Hi Everyone,

I've made a couple of new 6502-based projects lately you can check out online at http://www.6502.org/users/mike . Right now I'm working on a robot project. I welcome any comments or suggestions for anything on these pages.

I know that a number of you have made some interesting projects. I'd like to encourage you to publish them on the web. Space on the www.6502.org server is always available free for anyone who wants to put their projects and info online.

_________________
- Mike Naberezny (mike@naberezny.com) http://6502.org

Mike: that users/mike link has an errant period attached to the end.

Gilbert's brain looks great! Does he have a body yet?
What are you using for motors and drive train?

On my VICbot project, I'm getting ready to replace the screwdriver motors with something smaller and with more gear reduction.
The screwdrivers have turned out to be a real hassle:
- I can't get them to run slow, and still have any torque, while using PWM.
- They make big current spikes in my power system - yesterday they cooked part of my motor-control logic.
- They run down batteries quickly.

Pete

I'm working on a 6502 project right now. Nothing special, just an "ordinary"
computer very much like Chris Ward's. *g* It will most probably have 128k
SRAM which will be seperated into 4 32k banks switched by a 6522. It will
also have an LCD and perhaps a 3 1/2" floppy disk drive. Does anyone know
how to interface and program a standard PC 3 1/2" disk drive? I have no
clue at all... (

Thomas

>Does anyone know how to interface and program a standard PC >3 1/2" disk drive? I have no clue at all... (

I've never tried it, but I think it's messy/difficult unless you use one of those dedicated floppy-disk controller chips to handle the details for you.

If you just want some non-volatile storage, consider using Flash chips. They are relatively cheap, and you can interface to them via bank-switching just like you're doing with the SRAM.

Pete

But I don't want Flash memories, at least not exclusively.

I have an old disk drive and am thrilled by the thought of
having a small computer that boots from a floppy. It would
certainly be a good practise.
But perhaps I should build the actual computer first and then
bother with the more complicated peripherials.

Thomas

Is it safe to assume you're using a 6502 as you say you are, or are you actually using something else?

Your goals aren't clear in this context, but it seems to me that if you use a reasonably fast 6502 you should have no trouble bit-banging the interface to the FDD with no special support circuitry.

The ram can easily be implemented as you suggest, using SRAMs from an old '486 board's cache. Since you're going to have lots of memory, you won't have much trouble with software drivers to replace the now difficult-to-find LSI's you would have used back in the '70's.

PC-style FDD's have a number of features similar to 8" FDD's though these characteristics are not universal. Most, I believe rotate at the same rate as 8" FDD's though some rotate at 300. Far and away the easiest way to handle mass storage, however, is to find yourself a (not easy to find) Western Digital WD1002-05 board. This is a hard-disk/floppy disk interface bridge with a very simple host interface which will easily handle both the hard disk and a floppy disk, though it doesn't match up exactly with the current-generation PC's FDD's in that it expects 300 RPM drives. You shouldn't have any trouble getting them, however. Any old 360K drive will do, and with the capability of adding a hard disk, an MFM hard disk, for sure, but a hard disk nonetheless, you won't need to worry the FD capacity.

If you can't find the board I mentioned, you could easily use one of the many ISA boards that do the same thing, e.g. the PC FDC. It's a lot of bother building up a floppy disk interface and it's much easier to handle this by building an ISA bus interface and plugging your CPU into it, along with whatever other boards you can use. You can fabricate your own ISA backplane by using a band-saw on a defunct XT clone motherboard. Be sure to leave room for mounting holes, and be sure, as well, to remove any unneeded IC's. You could utilize an old "AT" motherboad as a passive backplane, having removed the unneeded parts, which is most of them, in which case you could use one of the PC/AT controllers, which, by the way work very much like the WD board, as they use the same LSI's.

Give this some thought, since you can use the ISA quite contstructively with a really fast 6502 at upwards of 12 MHz.

good luck!

Uli

Well, but still there's one problem: I will probably be using a 2Mhz 6502.
Which part numbers support higher frequencies?

The CMOS 65C02 is theoretically available at up to 14 Mhz I think. Finding one might be tough, unless you buy a bunch of them from the maker (Western Design Center).
I bought some a few months ago from Jameco, and they were 10 Mhz parts.

But, if you run at those speeds, you might have an even bigger problem finding 65C22's and things that will run at that speed. Also, a 10-14 Mhz 6502 board requires other things to be well-designed. If this is your first project of this type, then I don't recommend it. Stick to 2 Mhz or less.

