N64 Stick Converter PCB v2.2

[AlmostOriginal]

I got mine today! Money well spent! German quality.

[Edhel]

Having issues with the two I got. Calibrated fine and they work perfectly when I do that. However if I have my n64 powered off for a number of hours, the sticks stop working and I need to calibrate again.
Jumper is not inserted, so it can't be resetting to calibration mode and there's no bridging between those two pins. Any idea what else could be causing the pcb to -forget- it's calibration after being powered off for awhile?

[micro]

I really don't have an idea what might cause this strange behaviour. Can you post some (good & sharp) pics?

[Edhel]

Here we go. It's the best I can do at the moment with the current camera I have. Not the best soldering, especially as I tried to go over it several times to try and figure out what the issue was. Strange thing is I can leave it powered off for awhile and it still works, but if I come back to my n64 pad later I need to calibrate it again fully.

https://dl.dropboxusercontent.com/u/24863711/n64_solder/IMG_20140306_134947.jpg

https://dl.dropboxusercontent.com/u/24863711/n64_solder/IMG_20140306_135024.jpg

[micro]

Very strange indeed... Does the problem occur with both PCB's you got?

I suggest you clean up your PCB with q-tips and isopropyl alcohol, then take again some sharper pics.

[Edhel]

Yes, it happens with both pcbs I have. Unfortunately I don't have access to a camera that gives sharper images at the moment. I'm still puzzled what could be causing it. If a solder bridge somewhere could be and where likely that'd be on the board, though the analog works perfectly when I do have it calibrated, or if the pcb is faulty. I seem to be the only one having this issue, so the later seems unlikely I think.

I'll try and clean it up further later and see if that fixes anything. Any more suggestions would be appreciated. Spent far longer than I should on this.

[micro]

I don't think the PCB's are faulty. Also if they did work at some point, that means they're flashed with the right code. I've flashed all the PCB's with the same code in one big session. The values calculated during the calibration are stored into the EEPROM which remains its data even if you cut the power. You successfully execute the calibration once and that's it.

I guess your problems could be caused by bad solder joints. But to be really sure I need to see good pictures of your PCB's, not blurry ones...

[abduct]

Yes, it happens with both pcbs I have. Unfortunately I don't have access to a camera that gives sharper images at the moment.

Attempt to take the picture at a 45 degree angle rather then at 90 degrees. Also if you are zooming in via the camera, zoom all the way out and move your camera closer. Also avoid flash and try to take the photos in more natural lighting.

And if you are using a better quality camera, turn your ISO down and open up your aperture (lower f/stop number).

Normally just taking the photo at 45 degrees and holding it closer makes the world of difference, even if the photo is a bit blurry when you take it, since you are physically closer you can reduce the image resolution in an image editor and make it a bit sharper.

As for your problem if you haven't already I'd check your points with a continuity tester.

[AlmostOriginal]

Having issues with the two I got. Calibrated fine and they work perfectly when I do that. However if I have my n64 powered off for a number of hours, the sticks stop working and I need to calibrate again.
Jumper is not inserted, so it can't be resetting to calibration mode and there's no bridging between those two pins. Any idea what else could be causing the pcb to -forget- it's calibration after being powered off for awhile?

I got scared when i read this! So i tested mine again! I works flawless! *phu* If i where you i would suck up solder and resolder the point near INV.Y. It looks really dull.

I hope you get this going soon.

[Edhel]

Thanks for the suggestions. Redid all joints again on both boards and so far so good. Just hoping the calibration settings continue to stick. Still find it odd it was doing it for both. Thanks to Micro for the boards.

[AlmostOriginal]

I also thought that was wierd. 
Bad components aren't that common usually they work or dont't. I was thinking of ESD damage, but that usually kills it completely.

[flamingleg]

Hey just wanted to say that I installed the pcbs I ordered and they both work perfectly! There is a noticeable difference between this pcb and the standard one you get with replacement sticks. I was very impressed with the build quality, the layout of the pcb as well as the ability to calibrate the stick once it's all put together.

thanks again!

[Sjb266]

All 4 of mine done and working perfectly...

Much prefer these to the original analog sticks... Even the brand new original controller I have!!

Thanks again micro.. Total professional from start to finish!

