Sony CDP-101 Introduction
This is one of my oldest Hi-Fi friends and the one single thing that started me on the road of my interest in high-end Hi-Fi and digital electronics, my first CD player ever, the first CD player ever, the Sony CDP-101 circa 1983! Before anyone in the U.S.A. even knew what CD stood for where CD was not introduced until later in 1983, I had my first CD system working in the summer of 1983 in England on August 5th! This then lead to the purchase of my first Technics SE-A5 and my Quad ESL-63 speakers, a totally killer system that with the exception of the actual Sony CDP-101 has lasted me a lifetime and is still in daily use! The whole system was operational ready for me to get my first CD Christmas present that year! If anyone had a CDP-101 before me, please let me know! First introduced on October 1st, 1982 in Japan, the Sony CDP-101 meaning Compact Disc Player model five in digital binary notation, was so far ahead of its time in terms of features and build quality, that apart from it's obvious D/A digital limitations due to its age, it still looks like any top of the line player that has not yet even been invented! This product in it's day or any day for that matter shows the true inventive power of the Sony CBS corporation and along with Phillips and Polygram, without them we may still have been here now listening to vinyl and singing the praises of the Technics SL-1200! I think it would be nice if Sony were to release a 25th anniversary version of the CDP-101 in gold plate like Technics have done with the SL-1200 Gold edition, although of course, while the SL-1200 is still very well known and is still shipping, the old Sony CDP-101 has long since faded into the annals of audio history and now everybody just assumes that CD has existed forever! The "Goronta" shown to the right is the prototype predecessor to the CDP-101 and as you can see has the disc is mounted vertically and this is how all other first generation CD players were made with the exception of the Phillips designs that were top loading like old laservision machines. You can read more about this from Sony here. Trust Sony to come up with the sliding tray disc loading idea that again everybody just assumes was always there. And just when you were thinking that was it, there is more! Sony added an infrared remote control! Again today this is just taken for granted, but most TV's in 1983 didn't even have a remote! As far as I know, the Sony CDP-101 is the first ever widely available consumer electronics audio device to have a remote. Wow, for me that made selecting the CDP-101 as my first player a slam-dunk compared to anything else. There were 2 or 3 other first generation CD players to choose from by the summer of 1983 and the other main one I was looking at was the Technics SL-P10. Of course this was also a vertical loading player and looked very good, but the remote control feature on the Sony just eliminated the SL-P10 from the picture immediately! So after deciding to go for the Sony, I headed to Tottenham Court Road in London as that is where all the top Hi-Fi shops in England are clustered and I set about finding the best price. Well, I was amazed to find that not one single store had a single CD player in stock and this was in August 1983. The sales guys in most stores just laughed and said "you will never get one, and if you did, it would cost thousands"! I kept looking around and in one store, Arena Electronics where I was getting the same story, there were a couple of Indian guys behind the counter furiously talking in Hindi and just as I was about to leave, they called the sales guy over and changed his mind for him. They told me the the player would be waiting there for me the next day! I showed the sales guy 4 CD's that I had just purchased from the Virgin Mega Store in Oxford Street and he told me to be sure and ask for him the next day. Sure enough the next day the player was there and it was all hooked up for a demo & the sales guy was just bursting to see what CD's I had for him. No words can describe his disappointment when I showed him the only CD I had on me that day, Vivaldi's Four Seasons by the English Concert Orchestra, but it played fine! The Sony CDP-101 cost me 550 GBP that day which was about $850 at the time, and for 1983 that's quite a pricey piece of gear! You can read an original review of the CDP-101 from stereophile here.
