6502.org Forum  Projects  Code  Documents  Tools  Forum
It is currently Sun Jun 16, 2024 9:49 pm

All times are UTC




Post new topic Reply to topic  [ 38 posts ]  Go to page Previous  1, 2, 3
Author Message
PostPosted: Mon Nov 02, 2015 11:06 pm 
Offline

Joined: Mon Oct 12, 2015 5:19 pm
Posts: 255
oops; I still almost made the same error; I should price/shop for for 74AC as well; and get pdfs and read them too!

I am reading all the old (last months) posts a second time; I am not known for shirking other people's good advice; at least without good reason; this stuff is complex, and I must let it "stew" and "percolate' in my brain.

A few of my posts indicate I will stress what is physically possibly, given the tools of the 6502 hobbyist. So, 74AC chips would be wise to have around. (I don't think my decoder circuit will have TTL logic 74xxs, so ACT or HCT should not be necessary?!).

One of my books has a 3 X 74318 and 1 X 74319 circuit wired as a 5 to 32 decoder. I don't think this is necessary, but maybe it is? The 64816 has a lot of address space????!! If I look for bigger )total Bytes) RAM chips, then I might need less decoding? I dunno? It's all new to me!


Top
 Profile  
Reply with quote  
PostPosted: Tue Nov 03, 2015 12:16 am 
Offline
User avatar

Joined: Fri Aug 30, 2002 1:09 am
Posts: 8458
Location: Southern California
randallmeyer2000 wrote:
And wire wrap? I am not sure I will take that advice, though it is probably good advice. I don't have the tools

WW tools are pretty minimal—just the WSU-30M and stripper. I think you can probably do a better job with this manual tool than with an electric one, and almost as fast.

Quote:
still not sure if I want VIAs or PIAs? I must read more.

The VIA is more capable, and is less clumsy at some jobs. The only reason to use a PIA is if you're trying to protect a software investment.

_________________
http://WilsonMinesCo.com/ lots of 6502 resources
The "second front page" is http://wilsonminesco.com/links.html .
What's an additional VIA among friends, anyhow?


Top
 Profile  
Reply with quote  
PostPosted: Fri Nov 06, 2015 9:42 pm 
Offline

Joined: Mon Oct 12, 2015 5:19 pm
Posts: 255
Have spent the afternoon researching focal plane arrays (again). I have gotten out of the habit, but I should try to stay abreast of the subject.

I mention this, because my reply will undoubtedly focus, obsessively, on technologies I cannot possibly grasp adequately. I cling, in desperation, to your (collective/6502-group) hardware wisdom!

I guess I almost chuckle, when I think of using the 6502 to make something worthwhile. The CMOS sensors I purchased from Kodak (Kodak-ISS out of business; spun off, venture-vultured, then sold to ON semiconductor, who seems to have kept the CCD line and updated? and renamed? the CMOS line?), as dimly understood as the were by me, probably have equivalent processing power to the good ole 6502!

But I digress. I probably read half your primer, but my thoughts about it are still disorganized. I think address decoding and memory map are coming into focus. Had the old Wakerly "digital design" text out today, to refresh my memory (pun intended.... was studying DRAM and SRAM two nights ago) about propagation delay; 1 to 5 ns (I think?) per foot of PCB trace. Seems decent, given the prop. delays in chips, but I note that one must be cautious and read the pdfs carefully to understand "chip-delay" (several #'s are cited, depending upon the function of the chip, for the task at hand).

I think, if the decoding is not too complex, I will settle on 24KB RAM ($0000 - $5FFF) ; 16 KB I/O ($6000 - $9FFF); and 24 KB ROM ($A000 - $FFFF). This will follow the rules for IRQ/NMI and RESET stuff, and for stack in RAM and it is an approximately equal division, so that no aspect of computers function is too neglected.

So, I still want to design for immediate upgrade capability (i.e. to fully 65816 functionality), but was a bit stymied by the digikey offerings of DIP Asynchronyous Static RAM. 16MB addressable space, but the RAM chips (in DIP) are quite small?

Did I find a DRAM last night, of 512 KB? 32QTY of those would fill the space (or, 20QTY or so, leaving a little more for ROM or I/O)? Maybe I should see what Mouser offers?

I will keep wire wrap in mind. In fact, I might begin to "swear by it". I had difficulty when trying to design my focal plane array (5 or 6 or so years ago), as things were mighty crowded. At one point I decided to saw through the board and slide perpendicular boards into place. That was maybe a a mistake, though if I followed through I could have made it work (a couple of screws to physically fasten it in place, then solder the pad to the pads, or more liekly with a jumper wire between the two).

