It seems to shorten the 7220 clock output at reset. Hope the clock is not damaged, but it stills seems to work.

Sounds like a powerregulator gets too hot and shuts down. Are you shure you still have the correct powersupply voltages on the chips? Maybe you made a short somewhere when removing the stuff.

Or it is just wear: bad connecting due to age of the soldering which appears when a component gets hot. Always check the powersupply regulators in older philips gear. Might be just that, triggered by working on the pcb.

[tubee]

Thanks for reply Guido.

Indeed this 640 is a very old and much used player, i used it every day 1 or 2 hours for 18 years now! Have a spare 660 at hand from marktplaats.
Solved in time a few problems already (regulator, connectors broken, 1 transistor) Indeed a chip gets i little hot, will check the regulators/voltages

The transistor looks like a muting device indeed, in the serv. manual are asteriks by the parts, so not all models have the transistor. The XO still works i guess.

Cheers:

[tubee]

Always check the powersupply regulators in older philips gear.

Checked every voltage, also on the chips itself, all within normal range.
Reversed mods too: 7220 back, XO clock on 7220 Xin and restored clock conn. 7220 to 7210. Plays ok again after 2 startups, but after 10 min. again the 640 stopped and cannot be started, no disc recognized.
The 7220 gets awful warm (about 35 degrees) so this chip would be dead i think.

Have a good sounding standard 16 bit 304 mk2 also, and in the 10 minutes of comparison Non-os(640) and Os(304) my conclusion about non-os is: sounds good, gets out of speakers more directly, pretty detailed, more dynamic sound, better bass, but.... the depth in soundstage is overshadowed by the instruments on foreground. Also the non-os 1541 "stands on your ears" more, ears do the filtering, and they don't like that when they have to filter the whole evening(have experienced non-os for about a 1/2 year)
Second but.... my two daughters, one 4 other 6 years,
with the 304 they kept drawing paintings....
same disc in non-os player: they started to DANCE on the music........

[Jocko Homo]

Who ever said that the '7220 won't get warm? In any case, buy a can of freeze spray, and start blasting the chips.

Jocko

[guido]

Originally posted by tubee
Reversed mods too: 7220 back, XO clock on 7220 Xin and restored clock conn. 7220 to 7210. Plays ok again after 2 startups, but after 10 min. again the 640 stopped and cannot be started, no disc recognized.

So it could still be the XO (damaged by that mute transistor).
Try putting in the original xtal and stuff around it.
Got a scope to look at the clock line?
Otherwise try a digital meter: ~2.2V=clock, 0 or 5 is dead

As for the 7220, it gets hot indeed. Worked on my CD80 and i had it up side down for a while (powered on). At a certain moment i fixed the faults (powersupply, what's new). But the 7220 had died Could have been the powersupply or it got too hot from being up side down

[tubee]

7220 won't get warm

It did get warmer then usual, so mor than 35 degrees i guess

Otherwise try a digital meter: ~2.2V=clock

Both the XO and Kwak give such values, no scope at hand now, a friend has two but is on vacation.

The 640 worked again with the Kwak on the 7210, then after 3 minutes dead again.
Lets try Jocko's freezing method. It is not a bad connection on wires, because it always happens after 5 minutes, its the uP-chip i guess, lets freeze this one first. Have no uP at spare so the the 640 will be for parts, also because i am getting tired of searching

7220 and almost everything is upside down in the 640, the project i am building for almost a year now will have the PCB's all with solder side down (four 1541's parallel with shift registers, still thinking with or without 7220. (640 would be a handy testmachine then, Jocko's I/V or passive tube I/V)

I have read somewhere here on diyaudio, he liked the sound of one 1541 more then 2 parallel, so my conclusion would be then that each of the 4 outputs should be buffered and then the voltage summarized together as voltages, this to reduce negative interactions of 4 1541 outputs.

Thanks Jocko & Guido for comments and sorry for the long story.

[Jocko Homo]

When I was building CDPs full-time, I only replaced one '7220. Lots of '1541s. One each radial/servo and photodiode chip. No micros. A few front panels on '582 machines that did things so strange I won't even attempt to relate. One CD-80 had a problem with the display. Worked, you just could not tell track and time.

Does that give you any more clues?

Jocko

[Elso Kwak]

Im some of the antiquated Philips players of mine was a kind of clock mute SMD transistor. Maybe the problem...

[tubee]

Worked, you just could not tell track and time

I know... the saa8461p is not giving any signal to the drawermotor anymore since one time a voltage in reverse direction blew the output stage of the 8461 plus the motor-transistors, due to a broken connection. So have to move drawer manually now to load cd.

The 660 in not often used (less wear on disc clamp) and not tweaked a lot as i did with the 640, and the 660 still works(for now at least)

The 640 was my first bought cd player in 1987 from my sour earned $$$, much later tweaked it a lot but now it has become a sort of study-object. At that time (1987) i was a student and because i owned a CD player i was called a "decadent student"!

clock mute SMD transistor

Indeed between the 7210 and 7220, i have removed the connection from clock line to the collector, after i noticed the CD640 didn't recognize the XO, (cd spinned fast)
But still i think it has something around that transistor, or because when the power is off for a while, the 640 works again for about 5 minutes and stops again, so maybe a logic gate somewhere around the removed 7220 is floating and has to be connected to +5 or Gnd somewhere, but this is just a guess. It lookes like the uP misses something after some time. Looked in the serv. man. but can't find a clue.

Thanks