› Foros › PlayStation 2 › Modchips
Finally... I finished my part of the puzzle.
Maybe not the way I wanted it to be, but it's simple yet elegant!
We must thank the Matrix team for giving us the first lead to the problem, although I think their diode-fix is worthless. Then thanks to the Ripper team for really giving usefull information about what can go wrong with DVD bus patching and that a Mechacon lockup actually causes the coils to fry. The Ripper team described a safety circuit that should be built that would current limit the coils and another circuit that should detect the lock-up and reset the Mechacon. The Matrix team did it... with the PIC-Fix they made it possible to detect the lock-up and kindof reset the Mechacon... and now here it is... the current limiter a.k.a. "the vanDrunen fix" :
http://vandrunen.net/gallery/PSTwo_vanDrunen_Fix
This is probably the most simple current limiter existing and at first I didn't think that just this would do the job... but appearently it does... I've tested this thorougly even during a Mechacon lock-up!
This current limiter will allow more current through your coils than my initial designs, but it's low enough for your coils to "hold out"... and high enough for the Actuator to still operate within specifications.
In combination with the Matrix PIC-Fix this should definately protect you coils on short and long term. More testing to verify "playability" should be done, but with no or hardly any adaptations this combo can be the definite solution.
Instead of writing the full explanation of the "device" in this thread I'll make a PDF or something... because I think my photo gallery will be overloaded if I don't... and a photo-tutorial can be a bit unhandy!
About the 4x 10 ohms resistors "in series"... OOPS... it's parallel!! I'll fix it.
And I used 4x 10 ohms to make 2,5 ohms, because it needs to be able to dissipate a maximum of 0.75 Watts power. The standard small resistors are designed for 0.25, so that's why I took four. Using 2x 5 ohms or 1x 2.5 ohms is perfect too if you take the "bigger" ones... 2x 0.5 Watt, 5 ohm or 1x 1 Watt, 2.5 ohm.
Thanks for telling me Rebo.71
WilliamG... if your PS2 crashed after the first boot of a DVD-R ONLY and you have a DMS4-Lite chip, then I doubt it was the dreaded Mechacon crash. I measured the possible coil currents during a Mechacon crash with and without diode fix and it didn't make much difference, if your coils would've survived the 2A+ currents you were really lucky.
It came to my knowledge that "other chips" can cause these lockups easily because they don't overcurrent the DVD bus while patching. That's probably the reason why ot was the matr1x team that found a way to "recreate" the lockup so easy... and provide us with this good PIC-Fix .
I didn't say I would make an easy fix, but just one that works and is better than the diode fix and the romeo2 mod... and even this one is not exactely the way I wanted it. It would for instance be better to isolate the supply pin for the part of the actuator that drives the coils only and apply current limitation to only that pin... and put it at 100 mA. My first designs were two current limiters in between the coils themselves, but they are extremely complicated and hard to build... I only used them for testing purposes.
One advantage of this fix over the diode fix is that it doesn't degenerate the Actuators performance... read the ripper-team's documents on this if you want to know why.
One advantage of this fix over the romeo2 mod is that this dynamically lowers voltage to limit currents. You'll have full Actuator performance with full current limitation.
This fix together with the Matr1x-PIC-Fix could be soldered onto a nice PCB that can be easily installed and sold online as a quick-and-easy (tm) solution.
A tutorial on PIC programming will come later this day
2.5 ohm doesn't have to be exactely 2.5 ohm... something between 2.5 and 3.5 is ok. We don't really need 500 mA, but it's not already too much for the coils.
When you look at my install pictures you see that I unsoldered a fuse first. When you solder two wires to the two points where the fuse used to be and hook a current meter up in between, you'll measure currents between 10 and 600 mA with normal operation. When the Mechacon has crashed and the coils are driven at full current the current to the RS2004FS can go up to 2 Amps, it's just not enough to make the fuse PS11 blow. About 200 mA would go to the spindle motor and the power electronics (I don't have exact measurements, I didn't have a scope for this) and the rest would go through one or both coils.
When you would limit the current at less than 300 mA you would see a lot strange responses from the sled and spindle motors, especially during spin-up and disc seek. 400 mA works fine, but gave me some strange noises from the sled motor on a rare occasion... so I kept it at 500 mA... no real "scientifically proven" value, but ok, feel free to experiment.
When the Mechacon crashes and the current is limited at 500 mA I could measure 180 up to 240 mA through the coils when the spindle motor is off, 180 mA is about the same as maximum during normal operation and 240 is the peak during normal operation. With the spindle motor still on at high speeds these currents are lower.
About if it is overkill to do my fix as well as the Matrix fix... the answer is YES . But what if the PIC can't reset your console during a lockup... and what about those 1-2 seconds that your coils experience FULL current before the PIC can reset your PS2.. it all adds up!
Just think of it this way:
I mod quite a few consoles and a lot of those are going to be PSTwo's in the coming weeks / months. I ask about 100 euro's for a DMS4 install, which is a lot, but I give service and a full 3 months warranty. I will apply any patch necessary at no extra cost and at low expense if it will prevent my customers from becoming unhappy customers.
ok.. about using only my current limiter and not the PIC... you could do that, but don't you hate it to get up from your chair and having to unplug the power supply to get it to turn off?
