Okay, I’ve checked the cable, and now I am confident there is nothing wrong. The DC is not piggybacked, and I took out the small microchip. when I plug it in, it only shows up about the same amount of times, something like 1/10 or 15. I also retested the UPCB solder points, and it is all okay. Anything to double check to find out why this isn’t working now? Sorry if I’m blowing up your thread with technical questions. It’s just weird that I had this working last week and now it wont show up as working, I keep getting an error message…
Don’t worry about blowing up the thread with technical questions; more than anything, that’s what this thread should be for. You have my word, as long as you’re willing to troubleshoot it, I’m willing to help.
What kind of error message?
What difference, if any, is there between plugging it in in normal mode (no buttons pressed) versus bootloader mode (start+select, or program, pressed)?
What changes or devices show up in the Device Manager, for either mode?
-Go ahead and leave the 4066N chip out for now. We have no reason to put it back in until the USB is working right.
-First thing I’d like you to double check is the orientation of the 470nF capacitor. The Instructable for the UPCB assembly described the orientation it should be, no matter which type of cap you have. This cap is needed for the internal power conversion from 5v to the 3.3v the USB requires. That power conversion will not work if the cap isn’t installed right, causing problems.
-Second thing. We know the button select USB cable is good if it passed all of the tests with the multimeter like you said it does. Go ahead and plug the USB cable into the output DB-15, and the output DB-15 into the UPCB. Do not plug the USB into anything for this. Locate the two 22 Ohm resistors. The end of the 22 Ohm resistors closest to the PIC are going to be our test points. Test the resistance from those to point to each other, to the GND_TEST point, and the VCC_TEST point. Use the maximum setting on your multimeter. On each of those, the resistance should be infinite; those four points shouldn’t be connected at all. If any of them are connected, pop the PIC out of the socket and test again. If still connected, remove and disconnect the USB cable, but leave the ‘output to DB-15’ cable. If still there, disconnect the ‘output to DB-15’ cable and test again. If removing a cable causes the short to go away, you know where the short is. If the short remains after all of the cables and PIC are removed, you know the short is on the board somewhere. If there is no short with everything connected, we’re happy.
If any are still connected, post back which one and what the resistance was. If the resistance looks really low on the highest setting, crank it down a notch and try again until you get an accurate reading.
-Test without the ‘output to DB-15’ cable. If you have a female DB-15 connector handy, unplug the ‘output to DB-15’ cable, and put the female DB-15 directly on the end of the board on the pads next to the ‘output to DB-15’ header. You dont have to solder it in place, it should be plenty snug. Plug your USB cable into the female DB-15 and plug it. Any change?
well, I got snes working for a little bit after rewiring the console cable. I think somewhere, somehow, it’s my wiring of the console cable that is fucking it up. Ill get back to ask for any help when I am 1000% sure of the wiring.
i found out what my problem was. the two 15 pf capacitors connecting to the crystal were not connected to it, but were rather on the outside. I am an IDIOT. I have moved them, but do you think have used it before these were put in right will have damaged anything else?
I’m glad you found the problem. I assume the tests since have been all ok? No, if you put the capacitors into the wrong holes (I assume you mean the legs weren’t straddling that equal sign looking thing), you could not have damaged anything. Those capacitors work with the crystal to produce a nice clean square wave for the PIC to work off of. If they were in the wrong holes, no parts would have been damaged, but the timing was off enough that USB wouldn’t work properly and/or the PIC kept resetting itself.
I’m been trying to come up with a way to put in a self test to the UPCB, but everything I find says that the initial boot up has to use the oscillator option given in the configuration. I’d like to be able to boot off of the internal oscillator and then check the crystal to make sure its good, but there’s no way to do that that I can find.
Any idea on what I can use to read the SX28AC/DP? If I can read it, I just might use it as the base of my universal PCB.
Hey, I have a fully assembed UPCB kit with unflashed pic18 chip and playstation cable… anyone wanna take it off me?
I have no idea, I’ve never dealt with the SX types before. I will warn ya though, every chip has a way to protect the code so it can’t be read, to prevent exactly this sort of this. Read through the datasheet for the chip if you can find out how to read/write to it.
Heh, you went through all of the hassle of assembling it, and haven’t even tested it? I thought your friend could flash PICs?
Nah… its inability to do 360 without a shitload of work really puts a damper on things for me.
hehe. That’s a problem with the 360 pads, not the UPCB. The work needed to piggyback a 360 pad is less than using a 360 in project box. The work’s the same, but you don’t need the pullup resistors.
Please let us know if you come up with a better alternative; many people, UPCB users or not, would love to have one.
say toodles what ever happened to that flash comeback project… i know id still buy some. do yo uwant me to test a project one? and do you still need plastic help or something?
