PS Audio Lambda Drive Disaster

I recently purchased a CDM 9/44 with the plan of transplanting the laser pickup assembly into my ailing Lambda's CDM-9Pro (9/60) I have disassembled the tray mechanism, laser and controller board in the past with no difficulties. I have a wrist strap and static controlled environment.

I disassembled the 9/44 which was in the factory wrapper, and carefully removed the two torx head screws that retained the pickup assembly in the mechanism. I then did the same with the old pickup and installed the new one in its place.

I reassembled everything, and voila... Nothing.. Would not read the disk, would spin for endless ages and once in a while read the TOC. Will not play the disk.

I reinstalled the old pickup which was working prior to disassembly.

Again Nothing!

I have had the whole thing apart repeatedly and have done progressively more damage with each passing attempt. I checked for cold solder joints, I checked supply voltages. I got the data sheet for the TDA8808T which this mechanism uses. I looked at the ac signal at the output of the diode amplifier, looked ok?

Now after another ill-advised attempt it gets to the outer edge of the disk and sometimes doesn't know when to stop, with horrible results. Also the spindle motor stops and then runs backwards when this happens. This behavior is new and probably the result of some of unseen damage done today as I worked on it.

I have 25yrs of engineering experience, some level of patience, fast scope, but no schematics, and they aren't seemingly available.

Anyone have any ideas whatsoever? Schematics, service notes, even for the theta version??

So far it does know there is a disk present, it does correctly detect the disk size, although that funky focusing behavior at the disk edge. (Shouldn't get there should it?) The motor runs ok.

I am actually going to buy something else, as I said in a previous thread I have just about had it with this transport. A shame as it has always been the best sounding one I have owned.

Thanks, Kevin

:bawling: :bawling:
 

kevinkr

Administrator
Paid Member
Here is a little more information to go on.. I carefully cleaned the board tonight with some alcohol to remove flux residue from the board after resoldering a few connections.

The mechanism is about back to where it was last night. The motor spins up when I insert a disk in the tray, runs for about 10 seconds and returns an error to the display.

I removed the disk and position the pickup so that I can see it, when play is pressed it goes through the normal focusing routine and I can also see the laser output using a white card.

Supply voltages all appear to be correct and the circuitry controlling the spindle motor, focusing coils, tracking coils, and laser all appear to function in what appears to be an expected manner. I was able to observe a nasty looking waveform on the HF output of the diode amplifier which is a TDA8808T, basically a 0.5Vp waveform that appears to be the processed diode output..

I know some of you have encountered problems with the lambda and theta drives of this vintage, and was wondering whether any one has any experience resolving these issues.

The board on the mechanism is not made by Philips and has a moderate component count. More details to follow if warranted.

Kevin
 

Extreme_Boky

Member
2003-12-07 11:57 am
Supply voltages all appear to be correct and the circuitry controlling the spindle motor, focusing coils, tracking coils, and laser all appear to function in what appears to be an expected manner. I was able to observe a nasty looking waveform on the HF output of the diode amplifier which is a TDA8808T, basically a 0.5Vp waveform that appears to be the processed diode output..
You are talking 0.5V pp of an aye-pattern signal, aren't you?

This signal (could anyone with a circuit diagram for this model confirm this?) should be around 1.3 - 1.7 V pp. To adjust - turn the little trim-pot located on the laser pick-up PCB clock-wise.

If you can’t obtain the proper info on the proper pp level, try anyway… If it does bananas – set it back to what it was before.

Good luck!
Extreme_Boky
 

kevinkr

Administrator
Paid Member
Hi Boky,
According to the TDA8808T data sheet the HF output should be over 1.0V minimum, and it is definitely much less than that. There are definitely no adjustments on the board, however I have not checked the pickup assembly that closely - however I believe that is also the case. I can take a look at the old pickup to see whether there is a pot.

Schematics are almost impossible to come by for this unit as PS Audio and Theta Digital (Theta made the PSA digital components) were owned by a holding company at the time until around 1999, when it went bankrupt someone destroyed the design information. I have talked to both manufacturers about this.

