Archive for the ‘Uncategorized’ Category

Teasy Strike + SparkyPWM + Arc Eye 2 = Awesome

Ok, so I’m proud to say that I think I’ve got the first (of what might be many) TEasy Strike + SparkyPWM + Arc Eye 2 set up complete.

Let’s start with the basics:

The Sparky PWM, soldered together to my liking

However, there’s a slight mod that’s required in order for this set up to work. 32teeth is already working a plan for how to allow people who don’t have soldering abilities to make this mod happen. This mod is only required for 20P ribbon cable powering (via TEasy Strike). This is also being taken care of in the next version of Sparky from what I understand.

That’s the power signal from the 20P header being patched to the main screw terminals, It’s also required to do the same for the Ground position (Pin 15 of the 20P header). I snagged a picture of just the 1 wire (VCC), but there needs to be 2.


The SparkyPWM mounted, the Basic Breakout board was populated with a 6 pin female header for modular connectivity. If I ever want to remove it I can. The connected wires are for an advanced modification (shown below).


Basic set up complete. After the little mod and prepping for arc eyes, connecting everything was really quick. No pics of the lights, sorry I didn’t take one…but be assured that all my buttons are glowing beautifully 🙂

A quick picture of my buttons with the Arc Eye 2’s in place. As you can see, the buttons are drilled to allow the headers to poke through, making the basic version of this mod 100% solderless when all 3 of these boards are used together 🙂

The only difficulty now is closing the top panel, the right side of things is rather spongy with wires, even keeping them as neat as I can. I may eventually get rid of the ground wires from the QDs and daisy chain just 1 ground connection to alleviate some stress, but for now it works.

—–
Now to the advanced mod. The FTDI serial to USB adapter is required for sparky configuration and firmware updates. Well…I didn’t want to have to plug in a second cable every time I wanted to configure, so I went for a single cable full operation.

I did this by expounding on the data lines, both the FTDI USB data lines, as well as the TEasy Strike’s USB data lines run to an Imp board, which is controlled via the RS switch. When RS is activated, I’m in FTDI mode and can configure the SparkyPWM. I’m currently running into some weirdities with configuration, but I’ll get those sorted soon with the help of 32teeth.


I also put a piece of tape on the 1x pins for the AE2’s, this is so that I can remove them all at once, and put them back on all at once without having to re-figure out which goes to which button.

Hope this helps get an idea of what’s to come.

 

Update to this:

I ran into an issue with my SparkyPWM, 32teeth and I have been discussing why this might happen but in the meantime I’ve got a workaround for anyone who experiences this issue. After about 30 minutes of play the SparkyPWM v1 seemed to have died on me, it turns out that the resonator (tiny chip at the front of the board that creates the clock signal) had died.

We’re not 100% sure of what caused it, but 32teeth is adjusting and moving to a new package for the resonator that I’ve suggested. For now, I’ve currently got this in my set up, the set up works like a champ now, no problems whatsoever. I removed the old SMT resonator and replaced it with this THT resonator, and if you do this just get any like this that run at 16 Mhz. This style of resonator is built like a tank, and if you’re searching on digikey or something just make sure you get a 16 Meg version with internal capacitors, you should be OK.

Madcatz 360 TE PCB Fixes

I have something a little interesting to post about at the moment. These two things stem from issues customers of mine have had. Neither issue is caused by my product in the least, but it’s things that may come about while owning a 360 TE and it might be interesting to know a workaround to these issues.

First up is a dead button or joystick signal, and by dead I mean acts as though it’s constantly pressed. I had a customer tell me that LEFT was always active, and I couldn’t explain why until I had him take some voltage readings. It turns out the 360 PCB’s LEFT circuitry was damaged or dying, because the voltage on that signal alone was less than the others. This can be caused by a few things, but in the case of a simple game controller, it’s more than likely the MCU internal pull up acting up.

His readings were something akin to this:

Without the TEasy Strike connected :: All other pins = ~3.66v, LEFT = ~1.32v

With the TEasy Strike connected :: All other pins = ~3.66v, LEFT = 1.48v

I took those readings as a confirmation, the LEFT pull up on the 360 side was damaged in a way that it wasn’t drawing the voltage on the line up enough, and when connected to my board, the added pull up increased the voltage a little bit, but not enough. As a matter of fact, it drew the voltage up enough to behave normally on the 360 side, but the PS3 side didn’t like the lower voltage. So I needed to come up with something he could do to draw it up enough for both sides to be happy.

I scratched my head over a way to fix this in a way that wouldn’t require soldering. He had purchased TEasy Strike, and wasn’t comfortable with soldering at all. Then it hit me, I remembered seeing that people used to be ok with shoving bare wires into the connectors and do a ghetto solderless connection.