Inside the Sony CDP-101
When you open up the CDP-101 and compare it with a modern day player, the first thing that strikes you is just how much of it there is! Even though it contains the first LSI (Large Scale Integration) microchips in any consumer electronics device that replaced over 500 ordinary integrated circuits in prototype CD players, it is still stuffed full in there and it is a wonder it ever worked reliably, but work it did. As the digital to analog converter was of course one of the most complex and costly components to produce back then, the player was setup in such a way that the single converter was time shared between the two stereo channels, first calculating the output level for one channel, and then working on the other channel and so on. You can see the logical arrangement in the diagram to the left. These days, if a single DAC were used, there would be sample & hold buffers to get the samples for both channels in synch. So with CD having a sampling rate of 44.1kHz per channel and two stereo channels, that means a total of 88,200 samples per second, so dividing 1 second by 88200 gives about 11.34 micro seconds for each sample. So this means in fact that one channel is always about 11 micro seconds behind the other. Although very slight, this would causes an effect like you can hear when you adjust the azimuth on a tape deck although very much harder to detect. Comparing this problem with the effect you get by playing an LP with anything other than a linear tracking turntable, the effect is negligible and so in those days it was acceptable as the main impact of CD was the fantastic dynamic range and signal to noise ratio. The only time you can actually really hear this effect is if you are playing a Dolby Pro Logic surround sound disc and playing it through a surround sound processor to split the sound into 4 channels (left, right, center & rear surround) where exact phase is very important. Of course, nobody was doing this in 1983 with CD's so it was really a bit of a non issue. Having said all that though, there is some indication in the circuit that some rather crude effort was made to solve this problem. Look carefully at the circuit diagram of the Low Pass Filter section of the output stage shown to the right. Notice that for the left channel the feedback resistor (R530) is 15k ohms but for the right channel (R541) it is 16k ohms. That 1k of difference with the 75pF capacitor would give rise to an additional delay of about 75 micro seconds on its own, but we have to assume that in combination with the complete circuit it somehow resolves to around the 11.34 micro seconds required. This issue though is why it is important to have properly synchronized parallel phase converters in a laserdisc player for watching surround sound films. The way I solved this in the end for the CD case was to replace the CDP-101 with a Technics SL-P1000 in 1988 that has separate single chip converters for each channel. Still, it's always a great way to test any self pronounced audiophile experts that happen to be in my living room to see if they can tell the difference... The tiny difference in phase that still exists is probably the only way to tell this CD player from any with parallel converters apart from the usual discussions about the tonal differences.
Clearly by adding an elegant infrared remote control handset to the CDP-101, Sony were blazing the trail for the CD concept in every aspect! Like the player, the handset is as futuristic as anything you could imagine! With a slightly squareish profile by today's standards the handset is built tough with a stylish painted brushed aluminum clad top and soft rubber buttons like any remote of today, but it has a weight & feel that no modern day remote can touch. On the sides of the remote, there are some sliding covers that run almost the whole length of the handset and when you remove them, they reveal nothing but the slot from which they came. I have always presumed since first seeing this that Sony would soon make a receiver or something with a similar design and provide a fluted slide that could join two remotes together. As far as I know they never did and the design was abandoned. When you press a button, there is a small light that blinks on the player, just like a 1980's TV. There is also a switch on the bottom of the player that when switched on causes a little "beep" sound at every button press. The sound is actually quite smooth & cute and not like the cheep sounding beep of modern electronics and shows just how hard Sony were working on every detail. As one of the primary challenges in the integration of the large audio & video setup's I have built over the years is the detailed understanding of infrared remote control, it is fitting that I should describe the detail of this handset first, as first it was indeed, although of course it is no longer used in my primary system. This may also be vital information if you are trying to build a replacement for the Sony RM-101 Remote Commander or program some other device to use with a CDP-101!
Just like all remote handsets of today, the Sony RM-101 uses a modulated pulse position protocol to signal the CDP-101 player. The player looks for pulsing light in the area of the modulating carrier frequency which is 40kHz, and by looking at the length of each modulated burst, can detect binary 1's and 0's, so this too is a simple digital device that uses a kind of pulse position modulation. Referring to figure 1 below, this shows part of a PCRemote screen shot and is the display of the RM-101 Play button. As you can see, there is one wide pulse to start with, which is a synchronization start pulse, then 12 following pulses, some wider than others. The thin pulses represent a binary value of 0, and the wider pulses a binary value of 1. So going from left to right, the 12 bit value is 010011010001. By looking at the codes from all the other buttons, I have established two things: First, the data appears to be sent with the least significant bit first, but the PCRemote display is merely concerned with time going from left to right, so for us to read it as a number it needs to be reversed as 100010110010. Second, I can see that all the pulses start with 100010, so it is possible this is a 6 bit device code, and the remaining 6 bits 110010, are the button code, so this would give 64 possible devices with up to 64 buttons each. In fact, further research into this indicates that Sony uses (or used to back in the 80's) a 5 bit word code and a 7 bit data code. The total data packet is about 20 milliseconds long and there is another 20 millisecond gap at the end before the pattern is repeated. The repetition goes on and on for as long as a button is held down.