The board under the mechanism appears to have been made by Daisy Laser based on the markings, but they are not helpful either. The board has got relatively few components - some driver circuitry for the hall effect spindle motor, tracking and focus coils, laser driver and diode amplifier.. Everything else from the radial error correction IC (TDA 8809T) and SAA7310 are on the large board along with the master clock, mpu..etc

Kevin
 
I have the same problem.....

I have the same problem with a Lambda I have been trying to sort out. I also have a DATA. If I swap the loader assembly including PCB between these two machines (the Theta is fine) the problem stays with the lambda. I therefore do not think the fault is on the small pcb but the theta designed larger pcb with the dig outputs and SAA7310GP. I have replaced all the regs and lytics on this board without joy. I have also swapped the 7310 and 8809 but it came from a non working player so I do not know if they were good chips. I have swapped the 3702 and 5664 chips as well.
I have given up for now.....

I really do not know what to say except I hope you find the fault and tell me what it is....

Guillaume
 

kevinkr

Administrator
Paid Member
Hi Tubenut,
I hope eventually to get some schematics for this thing. I suspect the problem in mine is also perhaps in the main board. I have found sources for most of the ic's in the machine, all of course are obsolete..
The drive is redundant now as I recently purchased a low hour theta data basic. (I hate Philips, but I can't afford something a whole lot newer so my hands are tied. LOL)

Kevin
 
Well I got some bad news today, looks like the theta data I was getting has the dreaded drawer gear disease and I had to cancel the deal so I am back to square one. I thought that the mechanism and loader would be useful to trouble shoot the lambda as well as giving me a decent source to use in the interim..

I am going to look for other options after the holiday and see what I can come up with. In the meantime I am not experimenting too much with the drive as I don't want to make it any worse than it is already.

I have heard quite a lot on a board located in Singapore about having to fiddle with the HF filter response when replacing the pickup assembly with the 9/44. It may be that the pickup for the 9/65 was not exactly the same afterall. The comments seem to apply to both the theta and the psa drives.

Unlike earlier drives there are absolutely no adjustments anywhere on the pickup assembly or any of the boards.

My suspicion is that a lot of passive components have "aged" over the 12 yrs or so since the boards were manufactured. The drives were manufactured over a range of years, but the few I have seen have many components that date to as early as 1992. Mine was made in 1995 according to the guy I bought it from 8yrs ago, and the newest component date codes as well as the board date code bear that out.

The IC's are all discontinued but most are available from several sources in the UK - I have not had much luck finding them in single lot quantities elsewhere.

The Lambda design leaves EVERYTHING powered except the laser itself even when the unit is turned off, and unfortunately if you know anything about component life that means that EVERY component in the thing is now well beyond end of design life.
Mine has been continuously powered for most of the last 8yrs and probably for most of the 2yrs before that - that equates to 88,000hours! Way beyond the design life of every resistor, capacitor and semiconductor/IC in the unit. IC's and semiconductors have an MTBF, although I do not know it for any of the devices in this unit for consumer grade stuff it probably is much less 100K hours.

It was an unfortunate design choice, but PSA claimed that the Lambda took 72 hrs to reach sonic perfection after power was applied and that may have been the motivation for powering everything on a continuous basis. (I think they could have accomplished much the same thing by just powering the master clock continuously.)

The Sony SACD player, Sony DVD player and Panasonic LD Player I own all kill the power at the primary of the power transformer. None of them sound anywhere as good as the PSA Lambda Drive/Ultralink II combo however!

The first owner of a theta/PSA digital product of course would not care because in the USA at least no one keeps them for long because only the toy boys have that kind of money to invest - it's us second tier semi-impoverished owners who bear the brunt of the problems with these products. Some of us are likely to be able to fix them in at least some cases as well.

