Lessons Learned...total failures

[cbmeeks]

We have a great thread on "best practices". But we should have a thread (sticky) where dummies like me can tell the world were we messed up.

I messed up pretty good this weekend. I have some of those permanent protoboards from Adafruit (https://www.adafruit.com/product/590). I really love the quality of them.

So, recently, I started to build a very minimal 65C02 computer using a couple of them. Thinking I was being really smart, I managed to get the CPU, RAM, ROM and a single NAND on this one board. Basically a computer minus I/O.

While I did get it to free-run (see my other post), I made a few bone-head mistakes.

1) For whatever reason, I decided to leave the bottom two rows empty. I still don't know why I did this. That protoboard has 60 rows. So, now I have 58 to work with. This caused a MAJOR problem.

2) Because of #1, I then decided to put the CPU and NAND gate next to each other with no gap between the two. I didn't anticipate that the IC's would need a little space OVER the sockets. I was assuming the sockets represented the vertical space needed for the IC (no, I didn't dry-fit them before hand). :-/
So this makes the NAND gate actually bump into the CPU. With a little force, I was able to nudge the NAND in its socket. But it's very tight and not completely in the socket (although, it's in there good enough to work...but over time it may drift back out).

3) I left ONE row of space between the CPU and the ROM. I left this space for the decoupling caps. I thought I was being smart here. But again, I didn't anticipate the space needed for the actual IC's. Fortunately, it actually fits right in with no trouble inserting the chips. The trouble comes when trying to pull the ROM back out to program it. NO I didn't use a ZIF socket. And no, I can't program in-circuit because of the programmer I have. Well, I could probably wire something up temporarily but I don't have much space left over. So when the CPU and ROM are inserted, there is NO ROOM to put my chip puller in. I have to pry it out on one side only. Or, pull out the NAND and CPU just so I can pull out the ROM.
I about destroyed the ROM trying to pry it out.

So, that's my woes for today. LOL

I have a large bucket of my "failed" projects. Tons of boards I've messed up. But, I guess that's how you learn. And boy, have I done a lot of learning!

I guess I shouldn't call this adventure a full failure since it does technically work. But I don't know if I will try to salvage it.

See, the thing is, I can solder about 5 times faster than I can wire-wrap. Despite my poor eyesight. But, I have trouble sharing 2-3 wires on the same pin while I solder. Which is why I like these permanent protoboards.

So for the beginners out there...the lesson for today is....

1) Dry fit all components before you put solder to it!!

2) Leave enough working space around each component. Doesn't have to be much. But enough room for a chip-puller is a good idea.

3) Don't try to be uber-clever like I was (trying that is). In other words, unless you're a master maker, don't cram 4 components on a small board if you then have no room to work on them.

4) Use ZIF sockets for chips that you will remove frequently!

[BigEd]

Good advice! There's no total failure, if you learn something. This might be why it's so helpful to start small and tackle a succession of challenges, rather than every idea at once. It might even take less time and money, if it means a succession of successes.

I'm in the camp, most often, of not even trying the small challenges, in case I fail. So most of my triumphs, small as they are, are in making some contribution to an existing project. (Which you could say, is an incremental approach.)

And so I claim my failures: failure to even start!

[floobydust]

In general, experience is the best teacher, however, it can get expensive from time to time, so attempting to learn from other people's "teachings" is preferable in many cases.

When I got back into the 65xx hobby some 4+ years ago, I opted to make a board set: one board with the CPU, ROM/RAM and glue logic with a simple bus connector and an I/O board (6551/6522). This has worked out quite well using a short ribbon cable to attach them.

If I were to do this again, I would do it a bit differently;
1- A different pinout for the I/O connector making it more flexible for using other I/O devices
- 32-bytes wide (rather than 16-bytes)
- Positive going Reset line
- Cleaner signal distribution on the 30-pin connector
2- Use a CPLD for glue logic so I can alter the ROM/RAM-I/O addressing
3- Provide additional logic for a NMI trigger
- My monitor supports a NMI panic routine, so having a way to invoke it would be nice
4- Likely a 4-layer board for better shielding and higher clock speeds
5- Onboard serial console via a USB connector
- Power switch w/center off allowing power select from USB or dedicated connector

Again, live and learn, but here's a pic of my CPU board, of which I have three running for the past 4+ years.

