Reinventing the Swing Arm Mechanism - looking for allies!

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi Folks,
Why don't we try to reinvent one the mechanically most reliable mechanisms
In CD history, the swing arm? I did a long research in what we can get today
as traverse mechanisms without satisfying results, they all would not last long.
Aim is to develop a reliable swing arm which might even give the ability for swapping the diode and its housing for future repairs.
As far as I know, Patents have run out, technology has advanced with hard disc drives, and there are still enough players out there from the eighties for reverse engineering.
I had short contact with the developer from Philips about the question why swing arms were not build any more and to him, there was not technical reason for giving up this more reliable mechanisms.
I am not a technician at all, I barely understand how a tranny works. But I have access to the mechanical aspects of building a swing arm. For developing the servo mechanisms, for finding suitable lasers and anything related to electronics, I am looking for dedicated folks from the forum.
Aim is:
Using a sturdy, brushless motor for the disc
Swing arm with reliable bearing which does not need to be serviced.
On the - so to say - tip of the arm a detachable laser diode for future servicing/replacement.
We all know working phonographs from Edison's time. This mechanism should be the phonograph of the future.
Again, I do not have the knowledge for building this. Financially, I have some limited means. But maybe as a combined effort, we could establish a new High-End transport!
 
I have the impression that the CD format is dying. There are less and less CD stores. Media Markt closed its CD section in favor of the DVD section in my city. We have to listen to our ancient collections. The next generation listens to mp3 from online stream or from SD media, purchased online. As for me, I have a player with a swing arm mechanism, and stocked two other as donor for the same mech. So I feel safe for my life ;)
 
Hello lcsaszar,
I would propose not to discuss if CD-Players are needed or not, if CD is dying or not but to discuss
the possibility if those mechanisms can be built, to find solutions and to develop one as a group effort. My 5 cents:

I do not think that Media Markt is a good example. It has ever been a source of good music. Their customers were simply consumers of mainstream pop and Rock - easy to download. I only bought one CD there within 15 years - some Jazz they simply could not sell.

1)Especially unknown independent artists still produce CD - to them, streaming makes no profit. Also known bands complain. So for us music lovers who look for special music, CD´s will be always around.
2) My very first CD from 1983 still works like a charm - two 2,5" drives with media died after two years.
3) Even with Records: Many people buy to have something physical in their hands. The majority of record collectors simply collect - but listen to the download.
4) We could built a transport, that some national archive would be happy to use some 100 years from now:)
 
I guess we could only lobby Sony if we took the CEO as hostage - and he is Japanese, thus the individual does not count.:D
And Philips sold consumer electronics in 2014 - we would now have to ask Gibson Guitars - no joke but they also bought Onkyo.
I was in short contact with the co-inventor of the CD and inventor of the Compact Cassette Lou Ottens - he is very old now but was so kind to tell me in one short mail that he never understood why Philips gave up producing the swing arm - I guess simply because outsourcing the production would be easier.
Long story short - no, no help from the inventors and I also assume, the production lines have already become tin cans. But maybe, some retired Philips-employees are in this forum to give some clues?
There are many, many players with swing arms still around and many service manuals available from this era thus reverse engineering would be possible,
Again as far as I have researched, patents have expired.
 
It's a nice idea I guess, but your list of what you'd need to do is far from complete. You've left out the tracking drive, the prism, the photodiode array, the objective and focus lenses, the focus lens suspension and drive, as well as all the electronics needed. Check out the service manual for a CD100 to see just how complex it is to develop a CD player servo without custom ASICs.

I think there are better, more feasible, ways to go. A project I considered before I lost interest in CDs was building a CD Pro equivalent for a CDM-4. There are plenty of used working CDM-4s left, the TDA8808 / TDA8809 servo chipset is still obtainable, as well as the SAA7310 decoder. The only thing that put me off was the software side, I'm a hardware guy, I didn't even know where to start to write code to control a CD mechanism with a modern microcontroller.

Oh, and I think I heard somewhere that ditching radial arm CD mechanisms had something to do with them being not as suitable for some aspect of high speed CD-ROMs, but I'm not sure.
 
