Arcade Otaku - アーケード オタク

leonardoliveira wrote:Have the CPS2 power the arduino and all your problems will be gone in an instant. That's probably why I have 100% success in this stuff.

Having the A board connected or not doesn't matter because the system is kept held down by reset signal being asserted. If you have the A board connected you can power the B board properly by the supergun which will make your life a lot easier. Even more if you leech power from the CPS2 B board to power up the Arduino.

What voltage do you leech from the B board to power the Arduino?

leonardoliveira wrote:Have the CPS2 power the arduino and all your problems will be gone in an instant. That's probably why I have 100% success in this stuff.

Having the A board connected or not doesn't matter because the system is kept held down by reset signal being asserted. If you have the A board connected you can power the B board properly by the supergun which will make your life a lot easier. Even more if you leech power from the CPS2 B board to power up the Arduino.

pictures and video of the whole proccess:

pins soldered to to the "B" 93646B-4:

pcb


arduino


video

copados - what is powering the arduino? Try pulling a 5V line off of CN7. If A25 and B25 should both be 5V, and A23 and B23 should both both ground. Check with your multimeter, whichever pins you don't have pulling power and grounding for the board.

Unless leonardo chimes in with a better suggestion.

Going to 5V for power is what fixed it for me, and using the unused pins on CN7 would also satisfy leonardo's suggestion to leech power from the board.

Is it a genuine Arduino UNO you guys are using or one of those clones that can be found on eBay or Aliexpress for a couple of bucks? Could that be the reason why it doesn't quite match the specs found on Arduino's site?

nem wrote:Is it a genuine Arduino UNO you guys are using or one of those clones that can be found on eBay or Aliexpress for a couple of bucks? Could that be the reason why it doesn't quite match the specs found on Arduino's site?

Mine is genuine. Purchased directly from Arduino. LCD screen purchased directly from Sainsmart.

NoAffinity wrote:
Going to 5V for power is what fixed it for me, and using the unused pins on CN7 would also satisfy leonardo's suggestion to leech power from the board.

are you suggesting to use a 5v. power supply instead of the 9v one?

I have 4 arduino boards, 3 original " made in Italy" and one chinese, I have never used the chinese clone for this. may give it try if that changes something.

My suggestion is leech 5V from the CPS2 and not use the barrel plug jack on the arduino to power. Just connect the 5V pin at the shield to the CPS2 board and when you power on the supergun or cab the arduino will be powered by the system power supply.

From which place/point on the B board do you recommend to leech the 5v?

copados33 wrote:From which place/point on the B board do you recommend to leech the 5v?

Cn7 A23 or b23 for ground. A25 or b25 for 5v. Power the board and test before connecting to arduino.

Just for information sake, I'll stress that boards with CN9 installed have the encryption programming pins at CN2 shorted to GND. Lest you damage your arduino trying to program keys on newer boards that way.

leonardoliveira wrote:Just for information sake, I'll stress that boards with CN9 installed have the encryption programming pins at CN2 shorted to GND. Lest you damage your arduino trying to program keys on newer boards that way.

That's good to know. Using CN9 seems easier, so pretty sure that should be the default, but who knows, there probably is a board out there with a damaged CN9 that would cause someone to try to program through CN2.

I'll throw my hat in the ring and say I'm having Rev. 3 & 4 problems as well.

leonardoliveira wrote:Have the CPS2 power the arduino and all your problems will be gone in an instant. That's probably why I have 100% success in this stuff.

Having the A board connected or not doesn't matter because the system is kept held down by reset signal being asserted. If you have the A board connected you can power the B board properly by the supergun which will make your life a lot easier. Even more if you leech power from the CPS2 B board to power up the Arduino.