[rengg]

Never soldered before, maybe someone can answer some general questions i have:

what type of tip should i use?
what temp/watt should the solder iron be?
is 0.32 60/40 rosin core solder what i need?
are the fumes from the flux anything to worry about?

still another week+- before the pcbs get to me but i want to prepare before hand 

[Grambo]

what type of tip should i use?

For the purposes of this project, anything with a relatively fine tip will do.

what temp/watt should the solder iron be?

I like to work at around 550F for work like this. I find it to be a nice balance between transferring heat effectively and not burning up all the flux. The wattage isn't as important as a good quality soldering iron with an adjustable temperature, imo. I've never come across an iron too shitty/cold to make proper joints (for small electrical jobs such as this), but there are LOTS that are too hot and burn up everything in sight.

is 0.32 60/40 rosin core solder what i need?

Yup.

are the fumes from the flux anything to worry about?

Not really. If you were doing 8 hours a day, 5 days a week or sticking your face in the fumes and intentionally inhaling, it would be cause for an extractor. Just be mindful of not breathing the fumes too heavily and you'll be ok.


It is of my opinion that oxidization is the #1 most important and misunderstood factor of soldering. Transferring heat to the components properly is #2. Once you understand how they work, you'll understand how to solder well in no time.

Here's a fantastic video
https://www.youtube.com/watch?v=I_NU2ruzyc4

Watch it a couple times. It has a lot of incredibly important facts and visuals delivered at an incredibly quick rate.

[rengg]

Got 'em in the mail today (Canada), took about a week 

@Grambo thanks for the info!

edit: Did 1 PCB today...oh man those vids make soldering look easy. Opening up the GC case was very hard, broke one tab but eventually managed to get a clean break with minor scratches on the casing (the glue was thicker than I thought, had to cut deep). Then when I was soldering I ended up with two questionable connections that I couldn't de-solder with braid (oddly all of my solders were shiny though), and the board wasn't completely flat with the potentiometer and I wasn't about to re-solder it again. Luckily calibration worked fine and I compared it to my original controller and they are identical in execution. I should be able to do my second PCB perfectly now that I know what to expect 

Side note, I had one of those $10 hong king controllers laying around and I took out the A+B+C's button pad and put it into my original and now I don't have any weak button press issues. The quality of the pads in those controllers are the only good thing about them.

[Shadow_Zero]

Side note, I had one of those $10 hong king controllers laying around and I took out the A+B+C's button pad and put it into my original and now I don't have any weak button press issues. The quality of the pads in those controllers are the only good thing about them.

Nice one!  XD

[cumberworld]

I'm pretty annoyed with myself right now. I cut the wires down to 5 cm. When I was stripping them, I fucked up a couple of times and had to cut them a little more. When I had soldered them and put everything back together, I discovered that they were too short to reach plug on the controller. I'm sitting here now with a useless thumbstick because I'm missing half a centimeter of wire. It seems to me like my options are to (a) order a new thumbstick, or (b) cut off the wire from my original thumbstick and use that (which I really don't want to do). Anyone got any other suggestions? Also, I'd suggest changing the guide to 6 cm so there's a little slack for dumbasses like me.

[micro]

Don't worry! You can simply extend the wires:

How To Solder - Intro/Joining Stranded Wires - Part 1

Of course this might also be difficult for you if you've never done it before...

[Sjb266]

Cut the wires off the original n64 stick.. I did... You'll never use it again anyway..

Besides that the wire is slightly thicker and the plug will fit perfectly back into the controller PCB..

[micro]

If he decides to use the original stick's cable I suggest he should desolder the cable instead of cutting it off. IIRC the original cable is quite short. It would be a bad surprise if he cuts off the cable and ends up with the same problem. 

[cumberworld]

Don't worry! You can simply extend the wires:

I'd need heat shrink tubing to do that, which I don't have.

Cut the wires off the original n64 stick.. I did... You'll never use it again anyway..

I'll probably do this, although the only controller I have with me at the moment is the only one I have that's still semi-usable. I'll wait a week until I can get my hands on one of the absolutely ruined controllers I have.


By the way, I felt like I did I pretty poor job on the soldering. I've attached an image. Is it good enough? Any tips on what I could do better?

[micro]

Yeah, that really looks bad... If you look at the previous pages you'll find a video showing how to solder properly. But you clearly lack practice. The first solder job always looks like shit...

Heatshrink tube is extremely common and cheap to get!