When I realized how ridiculous my sawing was, I purchased some molex (ribbon) connectors. And the project stopped there; life intruded, so i shelved it. Then Kodak/digikey stopped selling; I needed 9 CMOS sensors; I had 6 of the same kind and 3 or 4 more of different type.

I noted your advice against "running the bus out to the backplane", and then noted your exception to the rule. I will follow the thread/link--to the guy who did it properly--but haven't had a chance to, yet.

I might give my old "focal plane array" project a second try, since the CMOS sensors have come down in price (5$ per, in some instances). maybe something of the Kodak-sensor project can still be salvaged? I don't know.

I have some colloidal silver solution (or was it graphite?) for making plated through holes, but never got to that step last time. Also, if I remember right, there was alot of stuff about "deburring the holes" before electroless plating. I don't know. Maybe I was in over my head, for a hobbyist.

I have been searching through my materials, and I realized I have alot of materials and tools ready for the upcoming projects. I just need a little more organization, patience, and elbow grease.

Tell me (Garth or anybody else with an opinion) what you think of "molex-ing", i.e. tiny ribbon cables, the entirety of the CMOS image sensor (PLCC 48?-pin?) pinouts to a perpendicular board (literally "backplane" as the "frontplane" will need to be clear, so the lens can throw the image at the sensors)? How about connectors like a DIMM or SIMM has? You know, those heavy-duty, black-plastic connectors, made to handle those wide land-pads?

You're right Garth; Wire wrap is looking better all the time. Any drawbacks, in terms of AC operation? High frequency digital signals?

I wish I had the skills (and cash) to design using CCD; maybe even 3-CCD, trichroic prism assembly! Oh well; maybe next time.

Thanks for the info on the VIA; I am not a very "proprietary" guy; nor am I very "software".... YET!

Let us hope, someday, I can make a living as a scientist, and leave all this "$-talk" behind me! I very much want to push the ideas, instead of letting the $-ideas push me!

P.S. wire wrap needs me to buy posts and wire and different solder (I think?) and some connectors/sockets and such? Although, I did like how neat it looked, and I also liked those "stacked RAMs" you had. However, as previously mentioned, I might even have a problem making a doublesided board (unless I can drill holes, and run jumpers from a land on one side of the board, to a "land+trace" on the other side of the board? Can that be done? Or is that an "intermittent-electrical-problem" waiting to happen?).


Top
 Profile  
Reply with quote  
PostPosted: Fri Nov 06, 2015 9:48 pm 
Offline

Joined: Mon Oct 12, 2015 5:19 pm
Posts: 255
Whooops; idiotic "wire wrap = no solder" . My bad! Spoke/wrote too fast!


Top
 Profile  
Reply with quote  
PostPosted: Fri Nov 06, 2015 9:57 pm 
Offline

Joined: Mon Oct 12, 2015 5:19 pm
Posts: 255
Ever hear of "tyranny of wires" .... I think it is a known phrase. I am sure it has "made the rounds" among "computer circles". (Come to think of it, I think my father's first job, as a teen back in the early 60s, was to take the piecework off the end of the bench on production line; I think it was mostly women in the factory, wiring up the backs of RCA TV sets? funny story!)


Top
 Profile  
Reply with quote  
PostPosted: Fri Nov 06, 2015 10:13 pm 
Offline

Joined: Mon Oct 12, 2015 5:19 pm
Posts: 255
OK, last night I was looking at some SRAM: Alliance memory Inc. ;

"The AS6C4008 is a 4,194,304-bit low power CMOS static random access memory organized as 524,288 words by 8 bits. It is fabricated using very high performance, high reliability CMOS technology. Its standby current is stable within the range of operating temperature.
The AS6C4008 is well designed for very low power system applications, and particularly well suited for battery back-up non-volatile memory application.
The AS6C4008 operates from a single power supply of 2.7V~ 5.5Vand all inputs and outputs are fully TTL compatible"

Does this seem suitable? I note the 55 ns propagation delay, and I noticed one of your posts says that is not good enough (perhaps if I run the clock at 1 MHz?)? ( viewtopic.php?f=4&t=573&p=32004#p32004 )


Top
 Profile  
Reply with quote  
PostPosted: Fri Nov 06, 2015 10:45 pm 
Offline
User avatar

Joined: Fri Aug 30, 2002 1:09 am
Posts: 8458
Location: Southern California
randallmeyer2000 wrote:
Tell me (Garth or anybody else with an opinion) what you think of "molex-ing", i.e. tiny ribbon cables, the entirety of the CMOS image sensor (PLCC 48?-pin?) pinouts to a perpendicular board (literally "backplane" as the "frontplane" will need to be clear, so the lens can throw the image at the sensors)? How about connectors like a DIMM or SIMM has? You know, those heavy-duty, black-plastic connectors, made to handle those wide land-pads?