Well I guess I am just ranting now. :D

I am going to fix this thing eventually, it is too good and I am too cheap not to, however it may take a while.. :bawling:

Kevin
 
Kevin. I have done a few laser swaps on Theta Data units with no problems. The Lambda is the first that did not like it.... The problems I had after the swap were the same as your Lambda...
I must llok at the 2 boards again next to each other and see if I spot any differences anywhere... Except for the fitted AT&T
I can not get the Data loader to work in the Lamda but I can get the Lamda loader to work in my Data...

I think you may be on to something with the HF story but I have replaced all electrolytic caps on the Lambda as well as regs so I am not so sure about component ageing.

PM me!!! The Data you turned down may be rescueable!!!
 
Hi Tubenut,
Your email is not available, but you can email me through my gmail account if you would like. direct.heated@gmail.com
The owner of the theta wanted to get it repaired prior to sending it to me, depending on its feasibility I may still get it.

The real issue is that I need something now, and I already have an ancient and unreliable Lambda and I think perhaps I need something just a little bit newer than my teen-age niece.. LOL

The best of all worlds is I would find the problem with the Lambda and fix it. I am waiting to get my hands on the schematics before I proceed to the next phase, inspection of the schematic may make it obvious as to where the problem is most likely to be, and I won't have to probe every single connection in existance to know what it is..

There must be some specific difference in the design of the lambda board and the theta? I haven't seen the theta so I don't what the differences are if any.

Kevin
 
The repair should be simple and just the tray gear on the Data. It is getting the gear that is fun..... Check your e mail.
I will try and send a decent Photo of the Theta if you want but I need a few days for this.
Perhaps it is something really small like resoistor values setting HF gain that are different then the ones on the Lambda.
I have found that many of the 9/44 units sold now are refurbs or rebuilds or copies. They have no Philips stickers anywhere and do not read CDRW. CD and CDR no probs. Perhps you have one of these refurb units (the one used in my Lambda is) and the lower HF gain is not enough to get this lesser laser going?
I have had mine playing once or twice but after skipping a few tracks or stopping and tryiong again it would fail to read again....

Hope this helps.

Guillaume
 
Hi Tubenut,
I supposedly purchased an original brand new 9/44, the mechanism itself was brand new and has a Philips label on it. The box said made in Japan on it and has other oem markings. The pickup itself is not marked but is absolutely identical to the one that was original to the Lambda. It had the anti-static shorting bar installed and it was obvious that it had not been plugged into a FPC connector either. The guy I bought it from has a good reputation and it was represented as new and not a refurb. Neither the old pickup which still worked or the new one function properly.

I think that the board itself is hosed as it does not appear that any valid HF is now getting to the main board. I may have ruined the FPC connector that the pickup plugs into and I am looking into the possibility of getting another - so far not encouraging as the pitch and configuration is quite obsolete. I have looked at Hirose, but not yet at Samtec types. Others will undoubtedly come to mind.

I cut what appeared to be a dendrite to one of the pins, it was not a proper ground connection, but apparently that was in fact what it was. I removed the connector to investigate as I have no information at all as you know. The player would not even spin up with that connection cut, so I resoldered it to ground and it is behaving nearly the same as before now.

I have heard a lot about that gear issue in the theta, actually the psa drive and the sonic frontiers drive should have the problem too, but apparently don't. (They use similar mechanisms) Something about the lubricant used rotting the plastic gear teeth.

Gears for the Theta are unobtainium, Theta has no more in stock and the once viable substitute talked about on other forums is no longer available either. I just don't want the hassle of looking for parts for superannuated machines.:D Spare parts for me are hard enough to come by. ;)

It would be cool if I could get a working board from another mechanism though..

I used to stick a little pouch with a schematic inside some of my projects so that years later I would have it if I needed it. Saved me a few times. ;)

Kevin
 
Theta gears are not quite unobtainium... Got some fresh a month or so ago via a local philips agent. Order it as spare for the Philips CD930. Do a search for CDM9 on the forum and you will also find the Philips part number... Did explain morte in PM sent to you.
I know of stripped Lambda ggears too. The new gear I got is a lighter coloured plastic and seem less brittle, even the parts of it not in touch with the grease.

