Dried out caps anywhere are a problem. There won't be any in the RAFOC itself but there are plenty of course on the all the other boards. Failure of a particular 33uf electroylitic is a common issue on many Philips players using swing arm pickups. It must be 33uf too. You'll find many references to that problem on the forum such as,
http://www.diyaudio.com/forums/digital-source/189859-marantz-cd-54-a.html
http://www.diyaudio.com/forums/digital-source/189859-marantz-cd-54-a.html
You should find out the reason for the -7V. If you have a tracking problem, and an incorrect voltage in the tracking circuit, then it makes sense to investigate.
Also establish for sure whether or not focus is OK on CD before it gives up trying to read. It must focus before it can track or lock to the right spin speed. You say it passes service loop B stage 1, and that includes focus and speed.
You say that the voltage across the radial motor changes when play is pressed on CD. In what direction is that change? Why does it change? Where does it hope to go to if it doesn't know where it is? Does the CDX do the same?
You can swap mechs, boards, processors, perhaps? (due consideration to laser current settings)
My CD104 sits and spins forever on switch-on with disc present, until I tell it to do something else. I don't think it moves violently.
Without a limit switch it can assume it's at the limit only because it maintains a current through the coil to drive it there. You measured that. How does it compare with the CDX? Does the CDX also maintain a current in stop mode? That's unnecessary because then it doesn't need to know its position. Are the servo processors the same versions? They are usually stamped or stickered.
Be wary of super-low ESR caps in the preamp board and early stages of processing. I don't know how tolerant the CDM1 is but, as Mooly says, some Philips mechs have a reputation for being choosy. You'll get away with higher voltage and higher temperature, lower-ESR normal electrolytics, but avoid exotics. Did you recap the pre-amp board? That's as close as caps get to the laser. I guess CDs have sharper transitions, and so more very high frequency noise.
Also establish for sure whether or not focus is OK on CD before it gives up trying to read. It must focus before it can track or lock to the right spin speed. You say it passes service loop B stage 1, and that includes focus and speed.
You say that the voltage across the radial motor changes when play is pressed on CD. In what direction is that change? Why does it change? Where does it hope to go to if it doesn't know where it is? Does the CDX do the same?
You can swap mechs, boards, processors, perhaps? (due consideration to laser current settings)
My CD104 sits and spins forever on switch-on with disc present, until I tell it to do something else. I don't think it moves violently.
Without a limit switch it can assume it's at the limit only because it maintains a current through the coil to drive it there. You measured that. How does it compare with the CDX? Does the CDX also maintain a current in stop mode? That's unnecessary because then it doesn't need to know its position. Are the servo processors the same versions? They are usually stamped or stickered.
Be wary of super-low ESR caps in the preamp board and early stages of processing. I don't know how tolerant the CDM1 is but, as Mooly says, some Philips mechs have a reputation for being choosy. You'll get away with higher voltage and higher temperature, lower-ESR normal electrolytics, but avoid exotics. Did you recap the pre-amp board? That's as close as caps get to the laser. I guess CDs have sharper transitions, and so more very high frequency noise.
Hi,
The Beogram behaves differently I've found, because it has a revised servo board, like the 104-b revision. So when the disc is inserted it reads the TOC and stops. The swing arm is idle, receiving no power at all. I cannot swap mechs easily because the CDX is difficult to pull apart in true B&O fashion!
I agree about exotic caps and use off-the-mill, good quality caps from Panasonic, Elna or Rubycon. I do not like ultra-low esr caps either.
I guess focus is fine. The eye-pattern is fine and stable.
Tomorrow I'll try renewing the five preamp board 33uF caps, just to discard that if anything. Then I'll check on that -7V again.
Jacques
The Beogram behaves differently I've found, because it has a revised servo board, like the 104-b revision. So when the disc is inserted it reads the TOC and stops. The swing arm is idle, receiving no power at all. I cannot swap mechs easily because the CDX is difficult to pull apart in true B&O fashion!