If you’re dying to work on Dreamcast, I did some hunting and kinda settled on the dsPIC30F2011 and the 24LC1025. The 30F2011 runs on 5v, can go up to 30 MIPS which should be way more than plenty, and is only 18 pin so a surface mount version should be able to fit into a D-Sub hood just fine. The 24LC1025 is a 1MBit i2c serial EEPROM. 1MBit (128Kbytes) is exactly the storage size of a VMU, so you should be able to handle fetching and writing to it like a VMU with very little code. The i2c lines on the dsPIC are separate from the UART lines, so I was expecting to have the two Dreamcast lines going to whatever pins are handy, the I2c to the two i2c lines, and the two UART lines connecting to the UPCB. Just have the UPCB send the state of the buttons and stick in a couple of bytes at high speed over the UART lines, and the dsPIC reading the button information when it can, and juggling the dreamcast stuff with the rest of its time.
Also, both are available in DIP versions, so prototyping will be easier. Package codes are DSPIC30F2011-30I/P and 24LC1025-I/P for the DIP versions in case you want to sample them and try your hand. Make sure to get the ‘-30’ version of the dsPIC to make sure it can do 30MIPs.We probably wont need to go that fast, but its there if we need it.
Using a CPLD would be less of a current drain, but I’m not sure if I can pull that off yet.
If only they’d come out with a just barely faster version of the 4550, we wouldn’t have to use another chip. But only 6 instructions per bit, that’s just too close.
Just so everyone knows, no, I’m not dead. I am juggling a 17 credit courseload at the moment so I might as well be.
First, good new. Jaxel says there is a brand new EBX or EB Games branded ‘micro’ sized controller for the Xbox360 that uses common grounds. I haven’t had a chance to test it yet, but if so, piggybacking this controller would be just as easy as piggybacking a dreamcast controller. There isn’t a line available for the Guide button, but I can certainly leave an option for an additional little OR gate chip in the next UPCB board revision so the Guide button would be just hitting both Select and Start. I’d really like to get more confirmation on this before I finish the next design, especially so I can get an idea about how to interface with the triggers.
Sadly, there is bad news. The existing Genesis and Saturn support doesn’t appear to work properly on real hardware. A local game store is closing out and they hooked me up with enough pieces and parts to assemble some older console, giving me my first Geny and Saturn to test on. I have been using the Innovation adapter for my Saturn testing, and it still works awesome, but the real hardware its twitchy. The Geny acts like you’re constantly pressing both left and right, so I think that the consoles are switching between banks and reading the outputs far faster than I expected. I have some ideas for how to streamline things so it works better, but I may have to learn how to use my Oscope to fix it. On the bright side, I have SF:NC for the Genesis and can work on getting the 6 button Genesis support done as soon as the 3 button is done.
There’s not much point in me making the xbox360 piggyback helper board if there exists a cheaper, easier to use PCB instead of the DOA4/Hori EX2 PCB’s. Please, if anyone gets one of these, let me know. I’ll be looking for one locally as well.
Glad to hear that you are still working on the project, I would hate to see it die cause UPCBs are awsome.
Any word yet on getting the joystick to work as xy axis on with the usb cord yet? I dont use my stick much right no since I have an 18 credit course load myself, but I am just curious if you have had time to read up on it at all.
With PS3 shitting the bed on diagonals, there’s no reason not to. If I get one or both of the genesis and saturn working properly on the real hardware, Ill make a new release and set the current USB code to report the XY instead of (or along with?) the POVhat report. That should take care of the doujin PC game problems until such time as I get the USB code separated for PS3 and PC.
Just confirmed the common ground Xbox360 pad. So, should I bother trying to make the piggyback helper boards for DOA4 pcb’s, or just put in an optional chip onto the UPCB itself so the piggybacking is a straight wire job? Frankly, I trust the look and feel of the DOA4 pcb over the rebranded madcatz job, but I know y’all want this shit easier and quicker.
tbh, i rather make it easier. theres not a whole lot of room in the doa4 stick. and i would want a second stick in a custom case, in the long run, i think making it easier with using the mad catz controller would be more beneficial, even tho i do like the doa4 pcb =)
as soon as the revision comes out, i am buying one or 2
Hows the project coming along I was wondering if Xbox 1 support is going to be worked soon… a lot of what i play is on the xbox (because of emulators). also please dont let this awesome project die~!!! i think this is great that you are doing this.
No way I’m gonna let it die. That’s the whole reason for releasing everything, so even if I get hit by a bus, it can keep going.
Xbox 1 support should be completely doable, IF I can get access to a USB protocol analyzer. The Xbox controllers are USB, but there are a few vendor specific commands and responses that I don’t have any information on. Until I can snoop an Xbox controller while its plugged in, I don’t know how I’d be able to get that information; I’m pretty sure its not covered in bunnie’s xbox hacking book.
I’ve been asking a few professors at the engineering school here, and none of them know of any on campus.
I do have a fall back plan, but it’ll take a while before I’m there. I have a really sweet FPGA development board, the Altera DE-2. It has both A and B USB sockets on it going to the FPGA. I’ll be taking a digital systems class next semester that uses this board, and I’m hoping to learn enough in the class to use it as a USB protocol analyzer.
It aint gonna die as long as I’m alive, but I’m the only one working on it, so its going kinda slow.