0

I have an old clock radio from the 90s that worked great until recently. Suddenly it stared running fast and every day it gets about 10m ahead. I am curious what could be going upon under the hood to cause this and how I might diagnose and fix it. Obviously easier to just buy a new one but I don’t want to throw it out and think I might learn something by fixing Thx

Chris Stratton
  • 33,491
  • 3
  • 44
  • 90
Rishi
  • 183
  • 1
  • 5
  • 2
    Without details of the implementation this is unanswerable. And indeed, repair questions which do not demonstrate a sound understanding of the thing being repaired are explicitly off topic under our rules. If you want to approach this, then being mindful of mains safety (if indeed it is mains powered) see if you can figure out what the frequency reference is: is it a crystal? Does it use the mains frequency? Then see if you can figure out what went wrong. – Chris Stratton Nov 18 '20 at 04:32
  • 2
    C'mon. Give us something to work with. At least give us a photo of the inside of the clock and we can spitball. – DKNguyen Nov 18 '20 at 04:42
  • Are you running it on an off-grid mains supply or an inverter? –  Nov 18 '20 at 14:39

1 Answers1

2

I'll make a guess or two. Old clock radios from the 1990s often used mains frequency for the clock reference. Many also had a battery backup that would use an RC circuit that might have 1% accuracy during a power failure. Also biplexed displays operating from center-tapped transformers, but that's not so important.

One possibility is that there is something creating spikes on your mains that are getting through whatever filtering they had in there. Either the spikes are new (maybe you added an appliance) or the filter has deteriorated for some reason.

Another possibility is that the mains clock has entirely failed and it is running on the backup clock source. Unlikely if the clock/radio has no place for a battery.

Another possibility is that the power supply filter capacitor has dried out and you are getting so much ripple on the supply that the clock cannot operate properly. Just look for the biggest electrolytic cap and replace it.

Spehro Pefhany
  • 397,265
  • 22
  • 337
  • 893
  • 1
    One thing that bothers me with all of these, is my reading that the OP is getting a near- consistent speed-up. The writing isn't clear but suggestive that it is consistent. It makes me wonder about some new torque issue. It may be entire coincidence and maybe I'm making way too much of precision in the error which doesn't really exist, but then 10 minutes in a day is a ratio of 1:144. 12 and 6 and 24 are all factors. I hate where my mind wanders when I see that kind of coincidence. – jonk Nov 18 '20 at 05:20
  • @jonk Torque? You're thinking it's a flippy plate clock? Had not considered that. Assumed VFD or LED. I think the flippy ones are older. – Spehro Pefhany Nov 18 '20 at 05:22
  • Yeah. That's what I was thinking about. A little bit of kick in the phase from something mechanical and a pole might "slip." The motor is old, dust mixes with oil, etc. I was trying to think of something that might account for that ratio and I started thinking about clocks with a mechanical display for that reason. I'm probably wrong. But it's where I went with my imagination for a moment. – jonk Nov 18 '20 at 05:25
  • @jonk I spent some years diagnosing technical control issues over the phone from all over North America, so I tend to have listening that can be a bit "selective". – Spehro Pefhany Nov 18 '20 at 05:41
  • 1
    I dedicated 2 hours each morning (8AM to 10AM) at Summa, handling technical support calls that our phone staff could not handle. (I was co-founder but considered this an important means to better know our customer base. We had 45 employees at the time and I worked until 8PM at night to compensate.) It was, however, solely focused on a product I developed. Not sure if that's your case, too. – jonk Nov 18 '20 at 06:03
  • I got calls from people who literally didn't know about plugging in a computer into a wall or that they needed to use a switch to turn it on. Instead, they'd call and complain that the software didn't work! Seriously. This was rare, I admit. But when it happened, you really had to have a creative mind to listen carefully and visualize the open possibilities and the follow-up questions to then ask. In a rare case or two, I was almost certain they could not feed themselves if left without "help." (Very rich kids in wealthy families, since the software was for securities analysis.) – jonk Nov 18 '20 at 06:06
  • @jonk Good point about consistent speed-up. Could be another appliance or lamp (perhaps PWM-dimmed?) on a timer that's adding pulses. I've run into IBM clocks whose master controller sent sync signals on the AC line to keep them all aligned. Created havoc in some sensitive equipment. – glen_geek Nov 18 '20 at 14:18
  • 2
    @jonk It was a mix of products, some I designed and others were ones I knew inside and out. This was industrial- engineers and maintenance guys so the people were not dumb, but that itself is an issue. They would tell you things that fit their mental model of what was going on, and omit others. Occasionally they would outright lie, esp. if they had done something careless like wire 240VAC to a thermocouple input causing an instrument to “just stop working”. – Spehro Pefhany Nov 18 '20 at 15:58