To update on this, if anyone has any suggestions;
I managed to buy a grey Playstation in the similar original style at the local Goodwill Computer Depot (Which has gone heavily downhill in the last few years). It cost me $25. It was a risk, but the crummy lasers from China cost about that much so I though I'd take a chance.
Not being terrifically familiar with the subtleties of these models, this one has a rather different motherboard and arrangement of the laser attachments. It is an SCPH-9001. It looks cost reduced, without the AV direct ports in the back.
I found that the laser had some extra plastic that made it impossible to fit in the debug. I trimmed that using some snips and fit it in. Unfortunately the laser worked exactly as poorly as the original laser in the debug. It basically could partially load games, like load screens and menus (some of the time) and it made it to a few full games exactly as the original laser did. And where the unit I just bought loaded retail games just fine, when the laser was transplanted, it did not want to load retail games.
SO, by elimination it seems like the problem is not the laser assembly. The only thing I feel comfortable doing is finishing the recap on the debug power supply (sadly I'll have to order just two caps from Digikey, unless I wait for the next project that demands caps to order them together to save a bit on shipping). Since the debug motherboard has 36 surface mount caps, often located close to other tiny surface mount resisters and components, I feel like I could do more harm than good if I tried to do that big of a job. I have successfully recapped other surface mount stuff like an Amiga 1200, but that is like half the caps with a lot more space to work.
Another flaw of note on the debug is that it has messed up sound. The start up sound is static, it beeps a lot, and the sound doesn't work for most games. If I can find a music CD, I 'll try one.
Any suggestions appreciated.
Unfortunately, swapping over the optical pickups on these units is not an entirely reliable way of determining if they are good. Specifically, the models with digital focus and tracking servos (start from the PU-18 board, SCPH-55xx series) are a lot more tolerant of poor signals than the older ones, so just because a pickup works in one of the later models shouldn't be taken to mean they will work on an older board. The only thing you can be sure of is that if it doesn't work on the newer boards then there is no chance at all of it working on the older ones.
If you have an older retail console around (I.E. SCPH-100x or the Japanese SCPH-3000) then that is a reasonable confidence test since they have the same analog servos that the debugs do.
You can also swap out the power supply between the retail and debug console for testing in this case (all the debugs use a 7-pin power supply) - but my general experience, even with very old consoles, is that the PSU is not typically the problem. One part of the circuit that is very sensitive to bad caps is the filtered 3.5V supply used by the RF amp chip and the pickup - if you have access to a scope then check the supply on pin 2 of the optical pickup FPC - it should have a very clean 3.5V supply on it - if you can see noise, this will screw up the CD reading for sure, and normally means you need to change the caps in the area near where the PSU plugs in.
The corrupted startup sound is unlikely to be related to the CD problems - the startup audio is generated by the SPU, which is downstream from the CD circuits. About the only thing they have in common is the 5V regulator - this provides the power for the audio and video DACs and is also sub-regulated down to provide the 3.5V_SER supply for the optical pickup I mentioned earlier. This comes from a 3-terminal regulator (78M05 type) mounted on the back on the PCB near the connector for the PSU.
If you find you can play audio CDs but nothing will boot, there are some other parts to check - if the drive doesn't seem to be getting to 2x speed then check the BTL driver (Rohm chip with heatsink tabs and a BA prefix mounted on the top of the board) - I've seen a number of them go bad and fail to supply enough current for 2x speed. If it plays audio and the drive goes to x2 but can't read data then suspect the CD interface chip (CXD1199 or CXD1815) and the RAM associated with it (on the back of the PCB) - especially if it's the original CXD1199BQ version.
Having said all that, these things are generally pretty reliable, and with the exception of the optical pickup there really isn't any part that has an obviously excessive failure rate - which is nice, but it does make suggesting possible faults harder.