I agree about exotic caps and use off-the-mill, good quality caps from Panasonic, Elna or Rubycon. I do not like ultra-low esr caps either.
I guess focus is fine. The eye-pattern is fine and stable.
Tomorrow I'll try renewing the five preamp board 33uF caps, just to discard that if anything. Then I'll check on that -7V again.
Jacques
It sounds about right. Often the amplitude is a little lower on CDR (and is very dependent on disc and burner and burn speed).
I still think you should try altering the laser power (as a one off for this player and fault). Lower it a little on a CDR and then try a CD. You can always reset it back if you note the amplitude on a known disc.
CDR and CD are fundamentally different in operation. CD uses the difference in "height" between the pressed pits and lands to detect the transitions (the 1's and 0's). CDR relies on a special dye being burned away to create an area of no reflectivity. Its a fundamental difference between the two formats.
I still think you should try altering the laser power (as a one off for this player and fault). Lower it a little on a CDR and then try a CD. You can always reset it back if you note the amplitude on a known disc.
CDR and CD are fundamentally different in operation. CD uses the difference in "height" between the pressed pits and lands to detect the transitions (the 1's and 0's). CDR relies on a special dye being burned away to create an area of no reflectivity. Its a fundamental difference between the two formats.
1Vpp is correct for CD. Compare CD with CDR using your CDX. If that's impossible then I'll check my CD104, but I haven't got your CDRs so it may not help. They vary a lot. CDs also vary quite a bit. I've never seen a CDR give a stronger signal than a good CD.
Maybe we have different CD104 service manuals? Test point 65, 1Vpp. Offset is 0.5Vpp, so the pattern sits on the 0V axis if the scope's set to DC.
Has your manual got annex IV, concerning k-factor, dealing with that part of the tracking error processing circuit where you have an incorrect voltage of -7V in stop mode?
My mistake.
I've just fixed what I hope is the last of many problems with my CDP101: a weak signal from one photodiode put the tracking balance just out of adjustment range. This I guess meant the pickup was slightly offset from the centre of the track during play. Some CDs played and measured perfectly, some wouldn't play at all, and some gave up half way, with the pick-up sounding like squabbling goldfinches. I didn't try it with CDRs.
Although the Philips system is very different, the principle of tracking, by moving to maintain the correct relationship between two signals, is similar. It could be that test points 46, 49, 50, 51 on the CD104 must be in some kind of balance for a similar reason, and the -7V suggests they may not be.
The CDP101 is less integrated than the Philips, and uses standard logic chips for much of the front end functions, including the preamp which uses a hex inverter, so there is a separate amp for each diode. The gain of each amp is set using its feedback resistor, and one of the two used for tracking uses a pot to set balance. I just used a resistor to extend the range of the pot. I don't know if the amp or the diode is at fault. Some day I'll change the amp to find out.
The CDP104 has no such adjustment so it probably does it automatically, so if it's wrong then it's likely because of a drifting resistor or other failing component.
The -7V may turn out not to be the cause of the problem, but either way it should be investigated.
Error in last post...I meant CD104, of course. Also:
Manual, Sub H, "Is the radial control funtioning properly":
"The offset circuit (d-multiplier) and AGC circuit (k-multiplier) are correction circuits. This means that under optimal conditions (new disc, minimal tolerance of components) the set may be working properly even if a fault is present in offset or AGC circuits". Or maybe not.
Could be that your -7V just means that it's doing lots of correcting. Who knows.
Manual, Sub H, "Is the radial control funtioning properly":
"The offset circuit (d-multiplier) and AGC circuit (k-multiplier) are correction circuits. This means that under optimal conditions (new disc, minimal tolerance of components) the set may be working properly even if a fault is present in offset or AGC circuits". Or maybe not.
Could be that your -7V just means that it's doing lots of correcting. Who knows.
Last edited:
Stupid me, they are in the service manual!
As expected, new caps haven't changed anything.