I can not speak for your player but my lambda has a fault on the main pcb as I can take the whole loader on its stilts and plug it in to the Theta and all is well so not likely a fault on the foil connector...... The fault symptom you describe is identical to mine.

I am sure it is something stupid and small and one day may figure it out...
 
Hi tubenut,
Thats a hot tip, I will look into it.. Who knows... LOL
The gear in my lambda is in excellent shape, there didn't appear to be much if any grease on it.

I hope it is something obvious, I don't think the flex cable from the pickup is bad, and I thought after inspecting the connector that is was still ok despite its removal.

It is quite possible that there is something wrong on my main board as the diode amplifier is dependent on the uP to set it up correctly amongst other things.

Kevin
 

Extreme_Boky

Member
2003-12-07 11:57 am
Hi guys,

It puzzles me that many people have troubles with this particular mechanism. It would be a good idea to start from pick up end and fault-find downstream. I suggest you to check conductivity between connector pins of all ribbon (and other) cables that run from pick-up to that custom PCB, and from there downstream to the main PCB. Check one pin at the time and check that there are no shorts between adjacent pins as well. Once you are happy and 100% sure that all connectivity is okay (do not use diode-check with beeper setting on your multimeters – always use resistance measurement setting and look for 0 ohm conductivity), start checking signals at appropriate pins for correct amplitude and waveforms. This HAS to lead to success.

Could you guys also e-mail me readable copy of circuit diagram - I'd like to have a look... (if you can get it....)

Regards,
Extreme_Boky
 
Hi Boky,
You are of course right, I don't yet have a copy of the schematic, and I don't know if and when that will come to pass. I have done what you suggest, however without knowing what the expected levels out of those photodiodes are I can't really get rolling.

Also the design is such that the mechanism can't be operated in positions conducive to debug. :xeye:

I can say that the HF signal does not look healthy in the least, and that it is the same with both extant pickup assemblies - it doesn't look like anything recognizable at this point. I am not sure that I didn't damage the FPC connector, but all other verifiable signals are getting to the pickup, focus coils, tracking coils and laser are all working. I am not sure what is going on with the diodes, but I suspect that they are connected as well since the mechanism senses the presence of a disk when inserted. I suspect an issue with the TDA8808, but until I have more information I am not going to do anything about that.

I have the spec sheet for the TDA8808 so at least I can identify which pins are the diode inputs and from the specification infer roughly what the diode levels should be...

Kevin
 
Hi all,
I did manage to get schematics for the main boards of the Lambda, (will scan to pdf when I have the opportunity) I don't however have the schematic for the board on the loader or of the pick up assembly itself. Either would be useful in my quest to fix this thing.. Ultimately neither the board nor the pick up are so complex that I could not trace it all out manually if I had to, just slows things down a bit..

Issue seems to point towards diode array not being connected properly to the diode amplifier (bad connector) bad diodes (horrors!) or a bad/malfunctioning TDA 8809T for whatever reason.

Kevin
 
Well the saga continues. Today I checked the 4 LF diode outputs from the pickup and the HF output and I am happy at least to confirm that the pickup appears to be good. I can see the HF on the 100K pullup resistor to the +5V supply as well during the initial toc search and at the issuance of the stop command as well. The HF looks pretty clean, and is > several hundred millivolts. It vanishes during the seek sequence, but I think that is simply because the pickup is in an invalid location.

I also discovered that the HF signal is not appearing at the input of TDA8808T at all and this could be due to the NPO ceramic smd (0805) cap being bad or it could be due to a breakdown between some substrate diode and the input. I don't know the value of this cap as it's not marked, I can get a range of appropriate values from work, anyone have input as to what a reasonable value for this cap might be? The data sheet does not include much application information and I have not had much luck finding apps notes on this chip. I am assuming that this input (HF) operates in the voltage domain, not current.

I ordered TDA8808T and 8809T from the UK about a week ago so I should have them soon and will replace the offending chip if the cap does not turn out to be the culprit.

Anyway that's the latest. As always any input or comments are welcome..

Kevin