!!!BRILLIANT!!! for some unfathomable reason @Daniela (as @Deb64 posts) remind me these funny 'words of wisdom' from Erasmus of Rotterdam which made me smile when I were in my sweet-teens: "Men were born for government affairs, and therefore they were to receive a few extra droplets of reason necessary to maintain male dignity..." Oh, Boy! playing with Discmans for a few years now I am not even at low 10% of Daniela and Debbie skills, no extra droplets were shed my way ...now I am totally gender-disoriented
This time I fixed the microcontroller board. Although the cassette and CD player were not working it is quite easy to solve. @Longman this time it was a failure as you suggested. The CPU sent a command and it was not executed by the mechanism. The only problem is to figure out what is the CPU trying to do and waiting is it for. A topic which the service manual does not point out. I describe what needs to be done. However, I am going to replace the buggy firmware anyway. The interesting thing is I have never seen a CD555 on sale which has not a hanging CPU. A quite common bug - but easy to solve and costs nothing. Already fixed the CDM2 of the cd-player but need to prepare the videos. Unfortunately the gear of the cassette lost a lot of teeth during testing so I have to replace it. Finally I can start my new resin printer for the very first time https://70cd555.com/fixing-the-microcontroller-board/
I have been recently thinking of another modification... to replace the analog tape counter with a digital one. Would add the possibility to automatically jump to zero or to a specific position. And well, here it is -> LINK if you want to read more Not finished but added a small OLED display with a button to reset or program the counter. Through the previous and next keys it is possible to jump to a position. The only thing which was hard to calculate was the elapsed time in minutes and seconds on the tape during fast forward/rewind. Haha, I love such things...
WOW!!! Fascinating effort... As a "fallen sideways" graduate from one of the top math schools of 'the Soviets', I can almost understand the problem..., but what is your math for FF/RW?
haha I studied Business Informatics in Moscow, Higher School of Economics, what a coincidence. Math in Moscow was like from another universe compared to the university in Austria. The fact that the professor has only spoken Russian since he could not speak English was just the icing on the cake. My knowledge in the Russian language is enough to survive. I wouldn't have had understand him even when he spoke English. No idea how I got through. My final exam was about to calculate the direction of solar panels of a satellite around an asteroid at a certain time and position. Pretty sure I crashed the thing since I haven't got an offer from Nasa or Roscosmos But that is another story... I decided to go the easy way. A real calculation of the elapsed time would need some sensors and right now I have not idea where to place them in the SER1 mechanism. It is very compact and there is basically no place to add them. If I am bored and I need a challenge I am going to do implement it. Found a patent which describes what to do -> LINK. Sorry, it is in German, but I think you get the idea based on pictures and formulas. As I said I am going the easy way. The rotation of the motor is fixed for play and fast forward/rewind. That is what I need to work with when there is no other data available. Another point is the different diameter of the reels. On the side with no tape the perimeter is smaller, the other side which has the tape has a higher diameter which means the transported mm per rotation at certain position of the tape is different when the rotation of the motor is constant. This can be easily calculated by the perimeter of the reel at the current position. And here is the problem, when you insert a tape we do not know the current diameters. The patent describes to start playing for a short time and calculate the difference between the speeds of the reels to determine the position and elapsed time. Since I do not have this info the design has some drawbacks to be able to calculate the times: It only works when you rewind to the beginning of the tape and when you reset the counter You need to enter the duration of the tape (for instance C60 or C90) The rotation speed oneeds to be known for play and fast forward/rewind. For the latter one this might be slightly different between different devices. But this data can be figured out. The thickness of the tape affects the calculation and leads to wrong values As far as I have seen some old cassette decks seem to use a similar method. Thus the math is quite simple, it calculates the current position in millimetres of the tape based on the diameters of the reels and the elapsed time frame since the last calculation. Than it recalculates the elapsed seconds based on the elapsed distance of the tape. What is given: Speed during playing = 4.76 cm/sec Perimeter of one rotation on emtpy (71mm) and full reel (150mm). Measure at a C90 tape I do have. Elapsed time since last calculation in milliseconds [elapsed] Factor for rotation per seconds during fast forward/rewind [FF_MULTIPLIER] Length of the cassette C60, C90 etc. [CassetteLengthInSeconds] The active reel is different for forward and rewind thus the direction [WindDirection] must be considered during the calculation. The values and code might need to get optimised and I am currently test the whole thing. Here's the C++ code: Code: float TapePosMM = (Position / 1000.0) * 47.6; // perimeter beginning of the tape = 71mm // end of tape 150mm // millimeter per full rotation float mmpfr = 71.0 + (79.0 * (Seconds / CassetteLengthInSeconds)); if (WindDirection < 0) mmpfr = 150 - (79.0 * (Seconds / CassetteLengthInSeconds)); // mm in elapsed time float mmet = FF_MULTIPLIER * (elapsed / 1000.0); if (WindDirection > 0) TapePosMM += mmpfr * mmet; else TapePosMM -= mmpfr * mmet; // new position Position = (TapePosMM / 47.6) * 1000.0; However, I might be completely wrong - since math is not my major and my knowledge of tape decks is just very basic - maybe someone has a better idea or can correct me if I am wrong? As far as I was able to test the method it it is quite accurate and works better than the manual tape counter.
