B&F Enterprises "Digital Clock" flakes out

More
7 years 4 months ago - 7 years 4 months ago #4816 by Ty_Eeberfest
To address your original question... I don't have any good ideas for specific components or voltages to look at closer. However, the PC board looks like it could benefit from some basic hygiene. The soldering job looks lousy!

First of all give it a bath in flux remover or at least isopropanol. Once the flux residue is all gone get it under mild magnification and find all the cold / crusty / rotting solder joints and re-flow them with your soldering iron.

That might not fix it, but then again it may. That flux residue say a lot about the (non-) skills of the person who assembled it. You can get away with leaving flux for a while but over a long time it's well known to cause trouble.

Look into it later when the dust is clearing off the crater.
Last edit: 7 years 4 months ago by Ty_Eeberfest.

Please Log in or Create an account to join the conversation.

More
7 years 4 months ago #4821 by Dr. Optigan
Yeah, whoever built this thing did a ham-fisted job of soldering it together. Appropriate, because I found it at a ham radio-related flea market...

I'll have to try cleaning the old flux off and check the joints when I go in there to re-seat the ICs. I'm not entirely convinced that the issues with the board are entirely due to the builder; my friend (who's an engineer) was cursing the size of the holes in the board, as well as the fragile condition of the wires connecting the switches and such to the board. I'll take some pictures of the component side of the board when I next remove it from its stand-offs.

The most annoying part of this clock's issues is that it seems entirely random; it'll run hours without a glitch, or it'll glitch up within a minute or so of being plugged in, and then run glitch-free for several hours. Also, there's no consistency to what it does during the glitch; it'll skip ahead several hours, or just a couple of minutes. In addition, tapping on the casing doesn't have any noticeable effect on how the clock works.

All in all, it definitely looks like I inherited someone else's problem. :unsure: I can't help but wonder if this glitching issue is what a previous repairman was trying to fix by replacing most of the ICs. As they say, you get what you pay for, and considering that I paid only $5 for this thing... :S
-Adam

Please Log in or Create an account to join the conversation.

More
7 years 4 months ago #4824 by Ty_Eeberfest

Dr. Optigan wrote: The most annoying part of this clock's issues is that it seems entirely random;

But of course... if it was a repeatable problem you could find it, fix it and move on to the next project... but where's the fun in that?? :P

Seriously, the intermittent nature of the problem(s?) is why I won't even guess at a specific part as a cause. It's also why I bothered to comment about flux and crappy assembly. I wasn't just punting; experience has taught me that poorly soldered dirty boards will often appear to be working fine when in reality signals are weak or noisy and the whole mess is just waiting for a cosmic ray or a change in room temperature or a stray freakin' neutrino to put it over the edge.

If you can get ahold of 1,1,1-trichloroethane it would be reasonable to soak the entire board in it.

Something else to worry about: if the board itself is as cheaply made as it sounds, you could have cut traces that are only cut at certain temperatures (or ???) and just a little bit resistive at other temperatures.

Giving the board itself sharp whacks with a wood dowel (or something) while the clock is operating wouldn't hurt anything and if you're real lucky it might help localize the problem.

Look into it later when the dust is clearing off the crater.

Please Log in or Create an account to join the conversation.

More
7 years 4 months ago #4825 by Dr. Optigan
Duly noted, thanks.

Noticed something odd during my latest test. Actually managed to record some of it in action, though not all of it. About as soon as I started recording, the time display jumped from 00:31:41 to 08:31:42. Then, when the minute ticked over, it jumped from 08:31:59 to 08:44:00. This was within a minute or so of being powered up.

Shortly before that, when it was first powered up, it displayed 00:05 and counting. When I held the Set switch and pressed the minutes set button, it immediately jumped to 00:10. No idea if that's related to any of the clock's other weirdnesses; just figured I'd mention it.
-Adam

Please Log in or Create an account to join the conversation.

More
7 years 4 months ago #4828 by Ty_Eeberfest
Still hard to say what's going on there. But the fact that it skips ahead - as opposed to missing "count events" - is certainly suggestive of noise. I stick to my basic premise that the board needs cleaning and inspection first, but I'm also wondering how the +5vdc looks on a scope and also whether or not 0.1uF "bypass" (or "decoupling") capacitors next to each chip were left out of the design to save a buck.

Since I have no schematic nor have I seen the component side of the board I have no idea whether or not the design included anything to debounce the setting buttons. What you describe sounds like contact bounce. With no debounce logic it might have worked okay when new but switches tend to get more bouncy with age.

Look into it later when the dust is clearing off the crater.

Please Log in or Create an account to join the conversation.

More
7 years 4 months ago - 7 years 4 months ago #4836 by Dr. Optigan
UPDATE:

I took the cover off of the case to try some tapping tests (haven't removed the board to take pictures of the component side yet, though I'm intending to do so soon). I used a Sharpie marker to tap the board in the area of each IC, as well as random places on the rest of the board, but it had no noticeable effect. I then left the clock sitting with the cover removed, and after awhile, noticed that the glitching had stopped. I left it like this overnight, and when I woke up, it was still keeping perfect time.

So far, it's been running for over 36 hours without a problem. *knock on wood* Could the problem have been as simple as some sort of overheating? I didn't think nixie tubes produced any kind of noticeable heat, and given that the clock was designed to be encased in an unventilated metal box from the start, I'm not sure how this could be happening. Any ideas? I haven't tried putting the cover back on to see if the glitching will return, though I'll probably try that tonight.
-Adam
Last edit: 7 years 4 months ago by Dr. Optigan.

Please Log in or Create an account to join the conversation.

Moderators: AccutronTy_EeberfestIan
Time to create page: 0.101 seconds

Search

Tube Suppliers

Go to top
JSN Boot template designed by JoomlaShine.com