Ditching the swing arm in favor of straight tracking mechanisms for high speed read out makes no sense to me. As far as I understand, the swing arm does the the "long" horizontal movements, whereas the lens does the "short" horizontal movements - just like in an ordinary mechanism. If the arm did the fine movements as well, I could understand that the arm was too much mass to keep accurate tracking @20x speed.
The swing arms were very quick in accessing tracks. I guess the reason was simply easier outsourcing.
Looking through patents from 1983 up to today, the prefaces/reason for a patent was always "cutting costs/simplification! Philips had simply more suppliers for linear tracking mechanisms than their own breed swing arm
I am aware of the obstacles and how much is needed.
This is why I started the topic, to gather experts and enthusiasts
Assuming that the mechanics have to be produced/milled, could it be done with today´s parts - not leftovers from the beginning? The optical assembly will be a problem, we can´t build this, it has to be still in production.

First question : Are single-beam optical assemblies a necessety for a swing arm because of the radial path?
 
I use a CD94 as a transport because it still tracks dodgy CD's better than other players. On the other hand, computer drives seem even more reliable, so a player based on a DVD drive would seem a better bet.

Don't forget that vinyl was written off too and most people do not realise how good CD can sound. There are still those of us who prefer dropping a CD in the drawer and pressing "play" to faffing around with computer menus, so good luck!
 
Last edited:
I use a CD94 as a transport because it still tracks dodgy CD's better than other players. On the other hand, computer drives seem even more reliable, so a player based on a DVD drive would seem a better bet.

CD94=CDM-1 swing arm, built around 1987 as far as I remember? Which DVD-Drive can cope with that?
Computer drives are not reliable - any more. Need them for work, many died. First generations were great, like one from Pioneer out of a Mac G4: Two laser units for CD/DVD, brushless disc motor, brass, steel and metal everywhere, still works after 15 years, bit the rest was built while cutting corners.

Don't forget that vinyl was written off too and most people do not realise how good CD can sound. There are still those of us who prefer dropping a CD in the drawer and pressing "play" to faffing around with computer menus, so good luck!

So lets prepare for the hype, when people love watch the shimmer of CD´s and want to play them in standalone players.
Back to the question:
Are single beam lasers a necessity for swing arms?
 
Computer cd drives are FAR from as reliable as a CDM-0 thru 4. I have the requisite experience to say with 100% certainty that there has yet to be a more long term reliable optical drive than the CDM0/1/2/4/1mkII. The 3's are too rare for stats to confirm, and cdm9's are *definitely* prone to sporadic laser failures. But failure among the other cdm's, that aren't actually down to electronic faults, like the infamous crappy blue caps, is EXTREMELY rare.
This is largely due to incredibly durable laser diodes(whereas I believe the cdm9 laser may have been first use of the dodgy cdm12-type 'hologram' laser diode units). The swing arm's contribution to reliability is of secondary significance. I have never researched who made the laser diodes for the early cdm's, but it would be essential to this potential project to find out, and if that co. is stil making laser diodes.
As far as the mech, the swing arm clearly is the best approach for reliability, as no other mech eliminates belts & gears without then introducing significant lubricant issues(linear motors are great, but lub is utterly critical).
The remaining factor in the mech is the platter motor/bearings. Of the cdm's, the CDM1mkII has definitely the most durable, maintenance free, brushless motor. This would be positively my choice on which to base the "new" platter system.
Top loading would then finalise a completely maintenance-free, bulletproof mech. I would personally recommend going with a clamper that is high-mass & captive in the lid, like the Meitner CD3, but not everyone is as easily annoyed by clampers you have to remember to put on as I am.
I had immediately thought about the singlebeam vs 3beam issue, since ALL cd lasers in the last 20 years have been 3beam. While 3beam is incompatible with radial arm tracking, I am pretty sure that the E-F photodiodes can simply be ignored, and a 3beam thus could be used as a singlebeam, given the right tracking servo.
Oh, and just point of information, the radial arm is the only tracking mech device, and does, indeed, do the fine movements that are in other mechs done by tracking coils. This is obvious just by looking at a cdm2, which plainly shows that the lens can only move up & down via focus coil.
 
