alexfree wrote: ↑January 13th, 2023, 9:58 am
After using these drive for a bit, I have some more info I can share.
My SCPH-1000 still works really well with it. There is something really weird about picking up the wobble on the super old console models that have manual Bias and Gain however. Sometimes it just doesn't pick up the SCEI signal. This is not that big of a deal though because you only need to do this once when you boot or spin up a Japanese psx disc. It again reads the disc contents perfectly, just the actual wobble is iffy.
All my SCPH-1001s just can't pick up the wobble at all with the drive. No amount of bias and gain tweaking has made it work once on 2 different consoles of the same model.
My PSone SCPH-101 works with the drive every time to pick up the wobble.
So I think this should be considered. If you console has manual Bias and Gain adjustments these drives may not be able to pick up the wobble. The early PU7 at least can pick it up most of the time. My PU8s have never picked it up.
Any Console with auto bias and gain however will work every time to pick up the wobble.
If your console is chipped, picking up the wobble doesn't matter in this case. If you have a 1001 or any USA or PAL console that can be unlocked, it doesn't matter as well I'd you use a Cheat Cart to boot into i.e. tonyhax international rom or use something like freepsboot since the drive just becomes unlocked immediately anyways.
I got the same problem on my 5502. I was really intrigued about that so I hooked up my oscilloscope and compared the differences.
I figured out the pin 5 on the original drive's connector called NC (which would stands for "Not Connected") is actually used by some motherboard revisions as the signal that will be filtered to get the wobble data.
This is also confirmed by other people (pages 7 and 8):
https://bbs.a9vg.com/thread-5214563-7-1.html
From Google Translate:
"the No. 5 line of ps1 is a genuine signal line. If it is connected to the line on 213c, it can run without IC."
"My understanding is that if you connect the genuine signal, you can run the game without ic"
The term "ic" designate a modchip.
Here is what this signal looks like (blue = pin 5, pink = 22kHz filtered Tracking Error, yellow = demodulated SCEx):
KSM-440-osc-capture-1.png
KSM-440-osc-capture-2.png
KSM-440-osc-capture-3.png
By looking at the 5502 service manual, this pin 5 is connected directly to the "Push-Pull" potentiometer which allows to set the proper filter threshold for wobble detection (too low = weak 22kHz, too high = saturated 22Khz + "noise").
That means, if the motherboard has the "Push-Pull" adjustment (at least PU-8 (RV702), PU-18 (RV703), PU-20 (RV703)), the signal on pin 5 is needed and the wobble is not detected through the Tracking Error signal made by the RF amplifier (E and F photodiodes subtracted).
This explains why the wobble detection works on a SCPH-101 but not on a SCPH-1001.
I tried to hook up the Tracking Error signal (CL709 on the 5502) directly on the pin 5 but the wobble is not well detected that way. It is sometimes detected when the "Push-Pull" potentiometer is set to the max but this is not reliable at all.
By checking some random Sony CD RF amplifier datasheets, I saw that the CXA2581N has another signal I didn't know before called Center Error which actually creates a sort of Tracking Error signal but only from the 4 photodiodes A to D: CE = (B+C)-(A+D).
CXA2581N-Center-Error.png
I started to think that this could be the signal on pin 5.
To verify that, I made a simple version of that schematic to try to replicate the CE signal and see if the wobble can be found.
After some gain tuning by choosing proper resistance values, I could see the 22kHz signal of the wobble, my 5502 detected it well and the game launched successfully, no hesitation!
KSS-213C_To_PS1_v2.0-osc-capture-1.png
KSS-213C_To_PS1_v2.0-osc-capture-2.png
No need to adjust the Push-Pull potentiometer on the PS1 since we can adjust the Center Error gain.
As we can see, the signal is not the same as the original and not perfect. This is maybe due to the (common and low-cost) operational amplifier LM324 I am using that is not fast enough.
There are faster opamps but I only have the LM324 on hand.
Or maybe the KSM-440 has a modified Tracking Error/Center Error circuitry.
Here is the schematic on the modified converter board and what it looks like:
KSS-213C_To_PS1_v2.0.png
KSS-213C_To_PS1_v2.0-prototype.jpg
The best value for the potentiometer RV1 is around the middle (50k).
I can't confirm if it works properly on others board revisions than PU-18 but I am almost sure it will!
You do not have the required permissions to view the files attached to this post.