You could do that, of course, I'll do some more endurance testing on some dummy coils to see how long they can stand a 180-240 mA current. Maybe even with my limiter they will still fry after leaving the PSTwo in it's locked up state for half an hour... I didn't get to test such extreme things, but I will.
And about the programmer... I think you're right, but we need the diodes to be connected this way to create a power supply by draining some current from the serial port. The zener diode has been inserted to keep the voltage at approx. 5 volts so you won't damage the PIC. Some current will flow, but those are limited (hard limited by the PC) and the zener will dissipate all excess power. Currents will be between 5 and 20 mA.
I actually built the JDM programmer from it's original diagrams a long time ago in a way I could use it with a pin header. I only rotated and swapped some components around and today I have redrawn it's diagram exactly like I built it on my circuit board. It's a bit easier to "recreate" it from my diagrams but nothing has technically changed
But really.. if you know some quick and easy fix to make the "power supply" more efficient and keep it functionally the same, tell me and I'll change my diagrams. I don't really feel like doing it myself because it works good enough now.
And for all the other people... ANOTHER TUTORIAL
Here's some pics in my photogallery showing you how to connect the JDM programmer and the PIC to the PC and how to use the program "IC-Prog" to flash the PIC with the "Matrix PIC-Fix"
http://vandrunen.net/gallery/PSTwo_Matrix_PIC-Fix
SEYMURX escribió:Esto pinta mejor, se avecina subida de precios de los instaladores, je, je.
psdos escribió:I will pass your mail to the Dev Team
Ok... if you would read and interpret their added info's right then this is what you would learn, 2 things:
(1) The Matrix PIC fix is only able to reset/turn off the PS2 when the Mechacon locks DURING operation.
These lock-ups during play are usually caused by chips that patch the DVD controller in a bad way... like the Matr1x 1nfinity (that's why they made their PIC fix, so they can still sell their bad chips) and the Ripper and DMS4 chips don't have this problem!
(2) The Mechacon can also crash during power on when plugging in the power cord. The mechacon chip gets some power when the outlet power is plugged in and the PS2 is still in standby. Reset and eject buttons are directly connected to the Mechacon.
The mechacon gets a "standby" power from the power supply and when you press the reset button the mechacon will switch fully on and all other chips will get power -> your PS2 boots, but Mechacon might go in the lockup state if voltage is unstable!. But of course this will ONLY happen when you try to power on the PSTwo when you JUST plugged in the power and the voltage is not stable yet.
I actualy use this "method" to test all kinds of fixes I have designed and implemented. I turned on a TL light which introduces huge glitches on my power supply... and then I also turned on my PSTwo.... 1 out of 20 times Iit would lock up immediately and I could test if the fix works. The PIC controller does not get a chance to turn off the PS2 again, because the PIC itsself does not get any power till after the PS2 is full turned on.
You can prevent this by adding some components following their "very hard to follow" diagrams. What they did is this make a delay in the Mechacon's reset/powerup line. You can delay the moment when the Mechacon initialises (when you plug in the power) by making a delay to it's reset line... this way you can prevent that the Mechacon gets in its locked-up state even before you turn the PS2 on. You can't even press the reset button to turn on till after the delay. And this delay should be enough for the voltae to become stable. Charging a capacitor with a resistor is something that is often used as a quite acurate delay timer. The reset line has to "charge" from a low to a high level.... let's say from 0 to a minimum of 2 volts.... and putting a capacitor in parallel will make this charging time much longer.
Ok.. this little addon seems usefull but not critical. Just don't power on the PSTwo when you just plugged in the power supply. I think I can join this fix and my current-limiter together in one big fix. The PIC code can be disassembled and checked.... the source code improved and put online... Someone could maybe think of better code. The PIC could be powered on too even before the PSTwo turns on (it can use standby power) so it does have a chance to turn off the PSTwo right after it boots if it locks up during boot.
The PIC and the current limiter could maybe use the same power supply line that will in term give the mechacon it's delayed power. The PIC even has 4 I/O pins left over, so writing a second program that runs in parallel that checks the power supply for stability could be inserted. Then the PIC is not only a watchdog for lockups, but also a "power-good" checker.
So many ideas ... but the thing is... I don't have a serive manual for V12 like I do for V1-V10... so i will take longer for me to find and trace the standby power lines, chip-enable and reset pins etc. If anyone has, please send it to me.
I'll "design" another power-on-delay and power-line filter, make some nice high-res pics, put a tutorial online and pot it up here again... like before.
Greetz, Edwin
sial escribió:Me.... le he puesto un diodo 1n4007 justo a la entrada de alimentación de la cónsola, con lo que consigo que la tensión de entrada se reduzca en 0,6 voltios y funciona ok. Espero que esto por lo menos haga algo. Poniendo dos diodos en serie (reducción 1,2v) no funciona, es para que os ahorreis de probarlo.
GAME-WORLD escribió:puedes poner los componentes k has utilizado
y como lo has echo