If we have the same copy of the manual, we've probably been driven equally mad. Mine's fragmented and barely legible, so the arcane logic of its organisation is almost impenetrable.
The CD304 mk2 (TDA1541) manual available from HiFi Engine | Download Free User/ Service Manuals, Amplifier, Receiver, CD, Tape, Tuner, Video is clearer and less fragmented, and the block diagram is more helpful. Looks like the same circuit up to the DAC.
You can see how TE1 and TE2 are derived from summing the two photodiode outputs on each side of the array, and how TE is a funtion of the difference between TE1 and TE2. What's not clear is under what circumstances the Automatic Gain Control block intervenes to correct the Xk block, or whether it is concerned with relative or absolute gain, and what the analogue switches (HEFxxxx) are all about. Inputs and outputs to two processors (decoder and servo) too. Not simple.
I wonder if that -7V is different on the CDX, and if it varies between CD and CDR?
Looking more closely, your -7V is the output of the AGC block, and input to the Xk block. It's a correction signal that increases gain of both signals d.(RE1+RE2) and RE2 equally, where d is the offset correction. The difference becomes the tracking error signal:
RE=k.d.(RE1+RE2)-k.RE2
Now, what I don't understand is that, if no correction is necessary so k=1 and d=1, then RE=RE1
I expected RE=RE1-RE2.
Anyway, it's likely the output from your photodiodes is weaker than it should be for a new machine. It's possible AFAIK that the HF signal is normal even though the mean pd output level is low, but the low level might explain some of the fuzziness of the eye pattern.
None of this quite explains why it won't read CDs, but all the evidence seems to point to a weak laser, I fear. It's still worth going through all the manual checks just to make sure there's no other reason for the AGC's correction signal.
I wonder, if you put the player into loop B with a CD and check that correction signal, if it becomes even lower than -7V when the machine tries to track? Maybe with a CD it just can't do enough correction, or the HF just gets too fuzzy if the light level is so low, so even if it can track it can't read.
RE=k.d.(RE1+RE2)-k.RE2
Now, what I don't understand is that, if no correction is necessary so k=1 and d=1, then RE=RE1
I expected RE=RE1-RE2.Anyway, it's likely the output from your photodiodes is weaker than it should be for a new machine. It's possible AFAIK that the HF signal is normal even though the mean pd output level is low, but the low level might explain some of the fuzziness of the eye pattern.
None of this quite explains why it won't read CDs, but all the evidence seems to point to a weak laser, I fear. It's still worth going through all the manual checks just to make sure there's no other reason for the AGC's correction signal.
I wonder, if you put the player into loop B with a CD and check that correction signal, if it becomes even lower than -7V when the machine tries to track? Maybe with a CD it just can't do enough correction, or the HF just gets too fuzzy if the light level is so low, so even if it can track it can't read.
Many thanks for not letting me down on this!
Unfortunately I can only do the static checks, due to a lack of complete equipment (HF generator). But the static tests report fine (ie k-factor). I'm going to try and check what you suggest then.
My guess has been a weak laser too, but how this affects CDR tracking I can't figure out, although your idea might just be the final solution.
Unfortunately I can only do the static checks, due to a lack of complete equipment (HF generator). But the static tests report fine (ie k-factor). I'm going to try and check what you suggest then.
My guess has been a weak laser too, but how this affects CDR tracking I can't figure out, although your idea might just be the final solution.
Last edited:
Odd. If it's -4.2V when the signal is normal, and -7V when it's weak, then -0.25V suggests it's too strong. When the machine is struggling to read a CD, what kind of eye pattern can you see?
Perhaps the monitor diode is weak. You found that the laser current is at the top end of it's range, and generally that would be because it's turned itself up because the laser is weakening. However, if the monitor diode weakens, then the circuit will adjust the laser current upwards unnecessarily.
Compare with CDX if you can. It may be that you need to adjust laser current downwards, as Mooly suggested.