Figure 1. PCRemote display of RM-101 Play button at scale 9 showing entire data packet
Looking more closely at the pulses in Figure 2 below we can see more detail about the pulse durations. The starting synchronization pulse is 2.4 milliseconds long and then each data bit is a 0.6 millisecond gap followed by a 0.6 millisecond pulse for a binary 0, or a 0.6 millisecond gap followed by a 1.2 millisecond pulse for a binary 1. So the duration of the shortest data packet with the value of 000000000000 would be 2.4 + ((0.6 + 0.6) x 12) = 16.8 milliseconds. The duration of the longest data packet with the value of 111111111111 would be 2.4 + ((0.6 + 1.2) x 12) = 24 milliseconds. Just for your information, the numbers on the top row of the PCRemote display are the pulse numbers starting from 0 and counting to the right, the second row of numbers is the actual count of the number of 40kHz cycles that is needed to fill the amount of time of the pulse or gap width. It is data at this level that PCRemote actually records and uses for playback of the pattern to emulate a button press. So in fact, PCRemote does not need to know anything about the details of the protocol we are discussing here, it just records and plays back the pattern of pulses it sees! This is in fact how all learning remotes work and why they use so much memory for learning something so simple! They are just not smart enough to learn the actual protocol like we are here! But at least with PCRemote the user can see what the application is recording and redo any recordings that are obviously wrong. That can sometimes happen if you have fluorescent and especially neon tube lights on in the same room as the receiver.
Figure 2. PCRemote display of RM-101 Play button at scale 40 showing detail of pulse sizes
The following table shows the binary values of all the buttons on the RM-101 The codes are shown as 12 bit binary numbers and as described above (least significant bit on the right but first to be transmitted), this makes the code for play 10001 0110010 and they are divided into a bunch of 5 and 7 just to make them easier to see and are listed in order of numerical code. This information can be used to manufacture a new RM-101 remote or you can convert the data into pulse width times for entry into some kind of other intelligent remote control transmitter device to use with a CDP-101 if you have lost your RM-101!
After a quarter of a century of usage, it is only fair that a CDP-101 should have a failure or two. Having said that, mine failed in about 1985 which I didn't feel was that good at the time! It therefore effectively suffered the same fate as every other piece of Sony equipment I have ever owned including the first ever Video 8 recorder with digital audio the EVS-700 and the first ever small head Video 8 HandyCam camcorder the TR-55. While all of these items did indeed work at one point and continued to do so through their warrantee period and were purchased at a time when each of them was incredibly difficult to make, they all failed after a relatively short time compared with other equipment I have. While it was great to have such futuristic equipment at the time, it's a shame that it all busted so quickly! So while Sony clearly has a significant leadership position in design concepts, I can't say as I'm that impressed with the equipment. Even within the last month, I have found two items of Sony equipment just discarded in the garbage at our condo (yes, I'm rooting through the garbage again!), an STR-DE915 receiver and a CDP-297 CD player. How come I never see any Pioneer or Technics gear tossed away there? After a little research I find that anyone who has had the misfortune to own an STR-DE915 receiver will never have anything good to say about Sony again and I think it is a shame that Sony puts their name on such poorly designed equipment, although the CDP-297 is up and running straight out of the trash can and works with the RM-101 remote! As far as my CDP-101 is concerned, it sat for many many years just gathering dust waiting for a service manual to come along. Well, now I have one, so let's see what can be done! What was happening with the player back in 1985, was that it did start to play the disc, but then started to skip and jump around. Then a knocking noise starts to come from the focus servo and it is unbelievable how loud the noise gets, so I would unplug it before it exploded! Now, there is also another problem, the drive for the loading draw has failed so you have to take the lid off to get the disc in!