As far as I know, Philips CD mechanisms from the CDM-0 to the CDM-4 all used the same laser diode, a Sharp LT022MC. They all have separate laser and photo diode array packages, oriented at 90° to each other with a prism or semi-transparent mirror. The CDM-9 uses a combined laser / photodiode array package, all in plane, but I think it's still single beam.

I'm not 100% sure that a three beam laser will work with a radial arm. The three beams sit at an angle across a track. With a linear sled that angle would be static, but with a radial arm that angle will change as the arm moves across a disc. This might be an issue, especially at the inner tracks of a disc, when this angle would be at its most extreme.

But hey, I'm not trying to rain on your parade. I think there is still a place for a high end DIY friendly CD mechanism, and I'm a huge fan of the Philips CDM series. I just think it's worth considering doing something like using a Philips CDM-4 radial arm assembly with everything else being custom.
 
The E-F photodiodes are the only factor in practical use of 3beam vs single. Theoretically, as long as you're not using the E-F signal, the signal from the ABCD diodes will be purely from the center beam, in practice.
The only added components of 3beam are the beam splitter optical plate and the EF diodes, so weight quite definitely was not a factor. Lots of 3beam pickups are far, far lighter weight than e.g. a cdm4 laser. Philips abandoned the whole radial arm/1beam approach purely out of naked greed. Those awful cdm12/vam12 series Daisy Laser units from China are probably 1/10th or less the cost of the cheapest radial arm unit.
 
Maybe, but you have to be careful about idealised diagrams vs. real life. Some three beam pickups that I've seen don't have all six photodiodes that a simplified diagram shows.

Anyway, I'm not sure where that leads anyway. Are you guys thinking of putting a current production three beam laser head on a radial arm? Are any of them reliable enough?
 
Maybe, but you have to be careful about idealised diagrams vs. real life. Some three beam pickups that I've seen don't have all six photodiodes that a simplified diagram shows.

Anyway, I'm not sure where that leads anyway. Are you guys thinking of putting a current production three beam laser head on a radial arm? Are any of them reliable enough?

AFAIK, 6 diodes in the pd array are mandatory for 3-beam tracking.

I think that, yes, the idea is for some of us to get together to try to actually produce a new transport with radial tracking. I'm not (yet) a full-on manufacturer of gear, but I am up for helping with the project.

But, indeed, reliability is the key issue, which is why I asked about the maker of the laser diodes for the cdm0-4. As far as current production pickups, no, none are as reliable as the cdm0-4. The nearest is likely the (real) Sony KSS210(how surprising is that?). Considering the truly VAST numbers of them out there, I'd estimate, based on my experience, that they have less than a 0.5% failure rate over the course of a >20 year sampling. Still, I would have a hard time, personally, accepting a KSS210 as the basis of this project idea. But there may not be better choices that are still produced, unless Sanyo is still making SF90's(I've never seen a bad one of those that I can recall).
 
In the last year, someone around here found the patent document for the Philips radial tracking servo. I never had time to read through all of it & digest it, and now have no idea where to find it again. Maybe one of y'all could dig it up again. The odd thing about the system is the 400Hz oscillator in the drive side of the tracking. Didn't read enough to be clear about the purpose of the osc, but you can see it's signal when scoping the trkg servo output line.
I wouldn't automatically assume that Sharp discontinued the LT022MC. They may well have kept making it for some industrial purpose. Needs to be researched.
The DCD1800 radial system was very different, and actually pretty badly conceived. As I recall, it used the very heavy Toshiba laser also used in the Yamaha CD1(still a very sexy looking player), plunked into Denon's weird radial carriage, which was, as I recall, some sort of rim drive on a passive pivot. Fair to say it was Denon's first cdp, since the DCD2000 before it was pretty much 100% Hitachi.
 