[cumberworld]

Yeah, that really looks bad... If you look at the previous pages you'll find a video showing how to solder properly. But you clearly lack practice. The first solder job always looks like shit...

It's that bad, huh? I watched the video a few times before I did it. When I was soldering, my biggest issue is that the solder would cling to the soldering iron, and not to the PCB. I got a little better as I progressed, but that was still my biggest issue. I've soldered a few times before, although only wire-to-wire.

So what can I do? Remove the solder with the soldering iron and try again? Or did I do it so poorly that I ruined the PCB?

Heatshrink tube is extremely common and cheap to get!

I'd rather just use the wires from one of my N64 controllers than have to solder the wires together again.

[micro]

No, the PCB isn't ruined but the solder joints are so bad that I can't guarantee that the PCB will work. I'd remove all solder with desolder braid and try again. I'd also watch the video again if I were you   Also make sure you're using solder with lead and flux.

[cumberworld]

No, the PCB isn't ruined but the solder joints are so bad that I can't guarantee that the PCB will work. I'd remove all solder with desolder braid and try again. I'd also watch the video again if I were you   Also make sure you're using solder with lead and flux.

Ok, thanks for the help. I'll buy some desolder braid then. Two questions:
1. Is there any difference in quality for desolder braid? Can I just buy the cheap stuff from eBay, and that will work well enough?
2. Can I just keep soldering and desoldering the PCB until I get it right, or am I damaging the components by doing this?

[dav3yb]

Are these still being made/available?  I've been getting back into N64 a bit lately and, like a lot of people, i've got some controllers that are pretty well ground out.  I'd like to replace some of them with the gamecube style stick, esp if those work better for some games that require a lot of movement. 

I think i've seen a PM, but it's hard to tell, and it's not showing up in "sent message"

[micro]

Please, see first post.

[dav3yb]

Must've missed that bit (was reading it on my phone). 

Well I'd love to pick up about 4 if there is another batch.  I'll check back every so often!

[MockyLock]

Hello
I ordered some of your boards and mounted them.
Everything works well but i have a strange issue, and as no one mentionned it, i wonder what i could do wrong.
I'll try to explain with my poor english:
I tried my fresh new pads on GoldenEye. It's the kind of game where you need to be very precise. Everything started well but when i change for "aiming mode" (you know, by pressing R), the target don't move anymore when it's in the very tiny corners (up right, up left, down right, down left); With regular pad, in aiming mode, when you go in the corner (up righ for exemple), your "aiming window" move to this direction (up right) until it reaches the sky.
WIth the new sticks, the "aiming window" don't move, the target is in the corner but it don't move the window.

Can anyone try this and let me know about the results ?
Thank you

[micro]

I don't own Goldeneye, but here's a little test you should try out:

Open your controller and temporarily remove the upper half of the replacement stick unit's case. Now that the octagonal gate isn't restricting the stick any longer, does your crosshair in aiming mode work like it does with a regular controller?

[MockyLock]

i'll give a try, hopefully today.
Anyway, i would like to add that I've calibrated the stick with the upper half in position (you could think i've done it without the octogonal form, and so now it's back in place,the stick can't reach the positions previously setted).
Thank you for keeping an eye !

[MockyLock]

Well, i tried. Without the octogonal shape, i can reach the maximum of each axis, and so in GoldenEye, you can reach the point where the target make move the aiming window in the extrem end of the corner.
As i told you before, i've done the calibrating operation with the octogonal shape in place.
How does the calibration work ? If it detects the maximum of each axis, and then calculate the "corner position", maybe the shape doesn't allow the stick to reach this point ?

[micro]

Yes, the shape of the octagonal gate is a little bit different from the original N64 stick one. And it doesn't let the stick reach its max values in the corners. You got 3 options to fix that:

1) calibrate the stick again:
The installation guide will tell you how to re-enable the calibration mode again. But instead of moving the stick in a circular motion as proposed in the manual, just move the stick from the center to the 4 corners:

center > top left corner > center > top right > center > bottom right > center > bottom left > center

This will make sure you'll reach the maximum even in the corners. But at the same time the stick will get more sensitive. And you will have a small dead margin when moving the stick up, down, left, right. For example if you move the stick only up, the stick will reach its maximum value some mm before it actually touches the restricting gate. This is inevitable.
You have to try out how much more sensitive the stick gets in the end.

