6502.org

I went through an ordering procedure for 1 x W65C816, 2 x W65C51N’s which weren’t available and 2 x W65C22S’s at http://nl.mouser.com/ and ended up at a price of €60,38. Naturally I declined.

However, my son studies at the university TU Eindhoven and I’ve read (I don’t recall where) that it’s possible to get some samples from WDC. Does anyone has experience with that?

Or does anyone living in the Netherlands or Germany know where to by WDC IC’s?

Please check the sticky (meaning it's always at the top) topic "65xx parts sources" under "General Discussions," and if you don't find it, we will try to update the head post when we find the answer you need.

Thanks!
Looks like Jameco ships worldwide at reasonable shipping costs

I have most of those parts in DIP (only 1 65C51N but have some 65C51S) and all of them in PLCC. £20.25 plus postage and packing (largely depends on the cost of international tracking ~£5-7) if you want them.

Andrew, may I add you to the head post of the 65xx parts sources topic? If so, should people just PM you here or do you want other info in the post?

Jameco is a good choice. I've done business with them for some 25 years and have never been dissatisfied.

If you order 65xx parts from Jameco be sure you know what you're ordering. For example, they stock the "65C02." There is the genuine WDC part, which would be the W65C02S-14 and is what you want (DIP or PLCC). There are also Rockwell surplus inventory (and some pulls) with the part number R65C02 that they've had since dirt was new, which you don't want. Ditto for some of the 65xx I/O parts.

People can post or PM me. I have some spares of most WDC parts bought from an official WDC supplier but not a huge stock.

I also have a number of older Rockwell and Synertek parts (probably pulls).

I’m looking for PDIP if possible.
Besides PDIP / PLCC there are N and S types of the WDC chips. For a “new” design I would want the S-types if I’m not mistaken?
For my system I’d like 1 x W65C816S, 2 x W65C51S and 3 x W65C22S.
It seems that the W65C51S isn’t available, would it be a problem to take the W65C51N instead?


@BitWise, I’ll send you a pm…

The W65C22S has a totem-pole IRQ output that can't be attached to a wired-OR circuit without an isolating Schottky diode. Be sure that you understand how to do that or else you'll have a non-functioning IRQ circuit.
The W65C51S is only available in quantity as dice. The W65C51N is what you want.

Regarding the W65C51 chips, I would inquire to the LOT number on the chips you're buying. I just isolated and confirmed a problem with the current LOT number on new chips with the Xmit flag being stuck in the status register. This makes using the chip problematic and limiting.

I also received a few older LOT number chips from the folks at WDC and am testing them now. I'll post results once I complete tests and confirm with WDC.

I certainly don't want to unnecessarily muddy the UART waters, but I don't recommend using the 65C51 under any circumstances. This device has had a long history of problems dating back to the early NMOS units produced by MOS Technology, and has what is in my opinion, a poorly thought out programming model. There are much better UARTs available. I use the NXP 26C92 in my POC unit and it works flawlessly. André Fachat used the 16550 (the UART that was used in many x86 machines) in his systems. Garth has mentioned some others with exemplary specs.

The only thing, again in my opinion, that the 65C51 has going for it is that it is a piece of cake to attach it to a 65xx system. The others are a little more involved, but not much.

BDD, somehow I don't think that's possible, I've already stirred things up enough already to get the mud to the top. Anyway, while I've not had extensive long term experience with the 6551, I've used them on and off a bit since the late 80's for simple stuff, but my only set of chips are Rockwell, Synertek and AMI. I've not really had any issues using them or getting them to work.

In any case, WDC was nice enough to send me some chips to test, 3 different Lot numbers from what is currently shipping. My initial test so far have resulted in getting them working but not without making some hardware changes to the async board. I've also discovered that they generate a large amount of noise on the power supply and the internal spiking causes jitter in the baud clock resulting in transmission errors, even with a can oscillator as I couldn't get the clocks to work reliably to do any tests with the crystal/cap/resistor.

I also am having a difficult time continuing down the path of the 65C51 as a result. While I have figured out how to make them work, redesign on the hardware is a requirement to contain the noise and for the current lot with the Xmit flag stuck on is a software change that results in limited applicability.

I would note that the folks at WDC have been working with me on this and have been great the whole time. I just wish my results were more favorable. I've been running the old Synertek chip with a 6MHz CPU board and it's been solid for months of continuous operation.

And thereby bolstering my argument that the 65C51 should be avoided.

Thanks for the effort everyone, but till I’m ready for a faster design, I’ll stick to the old NMOS parts I already have.

Good luck.