Heya all,
It's been a very long time but I'm planning on getting back into NES hardware.
This topic has cropped up a number of times over the years, but looking thru all the past posts, it seems like it always ends with 'use a 2.54 mm pitch connector, it's close enough if you're careful', or 'destroy a game genie'.
Has anyone ever found a source for a proper 2.5 mm pitch 72 pin connector (and dare I ask, for the narrower pcb thickness as well)? I did the rounds of various suppliers tonight with no luck (digikey, mouser, avnet, jameco, newark, etc..., even tried aliexpress on a whim).
Cheers,
Michael
I gave up on my search years ago. If we ever found a supplier, I'd be a buyer for a couple hundred of them.
Even though it's seems totally wrong, the 2.54mm (At least the Amp connector carried by Digikey) does seem to work. I seem to remember someone saying that NES clone systems are even 2.54mm (I don't have one to check, just a couple Famiclones which are naturally 2.54mm anyways). Not even "if you're careful", I purposely tried to short one out and couldn't make it happen. I can't say it'd be impossible though, I suppose progressively shaving the edges of a cart PCB until it fails would be one way to find out what the tolerances are.
Yeah teaguecl and I went back and forth on this ~1yr ago
I use the digikey connectors for various things and have never had issues. I use two of them in my
NES for the past year and would never go back.
Even if the pins on the very edge were to short out it's not going to do anything damaging like shorting Vcc to GND.
I did get a 72-60 pin converter from stoneage gamer and the pitch is off also, I haven't had issues with that either.
Excel to the rescue. To convince myself it was worth using the 0.1 in headers, at least for a prototype, I threw together a spreadsheet using info from the datasheet for the TE AMP connector (A31721-ND @ digikey) that has been recommended in previous threads.
Assumptions:
- The datasheet doesn't have the width of the actual pin contact, but it looks like 1.447 mm (blew it up, measured edge-to-edge of two pins to get image dpi then measured inner spacing).
- NES cartridge fingers are 2 mm wide normally (3.2 mm at the ends)
- NES cartridge total width of inserted portion of PCB is 93.5 mm (measured off my YAROM cart tonight, as I have a PCB handy).
Assuming you center the cartrige in the socket, the finger center point to to connector pin center spacing error is max. 0.7 mm, but that's not the interesting/important thing, it's the distance until an overlap. The minimum safe distance is 0.1165 mm, which is scarily close but not negative.
However, you'll get an error if you shift 0.1165 mm, and you have up to 0.24 mm of wiggle room in either direction to shift it. All the way to either side, and you'll have 4 pins touching the wrong pads on the opposite end!
So, assuming the data sheet and my recollection is right, and I didn't fubar something in Excel: Using 0.1" pitch headers is functional if you're careful, mildly risky depending on the application (can end up bridging pins in a way that will cause driver contention) and certainly not ideal.
Cheers,
Michael
For what it's worth I haven't been very careful. I just use it, it just works
The only place I've seen room for error is the connector I have on the motherboard, depending what you're doing that might not factor in. I aligned it to the mobo when I first installed it and haven't played around with it since. If you're trying to use this in your Nintendo you could your rig directly to the mobo. I also use them in Kazzo dumping devices I've built.
I can't speak from experience for every board manufacturer out there. But I play Nintendo, Tengen, colordreams, AVE, retrousb, one of Memblers' boards, and several PCBs I designed myself on a routine basis for over a year now with no issues.
I LOVE my 72 pin improperly pitched connector.
Yeah, another factor that I haven't measured is the outer plastic thickness from the socket to the inner dimension of the NES cartridge shell. It may be that there is basically no play, causing it to always be centered and effectively a non-issue for licensed carts in their cases.
/me wanders off to find a cart
100.54 mm ID on a cart shell, 3.86 in +/- 0.01 in OD on the connector datasheet, or 98.044 mm (98.298 mm if we're being generous). Nope, that's almost a pin worth of slop (2.242 mm).
Anyways, I figure I'll go ahead and order a few for now, though I'd really like to find the real deal some day.
Cheers,
Michael
joat wrote:
Excel to the rescue. To convince myself it was worth using the 0.1 in headers, at least for a prototype, I threw together a spreadsheet using info from the datasheet for the TE AMP connector (A31721-ND @ digikey) that has been recommended in previous threads.
Assumptions:
- The datasheet doesn't have the width of the actual pin contact, but it looks like 1.447 mm (blew it up, measured edge-to-edge of two pins to get image dpi then measured inner spacing).
- NES cartridge fingers are 2 mm wide normally (3.2 mm at the ends)
- NES cartridge total width of inserted portion of PCB is 93.5 mm (measured off my YAROM cart tonight, as I have a PCB handy).
Assuming you center the cartrige in the socket, the finger center point to to connector pin center spacing error is max. 0.7 mm, but that's not the interesting/important thing, it's the distance until an overlap. The minimum safe distance is 0.1165 mm, which is scarily close but not negative.
However, you'll get an error if you shift 0.1165 mm, and you have up to 0.24 mm of wiggle room in either direction to shift it. All the way to either side, and you'll have 4 pins touching the wrong pads on the opposite end!
So, assuming the data sheet and my recollection is right, and I didn't fubar something in Excel: Using 0.1" pitch headers is functional if you're careful, mildly risky depending on the application (can end up bridging pins in a way that will cause driver contention) and certainly not ideal.
Cheers,
Michael
Your math very closely follows mine - and has the same conclusion (see the argument infintelives and I had last year). I don't see any way these connectors can work, yet there are people using them without reporting issues - even when they deliberately try to short by pulling the carts to one side. I purchased these digikey connectors about 5 years ago and certainly did have shorts with them. People who have purchased them recently have not reported issues. Maybe the design of this part was modified, but the mechanical specifications in the documentation were not? Regardless, it's worth trying them out with all of the positive feedback from multiple reliable users here.