So the two problems are: Disc skipping which gets worse and worse the longer the player runs; Second, the loading draw is stuck. From all I have heard about CDP-101's, these are the two classic and most prominent problems. If your CDP-101 has not had either issue yet, one day it will. The CDP-101 Troubleshooting Guide lists the problems shown in the table above and the sequence of actions needed to solve them. Problem is, neither of these things are listed. The section about "Sound Skipping" just tells you to clean the sled drive and to replace the rack if needed. Hmm, well I did hear from one guy in Germany who said that cleaning and lubricating the sled drive did work for him, so I guess that would be worth a try. But from the amazing knocking and banging going on under the hood, I can tell you that an oil change is not going to work in this case!
STK6922 Causing Disc Skipping
Looking at the block diagram of the CDP-101 in the service manual and knowing that the knocking noise is coming from the sled or the laser assembly (or both), following back from the optical units focus & tracking servo leads to IC204, the motor driver. Following back from the sled motor and the chucking (the 4 lifting feet in the tray) motor leads to IC304, motor driver again. Each of these chips is an STK6922 and each has two drivers. The output pins for each are pin 2 and pin 16 (pin numbers are on the back board) and the +ve and -ve inputs for each driver are 7 & 6 and 11 & 12 respectively as shown in the diagram to the right. According to the service manual in idle conditions, all of these pins should show 0 volts for IC204 and IC304. Once warmed up, my CDP-101 would just start clicking and knocking of it's own accord with no disc loaded, so I just had to be looking between the right two pins at the right time to see what was going on. This test showed spurious voltages on the outputs of IC204 on both pin 2 and 16. A couple of times I also noticed the sled motor twitch but I never saw any confirmed results on IC304. Either way, this looked like the problem area and searching the net indicated others had changed both these chips, so I decided to replace them both. Once both chips were replaced, the player ran fine, at least so far! This is lucky, as I expected I would have to adjust the focus gain at the very least, but it looks like that was all that was needed. There are streams and streams of other adjustments and checks in the service manual and in some cases, other adjustments will be needed, so if you do this repair, just this may not be all that is needed. I picked up my service manual on eBay and this is where to look if you need one or I know that www.hifimanuals.com has copied versions of them here. While these chips are old and rare, here is one supplier I found who has a page for the STK6922 and sometimes you can find them on eBay. Later model CDP-101's used the BX1201 in place of the STK6922, so you could try that also. The missing pins on the BX1201 are not needed and were grounds on the STK6922 so the BX1201 is in fact a pin for pin replacement. Only problem is that the gain is a bit different so the focus and tracking servo gain will need to be adjusted if you substitute one chip for another. Good luck if you try this out! I have heard from many people who have had good luck with this repair with both chip types. More recently, Stephen Lafferty has published a paper that describes how to build a complete replacement for the old STK6922 using two LM1875 audio amplifier chips. Click here to link to Stephens CDP-101 page.
Oil Change Needed to Free Loading Draw
OK, I take back everything I said about this thing not needing an oil change! The reason that the loading draw was not working is that the grease used to lubricate the moving parts had dried up and actually gone solid in some places. In fact not only was the draw stuck, but I also noticed that the chucking mechanism that lifts the disc up and down was straining and slow. The best thing would be to go through the whole unit can change all the oil! Having said that, many parts do not really need oil and just cleaning them up with some Goo-Gone did the job for me. The particular part causing the draw to stick in my case is shown in the pictures below. You need to remove the bottom of the player and lift the bottom circuit board up to get access. This means removing the 5 screws that hold it and cutting some of the straps from the cabling above. That means removing the top of the player and removing the two screws that retain the top folding board. All quite easy though!
The Sony CDP-101 comes out for occasional usage to demonstrate some real digital audio history, but when I tell people that it is over 25 years old they just look blank... Try this: Setup a CDP-101 and tell people that you have just bought a new high end CD player that you want them to listen to... It works every time
Pictures & Reference
Here are some extra CDP-101 pictures I found on eBay. Mint working CDP-101's go for about $100 to $300 depending if they have all the parts with them. Remember, if buying on eBay, ask if the player can play all day when it's hot, or you can expect to be changing IC204/IC304 as described above! The skipping caused by the focus servo chip problem is time & heat related and just playing a couple of tracks is NOT a sufficient test!