I recently acquired a boxed WM-3 and i'm fixing it to be back to its original splendour. Beside easy fixable mechanical problems (needs new belts and a bit of lubrication) I have a weird issue with the Left channel amplifier: with fresh new batteries the L channel has a delay in turning on (1~2 seconds) but then it is OK; with exhausted batteries it does not work at all and i can only hear a hum noise. I thought about a power issue and then proceeded to check voltages on the IC's pins. Voltages on the right IC102 (power amplifier and power stabilizer) are fine, voltages on Q101 (pre-amp) are fine but on IC101 (pre-amp and EQ) voltages on pin 6 and 5 are way off: instead on 0.9 I get 1.77 and instead of 0.6 I get 1.2. Even considering production tolerances, those are way too off, and same pins in the other channel are more correct. My question is, why are they so off ? I do not see external components to set those voltages, they appears to be set by the IC itself. Pin 6 is the output of the pre-amp while pin 5 is the input to the diff amp (it closes the feedback loop of the diff amp and set the playback EQ). Any idea ? thanks
My wm-3 also has that left channel second delay before it kicks in to match the volume of the right. I'll be watching this thread!
My "finger in the air" guess would be electrolytic capacitors. it could be one of the other components but electrolytics are by far the least reliable as they contain liquid that can dry out, or start passing current when it shouldn't. The fact that capacitors take time to charge would explain the delay. For example if C105 or the one at the top right were out of spec they could mess your voltages up.
Electrolytic capacitors are my first bet too (or wishful thinking, capacitor are easier to find compared to IC's) but I cannot find a capacitor which could cause the DC bias of the pre-amp circuit to be so wrong. IC201 should be similar to an op-amp with pin 2 as positive input, pin 5 ad negative input and pin 6 as output (it is actually more complex than that as pin 2 and pin 5 have different DC bias but I have no idea how the ACG part works). Anyway all the resistors and capacitors between 5 and 6 are the feedback loop that set the gain and the EQ; there is also the switch that changes the tape EQ by modifying the resistance of the feedback loop and hence changing the frequency where the gain slope levels off. C205 is like an open circuit on DC bias and even if its capacity would be degraded it should not behave like a closed circuit. Bias on Pin 5 is set by the two resistors R208 and R205 and pin 5 bias is 2/3 of bias on pin 6. The point is, what is setting the bias on pin 6 ? But more important, is that really an issue or the delayed activation of left channels is caused by something else ?
Capacitors can develop electrical leakage. Another possibility is that a resistor has gone out of tolerance although it would probably be way out. A couple of suggestions for fault finding. Have you measure the voltages at switch on to see how they change. The brute force method of fault finding would be to remove the ICs from both channels, check the associated passives then refit the ICs the other way round. If the fault moves from left to right then the associated IC is faulty. Sorry I can only offer general fault finding advice here. I'm more familiar with digital circuits.
You should do measurements of audio signal path with oscilloscope. I would check both, IC input and IC output. You will see where the signal is lost. May be it is already missing on the IC input. You can record 315Hz tone at 0dB and play back that cassette. Power your walkman with lower voltage (2V fore example) because you mentioned that signal is lost when battery is low. If you do not have oscilloscope, download SW soundcard scope and use it instead.
Thanks for the suggestions, but unfortunately I do not have an oscilloscope; some years ago I could "borrow" one from work for the weekend but now it is being use 24/7 for test beds and cannot be removed. I use the sound card as oscilloscope but just for basic tape adjustment (level, azimuth ..) and I'm not sure it is so sensible to detect the input to the head pre-amp as the signal coming from the coils is very low. I think this problem goes beyond my ability with a soldering iron and I would need too many tools I do not have (a proper bench with lighting, magnifying glass, adjustable power supply, spare parts... ). I was hoping someone more experienced than me could spot the problem but I perfectly agree it should be properly investigated first. I will see what I can do.
Then use easier method. When you touch any pad on PCB, located on audio signal path, with small watch screwdriver, you will hear hum in the headphones (in L or R channel). You can trace signal path from the input to the output and see where it is lost. You can compare L and R channels. For that, you need service manual.
I tried the "screwdriver" technique, but it proved to be inconclusive. Referring to the below schematics: The expected behaviour should be as follows (R channel or L channel with high power voltage): 1) Q201 base -> strong hum 2) Q201 collector / IC201 pin 2 -> hum 3) IC201 pin 6 -> no sound 4) power amplifier input (not shown in the above picture) -> click sound. This is what I get on L channel with low power voltage: 1) Q201 base, hum noise 2) Q201 collector, only a click noise when the screwdriver head touches the pin. 3) pin 6 of IC201 no sound 4) input to the power amplifier cause only a click. 1) and 4) are almost correct, but the lack of hum from IC201 pin2 would indicate a problem in the IC itself. But if the IC has problem, why do I get a hum when touching Q201 base ? I am really confused ...
I was able to borrow an oscilloscope from work (it appears to be 20 years old, but it will do the job) and did some tests to see where the audio signal is lost. First with low voltage (~2.4 V, led almost off) and by comparing R and L channels, the audio signal is ok up to the input to IC201. On the output of IC201 (pin 6) the audio on L channel is 10 dB weaker. Then with normal operative voltage (3 V) the audio signal of L channel is ok also on IC201 output. Then I checked how the signal looks immediately after pressing play: R channel (the working one) on output of IC201 goes immediately to stable output amplitude; L channel has a visible amplitude ramp and takes seconds to reach a stable output. With low voltage the ramp stops pretty early, when the signal is still very weak. That could be the effect of a charging faulty capacitor, but it might also be a faulty IC (I hope not).
Marian has looked into the WM-3 and fixed it. Luckily it is a capacitors problem, not a faulty IC. With reference to the above schematics, the delayed start-up is caused by C102, while C109 can cause volume issue with low voltage power supply. In the end it is better to replace all electrolytic capacitors, typically if there is a faulty cap, many other will follow suit.