So I sent him the information, and told him to go to Radio Shack or Fry’s and grab a 10k Ohm resistor. He was going to shove one end into the LEFT position of the joystick connector, and the other in the closest access to the VCC line available.

Ignore the band color of the resistor, it’s a 220 Ohm and was only used as an example, if you try this DO NOT USE A 220 Ohm. At any rate, this worked, and it was able to bring the voltage up to a level that both the PCBs were happy with. The 10k was still too high, so eventually he used a 4.7k Ohm, but the fact remains that this might be a good fix if you find that one of your buttons or directional signals is magically “dying.”  The reason this works is because the internal resistors on MCU’s are generally very high in resistance (100k or greater at times, otherwise known as “weak pull ups”), so the lower resistance added in parallel allows for a stronger current pull, which in turn brings the voltage level up.

 

The second thing is something a little strange, but might be very helpful for some people. I had a customer who blew out the PCB of his SSF4 stick, and unfortunately when he was able to attain another one from Madcatz, they sent him the wrong version. They sent him the PCB for a Round 1/2 stick, which lacks certain things such as the added KGND (assuming it means Kill Ground), which adds the lockout option to the Start and Select lines, as well as the HOME.

Now, round 1/2 PCBs are MUCH easier to come by these days than the Super versions, so I used this opportunity to see if I could retrofit the TE-S stick to work with these older PCBs. Eventually, all I really had to do was connect the KGND and GND on both the control panel and the quick disconnect terminal section. Other than that, it was a matter of cutting off the shrouding of the headers on the PCB in the sections where Microsoft extended it by 1 pin for the KGND signal.

Now the grounds are connected entirely, and the cables go where they originally went. They just hang off the shrouded headers by 1 spot, but since that spot is GND/KGND it doesn’t matter as they’re already hacked together in other places.

The ONLY downside to this hack is that you will no longer have the lockout functionality. However, if you’re hard pressed for a PCB and have no way of attaining a replacement for your TE-S this might be something you want to explore. All it takes is a few solder joints and a few snips of wire cutters and you’re done.

 

Hope this helps some people out there! 🙂

TEasy Strike craziness

I haven’t made a blog post in a while here, been a bit backed up. So I’ll post up what I sent to the strike devices development team the weekend my first run of the TEasy Strike’s were finished.

aaaaaaand go.

This weekend was ridiculous. I spent the better part of saturday behind safety glasses while I dremeled, drilled, and soldered my way to a good position for production of the new boards.

Friday I was told the boards were ready for pickup, so I left work early and got to the assembly house about 15 minutes before they closed. I was presented with a gigantic box, which I later opened to find this:

Perfectly packed were my 150 boards, extra parts, and 50 bare PCBs (which turned out later to be more like 60 for some reason, I’ll need to count the full number of boards later). I nervously unwrapped one of each type of board. These guys supplied the PCBs, so I was really curious how the final quality would be. I was SERIOUSLY impressed, the silk screen, placing job, and overall quality was amazing.

The panel they made up for the job:

It’s a 6x configuration, 3 of each type of board. The ONLY issue I have is that they’re V-Scored, which is cheaper but leaves a rougher edge that I’ll want to sand down if I don’t want people telling me they got FR-4 splinters >.<

The color is fucking phenomenal, these guys are really good. I counted 10 panels, they must have given me extra because a few of the boards didn’t pass E-testing. The ones that don’t pass are marked with a big ass X in sharpie across them. When getting things assembled, having the PCBs E-tested saves you SO MUCH HEADACHE, so if they don’t include it, it might be a good idea to ask how much it would cost. These guys included it into the fabrication price 🙂

Now the fun begins. I needed to make a template for each version of the board, so like before I pasted 2 on top of each other, but this time around I needed to cut them and route out some parts so that when things are smashed together, the components don’t get in the way. Another good reason this needed to be entirely SMT.

The templates turned out a little convoluted, but made sense to me. They’re 3 layers now instead of 2 and I’ll show you why.

Top:

Bottom:

Side shot:

I had to cut out a bunch of the side of the board, making it really hard to get a good space to clamp the template down to the actual PCB I’m assembling. The top layer of the template is a large piece that gives me the area to clamp, as well as spreads out the pressure so things are evenly compressed. I used some of the E-Test failures for the top layer, the other layers are 2 boards are specific to the type of TS I’m putting together.

It sounds like a huge pain in the ass, and it really is, but there’s no other good way to populate the pogo pins. I’ve talked with the assembly house and they think they can handle populating them in my next order. So now all I need to do is sell all of these guys so I can eventually do that, lol.

Here’s the final product, not usually one to toot my own horn, but this thing is fucking awesome.

I also found some issues with my new test fixture when programming via ISP, so I decided to power-forward my ISP with a simple mod:

Now I can program the target board without needing an outside power source 🙂

Return top

Announcements

Hope everyone had a good time at EVO. I know I did!