I'm using the machined IC sockets from Mill-Max and can easily insert a ZIF connector into the ROM socket which makes removing the EEPROM quick and easy. I prefer not to solder these to the board as they are expensive and can easily be reused for other projects down the road.

[cbmeeks]

That's some good advice as well.

One thing I might try doing is putting a ZIF socket in the socket that I soldered. But only while experimenting. When I'm done, put the ROM back in and never take it out.

Problem is that those ZIF sockets usually have very small and fragile pins compared to the IC itself.

[GaBuZoMeu]

Perhaps you can stack 2 or 3 standard sockets to elevate your EPROM that little that's necessary to use your puller. If you use the so called precision sockets you can stack and solder them - but that would be one way. But stacking some sockets plus a ZIF socket on top would be the premium solution I think. IIRC Aries (Mouser) manufactures ZIF sockets with pins that fit into sockets.

[cbmeeks]

Actually, that's not a bad idea. In fact, I have some wire-wrap pins I could probably use.

Thanks!

[GaBuZoMeu]

Wire wrap pins are quite a bit fat for regular sockets! - You may afterwards have no good contact, once you will insert an EPROM directly. At least put one extra (standard) socket between. Or you get another lesson you probably better omit

[cbmeeks]

GaBuZoMeu wrote:

Or you get another lesson you probably better omit

Wouldn't be the first time. LOL

Actually, I have a WW socket that has a broken pin that I cannot get out in order to repair it (I've repaired other WW sockets).

What I might do is remove that cheapo socket that I soldered in (leaving the wires next to it) and just solder in that somewhat broken WW socket. I could leave it standing pretty high to clear the 6502.

[GaBuZoMeu]

If you can live with a ww socket there - everything is fine
Then you might try to fit the ww socket into the other one - but that might require very much force - so be careful.

[floobydust]

GaBuZoMeu wrote:

Perhaps you can stack 2 or 3 standard sockets to elevate your EPROM that little that's necessary to use your puller. If you use the so called precision sockets you can stack and solder them - but that would be one way. But stacking some sockets plus a ZIF socket on top would be the premium solution I think. IIRC Aries (Mouser) manufactures ZIF sockets with pins that fit into sockets.

Actually, that's the setup I've been using on my CPU board. I have a bunch of 28-pin Augat NOS gold pinned machined sockets from years ago. I have one of these plugged into the CPU ROM socket to elevate it past the jumpers and nearby chips on the board, and an Aries ZIF socket plugged into that, works like a charm.

[GARTHWILSON]

I heard nearly 40 years ago that the best kind of experience to have is someone else's. (Hopefully you can learn just as much from a thumb being smashed by a hammer if it was someone else's thumb!)

At about the same time, when the only test equipment I had access to to damage, I heard that one of Murphy's laws says that your experience is proportional to the number of dollars of test equipment you've damaged. I suppose the problem with that is that you lose experience through inflation!

Here's a five-minute video from the ultra-successful Jeri Ellsworth, "Secret to Learning Electronics - Fail and Fail Often": https://www.youtube.com/watch?v=xhQ7d3BK3KQ See lots of her videos at https://www.youtube.com/user/jeriellsworth/videos .

[BigEd]

That's a great video - thanks!

[BigDumbDinosaur]

GARTHWILSON wrote:

Here's a five-minute video from the ultra-successful Jeri Ellsworth, "Secret to Learning Electronics - Fail and Fail Often": https://www.youtube.com/watch?v=xhQ7d3BK3KQ See lots of her videos at https://www.youtube.com/user/jeriellsworth/videos .

Jeri's experiences are very good ones from which to learn. While she has a lot of native intelligence, it seems her success has come from a willingness to "crash and burn" every so often, and accept crashing and burning as part of the process. As one of my instructors from many decades ago once noted, failure serves a cosmic purpose: it eliminates that which won't work.

[BigDumbDinosaur]

cbmeeks wrote:

Actually, that's not a bad idea. In fact, I have some wire-wrap pins I could probably use.

Thanks!

Np, no, NO! You will ruin the socket into which you have plugged the WW socket.

[cbmeeks]

BigDumbDinosaur wrote:

Np, no, NO! You will ruin the socket into which you have plugged the WW socket.

The socket that I used for the ROM cost about $0.12 and I have dozens of them.