PS1 PU-22 Modchip Install (Causes disc drive to stop working)

Discussion in 'Modding and Hacking - Consoles and Electronics' started by deserada, Aug 1, 2015.

  1. TriMesh

    TriMesh Site Supporter 2013-2017

    Joined:
    Jul 3, 2008
    Messages:
    2,324
    Likes Received:
    750
    The problem with this test is that it can mislead you - I've seen quite a few of these units that had enough gain to read audio CD-Rs at 1x but not enough to read a data CD-R at 2x. It does work the other way around - if the pickup can't read an audio CD-R, then there is no way it will work with a data CD-R.
     
  2. deserada

    deserada Active Member

    Joined:
    Jan 26, 2014
    Messages:
    35
    Likes Received:
    1
    Just tested an Audio CD on both, and it appears to have the same symptoms as the first time I boot it after installing the chip.

    "The head tracks rapidly (significantly moreso than it would normally) for about 4 seconds, then it stops suddenly and returns to the BIOS/Main Menu."

    It does this on all successive boots, unlike previous experiments when it is only the instance where the console is first turned on after installing the chip. Again, with Retail discs and Backups it doesn't even attempt to read them. It spins at a slow rate for about 10s, then displays the "Please Insert a PlayStation CD-ROM" message.

    As before, the consoles behave normally when the chip is removed.

    Attempted to add shielding over the chip and under every wire, but this did not affect any of the symptoms.

    I'm going to try Mord.Fustang's port of MM3 next, and see if anything changes. Failing that, I'll relocate the chip. If there's an ideal placement for modchips in your experience, let me know.

    If none of that works, I'll do my best to probe all the signals and look for more clues.


    Edit: Fudged with my programmer's settings and when programming between 4.8 and 5V the chip will allow the console to read audio CDs. Before it was using "low voltage programming" and although I'm sure I could get that to work, I found a workaround. Also turned off "fast programming" to no effect.

    Still can't read retail and backups. Retail discs show up as a single track under the audio player, but I didn't "play" them or try backups there either. I assume the launch model exploits have no use here.

    If I install a MM3 chip in a socket wired for Mayumi 4.0, do I need to disconnect the external clock signal? I assume not, but to be safe I will disconnect it if no one can provide an answer.
     
    Last edited: Aug 12, 2015
  3. Mord.Fustang

    Mord.Fustang My goodness, it's nipley out!

    Joined:
    Feb 17, 2013
    Messages:
    818
    Likes Received:
    182
    I've done this before when testing chips in a socketed system and it should be fine. Either way I can't see it causing damage.

    Hopefully MM3 gives you better luck. As stated, I haven't tested done tests in that model specifically so I hope it works but there's a small possibility that something isn't playing nice. Trying another port or chip altogether is a good idea.
     
    deserada likes this.
  4. deserada

    deserada Active Member

    Joined:
    Jan 26, 2014
    Messages:
    35
    Likes Received:
    1
    Good news, it looks like the MM3 port is working better, although unstable.

    Retail discs will now boot, but it is once again showing symptoms of interference on the /DOOR line. The optical drive will get up to speed and slow down during reading without being prompted. The motor sounds as though it is "revving" its metaphorical "engine" when this occurs.

    This occurs on both consoles, and with four different chips. I don't believe the lot of chips I have is bad.

    Backups will not read, although they now display the PlayStation splash/logo, so it appears the code on the chips is doing at least that much. The console seems to remain on the PlayStation splash indefinitely. I will test discs made at a slower burn rate, although previous experiences with Dreamcast consoles showed little difference between 10x and 4x burns.

    I don't believe it should matter, but the backup I have been testing is NTSC-J and this is a NTSC-U console. Further tests will use a NTSC-U backup as well.

    Edit: After some research, I am lead to believe it either has to do with the strength of the laser, or my burning process. The /DOOR line interference can probably be mitigated if I disconnect the clock signal or isolate it, as the two come very close and cross at one point.

    Edit 2: One console has the laser diode more or less floored at 610Ohm, where the other is around 920Ohm. It doesn't look like I need to be adjusting those.
     
    Last edited: Aug 13, 2015
  5. deserada

    deserada Active Member

    Joined:
    Jan 26, 2014
    Messages:
    35
    Likes Received:
    1
    Well, I'm not sure what to say.

    Tried switching from TDK to Taiyo Yuden media, and after testing that everything seems to work.

    Are Taiyo Yuden discs capable of magic? The world may never know.

    This is a unsettling finding, however, and I still consider the configuration unstable.

    I'll do my best to run the consoles through their paces, and will report back with any findings.

    Thank you everyone who has helped! Hopefully this thread will be of use to someone in the future, although I would have liked to provide some more conclusive evidence as to the cause of the symptoms I experienced.
     
  6. master991

    master991 Enthusiastic Member

    Joined:
    Jun 4, 2012
    Messages:
    597
    Likes Received:
    28
    I thinks it's your burner (of chip and even CD-R burner) the real problem. Because I've used those modchips and they works without any problem....
    Plus you should set the right fuses (you must never use the low voltage programming for example)
     
    Last edited: Aug 14, 2015
  7. deserada

    deserada Active Member

    Joined:
    Jan 26, 2014
    Messages:
    35
    Likes Received:
    1
    I've been programming my chips at 5V, thankfully I caught on to that pretty early in the process.

    Were you using these exact ports of MM3 and Mayumi 4.0, for the 12F683 PIC chips? (Mord.Fustang's ports, that is)

    Some backups work just fine on one console, others have serious issues and will only boot about one in ten times. Some won't go past the opening cinematic, if they even make it past the PS1 boot splash.