I wonder if top-shelf Nak decks have linear digital counters: those guys put a lot of efforts into being the best. Maybe Tandberg 3014? Let me start with my lowly Yamaha: it has digital counter but I never checked if it is linear or not. I must confess that even the idea of actually using the timer while recording had never occurred to me. But I do have a strong preference for digital counters vs. mechanical. And yes, math and physics in the Soviets was something else: more of a 'selection' process than teaching
I've already fixed the CDM2 CD drive some time ago. Was quite easy this time. Also explaining the decoder panel which was working fine. I love the old TDA1540P chips. Wired 14 bit but producing a very good sound. Next task is to repair the tape mechanism. Need some spare parts and everything's closed in Austria due to the second Covid lockdown. Let's hope that we will get better times quite soon. Hope you are all well!
Fun to watch! I especially like your message "if I can do it, you can do it": watching what is being done I want to say (in Mike Myers words, and in his unforgettable style) "Yeah Baby!" Of course I can: Over the years I fixed about a dozen Discmans which played OK when held vertically or at an angle but skipped or played dead when placed horizontally. Each and every time it was a worn-out CD spindle motor, once replaced the problem was solved. Also, it looks to me that missing plastic part next to the lens does nothing during play and is there for when boombox is bumped around with CD inside, so that lens won't hit CD. But maybe I am wrong, my only experience with Philips swing-arm mechanisms was when I fixed Marantz CD-63 (re-badged Philips CD100) for a friend. I was too lazy to do full recap as you do, just recapped power and analog audio path. But the caps which actually died were in the swing-arm itself:
Cool project Daniela, you don't realize how complicated some of this equipment is and the mysterious unseen software makes it more interesting. My knowledge isn't that specialized but I find it odd that the chips, or the embedded software, is failing and it's a similar issue. How does the chip lose it's code, heat, cold, magnet, something else? The resin printer sounds interesting, I don't have one yet but I did get a 3Doodler but it's learning curve is pretty steep.
Oh behave! Like the movies Ah you are the one who is watching the videos Yes the CDM1 from the CD100 is very similar. The second generation, the CDM2 in this Sound Machine, is a "cheaper" version. Very special on this particular version of the CDM2/19 is that is has two PCBs on the panel - you can see this in your screenshot of the video. The capacitors you are talking about are in this case on the left, smaller PCB. Not sure why this is a special version, cannot believe it is just because of the vertical position since in Unit #2 a donor from a CD150 (horizontal player) is working perfectly. Only Philips will ever know. In my opinion it is always worth to replace all the capacitors and I am doing this before checking all other parts since a lot of issues were already solved by just a simple recap. I figured out the small plastic part helps the focus lens to stay even. It is placed on a thin metal foil and if it is missing the focus is getting crooked and cannot focus correctly anymore. I think you can hear this in the video - it rattles when it try to focus. Now the focus is fixed but it seems the weight of the tape I am using is too much for the coil to move the focus the whole way up when it is in a vertical position. Haha, the software will soon be published inclusive the KiCad project for the PCB and this will allow anyone to rebuild this modification. Unfortunately it is complicated and I failed twice preparing the clock for the CD-CPU. Both CPUs are using the same oscillator at 5.8MHz. The new CPU has 16MHz but the CD-Player CPU can only handle up to 6MHz. The build in timer of the AT1284P is able to deliver 1MHz which would be enough and it was working in tests but when I reassembled Unit #3 for more tests the CD-Player CPU sometimes hangs and is not stable. Thus I had to change the design again and I am now using two D-FlipFlops to divide the 16MHz clock to 4MHz. It seems that this is working now and the CD CPU clock is stable. In addition, both CPUs are now more or less "synched" which ensures a proper data exchange through the I/O pins. It's also a learning process for me - and fun. Since I am not a professional I need more attempts to get it finally done (Rev.F right now). I prefer a plug & play solution without any changes on the original PCB. This makes it easier for everyone to replace it. @Mister X: you should try a resin printer. Gives very good results and it is easier to use. I often had problems with the calibration of the filament printer. Do not know the 3D doodler but it looks much more complicated. Doubt it can be used tp print precise gears. But for quick and mobile solutions it's surely great. Here are some pictures of the current progress for the digital counter. The D-FlipFlop for the CD-CPU clock is outside on the experimental board and I am running some more tests. And there's a video about the first part of restoring the tape mechanism...