I made a wiring harness that is set up for all Revs. of boards (per the details in the Arcade Hacker tutorial). I've probably programmed a dozen Rev. 6 & 7 boards without a hitch. Incredibly simple. Rev. 3 & 4 are giving me fits. I have 3 known (by me) working Rev. 3 or 4 boards (1 MSH and 2 SZ2). My harness is set up to power the Arduino (via its USB port +5V, not its power jack) off the same +5V that is powering the B board. When that wasn't working (powering up/down the Arduino and B boards at the same time) after many attempts, I tried different power up/down sequences between the Arduino and B boards (first according to the Arcade Hacker tutorial and then other sequences) to no avail.

I've reached out to Artemio to see if he has any ideas. Leo, what type of supply are you using (arcade, PC, other) and when you apply/cut +5V, do you just have a switch on the +5V line or are you powering up/down your whole supply?

As Leo has mentioned, it's possible to clock bad data, and my guess is there are one or more spurious clock pulses happening during power down or disconnect of the Arduino, after the B board has been properly programmed. Am I correct in assuming when the programming process is complete, all 4 control signals go inactive/hi-z (so that there aught to be no state changes when the Arduino is disconnected)?
-ud

Try a completely external power supply for the Arduino - something that plugs into the wall. That's my best recommendation, based on my experience.

Power the board first, then the arduino. It really shouldn't make a difference, as long as both are powered when you actually hit the button on the arduino to start the programming. That's when the arduino sends the signal for the board to "unlock" and the board allows itself to be programmed. The arduino actually displays a message that says "unlocking", but it displays so briefly, you won't see it unless you are looking for it. If you have the B board connected to an A board during the process, and have a screen connnected via JAMMA connection to the A board, you will also see the screen go from its suicide color (blue, green, etc) to black in tandem with "unlocking".

Alright. I have a working setup for Rev. 3/4, now.

There are 2 notable differences between Rev. 3/4 boards and Rev. 5/6/7:

1. Not all Rev. 3/4 boards have a resistor added to CN2 pin A32. I own many Rev. 3/4 boards that never had that resistor added at the factory, so this is definitely a widespread problem. This resistor (4.7k ohm) is a pull-down resistor for the DATA signal of the decryption key programming interface and should be added for reliable programming (pin C32 is a convenient GND connection on CN2). Capcom added a surface mount pull-down resistor to their layouts on later PCBs.



2. Rev. 3/4 boards have much longer traces for the programming signals (they run almost the entire length of the board). This means your programming signals are ~10" + whatever your actual programming cable length is. Longer path means opportunity for noisier signals, so be careful with long cable length (especially the GND connection between the Arduino and the B board).

Notes about my setup:

- I am powering both my B board and the Arduino from the same supply (Toshiba power brick rated for +5V @ 5A, though Arcade Hacker's recommendation of 2A should be plenty) and am powering them up/down at the same time. I have found no benefit from powering the Arduino from a different supply or in a different order.
- I definitely needed a direct GND wire from the Arduino to the CN2 pin C32 as in the Arcade Hacker instructions. I verified programming both with and without on 3 boards. 100% fail without. 100% success with.
-ud

Just program the B board attached to the A board. Let the Arduino leech power from the cab or supergun. There's no reason to split the B board from the A board for programming since the programming process doesn't affect the A board in any way...

Exactly what DarkSoft kit is doing for key programming.

leonardoliveira wrote:There's no reason to split the B board from the A board for programming since the programming process doesn't affect the A board in any way...

I'm pretty sure most people don't want to have to solder/unsolder 4+ wires on every Rev. 3/4 board (no programming header) they wish to desuicide. I agree on Rev. 5/6/7, though.
-ud

Interesting info, about the resistor and the length of the traces. Thanks for adding those findings, ud. This thread is turning into quite a wealth of information.

:edit: anybody know what the wattage rating of that resistor is?

The resistor is being used as a weak pull down for the digital pin so it doesn't matter. Original uses 1/8 watt. Tolerance doesn't matter.

This is how they look like when new: (this price is ridiculous, don't bother. I am linking just for the picture)
https://www.radioshack.com/products/rad ... ors-5-pack