    When a console is struggling to read a disc, it tends to make that "revving" sound. I can't tell if it's successfully reading, or trying to read the same areas over and over (it rarely tends to get out of that "revving" loop). I noticed it can get into a loop on the loading screens of some games. At times it will just stop reading, which requires me to reset the console. Sometimes it won't read anything from the moment I boot the console, whence I have to reset it again.

    My second console seems to hate everything I feed it, other than audio CDs and retail discs. It will load parts of some games, but seems to always choke on FMV and CDDA. Before you say I need to make slower burns, these same backups worked on the other console.
     
    Last edited: Aug 14, 2015
  8. master991

    master991 Enthusiastic Member

    Joined:
    Jun 4, 2012
    Messages:
    597
    Likes Received:
    28
    i've tested on all motherboard from 550X to 900X using the 12f629 porting HEX.
    Have you tried with a decoupling cap?
     
  9. deserada

    deserada Active Member

    Joined:
    Jan 26, 2014
    Messages:
    35
    Likes Received:
    1
    No, I did try using a 5V supply instead of 3.5V off of the parallel port.

    Both of those supplies looked pretty stable, the 5V moreso.

    Which lines would you recommend decoupling caps on? Did you end up using any in your testing?
     
  10. deserada

    deserada Active Member

    Joined:
    Jan 26, 2014
    Messages:
    35
    Likes Received:
    1
    Success!

    Having wired up the same circuit on a 9000-series (PU-23) PS1, I found absolutely no issues.

    I surmise that the problems I was experiencing before either have to do with the placement of the chip, or the pinout for the install. On the 9000-series the chip is being supplied 5V and better isolated from sources of interference.

    Perhaps later I'll try relocating the chips on the 7500-series (PU-22) boards to test the theory.
     
  11. Mord.Fustang

    Mord.Fustang My goodness, it's nipley out!

    Joined:
    Feb 17, 2013
    Messages:
    818
    Likes Received:
    182
    Which chip and port are you referring to?
     
  12. TriMesh

    TriMesh Site Supporter 2013-2017

    Joined:
    Jul 3, 2008
    Messages:
    2,324
    Likes Received:
    750
    I still think it's simply that the CD front end amplifier is picking up noise from the wiring - I always made sure that I kept the wiring as far as possible from that RF amp chip for exactly that reason.

    This also makes sense of the better results with the TY CD-Rs - they are one of the last companies still using cyanine dye, and it tends to produce better read amplitude than the other dyes do, so there was more signal to get above the noise (if you're wondering why pretty much everyone else has changed to phthalocyanine despite the poorer signal, it's because it's more stable to UV light).

    Depending on how bad the interference was, the effects could range from merely a small reduction in read speed to a complete inability of the drive to maintain lock, forcing the drive to go through its initial disc acquisition procedure again.
     
  13. deserada

    deserada Active Member

    Joined:
    Jan 26, 2014
    Messages:
    35
    Likes Received:
    1
    @Mord.Fustang The configuration I finally had success with is your port of MM3 on the 12F683. I'd like to test Mayumi 4.0 today.

    @TriMesh I agree. Your advice was invaluable throughout the troubleshooting procedure and I'm very thankful you mentioned the location of that amplifier, or else I may never have been able to solve this. That last line concisely describes every single problem I experienced.

    Edit: Was unable to get either of the PU-22 boards to work with this configuration despite moving the chip to several different locations on the board. Same symptoms each time.
     
    Last edited: Aug 15, 2015
  14. TriMesh

    TriMesh Site Supporter 2013-2017

    Joined:
    Jul 3, 2008
    Messages:
    2,324
    Likes Received:
    750
    This is a bit of a necro, but I happened to be looking at a PU-22 today, and thought it might be useful to show the signal you get out of the optical pickup.

    This is a SCPH-7503 (Asian model - NTSC:J but with an English boot ROM and a universal PSU). It's never been modded, and it plays original discs fine - I tried booting a CD-R with a cheat cart, and it was very slow.

    So I connected up a scope and had a look at the signal out of the RF amp. I'm using an old analog scope because they work really nicely in this application - you can use a cheap DSO, but you have to fiddle with the persistence controls to get a good result. High end digital scopes work well too, but if you have one of those then I'm sure you already know how to use it.

    [​IMG]

    The first thing about this waveform is that it's only 800mV p-p - which is out of spec, since it should be at at least 900mV. Otherwise, it's pretty good. You can tell a lot about the condition of the pickup and transport from this waveform. The first thing is simply how big it is - and in this case, bigger is better. The second is the width of the scope traces - since this is a composite image of a large number of waveforms the trace is wider - and the width is proportional to the noise level in the system. You can also see that the traces become wider towards the right edge of the screen - this additional thickening is from jitter, and is a good indication of the condition of the motor and mechanics. The spaces in the trace on the x-axis line are the 'eyes' that the eye pattern gets it's name from - and as the noise and jitter gets worse, the size of the empty area in the eye gets smaller and smaller until eventually it closes completely - and the data becomes impossible to recover.

    For reference, the test conditions here were:

    x10 probe / 200mV/div (including probe factor) / no mod chip / original CD / game running in attract mode (2x speed)

    Swapping another optical pickup into this unit gave a 1.1Vp-p signal, and that one would happily boot CD-Rs using a gold finger cart.
     
    Helder, deserada and master991 like this.
  15. Mord.Fustang

    Mord.Fustang My goodness, it's nipley out!

    Joined:
    Feb 17, 2013
    Messages:
    818
    Likes Received:
    182
    Thanks for taking the time to post that TriMesh!
     
sonicdude10
Draft saved Draft deleted
Insert every image as a...
  1.  0%

Share This Page