Another mass-storage option: use an old Commodore 1541 or Atari drive, which already has the controller details built-in. On the web you can find all the details of the interfaces, I think.

Or, instead of a floppy, use cassette tape. The KIM-1, and many other boards, have the required hardware, and it's very simple.

Pete

The Western Design Center claims to have parts rated up to 14 MHz, and, from what I've seen, those are the only 6502's available anywhere. I am not sure, but I've assumed you mean what part number for the CPU. If you want to use LSI's you're probably stuck with the slow parts. Though I've used them all, I never have liked the 650x-family peripherals. I've used the 6502 more than any other CPU, particularly the fast CMOS ones from Rockwell. What's more, because they were always WAY cheaper than other CPU's they represent probably 75% of the volume that's been consumed over the years by all the CPU's I've put into various applications.

Of course you don't have to buy the fastest parts, but I don't think you'll find them anywhere else. In any case, these parts are generally too fast to run with any of the 650x-family-associated peripherals without some adjustment for speed.

At the speeds at which most of the popular (though I don't know why) peripherals operate, the 65SC02 at its maximum allows you to perform nearly any function the peripherals provided at much less cost and effort by generating the function in firmware. This means, for example, that you run a little routine to run a byte through a loop which shifts its bits out through a parallel I/O port bit, say, a 74HC273 register, rather than using a serial I/O port LSI. It also means you can, if you like, similarly bit-bang a floppy disk interface or the like, if you feel you can write clever enough code. The fast processor can, of course, be slowed down for I/O if you must use one of the old LSI's. The 6502 lends itself very well to clock-switching and clock-stretching. Both can be implemented in a single PAL, so you always have your choice.

Feel free to email me at <edick@idcomm.com> if you need details.

regards,

Uli

Hi All,

I've updated my "mini-page" at http://www.6502.org/users/mike

There are now a couple more projects and also pictures of my new robot.

_________________
- Mike Naberezny (mike@naberezny.com) http://6502.org

I wonder what's the significance of the "D" after the "6502A" on the chip?
I also notice it's a pretty late date code (1986).

How slow can you make it go using the R/C car platform?
Most R/C cars seem designed to go a zillion MPH...

Pete

Hi Pete,

The "6502AD" part number means "A" for 2 MHz, and I think the "D" is the packaging (Cerdip).

I got the old NMOS 6502 and 6522 chips from a junked Commodore 1541 5.25" disk drive. The date codes are pretty late, I believe the latest MOS date codes I have seen are 4xxx. In my PETs and VIC-20s I have early 0xxx codes!

As for using R/C cars for robot projects, I have had varied results with PWMing the motors. In most of my experiences I agree that these motors are geared for top speed, and trying to slow them down isn't very successful (no torque). Gilbert isn't very bad because he started out as a "truck" not a "car" and was geared a little lower. Still I can't recommend R/C cars for most projects, especially when tight control and a decent turning radius are required. Gilbert is just an explorer, so these things aren't very important to him.

On the subject of PWM, have you checked out Dallas Semiconductors' new DS1050 PWM modulator? It's an 8-pin DIP device that generates a PWM signal in 1, 5, or 10 KHz. You send it the duty cycle over a serial line. I've requested some samples and I'll be checking them out soon, it looks like a good way to offload processor time and VIA counters.

_________________
- Mike Naberezny (mike@naberezny.com) http://6502.org

Now, I could be mistaken here, but I think that the "4186" on your Gilbert chip means "41st week of 1986". I think that this scheme is semi-universal in the industry.
Many (most?) manufacturers will do the year first, followed by the week. I'm holding a 4GB SCSI drive right now (it was made in early 1995), and looking at the chips, I see date codes of 9440, 9439, 9448, 9444, etc.

So, an example of a really early 6502 date code would be "xx76", rather than "0xxx". For my 'collection', I'd like to have one of those real early ones, if anyone wants to trade...

Of the many VIC-20 chips I have, they're pretty much all in the 81 to 83 range; however, they all came from the same surplus dealer, who probably bought them from Commodore when they were shutting down.

Pete

You're right, I don't know how I missed that one. I guess I just never looked into the dates. I looked on my early chips are in the first weeks of '81. My best guess would be to look for early systems like the KIM-1 or maybe 2001 series PETs for these chips. You must really hate that ROR instruction, eh Pete?

_________________
- Mike Naberezny (mike@naberezny.com) http://6502.org