Administrator
Joined 2004
Paid Member
Hi Salar,
Why don't we try to reinvent one the mechanically most reliable mechanisms
In CD history, the swing arm?
Maybe a linear tracking system. The biggest reason the radial arm was a problem for Philips was that it was very labour intensive to align once built. Each one was aligned by hand through a reiterative process. The adjustments affect each other. I don't know how many of you ever attempted to set one up, but the glass CD itself was somewhere around $400 ~ $500 CDN at the time. You need a line light source and it is more difficult than setting up the tape path on a 24 track R-R. It is a nightmare.
AFAIK, 6 diodes in the pd array are mandatory for 3-beam tracking.
I agree stephensank. However, linear tracking can be done with either single or triple beam systems. :)
This is largely due to incredibly durable laser diodes(whereas I believe the cdm9 laser may have been first use of the dodgy cdm12-type 'hologram' laser diode units).
The single beam systems run at much lower laser output power. The optics in a three beam need that laser to run hotter - still not a deal breaker. Why are we afraid of what is essentially a maintenance issue. Motors will also need replacement and the mechanism will require cleaning and re-lubrication. Home owners will not maintain anything unless it stops working. Just make the job less painful. Keep in mind that the customer is getting a new machine back. If they never break, there won't be anyone around to repair them - right?
You've left out the tracking drive, the prism, the photodiode array, the objective and focus lenses, the focus lens suspension and drive, as well as all the electronics needed.
No way! The only way to do this is to buy head assemblies. The servo chip set can use standard amps and filters. The DSP chip is the only critical part. Even that can be done using standard parts (not nicely though!).
But, indeed, reliability is the key issue, which is why I asked about the maker of the laser diodes for the cdm0-4. As far as current production pickups, no, none are as reliable as the cdm0-4.
They had their issues. The early days were rough for everyone making CD Players, don't kid yourself.
The odd thing about the system is the 400Hz oscillator in the drive side of the tracking.
The oscillator causes the head to reach the extremes of the track width. They make the amplitude equal for each side and thus center the head in the track. There are some disc issues that the Philips system didn't like, others that tangential tracking systems didn't like.

There was only one machine that would track anything and everything. The unit used a single beam laser. Any ideas on who made the best tracking CD Player of all time?
As far as the mech, the swing arm clearly is the best approach for reliability, as no other mech eliminates belts & gears without then introducing significant lubricant issues(linear motors are great, but lub is utterly critical).
Linear mechanisms are not that bad at all. The radial arm is rough on flexible PCBs. Gears haven't been any trouble since they got those recordable CDs figured out. All machines stop when they ingest hair, fluff and things I don't even want to guess at.
Top loading would then finalise a completely maintenance-free, bulletproof mech.
Nope. The tray is by far the most reliable way to load a CD - bar none. Loading a CD directly onto the turntable places a downward force through the turntable and onto the motor shaft. This can damage the brushes, but mostly the table will eventually either become warped or travel downwards on the motor shaft. This messes up the focus and typically isn't noticed until some CDs get scratched. Avoid allowing the user to directly touch anything that is critical. The disc table and disc motor are things that can suffer a human's touch. Also, once they see the laser lens, cleaning will take place. Ever see someone scrub a lens? Kinda rough on the suspension. Believe me, you want users to wipe the outside of the magic box, and that is as close to the thing as they get.

I would suggest that you aim for a linear tracking system using a linear tracking motor and a three beam head. The first pick would be a KSS-151A system. If the heads were available, the NEC single beam head is by far the best choice. They are long out of manufacture, but a beauty to examine and hold. The optical path is set in a die cast assembly about 3 ~ 4" long with mirrors and prisms mounted properly in a very stable system.

You know what? Once you have a single play system working, it is easily adaptable to 5 disc, 500 disc systems. Set up properly, reading the TOC is a quick spin up and down. Two things to not allow as features are random play or random between CDs.

How about this idea Salar. How about an archival machine. It is intended to take the information off a CD and store it while also sending amazing audio out the jacks. The DAC can be an internal card so people can have whatever flavor they wish. External DACs not supported ... well, okay. The internal card idea allows the DAC to have jitter free signals in. Once the DAC is outside the machine you are already behind the "8 ball".

Sorry for the long post, but there is a lot of material in this thread already.

-Chris
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.