I finally finished my CPU replacement hardware. The CD-Player CPU is working fine with the new clock at 4MHz and does not hang anymore. Well the software mights till need some tests but it is working great and I am very happy with the result. It's Christmas and as a small gift I've updated the website with more details to this modification. The kiCAD project can be downloaded and the software is available at Github. So anyone who is fancy to rebuild this modification can do this now However I am happy to help and might have one set of hardware left... Unfortunately the tape mechanism in my 4th unit is as good as dead and the whole thing will become a donor for the other three. There are too many things broken and it's simply not worth and too expensive to repair it. And there is one more video... again, another boring video. And the second part of the tape mechanism will follow soon...
On Christmas Eve my little bastard lighted up like a candle during my long term test. The smoking signals were not a good sign and I switched off immediately. However I have not found any issue, everything was working. Some days later everything failed and it was not working anymore. Seems that some components got stressed during the fire and finally failed. I had to replace two transistors and two fusable resistors on the power supply panel to get it back working. Took me another day of desperation. Now I am running again a long time test since a couple of days and everything is working great. Think the machine and I do have a kind of love & hate relationship. Also finished the second part of the tape mechanism repair video. I contains an unprofessional way (for the poor wo/man) to measure wow & flutter and to set the speed. Anyway, the method was working very good, better than expected. A recorded tape on the CD555 sounds also very good on my Denon devices. Fun fact - I had an artefact during recording from whatever source I choose. I took me about 3 days of measuring just to recognise that the signal was coming from my magnifier LED lamp Artefact from the LED lamp on the audio line about 37kHz ish...
A few days ago we were talking about AM radio and how my wife still listens to Smooth Radio on it in her car. Coming back from shopping with her we pulled into the drive and started getting noticeable interference on the radio exactly in time with the LED Christmas lights on the other side of the garden.
I get the same thing, I listen to radio everyday and sometimes I get interference, I always thought LEDs were fine and flourescents were the issue.
Thank you folks pointing out the Evils of LEDs! At Linn and Naim chatrooms it was a lot of warnings about light DIMMERS being in the same room as your uber-TurnTable, so I used battery-powered LED light above my Naim-ed LP-12 (which I joked cost more than my Saab)... it was/still is fun to see dimmable lights offered for turntables at presumably respectable catalogs like AudioAdvisor and MusicDirect
Another day another video. Well, was busy with other things. And there will be a third part of the tape mechanism because I had to fix another thing on this '/%$("="&$ mechanics...
Look what I've found on eBay... Finally, after years of searching for the Battery Pack it showed up on eBay. Ordered without thinking twice. Well, the fifth Sound Machine, I must be crazy. But, although it is not working, it's complete with speakers and in very mint condition. Really happy. Now going mobile but need to train weight lifting before... Full Story and more pictures btw. CDM2 already repaired and tape controller replaced with my replacement CPU and tape fully working.
When you think "wow the original gear is great and works like a charm" and at the very same moment it crashes and loses almost all teeth... It is such a pain to repair the Philips SER1 tape mechanism. But finally fixed all issues - and the gear was the least problem.
Finally found some time to finish the videos. Everything is working great. Except the battery pack - well another design flaw prevents that a small switch (to enable or disable the battery pack) is getting pressed correctly. It almost killed my machine as it led to some white smoke and its typical electronic burned odour. Some components got stressed but everything is still working. Now the latest videos and except for the speakers - that's it. Thanks for watching and sorry for everyone to whom I have been a pain in the neck with this project...
it looks like now you can make a Wall Of Boom out of CD555!!! to raise stress level in your cat he(she?) is beautiful, BTW how did you solve the least of your problems, did you buy or made that orange gear yourself?