The connectors that press onto ribbon cables are called IDCs, for "insulation-displacement connectors." You have undoubtedly seen these inside PCs for disc drives in past years. They're super easy to mount. In a production environment where you want maximum productivity, you'd use a special tool; but for hobby applications, a bench vise works fine if that's all you have. IDCs are female, and the pin headers (which are male) go on the board. These usually have pin spacing of .100", so they fit nicely into prototyping board. This is not true however of SIMM sockets. Anything that can mount in a grid of holes that are on .100" centers is a big plus for home construction. OTOH, if you want to lay out your own boards, you can go for whatever spacing you want.

Quote:
You're right Garth; Wire wrap is looking better all the time. Any drawbacks, in terms of AC operation? High frequency digital signals?

It's not as good as a properly done PC board of course, but keeping to the guidelines in the primer, you won't have any problem. I won't repeat those all here.

Quote:
Let us hope, someday, I can make a living as a scientist, and leave all this "$-talk" behind me! I very much want to push the ideas, instead of letting the $-ideas push me!

Think of all the money you're saving by not having your hobby be a yacht, or even golf! :lol:

Quote:
P.S. wire wrap needs me to buy posts and wire and different solder (I think?) and some connectors/sockets and such?

Like your other post said, no solder. The IC sockets are rather expensive, but I think it's very much worth it, since you will need very few per year (and then probably go years before needing to buy more). If it were for production, the price would be a killer-- but if it were for production, you wouldn't use WW. You'd get PC boards made and have them assembled by machine. (I've done this many times for work; but for one-off stuff, whether for prototypes, hobby, in-house test equipment, etc., I often use WW.)

Quote:
Although, I did like how neat it looked

WW allows the highest density you can get with thru-hole digital ICs. High density makes for the shortest connections, and WW allows those to be direct, not having to snake their way around things.

Quote:
I might even have a problem making a doublesided board (unless I can drill holes, and run jumpers from a land on one side of the board, to a "land+trace" on the other side of the board? Can that be done? Or is that an "intermittent-electrical-problem" waiting to happen?).

I used to do it years ago, but now I just use the prototyping soldered breadboards shown in the primer for analog, or WW for digital. When I did the double-sided ones, I drew it out, drilled it first to make sure sides would line up, then put the etch resist on both sides, then etch. Vias were just wires soldered on both sides and cut short, since I had no way of thru-plating the holes. No reliability problems.

randallmeyer2000 wrote:
OK, last night I was looking at some SRAM: Alliance memory Inc. ;

"The AS6C4008 is a 4,194,304-bit low power CMOS static random access memory organized as 524,288 words by 8 bits. It is fabricated using [...]"

Does this seem suitable? I note the 55 ns propagation delay, and I noticed one of your posts says that is not good enough (perhaps if I run the clock at 1 MHz?)? ( viewtopic.php?f=4&t=573&p=32004#p32004 )

The AS6C4008 looks fine. I think 55ns is what I have in my 5MHz workbench computer. What's not suitable is when people think that 100ns for example is good for 10MHz. Not so; because although one cycle at 10MHz is 100ns, it takes the processor a little of that time to get the valid address out, then it takes the address decoding circuitry a little time to get its outputs valid from that, and then the processor needs the data it's reading from the memory to be valid a certain set-up time before phase 2 falls and the next cycle starts. What's left for the memory is well under 100ns. One has to spend some time in the timing diagrams and specifications.

_________________
http://WilsonMinesCo.com/ lots of 6502 resources
The "second front page" is http://wilsonminesco.com/links.html .
What's an additional VIA among friends, anyhow?


Top
 Profile  
Reply with quote  
PostPosted: Fri Nov 06, 2015 11:10 pm 
Offline

Joined: Mon Oct 12, 2015 5:19 pm
Posts: 255
Thanks a zillion (your site is neat! But I over-use it. More wires, less talk/type! I was looking at the user list, and noted that one could search by # of posts. I think I was 107th, with 38 posts! Far too much for guy who never ran a processor! Also, I will limit my discussion of image sensors, as that is off topic--that is, until I get my focal plane array up and running! Then, when I run into I2C problems, I will seek knowledgeable people!). I have TONS of work to do (preparation, workbench, purchase tools), before I can start working on the really interesting digital projects (I don't mean to "diss" the 6502, but I just clearly have a different purpose; vision science). Thanks a zillion.


Top
 Profile  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 38 posts ]  Go to page Previous  1, 2, 3

All times are UTC


Who is online

Users browsing this forum: No registered users and 12 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to: