Why would this clock run too fast?

I built this circuit (clock with alarm and timer). Everything seems to work well except that it runs fast, very fast...several hours in just 1 day. At first I rechecked all components and they seemed good, checking every diode as they seem to blow easily if not heatsinked well. All was well. Then after reading the
Datasheet for the LM8560 I thought I had found the problem... did I have J1 (Jumper 1) connected for 50Hz operation and not 60Hz operation. Nope! This was ok too! Now I am stumped as to why it runs so fast.

Anyone have any ideas/suggestions?

Schematic:
www.bakpakr.com/downloads/clock.jpg

Datasheet:
www.bakpakr.com/downloads/LM8560.pdf


Thanks

iONic said:

I built this circuit (clock with alarm and timer)...checking every diode as they seem to blow easily if not heatsinked well.

Click to expand...

Well, that seems to be a bit of a clue, there.

Sounds like major wiring problems to me.

What frequency is the mains where you live?

Is it stable?

Alvol:

The pcb was preprinted... a kit! The wiring is fine and after checking I believe the components are as well.

Hero999:

I'm in the US, the freq. is 60Hz, which is the default with no pin connection.

Question; How fast 4 hours 48 minutes? Always the same?

I made something like this in the 1970s. My problem was the clock was fine all night but ran fast in the daytime. I found every time the elevator motors turned on I got an extra count. (noisy power line) Check R2C4. The capacitor might be smaller than you think. You could make C4 larger to filter out noise.

Did you clean the board? Solder flux may be causing erratic operation.

i believe I did check solder connections and cleaned the board. What is a good substance to clean the board with...

You could use an external crystal oscillator for the clock.

You may have a bad clock IC, I saw one do this once however it was not an LM8560.

Why does every school kid need to make a clock?
In "the good old days", clock chips were sold and everybody made a clock. Now, clock chips are available only if you buy thousands of them. Millions (hundreds?) of counters and gates chips are needed to make a clock without a "clock" IC.

I bought some clock radios on sale for only $.99 CAN. The clocks work OK but the radios are crap. It probably cost more than I paid to ship them from China to me.

What is a good substance to clean the board with...

Click to expand...

You can clean the solder flux off with 99% isopropyl alcohol.

checking every diode as they seem to blow easily if not heatsinked well.

Click to expand...

Which diodes are getting hot? None of them should be.

The clock has provision for a back-up battery to keep it going when mains power has failed.
To keep time during mains failure it has an internal oscillator which I guess has its frequency determined by C3/R3.

So, if the 50/60 hz input to pin 25 is faulty in some way, it could be running on the RC oscillator, which could be (and probably is) way off frequency.

I have a clock which does just this, if the mains fails, the clock runs fast.

JimB

audioguru said:

Why does every school kid need to make a clock?

Click to expand...

For the learning experience, I suppose. No better way to gain a knowledge of gates and counters than to build a functional clock.

Actually, it only takes about 7 chips to make one.

**broken link removed**

kchriste said:

Which diodes are getting hot? None of them should be.

Click to expand...

That was my question back in the first reply. Diodes seldom need a heat sink (and if they did, where would you buy one?)

AllVol said:

That was my question back in the first reply. Diodes seldom need a heat sink (and if they did, where would you buy one?)

Click to expand...

They are usually just copper 'tabs' bent round the thick wires of the rectifier and soldered - with the tabs sticking up in free air.

However, it's pretty rare these days - and certainly no need for anything like that in a clock?.

RonSympson wrote:
Question: How fast 4 hours 48 minutes? Always the same?

The clock is aproximately 4 or more hours off in 24hrs, but i is not dependent on time of day or placement of the clock in a particular room.

---------------------------------------------------------------
kchriste Wrote:
Which diodes are getting hot? None of them should be.

The diodes are not geting hot by running the clock. I meerely heatsink them while soldering as I do every component!

---------------------------------------------------------------
JimB Wrote:
So, if the 50/60 hz input to pin 25 is faulty in some way, it could be running on the RC oscillator, which could be (and probably is) way off frequency.

There may be a problem with the IC but I haven't been connecting the battery backup.

---------------------------------------------------------------
AllVol Wrote:

That was my question back in the first reply. Diodes seldom need a heat sink (and if they did, where would you buy one?)

Must be I did not explain myself well in the first post... heat sinking was to protect the diode for soldering.

---------------------------------------------------------------

Bottom Line:
Although I have checked solder points, component values, and cleaned the poard with IPA, I will do this again for the hell of it.

When I said the diodes would blow if not heaksinked well I meant that when soldering the components to the board. I heatsink with a pair of hemostats.

I have never heatsinked anything that I have soldered with my temperature controlled soldering iron and nothing has ever been damaged.

I am looking for a way of getting another 4 hours in my day. Make you a deal!

The reason ronsimpson asked about being 4 hours 48 minutes off is:

24 hours X 60/50 = 4.8 hours fast.

Pin 27 should have a 900hz signal. That should not effect you problem but it is worth checking.

Pin 25 should have 60hz and I would put an o-scope on it looking for a reason why your clock is seeing >60hz. (noise)
The data sheet want 100k and .001uf, the kit wants 18k and .01uf. (very interesting) Can you find 100k and .001uf?

I think there are two roads to go down:
1) you have 50/60 problems. The error will all wise be the same.
2) pin 25 is seeing noise. The error will be different every day.

I am looking for a way of getting another 4 hours in my day. Make you a deal!

Well, cleaned the circuit board with alcohol and a toothbrush. I checked all solder joints for shorts or cold solder points. Still it runs about 4hrs 30min fast in a days time. it doesn't matter what room the clock is in either.

The heat sinking of the diodes I had refered to in my original post was during the soldering of the components to the board.