2) cut out the corners
It's possible to enlarge the octagonal gate in the corners by using a file/dremel/sanding paper. I guess it would be hard to enlarge all four corners equally. And of course this is not reversible!

3) modify the source code
You can also modify the source code (see 1st post of this thread). You could try out different max values. You could also eliminate the dead margin from 1) and tweak the sensitivity.
To modify and compile the program you need Atmel AVR Studio 5 (or 6). To flash the microcontroller on the PCB with your modified program you need an AVR programmer.

I'd definitely try out option 1) 

[MockyLock]

Thank you for your quick answer.
Well, the weird thing is that nobody mentionned this before.
I really would appreciate anyone else could try and tell me.
And about the solution, i think i'll try #1 first, then the second.
I definitly don't have the knowledge and gears for the 3rd option.
Something i thought about, i'll try to swap the octogonal shapes (genuine and 3rd part). I mean, i'll check if there is any difference, even very little, between the two ones, and use the smaller (if there is one smaller) for calibrating, and use the wider as stick holder cap.

[theeternal00]

Please, see first post.

Hello Micro,

I know these PCBs are not currently available, and that in another thread you may have said that you don't think there will be another batch, but I believe you have found the only lasting way of keeping the n64 controller alive and well without having sensitivity issues for some of my favorite games (especially super smash bros).

I would like to know the concerns you have for having more boards printed, and then I would like to try to help resolve those because it is much easier for me than learning how to design a circuit board and code a program to flash to the chip.

Some possible reasons for not printing more boards:
1. Lack of time (Are you still a student? Please tell me it is in the field of physics/electronics) Troubleshooting time, shipping time, testing time.
2. Lack of order quantity to get reasonable price on the boards. --I can see what I can do about organizing and confirming orders so there is no uncertainty. There are much better ways than reading forum posts saying, "I'll take 5".
3. Lack of upfront capital -- You seem like a very trustworthy and helpful person, I do not see a reason that I could not get you some upfront money to buy the boards.
4. Lack of interest in the project -- You probably have many projects going on, I know from your youtube channel. --If you are truly not interested in this, I would suggest transferring what you have done to someone who has more interest, and you should be compensated for this.

It is amazing work you've done, and I have a group of people here in the U.S. here that would like to make their controllers last, and I am determined to help them to stop using cheap third party original style replacements that get worn out or just don't work. Gamecube style replacements are the answer, and your fix for the sensitivity issue is the only way to make them work well enough to use for some of the best, and still most commonly played, n64 games.

Let me know if you are interested and we can talk more about the details of how I can help.

Thank you,

Steve

[Jaffa]

Oh my god... So this PCB is not working with golden eye? Golden Eye and Perfect Dark are the games why I need this mod... Now I am sad... I hope there would be some solution..

[MockyLock]

Actually they work, no doubt about that. There is a slight issue in the very end of each corner. It needs to be more tested before giving any description/solution.

[micro]

@MockyLock: I don't own Goldeneye anymore. But I've got Perfect Dark which shows the same behaviour with the aiming mode you've described.
Unfortunately it's a fact that the octagonal gate is different. You have to choose your top priority. If you want to have the same values in the four diagonal corners, then you have to recalibrate the stick while not touching the plastic gate all the way up/down/left and right.
I already told you one way to calibrate the stick. Another way would be to make the stick cap's shaft bigger. You could wrap some adhesive tape around the stick's plastic cap's shaft, for example 5 layers. After the calibration you can remove the tape again. May be 3 layers are enough, maybe you need 10. You really have to try.

Normally the stick get's calibrated in a way so it got about the same values as an original N64 analog stick when moving the stick all the way up/down/left or right. Calibrating the stick as described will make it (a little) more sensitive, it's a trade-off.

@theeternal00: When I'm sure I won't make another batch I'll release the PCB files. But I don't feel obliged to have these PCB's on stock for 365 days a year!
Also, I believe in 5 to 10 years everyone can use his old original N64 sticks again and just reproduce the expendable parts in a 3D printer...

[MockyLock]

I understood what you explained to me, no worry.
I was just telling that i need to test the different "fixing method" before choosing the best one.
Is there a way, by programming, that once the calibrating values are setted by moving the stick, the microcontroller adjust the values by reducing the max values.
Some kind of automatic trim ?